University of Groningen Linezolid in multidrug-resistant ... · Chapter 2 Pharmacokinetic drug interactions of antimicrobial drugs: a systematic ... tuberculosis (TB) is a bigger

University of Groningen

Linezolid in multidrug-resistant tuberculosisBolhuis, Mathieu

IMPORTANT NOTE: You are advised to consult the publisher's version (publisher's PDF) if you wish to cite fromit. Please check the document version below.

Document VersionPublisher's PDF, also known as Version of record

Publication date:2015

Link to publication in University of Groningen/UMCG research database

Citation for published version (APA):Bolhuis, M. (2015). Linezolid in multidrug-resistant tuberculosis. [S.n.].

CopyrightOther than for strictly personal use, it is not permitted to download or to forward/distribute the text or part of it without the consent of theauthor(s) and/or copyright holder(s), unless the work is under an open content license (like Creative Commons).

Take-down policyIf you believe that this document breaches copyright please contact us providing details, and we will remove access to the work immediatelyand investigate your claim.

Downloaded from the University of Groningen/UMCG research database (Pure): http://www.rug.nl/research/portal. For technical reasons thenumber of authors shown on this cover page is limited to 10 maximum.

Download date: 20-11-2020

https://www.rug.nl/research/portal/en/publications/linezolid-in-multidrugresistant-tuberculosis(f870bd42-858e-41fe-bfc5-645f0006b923).html

https://www.rug.nl/research/portal/en/persons/mathieu-bolhuis(d7869399-2910-4b0a-b14b-b0496c47e43d).html

https://www.rug.nl/research/portal/en/publications/linezolid-in-multidrugresistant-tuberculosis(f870bd42-858e-41fe-bfc5-645f0006b923).html

Linezolidin multidrug-resistant tuberculosis

Mathieu S. Bolhuis

Publication of this thesis was supported by the KNCV Tuberculosis Foundation, Stichting ter bevordering van Onderzoek in de Ziekenhuisfarmacie te Groningen (Stichting O.Z.G.), Stichting Beatrixoord Noord Nederland, Groningen University Institute for Drug Exploration (GUIDE), Stichting KNMP-fondsen, and UMCG Center for Rehabilitation Beatrixoord.

Cover Mathieu BolhuisCover image Pax Nicholas & the Nettey family, Daptone Recording Co.Layout Renate Siebes, Proefschrift.nuPrinted by Drukkerij Van Gorcum, AssenISBN 978-90-367-7611-0

© Mathieu S. Bolhuis, 2015

Copyright of the published articles is with the corresponding journal or otherwise with the author. No part of this publication may be reproduced, stored in a retrieval system, or transmitted, in any form or by any means, without the prior permission in writing from the author or the copyright-owning journal.

Linezolid in multidrug-resistant tuberculosis

Proefschrift

ter verkrijging van de graad van doctor aan deRijksuniversiteit Groningen

op gezag van derector magnificus prof. dr. E. Sterken

en volgens besluit van het College voor Promoties.

De openbare verdediging zal plaatsvinden op

woensdag 18 februari 2015 om 12.45 uur

door

Mathieu Sander Bolhuis

geboren op 11 mei 1981te Groningen

PromotoresProf. dr. T.S. van der WerfProf. dr. J.G.W. KosterinkProf. dr. D.R.A. Uges

CopromotorDr. J.W.C. Alffenaar

Beoordelingscommissie Prof. dr. D.M. BurgerProf. dr. F.G.J. CobelensProf. dr. G.M.M. Groothuis

Chapter 1 General introduction 7

Chapter 2 Pharmacokinetic drug interactions of antimicrobial drugs: a systematic

review on oxazolidinones

17

Chapter 3A Clarithromycin significantly increases linezolid serum concentrations 27

Chapter 3B Clarithromycin increases linezolid exposure in multidrug-resistant

tuberculosis patients

31

Chapter 4A Dried blood spot analysis for therapeutic drug monitoring of linezolid

in MDR-TB patients

47

Chapter 4B Clinical validation of the analysis of linezolid and clarithromycin in oral

fluid of multidrug-resistant tuberculosis patients

63

Chapter 5A Comment on: Daily 300 mg dose of linezolid for multidrug-resistant and

extensively drug-resistant tuberculosis: updated analysis of 51 patients

77

Chapter 5B Linezolid: safety and efficacy monitoring 81

Chapter 5C Linezolid safety and tolerability in multidrug-resistant tuberculosis

patients: a retrospective observational study

85

Chapter 6 In vitro synergy between linezolid and clarithromycin against

Mycobacterium tuberculosis

103

Chapter 7 General discussion and future perspectives 111

Summary 123

Samenvatting 131

Co-authors of manuscripts presented in this thesis 145

Dankwoord 149

About the author 157

Contents

General introduction

Chapter 1

Chapter 1

8

Tuberculosis

With over 1 billion deaths in the last 200 years, tuberculosis (TB) is a bigger killer than any other infectious disease in history (1). TB is responsible for more casualties than the plague, malaria, AIDS, or cholera (1). Although TB is now considered a poverty-related illness, it may affect any susceptible individual, whether male or female, young or old, rich or poor. TB has also claimed many famous victims. Although many people know the famous victims of TB such as George Orwell, few know that Ghanaian highlife superstar Azongo “Captain Yaba” Nyaaba and Senegalese Mandinga musician Kaouding Cissoko also fell victim to TB.

More recently, over 8 million people developed TB and approximately 1.3 million people died from the consequences of TB in 2012 (2). The incidence of TB is especially high in low-income countries. However, due to permanent and temporary immigration of migrants from high prevalence countries, TB is not only a low-income country problem, but affects countries worldwide (3).

Mycobacterium tuberculosis

TB is a potentially deadly infectious disease caused by the aerobe Mycobacterium tuberculosis. The pathogen M. tuberculosis is a species of the family Mycobacteriaceae, together with other pathogens such as Mycobacterium leprae, Mycobacterium ulcerans, and Mycobacterium avium. In contrast to many bacteria encountered in common hospital infections, M. tuberculosis replicates with a minimum in vitro doubling time of approximately 15 hours (4). It requires a host to duplicate and has no known environmental reservoir (5).

Typically, TB is transmitted through droplets in the air. Droplets are generated by coughing or sneezing by patients suffering from pulmonary TB. With its lipid rich cell wall, M. tuberculosis is able to survive in air droplets, forming an infectious aerosol. When individuals inhale this aerosol containing the aerobe mycobacteria, the organisms may reach the lower respiratory tract and the pulmonary alveoli. The mycobacteria are engulfed by macrophages, such as alveolar macrophages and bronchial dendritic cells. In most cases, M. tuberculosis is cleared by these macrophages (6). However, mycobacteria may survive intracellularly (7, 8). When the organism survives the host immune response, a ‘balance of terror’ may ensue: the organism persists in human macrophages, in a low metabolic, slowly replicating state, referred to as ‘latent TB infection’. Before this balance is established the organism may have spread through the body via the lymphatic and the circulatory system. In small children and

9

General introduction1

in adults with a failing immune system, infection may immediately become overwhelming, with miliary TB as a result. Most individuals however develop latent infection, and never develop active disease; some 10% of individuals develop active TB disease any time following infection. The pool of latently infected individuals represents the reservoir of the organism; as much as one third of the world population is considered carrier of TB (9). M. tuberculosis primarily infects the lungs, and established disease predominantly presents as pulmonary TB. As explained, TB may however affect virtually every organ and tissue.

Symptoms and diagnosis

Patients with latent TB infection are infected with M. tuberculosis without displaying any symptoms; the bacilli have low metabolic and replicative activity, and the numbers are so low that the organisms cannot be detected by microbiological methods. Only immunological tests are able to confirm the diagnosis. When the immune system fails to control M. tuberculosis replication and metabolic activity, active disease may ensue with increased numbers of bacilli causing symptoms of cough, weight loss, fever, chest pain, weakness or fatigue, night sweats and coughing up of blood (10, 11).

The clinical diagnosis of TB based on a combination of symptoms, history of prior TB infection, epidemiological factors and radiographic/laboratory findings can now be confirmed by microbiological methods. These microbiological methods comprises microscopy using acid- and alcohol-fast staining procedures, culture using specific culture media, and polymerase chain reactions demonstrating the presence of specific sequences of the DNA of M. tuberculosis.

Multidrug-resistant tuberculosis

A worrisome fact is that an increasing proportion of TB patients happen to be infected with drug resistant M. tuberculosis strains. The World Health Organization (WHO) suggests that approximately 3.6% of new TB cases have multidrug-resistant strains, with much higher levels – up to 20% – in previously treated cases (12). In multidrug-resistant TB (MDR-TB) the organism is resistant to at least the classic anti-TB drugs rifampicin and isoniazid (13). For the individual patient, MDR-TB is bad news: the treatment duration required to obtain cure is increased more than three-fold, and outcome is less certain. Drug resistance is the result of selective pressure to resistant mutants in the microbial population causing the

Chapter 1

10

infection; with inadequate therapy suppressing susceptible wild-type organisms, mutants may re-populate infection sites, and in the end all organisms are drug-resistant. It would help if fast molecular tests were available to timely detect and adequately target the offending organisms with a drug treatment combination that matches the susceptibility to the drug combination selected. Molecular testing using the Hain genotype MTBDRplus might give a swift preliminary result of the resistance of the isolate to rifampicin and isoniazid (14). The test is based on detecting the most common DNA mutations. As a proxy of MDR-TB, resistance to rifampicin only is now commonly used. The most widely used and studied test is probably GeneXpert-TB-RIF, a system that has been developed and validated in different settings (15, 16) and that can be used as a point-of-care test, also in settings with limited laboratory expertise available (17, 18). However, the gold standard to determine drug resistance is by analysing the minimal inhibitory concentration of a sample of mycobacteria isolated from the patient using the absolute concentration method (19).

To treat MDR-TB, the WHO recommends to design treatment regimens containing at least four drugs that are probably effective (13). Therefore physicians are forced to design treatment regimens using less effective and less well-studied ‘second-line’ drugs. Some are prone to elicit adverse effects, further limiting the applicability to treatment regimens.

Treatment options are divided in several WHO groups (13). Group 2 – the second line parenteral drugs or ‘injectables’ – is composed of aminoglycosides, e.g. amikacin and kanamycin. Group 3 are the fluoroquinolones, with very active and widely used drugs such as moxifloxacin; and the group 4 drugs including oral bacteriostatic second-line anti-tuberculosis drugs, e.g. prothionamide and cycloserine.

As a last resort, physicians are often forced to prescribe WHO Group 5 drugs such as linezolid and clarithromycin. Both drugs are described in this thesis. Linezolid and clarithromycin are drugs with unclear efficacy and are therefore not recommended for routine use in treatment regimens for MDR-TB (20). However, more knowledge on the efficacy, toxicity, tolerability, i.e. the clinical pharmacology might unleash their untapped potential.

Clinical pharmacology

New information on pharmacology in a clinical setting might contribute to the applicability of linezolid in the treatment of MDR-TB. The two main areas of pharmacology are pharmacokinetics and pharmacodynamics. Pharmacokinetics describes the movement

11


of the drug through the body (absorption, distribution, metabolism and elimination): it addresses the question how the organism handles the drug. Pharmacodynamics describes the action of the drug in the human host – and in case of micro-organisms – how the offending organism is targeted.

Therapeutic Drug Monitoring

These two principles, pharmacokinetics and pharmacodynamics, can be combined to perform therapeutic drug monitoring (TDM). TDM is defined by the International Association of Therapeutic Drug Monitoring and Clinical Toxicity as ‘a multi-disciplinary clinical specialty aimed at improving patient care by individually adjusting the dose of drugs for which clinical experience or clinical trials have shown it improved outcome in the general or specific populations (21)’.

TDM could perhaps improve the treatment of MDR-TB using the WHO group 5 drugs, such as linezolid. Therefore, we aimed to study the clinical pharmacology, with special focus on TDM of linezolid to optimize treatment of patients suffering from MDR-TB. More specific, we aimed to give a review of literature on pharmacokinetic drug interactions of anti-mycobacterial drugs. Furthermore, we aimed to study a potentially new pharmacokinetic drug-drug interaction between linezolid and clarithromycin and to analytically and clinically validate two new methods to analyse linezolid in oral fluid and dried blood spots obtained from MDR-TB patients. Finally, we aimed to retrospectively study linezolid exposure in relation to efficacy, safety and tolerability in the treatment of MDR-TB. Besides the aim to study pharmacokinetics of linezolid, we aimed to investigate possible pharmacodynamic interaction between linezolid and clarithromycin in MDR-TB isolates.

Outline of the thesis

Chapter 2: In this chapter, we have reviewed the literature on pharmacokinetic drug interactions of anti-mycobacterial drugs. These drug interactions are important since impact efficacy and toxicity of drugs that are part of a treatment regimen in which very few treatment options are open, especially in case of MDR-TB. Drug-drug, food-drug, and herbal medicine-drug interactions are described focusing on the effect of the interaction on the antimicrobial drug itself (antimicrobial drugs as victim) or on the effect of the co-prescribed drug (antimicrobial drugs as perpetrators).

Chapter 1

12

Chapter 3: In the third chapter, we studied a pharmacokinetic drug-drug interaction between linezolid and clarithromycin based on a remarkable clinical finding from a patient admitted at the Tuberculosis Center Beatrixoord (University Medical Center Groningen, Haren, The Netherlands). As a part of routine therapeutic drug monitoring, linezolid blood concentrations were analysed. We discovered a considerable increase of the patient’s linezolid concentration after co-administration of clarithromycin. In this chapter we further explored the role of drug-drug interactions of linezolid with clarithromycin.

Chapter 4: Therapeutic drug monitoring has gradually become a more widely accepted tool to optimize individual treatment regimens of anti-TB drugs. However, logistical problems with conventional drug sampling, and cold-transport of blood specimens to laboratories limit the use of TDM to research-oriented institutes. In this chapter, we aimed to develop and clinically validate a technique to overcome these problems: dried blood spot analysis of linezolid in patients with MDR-TB.

Despite “lacking the drama of blood, sincerity of sweat, and the emotional appeal of tears (22)”, another potential advantageous matrix with non invasive sampling is oral fluid. In the second part of this chapter we developed and clinically validated the analysis of linezolid and clarithromycin in oral fluid of MDR-TB patients.

Chapter 5: In this chapter, we strived to get insight in linezolid efficacy, safety, and tolerability in patients with MDR-TB, focusing on pharmacokinetics and pharmacodynamics. Published data were lacking detailed information on pharmacokinetic / pharmacodynamic targets; several recently published studies did not incorporate TDM in their study designs. In two letters to the editor, we advocated the use of TDM in clinical trials in order to generate data on an anti-TB drug that is known to show inter-patient variability in pharmacokinetics and to display drug-interactions.

This encouraged us to retrospectively analyse the data that was generated in the previous years in our TB Center. In order to enlarge our cohort, we included patients from both our Tuberculosis Center Beatrixoord (University Medical Center Groningen, Haren, The Netherlands) and from the Tuberculosis Reference Center for MDR-TB and HIV-TB E. Morelli Hospital (Sondalo, Italy). For this retrospective study, we planned to select patients that received linezolid as a part of their treatment regimen for MDR-TB and that underwent TDM. We aimed to relate linezolid efficacy, safety, and tolerability to linezolid drug exposure in MDR-TB patients.

13


Chapter 6: As a rule, pharmacokinetic interactions are considered disadvantageous and potentially harmful. In this chapter, we study a potentially beneficial pharmacodynamic interaction between linezolid and clarithromycin. With a lack of new anti-TB drugs emerging from the pipeline, an effort should be made to optimize treatment regimens containing existing drugs with activity against M. tuberculosis. One of these drugs, clarithromycin, has a controversial role in treatment regimens due to the fact that blood levels of clarithromycin measured in patients are often below minimal inhibitory concentration as determined in vitro in clinical isolates. One reason to incorporate clarithromycin in treatment regimens is in vitro synergy between clarithromycin and isoniazid, rifampicin and/or ethambutol against M. tuberculosis (23). In this chapter, we investigated whether linezolid and clarithromycin display in vitro synergy in clinical isolates of Mycobacterium tuberculosis.

Chapter 7 and 8: In the seventh chapter, we present a summary of the findings of previous chapters. In the General Discussion of this thesis (chapter 8), we discussed the clinical impact of the studies presented on the role of linezolid in optimizing treatment for patients with MDR-TB. We discussed the clinical pharmacology, with especial focus on TDM, of linezolid in the context of MDR-TB, and present future perspectives.

REFERENCES

1. Paulson, T. 2013. Epidemiology: A mortal foe. Nature. 502:S2-3.

2. World Health Organization (WHO). 2013. Global tuberculosis report 2013. World Health Organization (WHO), Geneva, Switzerland.

3. MacPherson, D. W., and B. D. Gushulak. 2006. Balancing prevention and screening among international migrants with tuberculosis: population mobility as the major epidemiological influence in low-incidence nations. Public Health. 120:712-723.

4. James, B. W., A. Williams, and P. D. Marsh. 2000. The physiology and pathogenicity of Mycobacterium tuberculosis grown under controlled conditions in a defined medium. J. Appl. Microbiol. 88:669-677.

5. Wang, J., and M. A. Behr. 2014. Building a better bacillus: the emergence of Mycobacterium tuberculosis. Front. Microbiol. 5:139.

6. Verrall, A. J., M. G. Netea, B. Alisjahbana, P. C. Hill, and R. van Crevel. 2014. Early clearance of Mycobacterium tuberculosis: a new frontier in prevention. Immunology. 141:506-513.

7. van Altena, R., S. Duggirala, M. I. Groschel, and T. S. van der Werf. 2011. Immunology in tuberculosis: challenges in monitoring of disease activity and identifying correlates of protection. Curr. Pharm. Des. 17:2853-2862.

Chapter 1

14

8. Prabowo, S. A., M. I. Groschel, E. D. Schmidt, A. Skrahina, T. Mihaescu, S. Hasturk, R. Mitrofanov, E. Pimkina, I. Visontai, B. de Jong, J. L. Stanford, P. J. Cardona, S. H. Kaufmann, and T. S. van der Werf. 2013. Targeting multidrug-resistant tuberculosis (MDR-TB) by therapeutic vaccines. Med. Microbiol. Immunol. 202:95-104.

9. Mirlekar, B., S. Pathak, and G. Pathade. 2013. Mycobacterium tuberculosis: approach to development of improved strategies for disease control through vaccination and immunodiagnosis. Indian J. Lepr. 85:65-78.

10. Pai, M. 2013. Diagnosis of pulmonary tuberculosis: recent advances. J. Indian Med. Assoc. 111:332-336.

11. Mayock, R. L., and R. R. MacGregor. 1976. Diagnosis, prevention and early therapy of tuberculosis. Dis. Mon. 22:1-60.

12. World Health Organization (WHO). 2013. Factsheet Multidrug-resistant tuberculosis (MDR-TB): october 2013 update. World Health Organization, Geneva, Switzerland.

13. World Health Organisation (WHO) (ed.), 2011. Guidelines for the programmatic management of drug-resistant tuberculosis. World Health Organization, Geneva, Switzerland.

14. Lacoma, A., N. Garcia-Sierra, C. Prat, J. Ruiz-Manzano, L. Haba, S. Roses, J. Maldonado, and J. Dominguez. 2008. GenoType MTBDRplus assay for molecular detection of rifampin and isoniazid resistance in Mycobacterium tuberculosis strains and clinical samples. J. Clin. Microbiol. 46:3660-3667.

15. Boehme, C. C., P. Nabeta, D. Hillemann, M. P. Nicol, S. Shenai, F. Krapp, J. Allen, R. Tahirli, R. Blakemore, R. Rustomjee, A. Milovic, M. Jones, S. M. O’Brien, D. H. Persing, S. Ruesch-Gerdes, E. Gotuzzo, C. Rodrigues, D. Alland, and M. D. Perkins. 2010. Rapid molecular detection of tuberculosis and rifampin resistance. N. Engl. J. Med. 363:1005-1015.

16. Boehme, C. C., M. P. Nicol, P. Nabeta, J. S. Michael, E. Gotuzzo, R. Tahirli, M. T. Gler, R. Blakemore, W. Worodria, C. Gray, L. Huang, T. Caceres, R. Mehdiyev, L. Raymond, A. Whitelaw, K. Sagadevan, H. Alexander, H. Albert, F. Cobelens, H. Cox, D. Alland, and M. D. Perkins. 2011. Feasibility, diagnostic accuracy, and effectiveness of decentralised use of the Xpert MTB/RIF test for diagnosis of tuberculosis and multidrug resistance: a multicentre implementation study. Lancet. 377:1495-1505.

17. Helb, D., M. Jones, E. Story, C. Boehme, E. Wallace, K. Ho, J. Kop, M. R. Owens, R. Rodgers, P. Banada, H. Safi, R. Blakemore, N. T. Lan, E. C. Jones-Lopez, M. Levi, M. Burday, I. Ayakaka, R. D. Mugerwa, B. McMillan, E. Winn-Deen, L. Christel, P. Dailey, M. D. Perkins, D. H. Persing, and D. Alland. 2010. Rapid detection of Mycobacterium tuberculosis and rifampin resistance by use of on-demand, near-patient technology. J. Clin. Microbiol. 48:229-237.

18. Vassall, A., S. van Kampen, H. Sohn, J. S. Michael, K. R. John, S. den Boon, J. L. Davis, A. Whitelaw, M. P. Nicol, M. T. Gler, A. Khaliqov, C. Zamudio, M. D. Perkins, C. C. Boehme, and F. Cobelens. 2011. Rapid diagnosis of tuberculosis with the Xpert MTB/RIF assay in high burden countries: a cost-effectiveness analysis. PLoS Med. 8:e1001120.

15


19. van Klingeren, B., M. Dessens-Kroon, T. van der Laan, K. Kremer, and D. van Soolingen. 2007. Drug susceptibility testing of Mycobacterium tuberculosis complex by use of a high-throughput, reproducible, absolute concentration method. J. Clin. Microbiol. 45:2662-2668.

20. Dooley, K. E., E. A. Obuku, N. Durakovic, V. Belitsky, C. Mitnick, E. L. Nuermberger, and on behalf of the Efficacy Subgroup, RESIST-TB. 2012. World Health Organization Group V Drugs for the Treatment of Drug-Resistant Tuberculosis: Unclear Efficacy Or Untapped Potential? J. Infect. Dis. 207:1352-1358.

21. IATDMCT Executive Committee Definitions of Therapeutic Drug Monitoring and Clinical Toxicity. http://www.iatdmct.org/about-us/about-association/about-definitions-tdm-ct.html. Date last updated: 2011. Date last accessed: December 31th 2014.

22. Mandel, I. D. 1990. The diagnostic uses of saliva. J. Oral Pathol. Med. 19:119-125.

23. Cavalieri, S. J., J. R. Biehle, and W. E. Sanders Jr. 1995. Synergistic activities of clarithromycin and antituberculous drugs against multidrug-resistant Mycobacterium tuberculosis. Antimicrob. Agents Chemother. 39:1542-1545.

Pharmacokinetic drug

interactions of antimicrobial

drugs: a systematic review

on oxazolidinones

M.S. Bolhuis, P.N. Panday, A.D. Pranger, J.G.W. Kosterink, and J.W.C. Alffenaar

Adapted from: Bolhuis MS, Panday PN, Pranger AD, Kosterink JGW, Alffenaar J-W. Pharmacokinetic Drug Interactions of

Antimicrobial Drugs: A Systematic Review on Oxazolidinones, Rifamycines, Macrolides, Fluoroquinolones, and Beta-Lactams.

Pharmaceutics. 2011; 3(4):865-913.

Chapter 2

Chapter 2

18

ABSTRACT

Like any other drug, antimicrobial drugs are prone to pharmacokinetic drug interactions. These drug interactions are a major concern in clinical practice as they may have an effect on efficacy and toxicity. The original article provides an overview of all published pharmacokinetic studies on drug interactions of the commonly prescribed antimicrobial drugs oxazolidinones, rifamycines, macrolides, fluoroquinolones, and beta-lactams, focusing on systematic research. However, in this chapter we present the data of the oxazolidinones. We describe drug-food and drug-drug interaction studies in humans, affecting antimicrobial drugs as well as concomitantly administered drugs. Since knowledge on mechanisms is of paramount importance for adequate management of drug interactions, the most plausible underlying mechanism of the drug interaction is provided when available. This overview can be used in daily practice to support management of pharmacokinetic drug interactions of antimicrobial drugs.

19

Review of pharmacokinetic drug interactions

2

INTRODUCTION

Antimicrobial drugs manifest a wide variety of drug interactions, which can differ greatly in extent of severity and clinical relevance. Not only co-medication, but also food and herbal medicine can interact with antimicrobial drugs and vice versa. The nature of these interactions can be of pharmacodynamic (PD) and/or pharmacokinetic (PK) origin.

A PD interaction consists of an alteration of a pharmacological response, through either agonism or antagonism, without affecting the kinetics of the drug. In cases of PD interactions physicians are advised to re-evaluate the benefit-risk ratio of the co-prescribed drug for each individual patient (1). PK interactions result in an altered disposition of a drug within a patient and can take place at the level of each of four processes influencing drug exposure, i.e. absorption, distribution, metabolism, and excretion, commonly described by the acronym ADME. Historically the relevance of drug distribution, particularly of protein binding, has been over-emphasized in the assessment of drug interactions, and nowadays the main cause of drug-drug interactions has been recognized to be modulation of the activity, i.e. inhibition or induction, of cytochrome P450 (CYP) enzymes and transporters.

Clinicians, prescribing the drug and pharmacists — often involved in medication review, therapeutic drug monitoring (TDM), or consultation on drug choice or dose — should be aware of clinically relevant interactions between antimicrobial drugs and co-medication, herbal medicine, and/or food in order to avoid toxicity, side effects, or inadequate treatment. PK interactions are in most cases manageable by adjusting the dose and by monitoring of drug levels (TDM) or vital signs. This review article will address PK interactions of antimicrobial drugs. The scope of the original article is to present an overview of PK studies on drug-drug and drug-food interactions of commonly prescribed antimicrobial drugs in daily clinical practice, i.e. oxazolidinones, rifamycines, macrolides, fluoroquinolones, and β-lactam antimicrobial drugs. In this chapter we present the data of the oxazolidinones.

EXPERIMENTAL SECTION

The Pubmed database was searched for PK interaction studies on drug-drug and drug-food interactions of antimicrobial drugs (Figure 1). The search was limited through the following selections: “Humans”, “Clinical Trial”, “Randomized Controlled Trial”, “Comparative Study”, and “Controlled Clinical Trial”. Only articles written in English were included. Per group of antimicrobial drugs, a separate search was conducted consisting of the name of the group,

Chapter 2

20

the name of the individual drugs, and the term “drug interaction”. When the Medical Subject Heading (MeSH) term “drug interaction” was used, indented terms such as “Herb-Drug Interaction” and “Food-Drug Interaction” were also searched. The search terms “NOT in vitro” and “NOT review” were added since this review focuses on original articles of studies with human subjects. Summaries of product characteristics or package leaflets were not consulted since these sources will only present a snapshot of the available information and will therefore not give a good overall impression of their use in clinical practice.

When a search resulted in only a few results, the query was expanded with the criterion “Case Report” and explicitly marked as such in this review since its contents have to be interpreted carefully because of the limited level of evidence.

All searches were conducted in March and April 2011. The relevant results were described per group of antimicrobial drugs. For each group, the drug interactions are divided into interactions affecting the antimicrobial drugs and interactions effecting the co-medication. The drug that is affected is identified by the term ‘victim’ and the drug that causes the effect by ‘perpetrator’. A table summarizing the most important drug interactions is provided for each group of antimicrobial drugs.

Since the scope of the original article is broader than the scope of this thesis, an adapted version is included. In this chapter, we only presented the drug interactions with drugs of

Figure 1 Scope of the original review and summary of the experimental section. The adapted version of the review, presented in Chapter 2, focuses on interactions with drugs of the oxazolidinone group. The gray area symbolizes the focus of this review, i.e. PK drug interactions of antimicrobial drugs.

Antimicrobial drugs

- oxazolidinones

- rifamycines (original article)

- macrolides (original article)

- fluoroquinolones (original article)

- β-lactams (original article)

This review:

Focus on PK interactions of antimicrobial drugs

Pubmed Search:

-Antimicrobial drug group name

OR

-Individual antimicrobial drug

AND

-Drug interaction

Limits:

“Humans”,

“Clinical Trial”,

“Randomized Controlled

Trial”,

“Controlled Clinical Trial”

“English”

“Comparative Study”

(“Case Report”)

“NOT in vitro”

“NOT review”

Co-medication Food Herbal medicine

21


2

the oxazolidinone group. Both interactions where the drug is the victim or the perpetrator are included in this chapter.

RESULTS AND DISCUSSION

Oxazolidinones

At this moment linezolid (LZD) is the only oxazolidinone authorized by the European Medicines Agency (EMA) and the US Food and Drug Administration (FDA). The number of properly designed drug-interaction studies with oxazolidinones is limited and the underlying mechanisms of some drug interactions are not yet fully elucidated. Furthermore, there is a lack of reviewed publications on drug interactions of newer compounds such as PNU-100480, posizolid (AZD2563), radezolid (RX-1741), torezolid and others, several of which are still being studied in phase I, II or III clinical research. A summary of LZD PK interactions is provided in Table 1.

Oxazolidinones as victims

Antimicrobial drugs: In an open-label comparative study of 8 healthy volunteers receiving 600 mg of both LZD and rifampicin (RIF) intravenously, a reduction of LZD plasma concentration was observed (2). An in vitro study demonstrated that LZD is not detectably metabolized by human CYP and did not inhibit the activities of human CYP isoforms 1A2, 2C9, 2C19, 2D6, 2E1, or 3A4 (3). Based on these observations along with the fact that RIF is a well-known P-gp inducer, the authors suggest LZD to be a P-gp substrate (2). This hypothesis was further supported by a case report of a patient with MDR-TB. This patient

Table 1 Summary of interactions of the oxazolidinone LZD with enzyme systems and/or food

Absorption Metabolism: Excretion:

Fat meal Antacids CYP P-gp Reactive Oxygen Species

LZD ↓ = - S* =

Downwards arrow (↓) indicates inhibition resulting in <50% decrease of AUC. “S” indicates the drug being a substrate, and “=” interaction is not relevant.* Mostly based on case reports: in need of further research.Note: Systematic research on newer compounds such as PNU-100480, posilozid (AZD2563), radezolid (RX-1741), torezolid, and others is not available.Since there were no interactions affecting displacement / distribution this process was not included in the table.

Chapter 2

22

received LZD and clarithromycin (CLR), a potent inhibitor of P-gp and a well-known CYP3A4 inhibitor. It was shown that co-administration of CLR with LZD resulted in a markedly increased LZD AUC (4).

The combination of aztreonam and LZD in an open-label cross-over study that included 13 healthy volunteers resulted in a statistically significant, although probably not clinically relevant increase of LZD AUC of approximately 18% (5). The authors suggest that the mechanism for this interaction is partly explained by a common elimination pathway, i.e. renal excretion. However, the definite mechanism remains unknown.

Food and antacids: In a two-phase single-dose open-label cross-over study of 12 healthy volunteers, a fatty meal caused a small but statistically significant reduction of mean LZD plasma concentration (6). The Cmax decreased by 23% and tmax increased from 1.5 hours to 2.2 hours, probably due to prolonged gastric residence time. An open-label cross-over study in 28 healthy volunteers tested the hypothesis that a disturbed balanced of reactive oxygen species might lower the in vivo clearance of LZD by supplementing dietary antioxidants, i.e. vitamin C and E, but concluded there was no significant effect on LZD Cmax and AUC (7). This is in line with current literature indicating that supplemented antioxidant vitamins have subtle effects on in vivo reactive oxygen species balance (7). A randomized open-label cross-over study of 17 healthy volunteers showed that the antacid Maalox has no effect on the PK of LZD (8).

Oxazolidinones as perpetrators

Serotonin reuptake inhibitors: A single randomized controlled trial (RCT) (9) and several case reports (10-23) describe LZD’s potential for drug interactions due to its reversible monoamine oxidase-A inhibitor activity. In case reports, serotonic toxicity was observed after co-administration of LZD with drugs that influence serotonin levels like selective serotonin reuptake inhibitors (SSRIs), tricyclic antidepressants, monoamine oxidase inhibitors, and other serotonergic agents such as citalopram, diphenhydramine, duloxetine, fluoxetine, paroxetine, sertraline, trazodone, and venlafaxine. However, one case report presented a depressed patient receiving co-administered mirtazepine and LZD being treated successfully without toxic signs (24). The RCT focused on the PK interaction of LZD with the over the counter (OTC) sympathomimetic drugs pseudoephedrine and phenylpropanolamine. A slight increase in blood pressure and a minimal effect on the PK of both co-administered drugs was found in 42 healthy individuals (9). The serotonin reuptake

23


2

inhibitor dextromethorphan was co-administered with LZD with no clinical effect: only a slight decrease of dextrorphan, the primary metabolite of dextromorphan, was observed (9).

CONCLUSIONS

The original article, presenting an overview of PK studies on drug-drug and drug-food interactions of macrolides, fluoroquinolones, rifamycines, oxazolidinones, and β-lactam antimicrobial drugs, can be used by physicians and pharmacists in daily practice to assist in preventing and managing PK drug interactions of antimicrobial drugs. This chapter, adapted from the original article, provides an overview of drug-interactions of oxazolidinones. The interactions presented vary in extent of severity and clinical relevance. Potential clinical problems can range from therapeutic failure due to low drug exposure to adverse events due to toxic drug concentrations. PK interaction studies in both patients and healthy volunteers are included. It has been demonstrated that PK characteristics of drugs can differ between healthy volunteers and patients (25). As a result of an underlying disease, physiological changes can influence drug PK, although the mechanism remains to be elucidated. In many critically ill patients extracellular fluids have increased, possibly resulting in a higher volume of distribution that might affect PK (26). One should bear in mind that findings in PK interaction studies performed in healthy volunteers might not be observed in clinical practice in specific patient populations.

Furthermore, PK interaction studies administering both single doses and multiple doses to study subjects were used in this overview. It need hardly be mentioned that multiple-dose studies will reflect best clinical practice. This is particularly true for PK interaction studies with biotransformation as possible underlying mechanism since induction of enzyme systems might require days to 2 – 3 weeks to develop fully (27). The interaction may also persist at a similar length of time when the inducing agent is stopped. Unlike induction, inhibition of enzyme systems can occur within 2 – 3 days (27).

Physicians and pharmacists should also be aware of the fact that some of the included studies used doses that are higher or lower than those used in daily clinical practice. Especially in non-linear PK, this makes PK interactions difficult to interpret.

Finally, drug interactions not only occur when two or more interacting drugs are administered, but can also surface when one of the interacting drugs is halted. Most electronic health record systems include a program that can routinely check for drug-drug interactions and

Chapter 2

24

could assist in preventing drug interaction-related adverse events. However, these programs rarely check for interactions that can occur when one of the interacting drugs is halted. Multidisciplinary vigilance of physicians, pharmacists, and other health care professionals remains necessary for adequate management of drug-interactions of antimicrobial drugs.

REFERENCES

1. Bruggemann, R. J. M., J. W. Alffenaar, N. M. A. Blijlevens, E. M. Billaud, J. G. Kosterink, P. E. Verweij, and Burger. 2008. Pharmacokinetic drug interactions of azoles. Curr Fungal Infect Rep. 2:20-27.

2. Egle, H., R. Trittler, K. Kummerer, and S. W. Lemmen. 2005. Linezolid and rifampin: Drug interaction contrary to expectations? Clin. Pharmacol. Ther. 77:451-453.

3. Wynalda, M. A., M. J. Hauer, and L. C. Wienkers. 2000. Oxidation of the novel oxazolidinone antibiotic linezolid in human liver microsomes. Drug Metab. Dispos. 28:1014-1017.

4. Bolhuis, M. S., R. van Altena, D. R. Uges, T. S. van der Werf, J. G. Kosterink, and J. W. Alffenaar. 2010. Clarithromycin significantly increases linezolid serum concentrations. Antimicrob. Agents Chemother. 54:5418-5419.

5. Sisson, T. L., G. L. Jungbluth, and N. K. Hopkins. 1999. A pharmacokinetic evaluation of concomitant administration of linezolid and aztreonam. J. Clin. Pharmacol. 39:1277-1282.

6. Welshman, I. R., T. A. Sisson, G. L. Jungbluth, D. J. Stalker, and N. K. Hopkins. 2001. Linezolid absolute bioavailability and the effect of food on oral bioavailability. Biopharm. Drug Dispos. 22:91-97.

7. Gordi, T., L. H. Tan, C. Hong, N. J. Hopkins, S. F. Francom, J. G. Slatter, and E. J. Antal. 2003. The pharmacokinetics of linezolid are not affected by concomitant intake of the antioxidant vitamins C and E. J. Clin. Pharmacol. 43:1161-1167.

8. Grunder, G., Y. Zysset-Aschmann, F. Vollenweider, T. Maier, S. Krahenbuhl, and J. Drewe. 2006. Lack of pharmacokinetic interaction between linezolid and antacid in healthy volunteers. Antimicrob. Agents Chemother. 50:68-72.

9. Hendershot, P. E., E. J. Antal, I. R. Welshman, D. H. Batts, and N. K. Hopkins. 2001. Linezolid: pharmacokinetic and pharmacodynamic evaluation of coadministration with pseudoephedrine HCl, phenylpropanolamine HCl, and dextromethorpan HBr. J. Clin. Pharmacol. 41:563-572.

10. Mason, L. W., K. S. Randhawa, and E. C. Carpenter. 2008. Serotonin toxicity as a consequence of linezolid use in revision hip arthroplasty. Orthopedics. 31:1140.

11. Das, P. K., D. I. Warkentin, R. Hewko, and D. L. Forrest. 2008. Serotonin syndrome after concomitant treatment with linezolid and meperidine. Clin. Infect. Dis. 46:264-265.

12. Packer, S., and S. A. Berman. 2007. Serotonin syndrome precipitated by the monoamine oxidase inhibitor linezolid. Am. J. Psychiatry. 164:346-347.

25


2

13. Steinberg, M., and A. K. Morin. 2007. Mild serotonin syndrome associated with concurrent linezolid and fluoxetine. Am. J. Health. Syst. Pharm. 64:59-62.

14. Strouse, T. B., T. N. Kerrihard, C. A. Forscher, and P. Zakowski. 2006. Serotonin syndrome precipitated by linezolid in a medically ill patient on duloxetine. J. Clin. Psychopharmacol. 26:681-683.

15. DeBellis, R. J., O. P. Schaefer, M. Liquori, and G. A. Volturo. 2005. Linezolid-associated serotonin syndrome after concomitant treatment with citalopram and mirtazepine in a critically ill bone marrow transplant recipient. J. Intensive Care Med. 20:351-353.

16. Morales, N., and H. Vermette. 2005. Serotonin syndrome associated with linezolid treatment after discontinuation of fluoxetine. Psychosomatics. 46:274-275.

17. Thomas, C. R., M. Rosenberg, V. Blythe, and W. J. Meyer 3rd. 2004. Serotonin syndrome and linezolid. J. Am. Acad. Child Adolesc. Psychiatry. 43:790.

18. Jones, S. L., E. Athan, and D. O’Brien. 2004. Serotonin syndrome due to co-administration of linezolid and venlafaxine. J. Antimicrob. Chemother. 54:289-290.

19. Tahir, N. 2004. Serotonin syndrome as a consequence of drug-resistant infections: an interaction between linezolid and citalopram. J. Am. Med. Dir. Assoc. 5:111-113.

20. Serio, R. N. 2004. Acute delirium associated with combined diphenhydramine and linezolid use. Ann. Pharmacother. 38:62-65.

21. Hammerness, P., H. Parada, and A. Abrams. 2002. Linezolid: MAOI activity and potential drug interactions. Psychosomatics. 43:248-249.

22. Wigen, C. L., and M. B. Goetz. 2002. Serotonin syndrome and linezolid. Clin. Infect. Dis. 34:1651-1652.

23. Lavery, S., H. Ravi, W. W. McDaniel, and Y. R. Pushkin. 2001. Linezolid and serotonin syndrome. Psychosomatics. 42:432-434.

24. Aga, V. M., N. E. Barklage, and J. W. Jefferson. 2003. Linezolid, a monoamine oxidase inhibiting antibiotic, and antidepressants. J. Clin. Psychiatry. 64:609-611.

25. Dickinson, L., S. Khoo, and D. Back. 2008. Differences in the pharmacokinetics of protease inhibitors between healthy volunteers and HIV-infected persons. Curr. Opin. HIV. AIDS. 3:296-305.

26. Gomez, C. M., J. J. Cordingly, and M. G. Palazzo. 1999. Altered pharmacokinetics of ceftazidime in critically ill patients. Antimicrob. Agents Chemother. 43:1798-1802.

27. Stockley, I. H. 2002. General considerations and an outline survey of some basic interaction mechanisms, p. 1-14. In I. H. Stockley (ed.), Stockley’s Drug Interactions, 6th ed., Pharmaceutical Press.

Clarithromycin significantly

increases linezolid serum

concentrations

M.S. Bolhuis, R. van Altena, D.R.A. Uges, T.S. van der Werf, J.G.W. Kosterink, and J.W.C. Alffenaar

Antimicrob Agents Chemother. 2010 Dec; 54(12)5418–9

Chapter 3A

Chapter 3A

28

TO THE EDITOR

A 42-year-old male patient was admitted at our hospital for treatment of smear-positive pulmonary tuberculosis (TB). Drug Susceptibility Testing (DST) revealed extensively drug resistant tuberculosis (XDR-TB) and the isolate appeared only susceptible to cycloserin, linezolid, clarithromycin and clofazimine. According to the WHO treatment guidelines for TB the treatment regimen was composed of these four drugs as no other options were available (11). Linezolid was the cornerstone of this regimen because of the high in vitro activity against M. tuberculosis (MIC of 0.125 – 0.5 mg/L) (1, 8). Linezolid is a toxic drug and its labelled duration of administration is therefore limited to 28 days to prevent peripheral neuropathy and anaemia. Dose reduction has been evaluated in TB patients as an attempt to reduce toxicity to allow for prolonged treatment for 18 – 24 months (6, 7, 10). The target of linezolid serum concentrations in our hospital is to maintain an AUC (the area under the concentration-time curve over 24 h) / MIC ratio over 100 and time in excess of the MIC of 100%. These conditions are generally reached with a dosage of 300 mg twice daily (2). Serum concentrations are analysed using a validated liquid chromatography tandem mass-spectrophotometer method (5). In this patient, we measured a considerable increase in the AUC of linezolid from 29 mg*h/L to 108 mg*h/L (Figure 1).

Figure 1 Linezolid serum concentrations over time before (solid circles) and after (open circles)

addition of clarithromycin.

Time post dosage (hours)

0 2 4 6 8 10 12 14

Lin

ez

oli

d s

eru

m c

on

cen

tra

tio

n (

mg

/L)

0

2

4

6

8

10

12

14

300 mg linezolid

300 mg linezolid + clarithromycin

29

Clarithromycin increases linezolid concentration

3A

This increase appeared to coincide with the start of clarithromycin (1000 mg once daily), a potent inhibitor of P-glycoproteins (3). We also found that the tmax of the absorption phase was delayed. Possible other drug-drug interactions were not expected as the patient received only clofazimine, domperidone, insulin, omeprazole, and vitamins. We did not observe any significant changes in his liver- or renal function to account for the sudden raise of linezolid serum concentrations. Based on the serum concentrations, the linezolid dose was decreased to 150 mg twice a day. After 6 months the sputum cultures and smear microscopy became negative. The patient was discharged from our Center in a good clinical condition. At follow-up his sputum cultures have remained negative for the last 12 months of his 18 months treatment, and his clinical condition has remained excellent. The timely reduction of linezolid dosage might have prevented toxicity such as time- and dose dependant myelosuppression (9). This case further strengthens the suggestion that linezolid is a P-glycoprotein substrate. From an earlier case it is known that the addition of a potent P-glycoprotein inducer rifampicin, resulted in a reduction of linezolid concentration (4). In our case the administration of the P-glycoprotein inhibitor clarithromycin resulted in a clear increase of linezolid serum concentrations. Based on our observations, a dose reduction of linezolid and therapeutic drug monitoring should be considered if linezolid is co-administered with clarithromycin in order to prevent potential toxicity. A prospective pharmacokinetic study may help to quantify the interaction that we describe.

Financial support

JWCA; this publication was prepared as part of the training in Clinical Pharmacology and was financially supported by the Dutch Society for Clinical Pharmacology and Biopharmacy.

Conflict of interest

None to declare.

Chapter 3A

30

REFERENCES

1. Alcala, L., M. J. Ruiz-Serrano, C. Perez-Fernandez Turegano, D. Garcia De Viedma, M. Diaz-Infantes, M. Marin-Arriaza, and E. Bouza. 2003. In vitro activities of linezolid against clinical isolates of Mycobacterium tuberculosis that are susceptible or resistant to first-line antituberculous drugs. Antimicrob.Agents Chemother. 47:416-417.

2. Alffenaar, J. W., J. G. Kosterink, A. R. van, T. S. van der Werf, D. R. Uges, and J. H. Proost. 2010. Limited sampling strategies for therapeutic drug monitoring of linezolid in patients with multidrug-resistant tuberculosis. Ther.Drug Monit. 32:97-101.

3. Eberl, S., B. Renner, A. Neubert, M. Reisig, I. Bachmakov, J. Konig, F. Dorje, T. E. Murdter, A. Ackermann, H. Dormann, K. G. Gassmann, E. G. Hahn, S. Zierhut, K. Brune, and M. F. Fromm. 2007. Role of p-glycoprotein inhibition for drug interactions: evidence from in vitro and pharmacoepidemiological studies. Clin.Pharmacokinet. 46:1039-1049.

4. Gebhart, B. C., B. C. Barker, and B. A. Markewitz. 2007. Decreased serum linezolid levels in a critically ill patient receiving concomitant linezolid and rifampin. Pharmacotherapy 27:476-479.

5. Harmelink, I. M., J. W. C. Alffenaar, A. M. A. Wessels, B. Greijdanus, and D. R. A. Uges. 2008. A rapid and simple liquid chromatography-tandem mass spectrometry method for the determination of linezolid in human serum. Eur.J.Hosp.Pharm. 14:5-8.

6. Koh, W. J., O. J. Kwon, H. Gwak, J. W. Chung, S. N. Cho, W. S. Kim, and T. S. Shim. 2009. Daily 300 mg dose of linezolid for the treatment of intractable multidrug-resistant and extensively drug-resistant tuberculosis. J.Antimicrob.Chemother. 64:119-1120.

7. Park, I. N., S. B. Hong, Y. M. Oh, M. N. Kim, C. M. Lim, S. D. Lee, Y. Koh, W. S. Kim, D. S. Kim, W. D. Kim, and T. S. Shim. 2006. Efficacy and tolerability of daily-half dose linezolid in patients with intractable multidrug-resistant tuberculosis. J.Antimicrob.Chemother. 58:701-704.

8. Rodriguez, J. C., M. Ruiz, M. Lopez, and G. Royo. 2002. In vitro activity of moxifloxacin, levofloxacin, gatifloxacin and linezolid against Mycobacterium tuberculosis. Int.J.Antimicrob.Agents 20:464-467.

9. Vinh, D. C. and E. Rubinstein. 2009. Linezolid: a review of safety and tolerability. J.Infect. 59 Suppl 1:S59-S74.

10. von der Lippe, B., P. Sandven, and O. Brubakk. 2006. Efficacy and safety of linezolid in multidrug resistant tuberculosis (MDR-TB)--a report of ten cases. J.Infect. 52:92-96.

11. World Health Organisation. 2010. Treatment of Tuberculosis: guidelines for national programmes. report 4th edition.

Clarithromycin increases

linezolid exposure in

multidrug-resistant


M.S. Bolhuis, R. van Altena, D. van Soolingen, W.C.M. de Lange, D.R.A. Uges, T.S. van der Werf,

J.G.W. Kosterink, and J.W.C Alffenaar

Eur Respir J. 2013 Dec; 42(6): 1614–21

Chapter 3B

Chapter 3B

32

ABSTRACT

The use of linezolid for the treatment of multidrug-resistant tuberculosis is limited by dose- and time-dependent toxicity. Recently, we reported a case of pharmacokinetic drug-drug interaction between linezolid and clarithromycin resulting in increased linezolid exposure. The aim of this prospective pharmacokinetic study is to quantify the effect of clarithromycin on the exposure of linezolid.

Subjects were included in an open-label, single-center, 1-arm, fixed-order pharmacokinetic interaction study. All subjects received 300 mg linezolid twice daily during the entire study, consecutively co-administered with 250 mg and 500 mg clarithromycin once daily. Steady-state serum curves of linezolid and clarithromycin were analyzed using validated methods and differences between pharmacokinetic parameters were calculated.

Linezolid exposure increased by a median of 44% (interquartile range: 23 – 102%, p=0.043) after co-administration of 500 mg clarithromycin (n=5) compared to baseline, whereas 250 mg clarithromycin had no statistically significant effect. Co-administration was well tolerated by most patients: none experienced severe adverse effects. One patient reported Common Toxicity Criteria Grade 2 gastro-intestinal adverse events.

In this study, we showed that clarithromycin significantly increased linezolid serum exposure after combining clarithromycin with linezolid in multidrug-resistant tuberculosis patients. The drug-drug interaction is possibly P-glycoprotein mediated. Due to large inter-patient variability, therapeutic drug monitoring is advisable to determine individual effect size.

33

Clarithromycin increases linezolid in MDR-TB patients

3B

INTRODUCTION

Multidrug-resistant tuberculosis (MDR-TB) is an infectious disease of major concern, especially in high TB burden countries (1-3). Treatment of MDR-TB poses challenges such as designing an effective second-line anti-tuberculosis regimen, entailing a combination of multiple drugs and a long duration of treatment (4). This translates in an intensive phase of at least 8 months and total treatment duration of at least 20 months as recommended by the WHO (5). In the intensive phase, treatment of MDR-TB should consist of at least four second-line anti-tuberculosis drugs likely to be effective. Additional drugs from Group 5, such as linezolid and clarithromycin, may be used, but their efficacy in the treatment of MDR-TB is unclear (5). Unfortunately, knowledge on the efficacy in the treatment of MDR-TB with these drugs is scarce.

Linezolid is a promising antimicrobial agent for the treatment of MDR-TB. However, evidence on the treatment of MDR-TB with linezolid is limited. Efficacy against Mycobacterium tuberculosis has been shown in vitro (6), in animals (7), and in patients (8-10). A recent meta-analysis confirms this efficacy, but shows the necessity of caution in the prescription of linezolid due to toxicity; almost 60% of all analyzed patients experienced adverse events (11). Adverse events, such anemia (38%), peripheral neuropathy (47%), gastro-intestinal side effects/symptoms (17%), optic neuritis (13%), and thrombocytopenia (12%), have all been reported and limit the use of linezolid (11). Reducing the dose of linezolid has been evaluated in an attempt to reduce toxicity (12). A dose of ≤600 mg linezolid daily resulted in lower frequency of adverse events than a dose of >600 mg daily (respectively 47% vs 75%), thereby enabling longer treatment duration (11).

Clarithromycin has a less prominent place in the treatment of MDR-TB. The minimal inhibitory concentration (MIC) of Mycobacterium tuberculosis was thought to be well in excess of achievable serum concentrations based on 12 strains of Mycobacterium tuberculosis (13). However, lower MICs have been observed (<2 mg/L) and clarithromycin shows concentrations in epithelial lining fluid that are often higher than in serum, enabling clarithromycin to be added to treatment regimens (14). Several Group 5 drugs, e.g. linezolid and clarithromycin, may need to be combined in a single MDR-TB treatment regimen, albeit that little is known on drug-drug interactions of these agents. Drug-drug interactions could compromise the efficacy of treatment regimens or could increase toxicity through reduced or increased exposure respectively.

Chapter 3B

34

Recently, we reported a pharmacokinetic drug-drug interaction between linezolid and clarithromycin that resulted in increased linezolid exposure (15). Increased serum linezolid concentrations could lead to toxicity, such as time- and dose-dependent severe myelosuppression and polyneuropathy. In a meta-analysis comparing a cohort treated with >600 mg linezolid per day with a cohort treated with ≤600 mg, there was a higher probability of anemia (60% vs 23%), leucopenia (17% vs 2%) and gastrointestinal symptoms (29% vs 8%) in the cohort that received >600 mg linezolid (11). Such toxicity could lead to the need to cease treatment with linezolid, severely limiting the treatment options left. Therefore, the aim of this prospective pharmacokinetic study was to quantify the effect of clarithromycin on the exposure of linezolid in adult MDR-TB patients hospitalized at the Tuberculosis Center Beatrixoord (Haren, the Netherlands).

METHOD

Study design

This study was an open-label, prospective single-center, 1-arm, fixed-order, interventional pharmacokinetic interaction study. The study was performed at the Tuberculosis Centre Beatrixoord (University of Groningen, University Medical Center Groningen, Haren, The Netherlands). All study subjects received standard care for their MDR-TB and co-morbidities. Treatment of MDR-TB was based on the World Health Organization (WHO) guideline (5) individualized for each included patient.

The primary objective was to quantify linezolid area under the concentration-time curve from 0 to 12 hours (AUC0–12h) without clarithromycin and with 250 mg and 500 mg clarithromycin once daily. Secondary objectives were to compare pharmacokinetic parameters of linezolid and clarithromycin between different dosing combinations and to describe tolerability and safety of co-administration of clarithromycin and linezolid in MDR-TB patients.

All patients gave written informed consent. The study protocol was approved by the Medical Ethical Review Committee of the University Medical Center Groningen (University of Groningen, Groningen, the Netherlands). The study was registered at clinicaltrials.gov (NCT01521364).

35


3B

Subjects

All study subjects were aged ≥18 years and were diagnosed with MDR-TB, confirmed with standard microbiological culture methods. The criteria for exclusion were based on the contraindications and known drug-drug interactions as mentioned in the Summary of Product Characteristics of linezolid and clarithromycin (16, 17). Subjects were excluded from the study if they were pregnant or lactating; had previously shown hypersensitivity to linezolid, any macrolide antibiotics, or any of the excipients of linezolid or clarithromycin; had hypokalemia; or concomitantly received P-glycoprotein modulators. Drug sensitivity testing (DST) was performed at the Dutch National Mycobacterial Reference Laboratory (National Institute for Public Health and the Environment [RIVM], Bilthoven, The Netherlands) using the Middlebrook 7H10 agar dilution method.

Treatment

All patients received linezolid 300 mg every 12 hours. In previous studies, we showed that this dose resulted in seemingly effective serum concentrations with a median (interquartile range, IQR) AUC0–12h of 57.6 (38.5 – 64.2) mg*h/L and AUC0–24h/MIC ratios of 452 (343 – 513) (12). Clarithromycin was added to therapy in a dose of 250 mg and 500 mg once daily consecutively during two weeks in a fixed order (Figure 1). From three cases at the Tuberculosis Center Beatrixoord, of which one case is published (15), it was expected that 500 mg clarithromycin would result in an approximately doubled linezolid exposure, matching the exposure of linezolid resulting of labeled dose of 600 mg twice daily.

Full linezolid pharmacokinetic curves were recorded at baseline (after one week of linezolid without clarithromycin), after receiving linezolid with 250 mg clarithromycin, and after

First blood curve (baseline): AUCInformed Second bloodcurve: Third blood curve:First blood curve (baseline):

LZD + 0 mg CLRAUC0-12h Trough

Informed

Consent

Second bloodcurve:

LZD + 250 mg CLR

Third blood curve:

LZD + 500 mg CLR

CLR 250 mg (once daily)CLR 500 mg (once daily)

LZD 300 mg (twice a day)

g y

LZD 300 mg (twice a day)

0 1 2 3 4 5 6

Week

Figure 1 Study design, showing dosing and sampling schedules of linezolid (LZD) and clarithromycin

(CLR).

Before, during, and after the study patients receive standard medical care and treatment for multidrug-resistant tuberculosis.

Chapter 3B

36

linezolid with 500 mg clarithromycin for two weeks (Figure 1). A trough sample was obtained after a washout period of one week, during which the patients only received linezolid besides their standard treatment, but no clarithromycin.

Sample size was derived from AUCs in a previous study in MDR-TB patients (12) and from the relative large increase of exposure observed in three cases ((15); two cases unpublished). To reach a desired power of 80%, a sample size of at least 5 patients was calculated using G*Power 3.1 (Heinrich Heine Universität, Dusseldorf, Germany). A drop-out rate of 15% was estimated based on previous studies at the Tuberculosis Center Beatrixoord (Haren, the Netherlands). To compensate for this estimated drop-out, seven patients were included.

Experimental procedures

The baseline linezolid pharmacokinetic curve and the trough after a one-week washout period were obtained at steady state, which is reached after approximately three days (16). Pharmacokinetic curves after co-administration of linezolid and clarithromycin were assessed at steady state after two weeks, allowing the pharmacokinetic interaction to develop fully (18). Blood samples were collected before and 1, 2, 3, 4, 8, and 12 hours after intake of medication. The second dosage of linezolid was given directly after this last blood sample. The patients did not receive standardized meals, but were allowed to eat a regular breakfast, reflecting common clinical practice, since food does not influence the linezolid exposure (19). Adherence was ensured through a directly observed inpatient treatment program.

Serum concentrations

Blood samples were drawn and after centrifuging serum samples were stored at -20°C until analysis. Linezolid and clarithromycin serum concentrations were analyzed using validated high performance liquid chromatography tandem mass-spectrometry methods (20, 21).

Tolerability and safety

The patients were clinically observed by nurses and attending physicians. Routine checks including blood tests were carried out at least weekly as part of continued standard care including monitoring for hyperlactatemia, haematological abnormalities such as thrombocytopenia and anemia. All patients received epoetine alpha (Eprex®) pre-emptively in a dose of 2000 IE twice a week to prevent anemia as part of standard care. Gastro-intestinal

37


3B

side effects were determined using the Common Toxicity Criteria (CTC) and were scored Grade 0 to 4 (22). Routine testing of neurotoxic adverse events through electromyogram (EMG) or vibration sense monitoring are not carried out during the study of 6 weeks, since these effects have been reported to occur after a median (range) of 16 (10 – 111) weeks (23). In case of clinical suspicion of peripheral neuropathy, a neurologist was consulted as is common practice at the Tuberculosis Center Beatrixoord. Furthermore, patients receiving linezolid were examined by an ophthalmologist once monthly, which is also common practice in this Center.

Pharmacokinetic and statistical analysis

The main study parameter, linezolid AUC0–12h and secondary study parameters clearance (CL), elimination constant (k) and elimination half life (t1/2) are calculated using trapezoidal rule in the Kinfit software (MWPharm 3.60; Mediware, Groningen, The Netherlands) (24). Pharmacokinetic parameters of linezolid and clarithromycin are described. Cmax was defined as the highest observed serum concentration and Cmin was defined as the concentration before intake of medication.

The hypothesis that the median of differences of AUC0–12h of linezolid at baseline compared to AUC0–12h after co-administration with either 250 mg or 500 mg clarithromycin equals zero was tested using the related-samples Wilcoxon Signed Rank test. Secondary pharmacokinetic parameters from the three curves were compared using the same related-samples Wilcoxon Signed Rank test. The non-parametric analysis of variances (ANOVA) Friedman test was used to test dose dependency of an effect of clarithromycin on linezolid exposure. All statistical evaluations were performed using SPSS 20 (SPSS, Chicago, IL, USA).

RESULTS

Patient characteristics

From December 2011 to October 2012, 16 patients with possible MDR-TB were admitted to the Tuberculosis Center Beatrixoord (Haren, the Netherlands). Two of these 16 patients were <18 years old, one patient was pregnant, one patient was participating in another study, for one patient the planned period of admission was too short, leaving eleven patients for formal screening. Four patients were not included in the study for various reasons. In one patient

Chapter 3B

38

DST revealed normal sensitivity, in another patient venous blood samples were not obtained due to venous access problems, rendering the collection of three full pharmacokinetic curves impossible; a third patient was included in another study; and the last patient was deemed psychologically too unstable to comply with the study protocol. Seven hospitalized patients were included in the study five of whom were suitable for evaluation. One of the included patients dropped out of the study in the fourth week due to medical reasons. The patient had a fever and was nauseous, probably due to an infected venous access port and possibly combined with side effects of clarithromycin and other anti-TB medication such as moxifloxacin. Another patient could not be evaluated due to a logistical problem with the study medication. The two patients that dropped out of the study were excluded from all analyses.

Table 1 Baseline demographics (n=5) and results from drug susceptibility testing

Parameter Value

Age – year† 35.0 (23 – 65)

Male sex – no. (%) 4 (80%)

Bodyweight – kg† 66.8 (55.2 – 78.5)

Height – m† 1.74 (1.67 – 1.82)

Body Mass Index† ¥ 22.1 (17.1 – 26.2)

Ethnicity – no.AfricanCaucasianAsian

3 1 1

HIV positive – no. 1

Isolate resistant to drug based on DST – no./no. totalEthambutolIsoniazidPyrazinamide∫

RifampicinStreptomycinCapreomycinAmikacinCiprofloxacinClaritromycin∫

Clofazimin∫

LinezolidMoxifloxacinProtionamide∫

Rifabutin

3/55/52/45/54/51/50/51/52/30/30/51/51/44/5

† Data presented as mean (range).¥ Body-mass index is calculated by dividing weight in kilograms by the square of the height in meters.∫ Drug susceptibility testing (DST) was not available for all isolates of the included patients.

39


3B

Patient baseline demographics and results from the drug susceptibility testing are presented in Table 1. The mean (range) age of included subjects was 35 (23 – 65) years and the mean (range) weight was 66.8 (55.2 – 78.5) kg. One of the patients was HIV positive and was treated with emtricitabin / tenofovir and raltegravir. Three patients originated from Somalia, one from Turkey, and one from the Netherlands.

Pharmacokinetic and statistical analysis

From all patients suitable for evaluation (n=5), three full pharmacokinetic curves in serum were available. The mean plasma concentration-time curves are shown in Figure 2. The baseline median (IQR) AUC0–12h of linezolid of 36.3 (33.2 – 46.3) mg*h/L in patients with a mean (range) body weight of 66.8 (55.2 – 78.5) kg, is lower than the median (IQR) AUC0–12h of linezolid of 57.6 (38.5 – 64.2) mg*h/L from a previous study with patients with a median (IQR) body weight of 58.3 (52.7 – 62.8) kg (12). Linezolid concentrations in serum increased after co-administration of clarithromycin compared to baseline, but display a large standard deviation. There appears to be no effect on time of Cmax, i.e. tmax.

Figure 2 Mean linezolid concentration-time curves in serum (n=5) for linezolid without clarithromycin

(solid circles with closed line), linezolid with 250 mg clarithromycin (open squares with dotted line),

and linezolid with 500 mg clarithromycin (open triangle with dashed line).

Standard deviations are presented as error bars. For visual purposes, error bars for linezolid with 250 mg clarithromycin are left out.

Time (hours after dose)

0 2 4 6 8 10 12 14

Lin

ez

oli

d c

on

ce

ntr

ati

on

(m

g/L

)

0

2

4

6

8

10

12

14

+ 0 mg clarithromycin



Chapter 3B

40

Pharmacokinetic parameters of linezolid and clarithromycin are presented in Table 2. Compared to baseline, the median AUC0–12h of linezolid increased statistically significantly after co-administration with 500 mg clarithromycin (p=0.043), but not after co-administration with 250 mg clarithromycin (p=0.686). After co-administration of linezolid with 500 mg clarithromycin, the median (IQR) AUC0–12h of linezolid increased by 44% (23 – 102%) compared to baseline. Furthermore, administration of 500 mg clarithromycin statistically significantly increased the Cmax of linezolid by median (IQR) 48% (35 – 103%, p=0.043), but not the Cmin of linezolid by median (IQR) 50% (44 – 189%, p=0.080) compared to baseline. There was no statistically significant difference in linezolid half life after co-administration of 500 mg clarithromycin with linezolid compared to linezolid alone (p=0.138). Linezolid clearance and elimination constant decreased statistically non-significant when linezolid and 500 mg clarithromycin are co-administered compared to baseline (both p=0.08). No dose-dependent effect of clarithromycin on the linezolid exposure could be detected using the Friedman test (p=0.091).

Safety / tolerability

Co-administration of linezolid and clarithromycin was well tolerated by most patients. None of the patients experienced severe adverse events, such as anemia, peripheral neuropathy, optic neuritis, or thrombocytopenia. One patient experienced CTC Grade 2 gastro-intestinal side effects three days after the start of administration of 500 mg clarithromycin once daily.

Table 2 Pharmacokinetic parameters of linezolid and clarithromycin (n=5)

Linezolid + 0 mg clarithromycin

Linezolid + 250 mg clarithromycin p-value

Linezolid + 500 mg clarithromycin p-value†

Linezolid AUC0–12u (mg*h/L)Cmax (mg/L)Cmin (mg/L)CL (L/h)kel (/h)t1/2 (h)

36.3 [33.2 – 46.3]6.0 [5.1 – 6.4] 1.2 [0.9 – 1.6]7.0 [5.4 – 8.0]0.17 [0.17 – 0.19]4.1 [3.6 – 4.2]

61.0 [34.6 – 63.9]8.0 [5.5 – 10.9] 2.1 [0.9 – 2.2]4.0 [3.5 – 7.8]0.14 [0.12 – 0.18]4.9 [3.8 – 5.7]

0.6860.1040.6860.6860.7850.686

67.2 [66.9 – 76.0]9.4 [8.9 – 10.5]2.6 [2.4 – 3.9]3.5 [2.7 – 3.5]0.13 [0.11 – 0.13]5.4 [5.4 – 6.5]

0.0430.0430.0800.0800.0800.138

Clarithromycin AUC0–12u (mg*h/L) N/A 8.2 [5.8 – 9.8] N/A 20.1 [14.0 – 23.6] 0.043¥

Data are presented as median [interquartile range]. † p-values comparing parameters from co-administration of linezolid with 500 mg clarithromycin to baseline.¥ p-value comparing AUC0–12h of 500 mg with 250 mg clarithromycin.

41


3B

DISCUSSION

In this study, we showed that clarithromycin significantly increased linezolid serum AUC0–12h after combining clarithromycin with linezolid in MDR-TB patients. After two weeks of co-administration of linezolid with clarithromycin 500 mg once daily, the Cmax of linezolid increased significantly by approximately 50%. Combining linezolid with clarithromycin in a dose of 500 mg once daily resulted in a significantly higher AUC0–12h of linezolid with a median of 44%. None of the patients experienced any severe adverse events. However, it should be noted that patients pre-emptively received epoetine alpha as part of standard care, potentially obscuring anemia as a side effect.

Besides our recent report on the interaction between clarithromycin and linezolid, there are no other reports on this pharmacokinetic drug-interaction. In fact, one of the few known drug interactions of linezolid to date is with rifampicin. Rifampicin, a well-known inducer of P-glycoprotein and cytochrome P 450 enzymes, decreases linezolid serum levels in critically ill patients (25). Another study confirmed this finding in healthy volunteers (26, 27). Gebhart et al. suggest the interaction to be mediated by P-glycoprotein, since an in vitro study has shown that linezolid is not metabolized by cytochrome P450 enzymes (28). The interaction of linezolid and clarithromycin could also be mediated by P-glycoprotein, since clarithromycin is a well-known cytochrome P450 3A4 inhibitor and a potent inhibitor of P-glycoproteins (29). P-glycoprotein is a membrane efflux transporter enzyme that is highly expressed in a variety of tissues including the intestine, liver, and kidney (30). Inhibition of the P-glycoprotein efflux pump by clarithromycin could result in the increased levels of linezolid, possibly a P-glycoprotein substrate, through inhibition of P-glycoprotein at the intestinal site as well as the renal site. P-glycoprotein polymorphism could explain some of the inter-patient variation that we observed. However, in a recent study Gandelman et al. refer to unpublished data on file from Pfizer suggesting linezolid is not a P-glycoprotein substrate (27). Their hypothesis for the observed interaction between linezolid and rifampicin is that a large increase in expression of cytochrome P450 3A (CYP3A) that typically has a small contribution (0.7 – 10.5%) to linezolid clearance, could cause a small decrease in linezolid exposure (27). Further research on the exact mechanism of the drug-drug interaction is needed.

Co-administration of clarithromycin and linezolid resulted in a near statistically significant decrease of clearance and elimination constant of linezolid compared to baseline. This might suggest inhibition of CYP3A or renal or hepatic P-glycoprotein efflux transporter pumps.

Chapter 3B

42

However, decreased clearance might not solely explain the observed increase of linezolid exposure. Unfortunately, due to the limited number of samples during the absorption phase, it is impossible to adequately compare data on absorption constant and tmax. Since patients did not receive intravenous linezolid, data on bioavailability is not available. It is therefore difficult to draw conclusions on involvement of inhibition of intestinal P-glycoprotein efflux transporters, which could result in increased absorption.

The increase of linezolid exposure after co-administration with clarithromycin has possible implications for clinical practice. The higher linezolid AUC0–12h could result in toxicity of linezolid, an agent that often causes adverse events, such as time- and dose-dependent severe myelosuppression and polyneuropathy. Severe adverse events often necessitate the cessation of effective anti-MDR-TB treatment, leaving few alternatives. Dose reduction of linezolid could prevent toxicity. However, care should be taken to assure adequate linezolid exposure and added information on whether linezolid exposure is too high, too low, or in the therapeutic range, could prove helpful. Therapeutic drug monitoring could help in assessing the linezolid dose after dose reduction (12), especially since the observed drug-drug interaction shows a large inter-patient variability. In limited resource settings, dried blood spot sampling could resolve logistical problems encountered with conventional therapeutic drug monitoring (31).

After evaluation of the combination of clarithromycin and linezolid in a larger population and during a longer period of time, clarithromycin could eventually even be used as a booster for linezolid, comparable to the use of low-dose ritonavir as a booster to improve protease inhibitor exposure in combined anti-retroviral therapy. The relatively cheap clarithromycin could reduce the dose of the expensive linezolid while the same exposure is maintained, thereby leaving the risk of toxicity unaltered. Since the highest prevalence of MDR-TB is found in countries with limited resources, such a booster strategy could make treatment with linezolid feasible for a larger group of patients. Such a cost reduction could even contribute to the call for making global MDR-TB control affordable (32). Further research on WHO Group 5 drugs, such as linezolid besides evaluation of new drugs such as delaminid (33) or old drugs such as co-trimoxazole (34), is of great importance.

In conclusion, we showed a 44% increase of linezolid AUC0–12h after co-administration of linezolid with clarithromycin in a dose of 500 mg daily in MDR-TB patients. The pharmacokinetic interaction between linezolid and clarithromycin is suggested to be P-gp mediated. Further research in a larger cohort is needed to provide insight in observed inter-

43


3B

patient variation, perhaps caused by P-gp polymorphism. Until effect size is predictable, possibly with help of P-gp polymorphism testing, therapeutic drug monitoring is advisable to determine individual effect size. The drug-drug interaction we showed in this study is an important step towards making the effective anti-TB drug linezolid available through cost reduction in less affluent settings where MDR-TB is highly prevalent.

ACKNOWLEDGMENTS

This study was financially supported by Stichting Beatrixoord Noord-Nederland (Groningen, The Netherlands). Reproduced with permission of the European Respiratory Society © Eur Respir J December 2013 42:1614-1621; published ahead of print March 21, 2013, doi:10.1183/09031936.00001913.

REFERENCES

1. Migliori, G. B., M. Dara, P. de Colombani, H. Kluge, and M. C. Raviglione. 2012. Multidrug-resistant tuberculosis in Eastern Europe: still on the increase? Eur. Respir. J. 39:1290-1291.

2. Jenkins, H. E., V. Plesca, A. Ciobanu, V. Crudu, I. Galusca, V. Soltan, A. Serbulenco, M. Zignol, A. Dadu, M. Dara, and T. Cohen. 2012. Assessing spatial heterogeneity of MDR-TB in a high burden country. Eur. Respir. J. 42:1291-1301.

3. Skrahina, A., H. Hurevich, A. Zalutskaya, E. Sahalchyk, A. Astrauko, W. van Gemert, S. Hoffner, V. Rusovich, and M. Zignol. 2012. Alarming levels of drug-resistant tuberculosis in Belarus: results of a survey in Minsk. Eur. Respir. J. 39:1425-1431.

4. Falzon, D., N. Gandhi, G. B. Migliori, G. Sotgiu, H. Cox, T. H. Holtz, M. G. Hollm-Delgado, S. Keshavjee, K. Deriemer, R. Centis, L. D’Ambrosio, C. Lange, M. Bauer, and D. Menzies. 2012. Resistance to fluoroquinolones and second-line injectable drugs: impact on MDR-TB outcomes. Eur. Respir. J. 42:156-168.


6. Prammananan, T., A. Chaiprasert, and M. Leechawengwongs. 2009. In vitro activity of linezolid against multidrug-resistant tuberculosis (MDR-TB) and extensively drug-resistant (XDR)-TB isolates. Int. J. Antimicrob. Agents. 33:190-191.

7. Cynamon, M. H., S. P. Klemens, C. A. Sharpe, and S. Chase. 1999. Activities of several novel oxazolidinones against Mycobacterium tuberculosis in a murine model. Antimicrob. Agents Chemother. 43:1189-1191.

8. De Lorenzo, S., J. W. Alffenaar, G. Sotgiu, R. Centis, L. D’Ambrosio, S. Tiberi, M. S. Bolhuis, R. van Altena, P. Viggiani, A. Piana, A. Spanevello, and G. B. Migliori. 2012. Efficacy and safety of meropenem/clavunate added to linezolid containing regimens in the treatment of M/XDR-TB. Eur. Respir. J. 41:1386-1392.

9. Lee, M., J. Lee, M. W. Carroll, H. Choi, S. Min, T. Song, L. E. Via, L. C. Goldfeder, E. Kang, B. Jin, H. Park, H. Kwak, H. Kim, H. Jeon, I. Jeong, J. S. Joh, R. Y. Chen, K. N. Olivier, P. A. Shaw, D. Follmann, S. D. Song, J. Lee, D. Lee, C. T. Kim, V. Dartois, S. Park, S. Cho, and C. E. Barry. 2012. Linezolid for Treatment of Chronic Extensively Drug-Resistant Tuberculosis. N. Engl. J. Med. 367:1508-1518.

Chapter 3B

44

10. Migliori, G. B., B. Eker, M. D. Richardson, G. Sotgiu, J. P. Zellweger, A. Skrahina, J. Ortmann, E. Girardi, H. Hoffmann, G. Besozzi, N. Bevilacqua, D. Kirsten, R. Centis, C. Lange, and TBNET Study Group. 2009. A retrospective TBNET assessment of linezolid safety, tolerability and efficacy in multidrug-resistant tuberculosis. Eur. Respir. J. 34:387-393.

11. Sotgiu, G., R. Centis, L. D’Ambrosio, J. Alffenaar, H. Anger, J. Caminero, P. Castiglia, S. De Lorenzo, G. Ferrara, W. Koh, G. Schecter, T. Shim, R. Singla, A. Skrahina, A. Spanevello, Z. Udwadia, M. Villar, E. Zampogna, J. Zellweger, A. Zumla, and G. B. Migliori. 2012. Efficacy, safety and tolerability of linezolid containing regimens in treating MDR-TB and XDR-TB: systematic review and meta-analysis. Eur. Respir. J. 40:1430-1442.

12. Alffenaar, J. W., R. van Altena, I. M. Harmelink, P. Filguera, E. Molenaar, A. M. Wessels, D. van Soolingen, J. G. Kosterink, D. R. Uges, and T. S. van der Werf. 2010. Comparison of the pharmacokinetics of two dosage regimens of linezolid in multidrug-resistant and extensively drug-resistant tuberculosis patients. Clin. Pharmacokinet. 49:559-565.

13. Truffot-Pernot, C., N. Lounis, J. H. Grosset, and B. Ji. 1995. Clarithromycin is inactive against Mycobacterium tuberculosis. Antimicrob. Agents Chemother. 39:2827-2828.

14. Honeybourne, D., F. Kees, J. M. Andrews, D. Baldwin, and R. Wise. 1994. The levels of clarithromycin and its 14-hydroxy metabolite in the lung. Eur. Respir. J. 7:1275-1280.


16. Pfizer. 2005. Zyvoxid. Product Information.

17. Sandoz. 2004. Clarithromycin. Product Information.

18. Stockley, I. H. 2002. General considerations and an outline survey of some basic interaction mechanisms, p. 1-14. In I. H. Stockley (ed.), Stockley’s Drug Interactions, 6th ed., . Pharmaceutical Press.

19. Islinger, F., P. Dehghanyar, R. Sauermann, C. Burger, C. Kloft, M. Muller, and C. Joukhadar. 2006. The effect of food on plasma and tissue concentrations of linezolid after multiple doses. Int. J. Antimicrob. Agents. 27:108-112.

20. Harmelink, I. M., J. W. Alffenaar, A. M. Wessels, B. Greijdanus, and D. R. Uges. 2008. A rapid and simple liquid chromatography-tandem mass spectrometry method for the determination of linezolid in human serum. EJHP Science. 14:3-7.

21. de Velde, F., J. W. Alffenaar, A. M. Wessels, B. Greijdanus, and D. R. Uges. 2009. Simultaneous determination of clarithromycin, rifampicin and their main metabolites in human plasma by liquid chromatography-tandem mass spectrometry. J. Chromatogr. B. Analyt Technol. Biomed. Life. Sci. 877:1771-1777.

22. Trotti, A., A. D. Colevas, A. Setser, V. Rusch, D. Jaques, V. Budach, C. Langer, B. Murphy, R. Cumberlin, C. N. Coleman, and P. Rubin. 2003. CTCAE v3.0: development of a comprehensive grading system for the adverse effects of cancer treatment. Semin. Radiat. Oncol. 13:176-181.

23. Anger, H. A., F. Dworkin, S. Sharma, S. S. Munsiff, D. M. Nilsen, and S. D. Ahuja. 2010. Linezolid use for treatment of multidrug-resistant and extensively drug-resistant tuberculosis, New York City, 2000-06. J. Antimicrob. Chemother. 65:775-783.

24. Proost, J. H., and D. K. Meijer. 1992. MW/Pharm, an integrated software package for drug dosage regimen calculation and therapeutic drug monitoring. Comput. Biol. Med. 22:155-163.

25. Gebhart, B. C., B. C. Barker, and B. A. Markewitz. 2007. Decreased serum linezolid levels in a critically ill patient receiving concomitant linezolid and rifampin. Pharmacotherapy. 27:476-479.


45


3B

27. Gandelman, K., T. Zhu, O. A. Fahmi, P. Glue, K. Lian, R. S. Obach, and B. Damle. 2011. Unexpected effect of rifampin on the pharmacokinetics of linezolid: in silico and in vitro approaches to explain its mechanism. J. Clin. Pharmacol. 51:229-236.


29. Eberl, S., B. Renner, A. Neubert, M. Reisig, I. Bachmakov, J. Konig, F. Dorje, T. E. Murdter, A. Ackermann, H. Dormann, K. G. Gassmann, E. G. Hahn, S. Zierhut, K. Brune, and M. F. Fromm. 2007. Role of p-glycoprotein inhibition for drug interactions: evidence from in vitro and pharmacoepidemiological studies. Clin. Pharmacokinet. 46:1039-1049.

30. Thiebaut, F., T. Tsuruo, H. Hamada, M. M. Gottesman, I. Pastan, and M. C. Willingham. 1987. Cellular localization of the multidrug-resistance gene product P-glycoprotein in normal human tissues. Proc. Natl. Acad. Sci. U. S. A. 84:7735-7738.

31. Vu, D. H., M. S. Bolhuis, R. A. Koster, B. Greijdanus, W. C. de Lange, R. van Altena, J. R. Brouwers, D. R. Uges, and J. W. Alffenaar. 2012. Dried blood spot analysis for therapeutic drug monitoring of linezolid in patients with multidrug-resistant tuberculosis. Antimicrob. Agents Chemother. 56:5758-5763.

32. Loddenkemper, R., G. Sotgiu, and C. D. Mitnick. 2012. Cost of tuberculosis in the era of multidrug resistance: will it become unaffordable? Eur. Respir. J. 40:9-11.

33. Skripconoka, V., M. Danilovits, L. Pehme, T. Tomson, G. Skenders, T. Kummik, A. Cirule, V. Leimane, A. Kurve, K. Levina, L. J. Geiter, D. Manissero, and C. D. Wells. 2012. Delamanid Improves Outcomes and Reduces Mortality for Multidrug-Resistant Tuberculosis. Eur. Respir. J. 41:1393-1400.

34. Alsaad, N., R. van Altena, A. D. Pranger, D. van Soolingen, W. C. de Lange, T. S. van der Werf, J. G. Kosterink, and J. W. Alffenaar. 2012. Evaluation of co-trimoxazole in treatment of multidrug-resistant tuberculosis. Eur. Respir. J. 42:504-512.

Dried blood spot analysis

for therapeutic drug

monitoring of linezolid in

MDR-TB patients

D.H. Vu, M.S. Bolhuis, R.A. Koster, B. Greijdanus, W.C.M. de Lange, R. van Altena, J.R.B.J. Brouwers,

D.R.A. Uges, J.W.C Alffenaar

Antimicrob Agents Chemother. 2012 Nov; 56(11): 5758–63

Chapter 4A

Chapter 4A

48

ABSTRACT

Linezolid is a promising antimicrobial agent for the treatment of multidrug-resistant tuberculosis (MDR-TB), but its use is limited by toxicity. Therapeutic drug monitoring (TDM) may help to minimize toxicity whilst adequate drug exposure is maintained. Conventional plasma sampling and monitoring might be hindered by logistic problems in most parts of the world that may be solved by dried blood sampling (DBS). The aim of this study is to develop and validate a novel method for TDM of linezolid in MDR-TB patients using DBS.

Plasma, venous DBS and capillary DBS specimens were obtained simultaneously from eight patients receiving linezolid. A DBS method was developed and clinically validated by comparing DBS with plasma results using Passing-Bablok regression and Bland-Altman analysis.

This study showed that DBS analysis was reproducible and robust. Accuracy and between and within-day precision from three validation presented as bias and CV were less than 17.2% for lower limit of quantification level and 7.8% for other levels. The method showed a high recovery of approximately 95% and a low matrix-effect of less than 8.7%. DBS specimens were stable at 37°C for 2 months and at 50°C for one week. The concentration ratio of DBS/plasma was 1.2 (95% CI: 1.12 – 1.27). Linezolid exposure calculated from DBS and plasma showed good agreement.

In conclusion, DBS analysis of linezolid is a promising tool to optimize linezolid treatment in MDR-TB patients. Easy sampling procedure and high sample stability may facilitate TDM, even in underdeveloped countries with limited resources where conventional plasma sampling is not feasible.

49

DBS analysis of linezolid in MDR-TB patients

4A

INTRODUCTION

Linezolid is used as a second line drug in the treatment of multidrug-resistant tuberculosis (MDR-TB) due to its efficacy in vitro (21), in vivo (9) and in patients (1, 2, 14, 17, 34) against Mycobacterium tuberculosis. The World Health Organization (WHO) classifies linezolid as a reserve anti-tuberculosis drug for the treatment of multidrug-resistant/extensively drug-resistant tuberculosis (MDR/XDR-TB) (33). Linezolid is usually added to a treatment regimen consisting of anti-tuberculosis drugs for which the Mycobacterium tuberculosis is still susceptible. However, treatment with linezolid may be limited by toxicity, such as time- and dose-dependent neuropathy or myelosuppression (17, 29), urging dose reduction or cessation of treatment with linezolid. Therapeutic drug monitoring (TDM) can be used to implement dose reductions to limit toxicity, whilst preventing inadequate exposure. Efficacy predicting pharmacokinetic / pharmacodynamic (PK/PD) parameters, such as the area under the concentration-time curve to MIC ratio (AUC0–24h/MIC), might be helpful in evaluating linezolid dosages (1, 7, 26, 32). The AUC0–24h/MIC has been shown to be the best predictive model in a murine model (32), but evidence from human data are lacking. Further PK/PD data from TB-programs or large studies are needed for the development of evidence based PK/PD parameters, such as an AUC0–24h/MIC ratio target.

Linezolid treatment has been evaluated for TB treatment, in several case series (17, 23). However, neither drug susceptibility testing (DST) nor drug exposure assessment was performed for linezolid, making it difficult to draw conclusions on efficacy (5). For instance, drug-interactions with other antimicrobial agents might have occurred and may have had an impact on linezolid exposure (6, 15). In addition, conventional drug exposure evaluation for TB drugs using plasma samples might have been hindered in these studies by logistical challenges (30). The use of dried blood spot (DBS) sampling may provide a helpful alternative to conventional plasma sampling through simplified sampling procedure and increased sample stability. DBS sampling has been applied in the treatment of other infectious diseases like malaria and HIV (30). Other advantages may include a lower required blood sample volume and lower biohazard risk of DBS samples compared to conventional plasma samples (12, 18, 30). Compared to conventional sampling, DBS sampling may be hindered by inter and intra-patient hematocrit (Hct) variation causing different blood viscosity yielding a proportional analytical bias with Hct value. Furthermore, Hct may affect the drug blood /plasma partition ratio complicating the comparison with conventional plasma samples. In the development of a bioanalytical method for linezolid using DBS analysis important

Chapter 4A

50

patient related factors like blood spot volume, Hct value (3, 24) and difference between capillary and venous blood, have to be assessed during validation (12, 18, 25, 30). To enable individualized linezolid treatment the aim of this study was to develop and validate a method for DBS analysis and evaluate it in MDR-TB and XDR-TB patients.

MATERIALS AND METHODS

Patients

From September 2010 to March 2012, MDR-TB patients (≥18 years) were recruited from the Tuberculosis Centre Beatrixoord, University Medical Center Groningen (Haren, The Netherlands). Eligible for inclusion were patients receiving treatment with anti-tuberculosis drugs for which routine therapeutic drug monitoring was scheduled. Patients with bleeding disorders were excluded from the study. The study procedures were reviewed and approved by the local Ethics Committee. Patients receiving linezolid were included after providing written informed consent.

Sampling was performed at least one week after the start of linezolid treatment to ensure the steady-state was achieved. Venous blood samples were obtained our before drug intake and at 1, 2, 3, 4 and 8 hours after dosing according to a previous study (2) and local procedures for TDM of TB drugs to be able to calculate drug exposure and other PK parameters. Venous dried blood spot (vDBS) specimens were prepared by pipetting 50 μL venous blood onto Whatman 31 ET CHR paper. The remaining venous blood was centrifuged at 3000 rpm for 5 minutes at room temperature to attain plasma which was stored at -20°C until analysis. DBS specimens were obtained through a finger prick by dropping the blood directly on dried blood spot paper. DBS samples were obtained before drug intake, 2 and 8 hours after dosing, representing low, high and medium linezolid blood levels respectively. Both the vDBS and DBS samples were left to dry at room temperature and stored in sealed plastic bags with desiccant sachets at -20°C until analysis.

DBS analysis

To quantify DBS samples an 8 mm-diameter disc was punched out of each blood spot. Extraction of these discs was performed by sonication with a frequency of 47 kHz during a period of 20 minutes using 500 μL of extracting solvent consisting of cyanoimipramine

51


4A

0.3 mg/L (internal standard) and EDTA 1 g/L in water. From this solution, a volume of 200 μL was added to 750 μL of acetonitrile. The samples were vortexed for 1 minute and subsequently centrifuged at 11000 rpm for 5 minutes. An injection volume of 5 μL was analyzed using a validated LC-MS/MS analysis method (16). The plasma samples were prepared and analyzed using the same method.

DBS method validation

The DBS analytical method was validated in accordance with the recommendation of US Food and Drug Administration’s (FDA) Guidance for Industry Bioanalytical Method validation (27). For the validation, blood was prepared by mixing plasma, red blood cell and linezolid stock solution to achieve blood at desire concentration and Hct. Subsequently, the validation DBS samples were prepared by pipetting 50 μL of blood onto the paper. Linearity was assessed with 1/x2 weighting over a concentration range of 0.05 – 40 mg/L. Clinical relevant concentrations were well within the range of the assay standards (2). The within-day and between-day accuracy and precision were evaluated on four validation levels of LLOQ (lower limit of quantification), LOW, MED and HIGH at concentrations of 0.05, 0.25, 15 and 30 mg/L, respectively. Each validation level was analyzed in fivefold on three consecutive days. The matrix effect and the recovery of linezolid from DBS were determined using a common method (18, 31). The stability of DBS specimens was assessed by storing validation DBS at ambient condition and 37°C after one week, two weeks and two months. As a worst case scenario the stability of DBS specimens was also assessed at 50°C after one day, two days and one week. The stability was evaluated at LOW and HIGH levels in fivefold by comparing the analytical result with the nominal concentrations. In addition to the criteria suggested in the FDA guideline (27), the impact on assay accuracy and precision due to the variations of Hct and blood spot volume were evaluated. For these purposes, Hct of 20, 25, 30, 35, 40, 45 and 50%, and blood spot volumes of 30, 50, 70 and 90 μL were assessed. During the method validation, blood spot volume and Hct were standardized at 50 μL and 35%, respectively. The set of Hct of 35% reflects the Hct in tuberculosis patients (3).

Pharmacokinetic and pharmacodynamic evaluation

Pharmacokinetic parameters were evaluated using a non-compartmental model of the KINFIT module of MW Pharm (version 3.9; Mediware, The Netherlands). The AUC0–12h was calculated using the trapezoidal rule from 0 up to 12 hours and the AUC0–24h by doubling the

Chapter 4A

52

AUC0–12h. The maximum concentration (Cmax) was defined as the highest observed linezolid concentration with tmax as corresponding time. The elimination half-life (t1/2) was calculated by dividing the natural logarithm of 2 (ln2) by the elimination constant (ke) as calculated by MW Pharm. The apparent clearance (Cl) of linezolid was calculated by dose/AUC0–12h. The volume of distribution (Vd) was calculated by dividing Cl with ke.

The drug susceptibility testing of the Mycobacterium tuberculosis isolates was performed at the Dutch National Mycobacteria Reference Laboratory (National Institute for Public Health and the Environment; RIVM) using the Middlebrook 7H10 agar dilution method (28). The AUC0–24h/MIC ratio, often used as a predictive pharmacodynamic parameter for efficacy, was calculated (32).

Statistics

In the method validation, the bias was defined as the difference (in percentage) between analytical result and the nominal concentration. The method was clinically validated by comparing the concentrations of DBS and vDBS with plasma concentrations using Passing-Bablok regressions and Bland-Altman analysis by applying the software tool Analyse-it 2.20 (Analyse-it Software, Ltd). Conversion factors, calculated from geometric mean (v)DBS/plasma concentration ratios, were used to calculate conversed DBS and vDBS concentrations(4). Subsequently, the conversed concentrations were used to calculate the AUC0–12h of DBS and vDBS. The agreement between AUC0–12h value of conversed DBS and plasma was evaluated using Bland-Altman analysis. Spearman correlation and Wilcoxon signed-rank test was applied to other comparisons.

RESULTS

Patients

Eight patients with a median (IQR) age of 29 (24 – 33) years were included in this study. The baseline characteristics are presented in Table 1. The median (IQR) of Hct was 37.4 (33.0 – 41.4) %. At time of the study three of eight patients received linezolid 300 mg twice a day and five patients in a dose of 600 mg twice daily. Isolates of seven patients showed resistance to first-line drugs isoniazide, rifampicin, ethambutol, pyrazinamide, and streptomycin. The isolate of one patient showed resistance to all first-line drug except pyrazinamide. All DSTs

53


4A

revealed resistance for rifabutin, whereas one isolate showed fluoroquinolone-resistance and three protionamide-resistance. None of the patients experienced significant discomfort from the finger pricks during DBS sampling which was supported by the fact that all completed the three consecutive samples in this study.

DBS method validation

The DBS assay method showed linearity over the analytical concentration range. The pooled correlation coefficient was r2 = 0.9947. The regression equation is: concentration = (0.1635 ± 0.0025) × response + (0.0001 ± 0.0003). Within-day and between-day accuracy and precision showed CVs within accepted range. Within-day CVs ranged from 1.6% to 13.8% and between-day CVs from 3.5% to 10.2%. The mean measured concentration was within 98.7%

Table 1 Baseline demographics of study patients (n=8)

Parameter Value

Age (y) (#) 29 [24 – 33]

Sex (n)Male 2Female 6

Bodyweight (kg) (#) 60.5 [55.8 – 61.5]

Height (m) (#) 1.69 [1.67 – 1.76]

Body mass index (kg/m2) (#) 20.1 [19.0 – 21.4]

Ethnicity (n)Caucasian 4African 3Asian 1

Co-morbidity (n/N)HIV 1/8

Hemoglobin (g/L) (#) 7.0 [6.4 – 8.8]

Hematocrit (v/v*100%) (#) 37.4 [33.0 – 41.4]

TB treatment other than linezolid (n)Moxifloxacin 7 Amikacin/Kanamycin 6 Ethambuthol 5 Cotrimoxazole 3 Clofazimine 2 Ertapenem 2 Pyrazinamide 1

(#): median [interquartile range].

Chapter 4A

54

to 106.3% of the nominal concentration. The bias caused by variable matrices, i.e. DBS and EDTA matrices, was less than 8.7%. The recovery of DBS extraction was between 94.1% and 97.2%. No significant linezolid degradation was observed after storing DBS at 50°C for at least one week and at 37°C or ambient temperature for two months as biases were less than 15%.

Variation of blood spot volume between 30 μL to 90 μL had a minor impact on the assay accuracy as the bias ranged from -11.6% to 7.1%. The variation of Hct from 20% to 50% yielded biases within -7.6% to 6.8% and -12.5% to 5.7% for MED and HIGH level. Larger biases of -17.8% to 11.9% were observed at the LOW concentration level (0.25 mg/L) (Table 2).

Comparisons of DBS, vDBS and plasma analysis

Significant proportional biases were observed in Passing-Bablok regressions in which the slope of regression line between DBS and plasma was 1.28 (95% CI: 1.13 – 1.44) and vDBS and plasma was 1.46 (95% CI: 1.40 – 1.54). The intercepts were -0.42 (95% CI: -1.72 – 0.17) and -0.67 (95% CI: -1.36 – -0.09), respectively (Figure 1). In Bland-Altman analysis, the geometric mean concentration ratios of DBS and vDBS versus plasma were 1.20 (95% CI:

Table 2 Summarized results of the validation of DBS analysis

Validation levels (n=5)

Validation criteria LLOQ LOW MED HIGH

Nominal concentrations (mg/L) 0.05 0.25 15 30

Reproducibility (#)

Accuracy (% Bias) 4.5 6.3 3.2 -1.3Within-day precision (% CV) 13.8 4.0 4.1 1.6Between-day precision (% CV) 10.2 3.5 6.1 7.7Overall precision (% CV) 17.2 5.3 7.4 7.8

Matrix effect (%) 2.9 8.7 1.9

Recovery (%) 95.5 94.1 97.2

Effect of blood volume (range of % Bias) (¥) -2.9 – 4.1 -11.4 – 7.1 -11.6 – 9

Effect of hematocrit (range of % Bias) (£) -17.8 – 11.9 -7.6 – 6.8 -12.5 – 5.7

Stability (€)

1 week at 50°C (% Bias) 6.7 -3.42 months at 37°C (% Bias) -10 -5.92 months at ambient temperature (% Bias) -2.5 -2.0

(#): data from 3 separated validation days; (¥): comparison with samples of standardized hematocrit (35%); (£): comparison with sample of standardized blood spot volume (35 μL); (€): present data from the last time of the period only.

55


4A

1.12 – 1.27) and 1.36 (95% CI: 1.32 – 1.40), respectively. The ratio of vDBS/plasma was higher than that of DBS/Plasma (Wilcoxon signed-rank test, n=24, p<0.01). 95% limits of agreement were shown with less than 5% of the values falling out of the ranges (Figure 2).

30

g/L

)

25mg

25

on

(m

20ati

o

20

ntr

ac

en

15co

nB

S c

10vD

B

10

BS

/vD

B

5

00

0 5 10 15 200 5 10 15 20

Plasma concentration (mg/L)

Figure 1 Passing-Bablok regression between measurements in DBS/vDBS and plasma. – – – – : vDBS-plasma Passing-Bablok regression: slope = 1.46 (95% CI: 1.40 – 1.54), intercept = -0.67 (95% CI: -1.36 – -0.09); : DBS-plasma Passing-Bablok regression: slope = 1.28 (95% CI: 1.13 – 1.44), intercept = -0.42 (95% CI: -1.72 – 0.17).

Figure 2 Bland-Altman plot of concentration ratios of DBS and vDBS vs. plasma. : Mean ratio; – – – – : Limit of agreement (mean ratio ± 1.96 × SD ratio).

1.8 1.8

1.6

1.8

1.6

1.8

1.4

1.6

tio 1.4

1.6

o

1.2

1.4

ma

ra

tio

1.2

1.4

ma

ra

tio

1.0

1.2

S/P

lasm

a r

at

1

1.2

Pla

sma

ra

tio

0.8

1.0

vD

BS

/Pla

sm

0.8

1

DB

S/P

lasm

0.6

0.8vD

BS

0.6

0.8DB

S/P

0.6

0 10 20

(vDBS+Plasma)/2 (mg/L)

0.6

0 10 20

(DBS+Plasma)/2 (mg/L)

0.6

0 10 20

(vDBS+Plasma)/2 (mg/L)

0.6

0 10 20

(DBS+Plasma)/2 (mg/L)

Chapter 4A

56

Table 3 Steady state pharmacokinetic parameters of linezolid in plasma (n=8)

Parameter300 mg linezolidtwice a day (n=3)

600 mg linezolidtwice a day (n=5)

AUC0-12h (mg*h/L) 50.9 [50.5 – 54.9] 126.9 [121.6 – 127.6]

Cmax (mg/L) 8.8 [7.8 – 8.9] 16.5 [14.4 – 16.5]

Tmax (h) 1.9 [1.9 – 4.8] 1.9 [1.7 – 3.0]

T1/2 (h) 4.6 [4.0 – 6.9] 7.5 [7.3 – 7.9]

Cl (L/h) 4.9 [3.8 – 5.1] 3.1 [3.0 – 3.1]

Vd (L) 32.6 [29.4 – 34.4] 34.8 [32.9 – 41.6]

Data are presented as median [interquartile range].


A median (IQR) plasma AUC0–12h of 50.9 (50.5 – 54.9) mg*h/L was observed following a dose of 300 mg and 126 (121.6 – 127.6) mg*h/L following a dose of 600 mg. Linezolid pharmacokinetic parameters are shown in Table 3. The concentration-time curves of plasma, DBS and vDBS are presented in Figure 3.

The AUC0–12h values of DBS and vDBS were calculated using the conversing factors 1.20 and 1.36 for DBS and vDBS respectively. The subsequent result showed a good agreement with

Figure 3 Concentration-time curves of linezolid in plasma, vBDS and DBS. Plasma and vDBS data are presented as mean and SD. For visual purposes, the DBS data are presented as mean without error bar.

Pl VDBS DBSPlasma VDBS DBS

30

L)

25

mg

/L

20n (

m

20

tio

n

15ntr

at

15

nc

en

10co

n

10

ag

e

5

ve

raA

v

0

-1 0 1 2 3 4 5 6 7 8 9 10 11 12 13

Time (hours)Time (hours)

57


4A

plasma. All the values were within the 95% limit of agreement (Figure 4). The individual data for each patient for AUC0–12h attained from plasma and conversed (v)DBS concentrations and the respective AUC0–24h/MIC values are presented in Table 4. Patients that received a linezolid dose of 300 mg twice daily (n=3) had a median (IQR) plasma AUC0–24h/MIC ratio of 236 (219 – 322) mg*h/L and patients that received 600 mg twice daily (n=5) had a median (IQR) plasma AUC0–24h/MIC ratio of 508 (486 – 1398) mg*h/L.

Table 4 Pharmacokinetic and pharmacodynamic parameters of linezolid using plasma, vDBS and

DBS concentrations

PatientDose (2d.d)

MIC (mg/L)

AUC0–12h (mg*h/L) AUC0–24h/MIC

Plasma vDBS(#) DBS(#) Plasma vDBS(#) DBS(#)

1 300 0.5 50.1 46.7 51.5 201 187 206

2 300 0.25 50.9 54.2 50.7 407 433 405

3 600 0.5 121.6 118.1 112.1 486 472 449

4 600 <0.125 127.6 130.9 114.2 >2042 >2094 >1827

5 600 0.5 126.9 132.0 140.2 508 528 561

6 600 0.5 66.6 69.4 64.2 266 278 257

7 300 0.5 58.9 46.1 42.4 236 184 170

8 600 0.25 174.7 183.1 189.6 1398 1465 1517(#): relative AUC0–12h and AUC0–24h/MIC calculated using conversion factors (i.e. 1.20 for DBS and 1.36 for vDBS).

Figure 4 Bland-Altman plot of AUC0–12h

from corrected DBS vs AUC0–12h

plasma samples. : Mean difference; – – – – : Limit of agreement (mean difference ± 1.96 × SD difference).

35

25/L)

25

g*

h/

15h (mg

15

0-1

2h

a)

(

5UC

0

sma

e A

U

pla

s

-5

en

ce

BS

-p

-15fere

DB

-15

Dif

f

sed

-25

D

ve

rso

nv

-35

40 90 140 190

(Co

40 90 140 190

Average AUC0-12h

(C d DBS l )/2 ( *h/L)(Corrected DBS+plasma)/2 (mg*h/L)

Chapter 4A

58

DISCUSSION

This study showed that DBS analysis is an easy tool to individualize MDR-TB treatment with linezolid. In addition, this report presents a novel, validated method of analysis of linezolid in dried blood spots, with specimens that proved to be very stable over time.

In previous studies on DBS analysis of other drugs, several technical factors were pointed out that have to be considered when interpreting DBS analysis, such as the effect of Hct and blood spot volume (12, 13, 18, 30). For the analysis of linezolid in DBS, the effect of Hct seemed to be of minor concern. In this study, biases fell within accepted ranges for Hcts between 20 – 50%. These Hcts cover an even broader range than clinical Hcts found in TB patients in literature, i.e. 35.4 ± 6.7% (3), and in this study 37.4 ± 4.4%. Based on these findings, the standardization of Hct at 35% during DBS validation is acceptable. Furthermore, variation of blood spot volume between 30 – 90 μL had little effect as biases were within 15%.

Despite the minor influence of technical factors, i.e. Hct value and blood spot volume, physiological factors are also mentioned in literature to possibly limit the applicability and interpretation of DBS analysis (13). Such a factor might be differences between plasma concentration and whole blood concentration. This study shows that concentration of linezolid is higher in blood than in plasma. This is caused by different binding capacity to plasma proteins and blood cells. Furthermore, concentrations of linezolid were higher in vDBS than in DBS. This might be caused by differences between the capillary and venous blood (13, 25, 30). Nevertheless, the concentration of DBS and vDBS specimens, both showed good correlation with plasma concentration. To compensate for these differences, we propose conversion factors of 0.83 (1 / 1.20) for DBS and 0.74 (1 / 1.36) for vDBS to calculate corresponding plasma values. After the conversion, good agreement between AUC0–12h of DBS and plasma was observed.

A meta-analysis showed that a ≤600 mg linezolid daily dose resulted in lower frequency of either adverse event or adverse events necessitating treatment discontinuation than the dose of >600 mg daily (8). Among the published data, the lowest rate of adverse effects was observed with a dose of 300 mg once daily (17). Nevertheless, lowering the dose clearly results in lower exposure to the drug (2, 19). In combination inter-patient variability and possible drug-drug interactions, under or overexposure may occur. Therefore, treatment with a fixed dose may be questionable (6, 11, 22, 26). The application of TDM for linezolid can help avoid under- or overexposure which may occur in 30 to 40% of the cases (20).

59


4A

In this study, all patients had Mycobacterium tuberculosis isolates with a MIC ≤0.5 mg/L for linezolid. With a dose of 600 mg (n=5) twice daily, very high AUC0–24h/MIC ratios were reached (10), so dose reductions could be implemented to prevent time- and dose-dependent toxicity. Furthermore, a high correlation of AUC0–24h/MIC values between conversed DBS and plasma (Spearman’s rho = 0.976, n=8) was observed. This suggests that TDM using DBS may result in identical interventions compared with conventional plasma sampling. Therefore, adaptive dosing of linezolid to prevent potential toxicity and to assure therapeutic exposure is feasible using DBS.

The high stability of DBS specimens can minimize the logistic burden of conventional sampling in limited-resource areas. With a simple instruction, the DBS samples can be performed easily and sent to equipped facilities for analysis by mail (12, 30). This could enable the TDM in TB-programs worldwide including resource limited settings where MDR/XDR-TB epidemic is a growing problem. TDM using DBS for MDR/XDR-TB should be especially considered in areas where HIV or malaria co-infections are highly prevalent as DBS has been successfully applied to monitor the treatment of such diseases (30).

Since treatment of MDR/XDR-TB is long and complicated by adverse drug reactions, TDM of linezolid with DBS could be used to optimize drug exposure during treatment. In conclusion, this study presents a novel, validated analysis of linezolid in DBS specimens that is suitable for optimization of linezolid treatment of MDR-TB. Advantages include a very simple, low biohazard risk sampling method using a finger prick, easy logistics and very good stability of DBS specimens.

REFERENCES

1. Alffenaar, J. W., J. G. Kosterink, R. van Altena, T. S. van der Werf, D. R. Uges, and J. H. Proost. 2010. Limited sampling strategies for therapeutic drug monitoring of linezolid in patients with multidrug-resistant tuberculosis. Ther. Drug Monit. 32:97-101.


3. Antony, S. J., V. Harrell, J. D. Christie, H. G. Adams, and R. L. Rumley. 1995. Clinical differences between pulmonary and extrapulmonary tuberculosis: a 5-year retrospective study. J. Natl. Med. Assoc. 87:187-192.

Chapter 4A

60

4. Bland, J. M., and D. G. Altman. 1999. Measuring agreement in method comparison studies. Stat. Methods Med. Res. 8:135-160.

5. Bolhuis, M. S., A. D. Pranger, and J. W. Alffenaar. 2012. Linezolid: safety and efficacy monitoring. Eur. Respir. J. 39:1275-1276.


7. Conte, J. E.,Jr, J. A. Golden, J. Kipps, and E. Zurlinden. 2002. Intrapulmonary pharmacokinetics of linezolid. Antimicrob. Agents Chemother. 46:1475-1480.

8. Cox, H., and N. Ford. 2012. Linezolid for the treatment of complicated drug-resistant tuberculosis: a systematic review and meta-analysis. Int. J. Tuberc. Lung Dis. 16:447-454.

9. Cynamon, M. H., S. P. Klemens, C. A. Sharpe, and S. Chase. 1999. Activities of several novel oxazolidinones against Mycobacterium tuberculosis in a murine model. Antimicrob. Agents Chemother. 43:1189-1191.

10. Dooley, K. E., E. A. Obuku, N. Durakovic, V. Belitsky, C. Mitnick, E. L. Nuermberger, and on behalf of the Efficacy Subgroup, RESIST-TB. 2013. World Health Organization Group V Drugs for the Treatment of Drug-Resistant Tuberculosis: Unclear Efficacy Or Untapped Potential? J. Infect. Dis. 207: 1352-1358.

11. Dryden, M. S. 2011. Linezolid pharmacokinetics and pharmacodynamics in clinical treatment. J. Antimicrob. Chemother. 66 Suppl 4:iv7-iv15.

12. Edelbroek, P. M., J. van der Heijden, and L. M. Stolk. 2009. Dried blood spot methods in therapeutic drug monitoring: methods, assays, and pitfalls. Ther. Drug Monit. 31:327-336.

13. Emmons, G., and M. Rowland. 2010. Pharmacokinetic considerations as to when to use dried blood spot sampling. Bioanalysis. 2:1791-1796.

14. Fortun, J., P. Martin-Davila, E. Navas, M. J. Perez-Elias, J. Cobo, M. Tato, E. G. De la Pedrosa, E. Gomez-Mampaso, and S. Moreno. 2005. Linezolid for the treatment of multidrug-resistant tuberculosis. J. Antimicrob. Chemother. 56:180-185.



17. Koh, W. J., Y. R. Kang, K. Jeon, O. J. Kwon, J. Lyu, W. S. Kim, and T. S. Shim. 2012. Daily 300 mg dose of linezolid for multidrug-resistant and extensively drug-resistant tuberculosis: updated analysis of 51 patients. J. Antimicrob. Chemother. 67: 1503-7.

18. Li, W., and F. L. Tse. 2010. Dried blood spot sampling in combination with LC-MS/MS for quantitative analysis of small molecules. Biomed. Chromatogr. 24:49-65.

61


4A

19. McGee, B., R. Dietze, D. J. Hadad, L. P. Molino, E. L. Maciel, W. H. Boom, M. Palaci, J. L. Johnson, and C. A. Peloquin. 2009. Population pharmacokinetics of linezolid in adults with pulmonary tuberculosis. Antimicrob. Agents Chemother. 53:3981-3984.

20. Pea, F., M. Furlanut, P. Cojutti, F. Cristini, E. Zamparini, L. Franceschi, and P. Viale. 2010. Therapeutic drug monitoring of linezolid: a retrospective monocentric analysis. Antimicrob. Agents Chemother. 54:4605-4610.

21. Prammananan, T., A. Chaiprasert, and M. Leechawengwongs. 2009. In vitro activity of linezolid against multidrug-resistant tuberculosis (MDR-TB) and extensively drug-resistant (XDR)-TB isolates. Int. J. Antimicrob. Agents. 33:190-191.

22. Sasaki, T., H. Takane, K. Ogawa, S. Isagawa, T. Hirota, S. Higuchi, T. Horii, K. Otsubo, and I. Ieiri. 2011. Population pharmacokinetic and pharmacodynamic analysis of linezolid and a hematologic side effect, thrombocytopenia, in Japanese patients. Antimicrob. Agents Chemother. 55:1867-1873.

23. Singla, R., J. A. Caminero, A. Jaiswal, N. Singla, S. Gupta, R. K. Bali, and D. Behera. 2012. Linezolid: an effective, safe and cheap drug for patients failing multidrug-resistant tuberculosis treatment in India. Eur. Respir. J. 39:956-962.

24. Spooner, N., R. Lad, and M. Barfield. 2009. Dried blood spots as a sample collection technique for the determination of pharmacokinetics in clinical studies: considerations for the validation of a quantitative bioanalytical method. Anal. Chem. 81:1557-1563.

25. Spooner, N., Y. Ramakrishnan, M. Barfield, O. Dewit, and S. Miller. 2010. Use of DBS sample collection to determine circulating drug concentrations in clinical trials: practicalities and considerations. Bioanalysis. 2:1515-1522.

26. Srivastava, S., J. G. Pasipanodya, C. Meek, R. Leff, and T. Gumbo. 2011. Multidrug-resistant tuberculosis not due to noncompliance but to between-patient pharmacokinetic variability. J. Infect. Dis. 204:1951-1959.

27. US Department of Health and Human Services Food and Drug Administration. 2001. Guidance for Industry. Bioanalytical Method Validation.


29. Vinh, D. C., and E. Rubinstein. 2009. Linezolid: a review of safety and tolerability. J. Infect. 59 Suppl 1:S59-74.

30. Vu, D. H., J. W. Alffenaar, P. M. Edelbroek, J. R. Brouwers, and D. R. Uges. 2011. Dried blood spots: a new tool for tuberculosis treatment optimization. Curr. Pharm. Des. 17:2931-2939.

31. Vu, D. H., R. A. Koster, J. W. Alffenaar, J. R. Brouwers, and D. R. Uges. 2011. Determination of moxifloxacin in dried blood spots using LC-MS/MS and the impact of the hematocrit and blood volume. J. Chromatogr. B. Analyt Technol. Biomed. Life. Sci. 879:1063-1070.

Chapter 4A

62

32. Williams, K. N., C. K. Stover, T. Zhu, R. Tasneen, S. Tyagi, J. H. Grosset, and E. Nuermberger. 2009. Promising antituberculosis activity of the oxazolidinone PNU-100480 relative to that of linezolid in a murine model. Antimicrob. Agents Chemother. 53:1314-1319.

33. World Health Organisation (WHO). 2010. Treatment of tuberculosis: guidelines. World Health Organisation, Geneva.

34. Yew, W. W., C. H. Chau, and K. H. Wen. 2008. Linezolid in the treatment of ‘difficult’ multidrug-resistant tuberculosis. Int. J. Tuberc. Lung Dis. 12:345-346.

Clinical validation of the

analysis of linezolid and

clarithromycin in oral fluid

of multidrug-resistant


M.S. Bolhuis, R. van Altena, K. van Hateren, W.C.M. de Lange, B. Greijdanus, D.R.A. Uges,

J.G.W. Kosterink, T.S. van der Werf, and J.W.C. Alffenaar

Antimicrob Agents Chemother. 2013 Aug; 57(8): 3676–80

Chapter 4B

Chapter 4B

64

ABSTRACT

Linezolid plays an increasingly important role in the treatment of multidrug-resistant tuberculosis. However, patients should be carefully monitored due to time- and dose-dependent toxicity. Clarithromycin plays a more modest role. Therapeutic drug monitoring may contribute to assessing treatment regimens, helping to reduce toxicity whilst maintaining adequate drug exposure. Oral fluid sampling could provide a welcome alternative in cases where conventional plasma sampling is not possible or desirable. The aim of this study was to clinically validate the analysis of linezolid and clarithromycin and its metabolite hydroxyclarithromycin in oral fluid of patients with multidrug-resistant tuberculosis.

Serum and oral fluid samples were simultaneously obtained and analyzed using validated methods, after extensive cross-validation between the two matrices. Passing-Bablok regressions and Bland-Altman analysis showed that oral fluid analysis of linezolid and clarithromycin appeared suitable for therapeutic drug monitoring in MDR-TB patients. No correction factor is needed for the interpretation of linezolid oral fluid concentration with a linezolid serum / oral fluid ratio of 0.97 (95% CI, 0.92 – 1.02). However, the clarithromycin serum / oral fluid concentration ratio is 3.07 (95% CI, 2.45 – 3.69). Analysis of hydroxyclarithromycin in oral fluid was not possible in this study due to a non linear relationship between concentration in serum and oral fluid. In conclusion, the analysis of linezolid (no correction factor) and clarithromycin (correction factor *3) in oral fluid is applicable for therapeutic drug monitoring in multidrug-resistant tuberculosis as an alternative to conventional serum sampling. Easy sampling, using a non-invasive technique may facilitate therapeutic drug monitoring in specific patient categories.

65

Oral fluid analysis of linezolid and clarithromycin

4B

INTRODUCTION

Tuberculosis is a mostly curable and preventable infectious disease caused by Mycobacterium tuberculosis. Approximately 3.7% of new tuberculosis patients and 20% of previously treated patients are infected with multidrug-resistant strains that are resistant to at least rifampicin and isoniazid (1). Treatment regimens of multidrug-resistant tuberculosis (MDR-TB) should consist of at least four anti-TB drugs for which the bacterium is susceptible (2).

The oxazolidinone linezolid is effective against M. tuberculosis and is increasingly used as a part of treatment regimens in patients with multidrug-resistant or extensively drug-resistant tuberculosis (3). However, patients should be carefully monitored due to time- and dose-dependent serious toxicity of linezolid, such as myelosuppression and polyneuropathy (4).

Clarithromycin has a less pronounced place in MDR-TB treatment regimens due to serum concentrations that usually do not reach minimal inhibitory concentrations (5). Nevertheless, sufficiently high local clarithromycin concentrations are reached in epithelial lining fluid and alveolar cells (6, 7). Furthermore, occasionally observed lower MICs [unpublished data], synergistic activity of clarithromycin against MDR-TB strains (8) and absence of severe adverse events (9) contribute to its place in anti-TB therapy.

Serum concentrations of linezolid have shown large inter-patient variability (10). Drug-drug interactions might further contribute to the observed variability in linezolid pharmacokinetics. For instance, clarithromycin has been observed to increase linezolid serum concentrations significantly (11). Therapeutic drug monitoring could potentially assist in identifying MDR-TB patients with too low or too high linezolid exposure. Conventional serum sampling is not always possible or desirable due to lack of venous access, complicated logistics, or due to the invasive character of the technique. Previously, we developed a dried blood spot analysis of linezolid (12) and clarithromycin (unpublished data) as an alternative to conventional serum sampling. A clinically validated method could be useful in patients that do not accept or tolerate an indwelling venous catheter, or who have difficult venous access. We therefore aimed to clinically validate the analysis of linezolid, clarithromycin, and hydroxyclarithromycin concentrations in oral fluid.

Chapter 4B

66

METHOD

From December 2011 to October 2012, patients were included from the Tuberculosis Center Beatrixoord (Haren, The Netherlands). Patients were ≥18 years old, were diagnosed with MDR-TB and provided written informed consent. The study protocol was approved by the local Medical Ethical Review Committee, as part of a previously published study. The prospective pharmacokinetic study is registered at www.clinicaltrials.org (NCT01521364).

All patients received oral dosages of linezolid 300 mg twice daily and clarithromycin 250 mg once daily. Full pharmacokinetic curves were obtained at steady state, after at least two weeks of administration of both drugs, using a practically feasible sampling schedule which resulted in adequate area under the time-concentration curves (AUC) in previous studies in our Center. Blood and oral fluid samples were collected simultaneously before and 1, 2, 3, 4, 8, and 12 hours after medication intake. Blood samples were drawn and after centrifuging serum samples were stored at -20°C until analysis. Oral fluid samples were collected using a small cotton roll on which the patients chewed for approximately two minutes (Salivette®, Sarstedt, Leicester, UK). Oral fluid samples were centrifuged and then stored at -20°C until analysis.

Linezolid and clarithromycin serum and oral fluid concentrations were analyzed using validated high performance liquid chromatography tandem mass-spectrometry methods (13, 14). Cross-validation between two matrices was performed by comparing calibration samples in pooled serum and non-stimulated, pooled oral fluid from six batches.

Pharmacokinetic parameters such as, most importantly, AUC were calculated using Kinfit software (MWPharm 3.60; Mediware, Groningen, the Netherlands) as described in a previous study using non-compartmental, trapezoidal calculations (10). Other pharmacokinetic parameters that were calculated using Kinfit software were Cmax, Cmin, apparent clearance (CL), elimination rate constant (kel) and half-life (t1/2). Of these parameters, CL, kel, and t1/2were determined using log-linear regression of the concentrations in the terminal period.

The method was clinically validated by comparing the linezolid, clarithromycin and hydroxyclarithromycin concentrations in serum samples with the concentration in oral fluid using Passing-Bablok regressions and Bland-Altman analysis (Analyze-it Software, Ltd.). The Pearson’s correlation and Wilcoxon signed-rank test were applied to other comparisons.

67


4B

RESULTS

Baseline patient demographics are presented in Table 1. Seven patients with MDR-TB, four male and three female, were included in the study. The isolates that were obtained from all patients displayed resistance for at least rifampicin and isoniazid. The patients had a median (interquartile range, IQR) age of 31 (25.5 – 33.5) years and weighed in at median (IQR) 71.5 (56.0 – 75.3) kg. Five patients were from Somalia, one from Turkey and one from the Netherlands. At the time of sampling, patients were on MDR-TB treatment for a mean (range) of 61.4 (33 – 149) days. One patient was HIV positive for which the patient was on cART with adequate virological and immunological response (before admission CD4 count: 380 cells/mm3).

Table 1 Baseline demographics (n=7) and results from drug susceptibility testing

Parameter Value

Age – year† 31.0 (25.5 – 33.5)

Male sex – no. (%) 4 (57%)

Bodyweight – kg‡ 71.5 (56.0 – 75.3)

Height – m‡ 1.73 (1.64 – 1.74)

Ethnicity – no. (%)AfricanCaucasianAsian

5 (71%)1 (14%)1 (14%)

HIV positive – no. (%) 1 (14%)

Isolate resistant to drug based on DST – no./no. totalRifampicinIsoniazidEthambutolPyrazinamide∫

StreptomycinCapreomycinAmikacinCiprofloxacinClaritromycin∫

Clofazimin∫

LinezolidMoxifloxacinProtionamide∫

Rifabutin

7/77/75/74/66/72/70/71/73/50/40/71/72/66/7

† Data presented as mean (range), ‡ Data presented as median (interquartile range), ∫ Drug susceptibility testing (DST) was not available for all isolates of the included patients.

Chapter 4B

68

Figure 1 Scatter plot with Passing-Bablok fit of serum and oral fluid concentration in mg/L. Identity lines are presented as dashed lines and regression lines as solid lines. A. linezolid (n=49): regression line of linezolid serum-oral fluid concentration has a slope of 1.05 (95% CI, 0.94 – 1.11) and intercept of -0.26 (95% CI, -0.52 – 0.05). B. clarithromycin (n=42): regression line of clarithromycin serum-oral fluid concentration has a slope of 2.67 (95% CI, 1.95 – 3.75) and intercept of -0.06 (95% CI, -0.18 – 0.21).

A li lid16

A. linezolid16

1414L

)

12g/L

12

mg

(m

10on

10

tio

rat

8ntr 8

en

nc

e

6on

6

m c

ou

m

4ruses

2

00

0 2 4 6 8 10 12 14 160 2 4 6 8 10 12 14 16

oral fluid concentration (mg/L)oral fluid concentration (mg/L)

B l ith iB. clarithromycin

55

4 54.5

4/L)

g/

3 5mg

3.5

n (

3on

3

ati

2 5tra

2.5

en

tc

e

2

on

cc

o

1.5m

.5

rum

1ser

1s

0 50.5

00

0 1 2 3 4 50 1 2 3 4 5

l fl id ioral fluid concentration (mg/L)g

69


4B

Comparison oral fluid and serum analysis method

Comparison of analysis in two matrices showed that linezolid, clarithromycin, and hydroxyclarithromycin had no significant differences of intercept and slope in serum and in oral fluid. The calibration curves in oral fluid were analyzed three times with all coefficients of variation (CVs) below 15%. All biases in concentration were <12% for linezolid and <8% for clarithromycin and its metabolite.

Passing-Bablok regression (n=49) of linezolid concentration in serum and oral fluid showed a proportional bias of 1.05 (95% confidence interval [CI], 0.94 – 1.11) and a constant bias of -0.26 (95% CI, -0.52 – 0.05) (Figure 1A). For clarithromycin, the Passing-Bablok scatter plot (n=42) showed a proportional bias of 2.67 (95% CI, 1.95 – 3.75) and a constant bias of -0.06 (95% CI, -0.18 – 0.21) (Figure 1B). There were 7 missing clarithromycin and 8 missing hydroxyclarithromycin values due to concentrations below limit of quantitation of the applied method of 0.2 mg/L. A linear relationship was detected using Cusum linearity test (p>0.1) between oral fluid and serum concentrations of both linezolid and clarithromycin. However, the Cusum linearity test detected a non-linear relationship (0.05<p<0.1) between serum and oral fluid hydroxyclarithromycin concentration, with Passing-Bablok showing a constant bias of 0.02 (95% CI, -0.20 – 0.24) and proportional bias of 2.00 (95% CI, 1.14 – 3.00).

Bland-Altman assessment showed good agreement between analyses of linezolid and clarithromycin in serum and oral fluid; 4.1% (2/49) of observations falling outside 95% limits of agreement for linezolid, and 7.1% (3/42) for clarithromycin; see Figure 2. The observed bias for linezolid (n=49) was 0.97 with 95% confidence intervals below and above one (95% CI, 0.92 – 1.02) (Figure 2A). For clarithromycin, the observed bias was 3.07 (95% CI, 2.45 – 3.69) (Figure 2B). Pearson’s test revealed that the analyses of linezolid and clarithromycin in serum and in oral fluid were correlated with r = 0.95 (p<0.01) and r = 0.80 (p<0.01) respectively.


Pharmacokinetic parameters of linezolid and clarithromycin in serum and oral fluid are displayed in Table 2. Paired sample Wilcoxon Signed rank test showed no statistically significant difference between medians of all pharmacokinetic parameters in serum and oral fluid, except for linezolid kel and t1/2 (p=0.018). However, the clinical significance of this observed difference is little since the AUC is the most important pharmacokinetic parameter for therapeutic drug monitoring of linezolid and clarithromycin in multidrug-resistant tuberculosis patients.

Chapter 4B

70

Figure 2 Bland-Altman plot of serum/oral fluid concentration ratio compared to average serum and

oral fluid concentration. The line representing the bias is presented as a solid line and the 95% limits of agreement as dashed lines. A. linezolid (n=49): bias is 0.97 (95% CI, 0.92 – 1.02), the lower and upper 95% limits of agreement are respectively 0.64 (95% CI, 0.56 – 0.73) and 1.30 (95% CI, 1.22 – 1.38). B. clarithromycin (n=42): bias is 3.07 (95% CI, 2.45 – 3.69), the lower and upper 95% limits of agreement are respectively -0.82 (95% CI, -1.89 – 0.24) and 6.97 (95% CI, 5.90 – 8.03).

A li lid1 5

A. linezolid1.5

1 41.4

1.3

o

1 2ati

1.2

d r

au

id

1.1flu

al

f

1ora

1

/ o

m /

0.9rum

0.9

ser

0 8

s

0.8

0.70.7

0 60.6

0 5 100 5 10

(serum / oral fluid)/2 (mg/L)( ) ( g )

B l ith i9

B. clarithromycin9

8

77

6io 6at

d r

a

5

uid

flu

4al

f

4

ora

3/ o

3

m /

um

2er

se

11

00

1-1

0 1 2 3 40 1 2 3 4

(serum / oral fluid)/2 (mg/L)(serum / oral fluid)/2 (mg/L)

71


4B

Pharmacokinetic and pharmacodynamic parameters of linezolid from all patients are displayed in Table 3. Isolates from all patients were susceptible for linezolid with a MIC of median 0.25 mg/L. Patients had a linezolid AUC0–12h of median (IQR) 63.9 (47.8 – 83.8) mg*h/L in serum and 62.1 (50.5 – 59.2) mg*h/L in oral fluid. All patients had an AUC0–24/MIC ratio of >100 in both serum and oral fluid. For patient 2, AUC0–24h/MIC ratio reached approximately 1000. Median (IQR) linezolid AUC0–24h/MIC ratio in serum was 277 (260 – 517) mg*h/L and in oral fluid 288 (262 – 594) mg*h/L. Paired sample Wilcoxon Signed rank test showed no statistically significant difference between AUC0–12h or AUC0–24h/MIC ratio in serum or oral fluid (p=0.296).

Table 2 Pharmacokinetic parameters of linezolid and clarithromycin in serum and in oral fluid (n=7)

Serum Oral fluid p-value†

LinezolidAUC0–12u (mg*h/L) Cmax (mg/L)Cmin (mg/L)CL (L/h)kel (/h)t1/2 (h)

63.9 [47.8 – 83.8]10.9 [6.8 – 11.5]2.2 [1.5 – 4.2]3.5 [2.4 – 5.9]0.14 [0.10 – 0.17]4.9 [4.2 – 7.9]

62.1 [50.5 – 89.2]10.1 [8.2 – 10.7]2.3 [1.7 – 4.2]3.6 [2.2 – 5.0]0.13 [0.08 – 0.16]5.2 [4.5 – 9.8]

0.2961.00.0840.0630.018¥

0.018¥

ClarithromycinAUC0–12u (mg*h/L) Cmax (mg/L)Cmin (mg/L)CL (L/h)kel (/h)t1/2 (h)

8.2 [6.2 – 12.2]1.7 [1.3 – 2.7]0.01 [0.01 – 0.04]28.5 [19.3 – 39.1]0.21 [0.19 – 0.23]3.3 [3.1 – 3.6]

10.7 [9.4 – 12.1]2.8 [2.0 – 3.4]0.03 [0.03 – 0.06]62.2 [52.8 – 81.0]0.64 [0.49 – 1.06]10.2 [6.4 – 13.5]

0.0910.0631.00.2370.6671.0

Data are presented as median [interquartile range]. † p-values comparing pharmacokinetic parameters between serum and oral fluid. ¥ Statistical significant difference between median of the parameter in serum and in oral fluid.

Table 3 Pharmacokinetic and pharmacodynamic parameters of linezolid

AUC0–12h (mg*h/L) AUC0–24h/MIC ratio

Patient MIC (mg/L) Serum Oral fluid Serum Oral fluid

1 0.25 34.6 42.1 277 3372 0.25 120.1 126.4 961 10113 0.5 61.0 62.1 244 2484 0.5 63.9 58.8 256 2355 0.25 33.0 34.5 264 2766 0.5 76.6 72.0 306 2887 0.25 90.9 106.3 727 850Total¥ 63.9 [47.8 – 83.8] 62.1 [50.5 – 89.2]± 277 [260 – 517] 288 [262 – 594]±

¥ Median [interquartile range], ± no statistically significant difference between serum and oral fluid (p=0.296).

Chapter 4B

72

Patients had a clarithromycin AUC0–12h of median (IQR) 8.2 (6.2 – 12.2) mg*h/L in serum and 3.5 (3.1 – 4.0) mg*h/L in oral fluid. One patient was inadvertently administered 500 mg, instead of 250 mg clarithromycin at the day of sampling. The samples obtained from this patient were included in the evaluation. The clarithromycin AUC0–12h after administration of 500 mg clarithromycin was 29.1 mg*h/L in serum and 15.7 mg*h/L in oral fluid, well above the AUCs of the patients receiving 250mg clarithromycin. After applying the correction factor of 3.07 as determined using the Bland-Altman assessment, patients had an adjusted clarithromycin AUC0–12h of median (IQR) 10.7 (9.4 – 12.1) mg*h/L in oral fluid. Paired sample Wilcoxon Signed rank test showed no statistically significant difference, but did show a trend towards difference between clarithromycin AUC0–12h in serum and in oral fluid after applying a correction factor of 3.07 (p=0.091).

DISCUSSION

The clinical validation performed in this study showed that oral fluid analysis of linezolid and clarithromycin are suitable for TDM in MDR-TB patients. No correction factor is needed for the interpretation of linezolid oral fluid concentration. However, clarithromycin oral fluid concentration should be corrected by multiplying by three to enable comparison to clarithromycin serum levels. After applying a correction factor of three in case of clarithromycin or no correction factor in case of linezolid, the pharmacokinetic parameter AUC0–12h as calculated from oral fluid samples are applicable for TDM and could assist in identifying patients with too high or too low exposure.

Unfortunately, analysis of hydroxyclarithromycin in oral fluid is not possible due to a non-linear relation with concentrations in serum. Nevertheless, the analysis of hydroxyclarithromycin shows good linearity over a range of 0.2 – 10 mg/L in both serum and oral fluid. A possible explanation for the observed non-linear relation between analysis of hydroxyclarithromycin in serum and in oral fluid might be the low hydroxyclarithromycin concentrations that were observed. In this concentration range, around the limit of quantitation, CVs are relatively high although within acceptable limits of <20%. This could explain a non-linear relationship between analysis of hydroxyclarithromycin in serum and oral fluid in this low concentration range. Possibly, analysis of a larger cohort or higher clarithromycin doses with corresponding higher hydroxyclarithromycin concentrations would reveal a linear relationship. Furthermore, this could confirm that there is no significant difference between clarithromycin exposure in serum and oral fluid

73


4B

(after correction), despite a trend towards statistical significance that was observed in our cohort.

The kel and t1/2 of linezolid showed a statistical significant difference of medians in oral fluid compared to serum. The parameters kel and t1/2 are calculated from concentrations obtained in the terminal period, i.e. in the last 3 – 4 samples in a relatively small cohort. The relevance should be confirmed in a larger cohort or using curves with more samples in the terminal period. In clinical practice, not the paramater kel or t1/2, but the pharmacokinetic/pharmacodynamic parameter, AUC/MIC ratio of linezolid, is used for therapeutic drug monitoring (10–11).

To date, there has been no comparison in the literature of the analysis of clarithromycin and linezolid between serum and oral fluid in patients with MDR-TB. However, there are several studies describing pharmacokinetics of clarithromycin and hydroxyclarithromycin in saliva (15, 16), but none describing pharmacokinetics of linezolid in oral fluid. Clarithromycin administered to 12 healthy volunteers in a dose of 500 mg twice daily, resulted in an AUC0–12h of 18.0±5.0 mg*h/L (15). We observed a similar AUC0–12h of 15.7 mg*h/L in the one patient that was administered 500 mg clarithromycin. Saliva-serum ratios of around two were reported, lower than the ratio of three observed in our study. However, no data were presented comparing results from the analysis of clarithromycin in serum and oral fluid, since the study aimed to describe kinetics, not to clinically validate the analysis in oral fluid (15). Another study, described pharmacokinetics of clarithromycin in saliva and serum after a single dose of 500 mg clarithromycin (16). However, the aim was to describe the penetration of clarithromycin into saliva, not to clinically validate the analysis of clarithromycin in saliva. The Summary of Product Characteristics (SmPC) of linezolid describes a linezolid oral fluid – plasma concentration ratio of 1.2, comparable with the bias of 0.97 that was observed in the Bland-Altman assessment (17). Our current study describes the method of analysis of clarithromycin and linezolid, cross-validation between serum and oral fluid, and most importantly clinical validation in MDR-TB patients. No statistically significant differences were found between AUC0–12h or AUC0–24h/MIC ratio of serum and oral fluid.

TDM could potentially assist in identifying MDR-TB patients with too low or too high linezolid exposure. Analysis of anti-TB drugs in oral fluid may be advantageous in patients without readily accessible venous access, hindering blood sampling. Furthermore, the non-invasive sampling could be suitable in children with MDR-TB, in whom indwelling intravenous catheters are no option, and in whom no study data on pharmacokinetics

Chapter 4B

74

are available to guide therapy. Oral fluid sampling in pediatric patients is preferred over conventional serum sampling by a majority of children and their parents (18). Oral fluid sampling might even reduce costs due to the higher level of trained personnel needed for blood sampling and since less time is needed (18). Oral fluid sampling might even take place at home. No children were included in this study. A clinical validation of oral fluid sampling in pediatric MDR-TB patients is urgently needed. Furthermore, the applicability of saliva and/or other collection devices than the Salivette® (Sarstedt, Leicester, UK) for pharmacokinetic analysis and therapeutic drug monitoring in MDR-TB patients should be clinically validated.

In conclusion, the clinically validated analysis of clarithromycin and linezolid in oral fluid could provide a helpful alternative if conventional blood sampling is not possible or desirable. Using a correction factor of 3.07 for clarithromycin oral fluid concentrations and no correction factor for linezolid makes the oral fluid sampling readily applicable in clinical practice and allows for easy interpretation.

REFERENCES




4. Xu, H. B., R. H. Jiang, L. Li, and H. P. Xiao. 2012. Linezolid in the treatment of MDR-TB: a retrospective clinical study. Int. J. Tuberc. Lung Dis. 16:358-363.


6. Conte, J. E.,Jr, J. A. Golden, S. Duncan, E. McKenna, and E. Zurlinden. 1995. Intrapulmonary pharmacokinetics of clarithromycin and of erythromycin. Antimicrob. Agents Chemother. 39:334-338.

7. Rodvold, K. A., M. H. Gotfried, L. H. Danziger, and R. J. Servi. 1997. Intrapulmonary steady-state concentrations of clarithromycin and azithromycin in healthy adult volunteers. Antimicrob. Agents Chemother. 41:1399-1402.

75


4B


9. Chu, S. Y., L. T. Sennello, S. T. Bunnell, L. L. Varga, D. S. Wilson, and R. C. Sonders. 1992. Pharmacokinetics of clarithromycin, a new macrolide, after single ascending oral doses. Antimicrob. Agents Chemother. 36:2447-2453.





14. de Velde, F., J. W. Alffenaar, A. M. Wessels, B. Greijdanus, and D. R. Uges. 2009. Simultaneous determination of clarithromycin, rifampicin and their main metabolites in human plasma by liquid chromatography-tandem mass spectrometry. J. Chromatogr. B. Analyt Technol. Biomed. Life. Sci. 877:1771-1777.

15. Burkhardt, O., K. Borner, H. Stass, G. Beyer, M. Allewelt, C. E. Nord, and H. Lode. 2002. Single- and multiple-dose pharmacokinetics of oral moxifloxacin and clarithromycin, and concentrations in serum, saliva and faeces. Scand. J. Infect. Dis. 34:898-903.

16. Wust, J., and U. Hardegger. 1993. Penetration of clarithromycin into human saliva. Chemotherapy. 39:293-296.


18. Gorodischer, R., P. Burtin, P. Hwang, M. Levine, and G. Koren. 1994. Saliva versus blood sampling for therapeutic drug monitoring in children: patient and parental preferences and an economic analysis. Ther. Drug Monit. 16:437-443.

Comment on: Daily 300

mg dose of linezolid for

multidrug-resistant and

extensively drug-resistant

tuberculosis: updated

analysis of 51 patients

M.S. Bolhuis, R. van Altena, and J.W.C. Alffenaar

J Antimicrob Chemother. 2012 Aug; 67(8): 2055–6

Chapter 5A

Chapter 5A

78

Sir,

We read with great interest the article “Daily 300 mg dose of linezolid for multidrug-resistant and extensively drug-resistant tuberculosis: updated analysis of 51 patients” by Koh et al. in which they describe retrospectively examined records of one of the largest case series of patients treated with linezolid for multidrug-resistant and extensively drug-resistant tuberculosis (MDR/XDR-TB) (1). They describe long-term outcomes of 51 patients, whereas the previous study only described the short-term outcomes of 17 patients (2). Their objective was to evaluate efficacy, tolerability and adverse events of a 300 mg daily dose of linezolid in the treatment of MDR/XDR-TB. Based on a favorable treatment outcome of 78%, compared to 60 – 100% in literature albeit in smaller case series, they suggest that linezolid is effective against intractable MDR/XDR-TB at a daily dose of 300 mg. In our opinion it is difficult to draw this conclusion from the presented data. The lack of a control group makes it impossible to attribute favorable outcome in patients to a single drug such as linezolid as it is only a part of an expanded treatment regimen. Favorable treatment outcomes could very well be caused by other drugs of the expanded regimen the patients received.

To implicate efficacy of their linezolid containing regimen, Koh et al. make assumptions on the minimum inhibitory concentration (MIC) for linezolid for their population and the linearity of linezolid pharmacokinetics. They assume their patients are infected with Mycobacterium tuberculosis with a MIC of 0.25 mg/L for linezolid. Koh et al. base this on a recent study that described wild-type MIC distributions for linezolid and 6 other second-line drugs in 78 consecutive susceptible clinical isolates (3). Although most isolates had a MIC of 0.25 mg/L for linezolid, the wild-type MIC distribution ranged from 0.125 to 0.5 mg/L and an epidemiological cut-off value of 0.5 mg/L was suggested (3). The fact that we found the MIC to be 0.5 mg/L in eight isolates, 1 mg/L in eight isolates and even greater than 1 mg/L in one isolate in a previous study in 23 isolates (4), may raise some doubt about the assumption that all clinical isolates have a MIC value of 0.25 mg/L.

Koh et al. assume the pharmacokinetics of linezolid to be linear. Unfortunately, the pharma-cokinetics of linezolid are not linear in TB patients as we have demonstrated in a previous study (5). Besides, substantial intra- and interpatient variability of linezolid in TB patients can be observed. We found the AUC0–12h of 300 mg twice daily to be 56 mg*h/L but with an interquartile range (IQR) of 38.5 – 64.2 mg*h/L.

We agree with Koh et al. that for linezolid, the in vitro MIC to area under the free concentration time curve (fAUC0–24h/MIC ratio) is often used as a predictive model for efficacy (5). Based

79

Importance of TDM and DST of linezolid

5A

on the data of Schon et al. they suggest that a daily dose of 600 mg linezolid would lead to an fAUC of 56 mg*h/L, resulting in an fAUC/MIC ratio of approximately 100 for a wild-type MICECOFF of 0.5 mg/L and of approximately 200 for the more common MIC of 0.25 mg/L. In both cases, the pharmacodynamic target of fAUC/MIC >100 is met. However, during the study period, DST as well as plasma concentration monitoring (therapeutic drug monitoring; TDM) were not performed for linezolid. This is very unfortunate since linezolid is the drug of interest in their study. As a consequence, it is unknown if the pharmacokinetic/pharmacodynamic target of fAUC0–24h/MIC >100 is met. Therefore, in our opinion it is not correct to assume linezolid to be effective without DST and TDM for linezolid or without a control group.

Finally, we also do not support the conclusion that Koh et al. draw from the presented data being that a daily dose of 300 mg linezolid may be associated with fewer neuropathic side effects than a daily dose of 600 or 1200 mg linezolid. It was necessary to cease linezolid therapy in 14 patients (27%) due to neurotoxicity, which is within the range of the highly variable incidence of neurotoxicity of 4 – 89% at daily doses of 600 or 1200 mg as presented by Koh et al. in an overview of literature (1). Despite the daily dose of linezolid being low, the duration of administration of linezolid is long with a median of 278 days. This concurs with the current notion that the risk of adverse events of linezolid increases time-dependently (6).

Transparancy declarations

None to declare.

REFERENCES

1. Koh, W. J., Y. R. Kang, K. Jeon, O. J. Kwon, J. Lyu, W. S. Kim, and T. S. Shim. 2012. Daily 300 mg dose of linezolid for multidrug-resistant and extensively drug-resistant tuberculosis: updated analysis of 51 patients. J. Antimicrob. Chemother. 67:1503-1507.

2. Koh, W. J., O. J. Kwon, H. Gwak, J. W. Chung, S. N. Cho, W. S. Kim, and T. S. Shim. 2009. Daily 300 mg dose of linezolid for the treatment of intractable multidrug-resistant and extensively drug-resistant tuberculosis. J. Antimicrob. Chemother. 64:388-391.

3. Schon, T., P. Jureen, E. Chryssanthou, C. G. Giske, E. Sturegard, G. Kahlmeter, S. Hoffner, and K. A. Angeby. 2011. Wild-type distributions of seven oral second-line drugs against Mycobacterium tuberculosis. Int. J. Tuberc. Lung Dis. 15:502-509.

Chapter 5A

80

4. Alffenaar, J. W., T. van der Laan, S. Simons, T. S. van der Werf, P. J. van de Kasteele, H. de Neeling, and D. van Soolingen. 2011. Susceptibility of clinical Mycobacterium tuberculosis isolates to a potentially less toxic derivate of linezolid, PNU-100480. Antimicrob. Agents Chemother. 55:1287-1289.



Linezolid: safety and

efficacy monitoring

M.S. Bolhuis, A.D. Pranger, and J.W.C. Alffenaar

Eur Respir J. 2012 May;39(5):1275–6

Chapter 5B

Chapter 5B

82

We read with interest the article “Linezolid, an effective and cheap drug in MDR-TB treatment failure patients in India” by Singla et al. (1), which described the treatment outcome of 29 (pre-)extensively drug-resistant (XDR) tuberculosis (TB) patients from Delhi, India. All patients received linezolid as part of their anti-TB regimen. The high percentage favorable treatment outcome in the study led Singla et al. (1) to conclude that “linezolid could have played a key role”. However, in our opinion, conclusions on the role of a single agent, such as linezolid, are difficult to draw from a series of cases without controls, in which every patient received linezolid in addition to an injectable and a fluoroquinolone. Indeed, the important role of later-generation fluoroquinolones is addressed, but neither drug susceptibility testing (DST) nor drug concentration monitoring for linezolid, is performed. Therefore linezolid treatment itself could even be sub-therapeutic (2).

Singla et al. (1) conclude that “an aggressive, comprehensive management program using linezolid along with other drugs can favorably treat significant number of patients”. Although we concur with this statement, a closer look at the management program applied in this study suggests that the program might not be too aggressive or comprehensive. For instance, no Directly Observed Therapy (DOT) is applied, nor did the patients receive nutritional or good psychosocial support. Compliance was only assessed indirectly by checking empty blisters. Since non-compliance could lead to treatment failure and increase of resistance against the few drugs that are still effective in (pre-)XDR-TB treatment, we would strongly advise to abandon DOT only in exceptional cases where compliance is highly probable (3). Therapeutic drug monitoring (TDM) can be recommended to ensure adequate drug exposure during treatment. In rural areas dried blood spot analysis may enable TDM by offering an affordable tool for drug concentration measurement in a centralized laboratory using stable easy-to-obtain samples (4).

The very low incidence of major adverse events (AEs) of 10.3% as reported by Singla et al. (1) is in contrast with findings in literature where 41.2% of 85 MDR/XDR-TB patients treated with linezolid experienced major AEs (5). The authors provided no explanation for low AE incidence found in their study. Perhaps the fact that temporary discontinuation of linezolid is scored as a minor adverse event along with the absence of DOT resulted into a lower score of AEs in the study of Singla et al. Unfortunately, the authors provided no information on the manufacturer of the linezolid. Only the low cost of linezolid of less than $1 per tablet is mentioned, compared to approximately $80 per tablet in The Netherlands. It is well established that counterfeit drugs pose a great threat and counterfeit drugs sometimes contain little to none of the claimed drug (6). Although there is no evidence that the

83

Linezolid safety and efficacy monitoring

5B

administered drugs in this study are counterfeit, it can also not be excluded based on the information as provided by Singla et al. (1). This, combined with the absence of DOT and TDM, could very well be a reason for the low incidence of major adverse events as observed in the study by Singla et al. (1).

In our opinion, only a randomized controlled trial of linezolid vs. placebo as add on to an adequate background regimen using DST and TDM will provide comprehensive results on efficacy and safety of linezolid as potential drug for (pre-)XDR-TB treatment regimen.

Statement of interest: none.

ACKNOWLEDGMENTS

Reproduced with permission of the European Respiratory Society © Eur Respir J May 2012 39:1275-1276; doi:10.1183/09031936.00200911.

REFERENCES


2. Alffenaar, J. W., J. G. Kosterink, R. van Altena, T. S. van der Werf, D. R. Uges, and J. H. Proost. 2010. Limited sampling strategies for therapeutic drug monitoring of linezolid in patients with multidrug-resistant tuberculosis. Ther. Drug Monit. 32:97-101.

3. World Health Organisation (WHO). 2010. Treatment of tuberculosis: guidelines. World Health Organisation, Geneva.

4. Vu, D. H., J. W. Alffenaar, P. M. Edelbroek, J. R. Brouwers, and D. R. Uges. 2011. Dried blood spots: a new tool for tuberculosis treatment optimization. Curr. Pharm. Des. 17:2931-2939.


6. Bate, R., R. Tren, L. Mooney, K. Hess, B. Mitra, B. Debroy, and A. Attaran. 2009. Pilot study of essential drug quality in two major cities in India. PLoS One. 4:e6003.

Linezolid safety and

tolerability in multidrug-

resistant tuberculosis

patients: a retrospective

observational study

M.S. Bolhuis, S. Tiberi, G. Sotgiu, S. De Lorenzo, J.G.W. Kosterink, T.S. van der Werf, G.B. Migliori, and J.W.C. Alffenaar

Submitted

Chapter 5C

Chapter 5C

86

ABSTRACT

Objectives

Linezolid, known for its toxicity, is a promising drug for the treatment of multidrug-resistant tuberculosis (MDR-TB). Dose reduction has been studied attempting to limit toxicity, but concerns exist that dose reduction could result in inadequate linezolid exposure.

Methods

We aimed to investigate linezolid safety and tolerability in relation to linezolid exposure in a retrospective study at two tuberculosis centers in the Netherlands and Italy.

Results

A total of 58 MDR-TB patients was included. No correlation was observed between microscopy or culture conversion and the area under the time concentration curve / minimal inhibitory concentration ratio. Patients that experienced peripheral neuropathy had received a higher median cumulative dose or received linezolid for a longer median period of time, compared to patients without peripheral neuropathy.

Conclusions

Treatment regimens containing linezolid were effective and well tolerated. Peripheral neuropathy seemed to be mediated by cumulative linezolid dose and number of days of exposure to linezolid.

87

Retrospective study of linezolid in MDR-TB patients

5C

INTRODUCTION

The prevalence and incidence of multidrug-resistant tuberculosis (MDR-TB) cases, caused by M. tuberculosis strains resistant to at least rifampicin and isoniazid, is increasing in high tuberculosis burden countries and is expected to keep rising in the next three years (1). The World Health Organization (WHO) recommends to prescribe an MDR-TB therapeutic regimen consisting of at least four in vitro active anti-tuberculosis drugs (2-6).

Linezolid is a promising anti-tuberculosis drug for the treatment of MDR-TB (7, 8) and may be added to anti-tuberculosis regimens requiring a Group 5 drug. Recently, two systematic reviews and meta-analyses pointed out its excellent efficacy (9, 10). However, linezolid toxicity may outweigh its potential benefits. Indeed, adverse events were notified in almost 60% of the treated cases, with a high incidence of severe events such as anemia, peripheral neuropathy, optic neuritis, and thrombocytopenia. Dose reduction has been studied in an attempt to limit the toxicity. Decreased doses were associated with significantly lowered toxicity (11). Nevertheless, concerns might exist about the loss of efficacy or the emergence of acquired resistance, since dose reduction might result in inadequate linezolid exposure.

Therapeutic drug monitoring (TDM) has increasingly been recognized as an asset in the field of tuberculosis treatment several years ago (12-14). TDM may be adjunct in assessing individual linezolid exposure, especially since linezolid pharmacokinetics show a large inter-individual variability (15) and important drug-drug interactions have been observed (16, 17). Defining predictors for inter- or intra-patient variability might help in identifying patients at risk for deviating exposure. Preliminary data show that area under the time concentration curve (AUC)/minimal inhibitory concentration (MIC) ratio may be the pharmacokinetic / pharmacodynamic (PK/PD) target for Mycobacterium tuberculosis in order to adjust dosages based on TDM results (18). Furthermore, the evidence on correlation between reduced linezolid exposure and lowered toxicity is limited. Therefore, we aimed to retrospectively investigate linezolid safety and tolerability in relation to linezolid exposure.

METHODS

Study setting and participants

A retrospective study was conducted at two tuberculosis reference hospitals: the Tuberculosis Center Beatrixoord (University Medical Center Groningen, Haren, The Netherlands) and

Chapter 5C

88

the Tuberculosis Reference Center for MDR-TB and HIV-TB E. Morelli Hospital (Sondalo, Italy). We selected multi- and extensively-drug resistant tuberculosis (XDR-TB) patients that received linezolid as a part of their treatment regimen from 2010 – 2012 in Sondalo and from 2007 – 2012 in Haren (4). Patients had been diagnosed by means of standard microbiological culture tests. Patients younger than 18 years, patients lacking data due to recent admission to either reference hospital, and patients of whom no TDM data was available were excluded from the study.

Drug susceptibility testing and sample analysis

Data from drug susceptibility testing (DST) and pharmacokinetic analyses was collected retrospectively. DST was performed by the National Mycobacteria Reference Laboratory (National Institute for Public Health and the Environment, Bilthoven, the Netherlands) for patients in Haren or by the WHO Supranational Reference Laboratory (Milan, Italy) for patients in Sondalo. Exact MICs were determined for linezolid (19).

In Haren, linezolid samples were analyzed using a liquid chromatography tandem mass-spectrometric method (20). Samples obtained from patients in Sondalo were analyzed at the Luigi Sacco hospital (Milan, Italy) using a high performance liquid chromatography method (21).

Data collection

Anonymous retrospective data were retrieved by two researchers. The Institutional Review Board of the University Medical Center Groningen waived the requirement for research subjects to give informed consent (METc 2013/492). Date of admission, gender, country of birth, WHO Region of birth were collected. We calculated body mass index (kg/m2) at time of admission. In case of diabetes mellitus type II co-morbidity and HIV co-infection, clinical information including treatment was recorded. Besides these variables, detailed information on the DST and the location of infection (extra-pulmonary and/or pulmonary, including radiographic findings) based on findings of the attending physician were noted. Treatment information consisted of treatment outcome, duration of treatment in months, and cumulative composition of the administered treatment regimen.

Pharmacokinetic data consisted of AUC0–12/24, the highest and the lowest measured concentration (Cmax, and Cmin, respectively). AUCs were calculated using trapezoidal, non-

89


5C

compartmental, no lag-time calculations. Date of sampling, number of days after starting linezolid, linezolid dose at time of sampling, co-medication during linezolid sampling were noted in regard to linezolid pharmacokinetics. In Haren, dose reduction was considered if the calculated AUC0–24h/MIC ratio exceeds 100 (12, 15, 18). After dose reduction, linezolid TDM was repeated at steady state. If full pharmacokinetic curves were available for different doses, the curve of the dose that was administered during the longest period of time was used for the analysis.

Treatment outcome was assessed besides interim status through collecting data on days to sputum smear (SS) or culture conversion, i.e. time in number of days between the first positive and first of two consecutive negative samples. To assess safety and tolerability, adverse events, e.g. leucopenia, peripheral neuropathy, and optic neuritis, were collected from the hospital records or laboratory data using local reference values. Patients in Haren, as opposed to patients in Sondalo, received prophylactic erythropoietine. Information on hemoglobin (Hb in mmol/L) was recorded at respectively day 0, 30, 60, 90, 120, and 150 after the first dose of linezolid, regardless of treatment duration of linezolid.

Statistical analysis

Statistical evaluation was performed using SPSS 20 (SPSS, Chicago, IL, USA). Baseline data of patients from both hospitals were compared using Student’s t test was performed for parameters that were normally distributed. Levene’s test for equality of variances was used to determine whether equal variances may or may not be assumed. Independent sample Mann-Whitney U tests were used for parameters that were not normally distributed. Frequency distributions were compared using Pearson Chi-square test for categorical parameters; in case the necessary assumptions were not met, Fisher’s Exact test was used.

RESULTS

In both centers, a combined number of 58 MDR/XDR-TB patients that had received linezolid and underwent TDM were included. Figure 1 depicts a schematic summary of the selection process of the patients at both centers. From 54 enrolled patients in Sondalo, 36 were excluded: 6 due to insufficient data due to recent admission and 27 due to absence of pharmacokinetic data of linezolid. Demographic and clinical characteristics of the individuals included in the study are described in Table 1. Due to the different geographical

Chapter 5C

90

Fig

ure

1

Sc

he

ma

tic

su

mm

ary

of

pa

tie

nt

sele

cti

on

.

58

pa

tie

nts

en

roll

ed

18

exc

lud

ed

4 y

ou

ng

er

tha

n1

8 y

ea

rs

4 n

orm

all

yse

nsi

tiv

e t

ub

erc

ulo

sis

7 n

oth

era

pe

uti

cd

rug

mo

nit

ori

ng

da

ta a

va

ila

ble

3 in

suff

icie

nt

da

ta d

ue

to r

ece

nt

ad

mis

sio

nto

ho

spit

al 4

0 p

ati

en

tsin

clu

de

d

54

pa

tie

nts

en

roll

ed

36

exc

lud

ed

3 y

ou

ng

er

tha

n1

8 y

ea

rs

0 n

orm

all

yse

nsi

tiv

e t

ub

erc

ulo

sis

27

no

the

rap

eu

tic

dru

g m

on

ito

rin

gd

ata

av

ail

ab

le

6 in

suff

icie

nt

da

ta d

ue

to r

ece

nt

ad

mis

sio

nto

ho

spit

al

18

pa

tie

nts

incl

ud

ed

Tu

be

rcu

losi

sC

en

ter

Be

atr

ixo

ord

(H

are

n, t

he

Ne

the

rla

nd

s)A

OV

V E

Mo

rell

iH

osp

ita

l(S

on

da

lo,

Ita

ly)

58

incl

ud

ed

in a

na

lysi

s

91


5C

setting and different policy concerning migrants of both countries and, thus, the influx of migrants, the WHO Region of origin of patients differed for both hospitals. In Sondalo, most patients were from Europe (n=14, 77.8%), whereas in Haren, most patients were from Europe (n=13, 32.5%) and the Eastern Mediterranean (n=12, 30.0%). In one case from Sondalo, radiology revealed extra-pulmonary tuberculosis. The referral center in Sondalo mostly received complicated infectious cases and does not admit extra-pulmonary TB cases for a long period of time.

Table 1 Socio-demographic and clinical characteristics in 58 patients with multidrug-resistant

tuberculosis treated at Sondalo (Italy) and Haren (the Netherlands)

All patients (n=58) Sondalo (n=18) Haren (n=40) p-value

Age (years; median, IQR) 30.0 (25.0 – 37.3) 35.0 (29.8 – 40.3) 29.5 (23.3 –33.5) 0.075

HIV status Negative Positive Missing

52 (90%)4 (7%)2 (3%)

17 (94%)1 (6%)0 (0%)

35 (88%)3 (8%)2 (4%)

Exposure to ART 3 1 2

SexFemaleMale

28 (48%)30 (52%)

7 (39%)11 (61%)

21 (52%)19 (48%)

0.337

WHO Region of originAfricaSouth East AsiaEastern MediterraneanAmericasEuropeWestern Pacific

5 (9%)4 (7%)14 (24%)3 (5%)27 (47%)5 (9%)

1 (6%)1 (6%)2 (11%)0 (0%)14 (78%)0 (0%)

4 (10%)3 (8%)12 (30%)3 (8%)13 (32%)5 (12%)

Weight (kg; median, IQR) 62.2 (55.6 – 69.9) 64.0 (58.0 – 70.0) 60.9 (55.5 – 69.4) 0.550

BMI (kg/m2; median, IQR) 21.2 (19.3 – 23.7) 21.4 (18.4 – 23.2) 21.0 (19.5 – 23.8) 0.455

Radiology findingsCavitary lesionsBilateral pulmonary involvement with cavitary lesionsBilateral pulmonary involvement without cavitary lesionsNoncavitary nonbilateral pulmonaryExtrapulmonary tuberculosis

16 (28%)7 (12%)

8 (14%)

19 (33%)

8 (14%)

7 (39%)4 (22%)

4 (22%)

3 (17%)

0 (0%)

9 (22%)3 (8%)

4 (10%)

16(40%)

8 (20%)

IQR: interquartile range; HIV: human immunodeficiency virus; ART: antiretroviral therapy; WHO: World Health Organization; BMI: body mass index; *Fisher’s Exact Test. The t-test for equality of means between Sondalo and Haren was used for normally distributed parameters, i.e. age, weight and BMI. Normality was confirmed using Shapiro-Wilk test. Frequency distribution of sex between patients in Sondalo and Haren is calculated using the Pearson Chi-square test.

Chapter 5C

92

Table 2 Treatment regimen composition in 58 patients with multidrug-resistant tuberculosis treated

at Sondalo (Italy) and Haren (the Netherlands)

All patients (n=58) Sondalo (n=18) Haren (n=40) p-value

Drug susceptibility of Mtb (Number of drugs; median, IQR)

Susceptible 6.0 (4.0 – 8.0) 7.0 (5.8 – 8.3) 5.0 (4.0 – 7.8) 0.071Resistant 8.0 (7.0 –10.0) 8.5 (6.8 – 9.3) 8.0 (7.0 – 10.0) 0.534

Drugs used:

Number of anti-TB drugs (median, IQR)

6.0 (5.0 – 7.0) 7.0 (6.0 – 7.3) 5.0 (4.0 – 6.0) 0.004

Group 1:Ethambutol 26 (45%) 5 (28%) 21 (52%) 0.080Pyrazinamide 14 (24%) 3 (17%) 11 (28%) 0.513*

Rifabutin 4 (7%) 0 (0%) 4 (10%) 0.300*

Group 2:Aminoglycosides 52 (90%) 16 (89%) 36 (90%) 1.000*

Kanamycin 7 (12%) 0 (0%) 7 (18%) 0.087*

Amikacin 45 (78%) 16 (89%) 29 (73%) 0.307*

Capreomycin 1 (2%) 0 (0%) 1 (3%) 1.000*

Group 3:Moxifloxacin 55 (95%) 17 (94%) 38 (95%) 1.000*

Levofloxacin 1 (2%) 0 (0%) 1 (3%) 1.000*

Group 4:Ethionamide † 18 (31%) 4 (22%) 14 (35%) 0.330Cycloserine 19 (33%) 17 (94%) 2 (5%) <0.001PAS 10 (17%) 9 (50%) 1 (3%) <0.001*

Group 5:Clofazimine 30 (52%) 8 (44%) 22 (55%) 0.457Amoxicillin-clavulanic acid 17 (29%) 15 (83%) 2 (5%) <0.001Thioacetazon 1 (2%) 0 (0%) 1 (3%) 1.000*

Clarithromycin 12 (21%) 0 (0%) 12 (30%) 0.011*

Other drugsMeropenem 13 (22%) 13 (72%) 0 (0%) <0.001*

Cotrimoxazole 8 (14%) 0 (0%) 8 (20%) 0.048*

Mtb: Mycobacterium tuberculosis; IQR: interquartile range; TB: tuberculosis; † consists of data of both protionamide and ethionamide; *Fisher’s Exact Test. To test whether the drug susceptibility and number of drugs used at any time during treatment was the same in both hospitals the Independent Sample Mann Whitney U test was used. Frequency distribution of number drugs between Sondalo and Haren are calculated using the Pearson Chi-square test. When assumptions for the Pearson Chi-square test were not met, 2-sided Fisher’s Exact Test was used.

DST revealed susceptibility to a median (interquartile range, IQR) number of 6.0 (4.0 – 8.0) drugs and resistance to 8.0 (7.0 – 10.0) drugs. Treatment regimens of 58 MDR/XDR-TB patients are displayed in Table 2. The median (IQR) number of different active drugs

93


5C

administered at any one time point during treatment was 6.0 (5.0 – 7.0). Patients in Sondalo received more different drugs compared to Haren (p=0.004), with a median (IQR) of 7.0 (6.0 – 7.3) drugs in Sondalo and 5.0 (4.0 – 6.0) drugs in Haren. DST revealed a median (IQR) MIC for linezolid of 0.5 (0.25 – 0.5) mg/L.

There were some notable differences in treatment regimens between the two centers: in Sondalo, patients received an aggressive treatment regimen including moxifloxacin, amikacin, linezolid, meropenem and/or amoxicillin/clavulanic acid if patients were very ill or DST revealed lack of active oral options. Two patients in Sondalo received bedaquiline as a part of a compassionate use program (22). In Sondalo, meropenem was administered to 13 (72.2%) patients, in contrast to none of the patients in Haren (p<0.001). Furthermore, WHO group 4 drug cycloserine was administered to almost all patients in Sondalo (94%), but was rarely (5%) administered to patients in Haren (p<0.001) due to differences in the availability of tuberculosis medication (23).

In Haren, identification of MDR-TB isolates by rapid molecular tests was followed by empirical treatment including kanamycin IV in doses adjusted using TDM, resulting in relatively low dosage of around 7 mg/kg. Usually, after 6 months treatment was switched to oral clofazimine; moxifloxacin (24); linezolid (12); and clarithromycin (12, 25). Cotrimoxazole was administered to 8 (20.0%) of the included patients in Haren based on DST (26, 27), compared to none (0.0%) in Sondalo (p=0.048).

Efficacy

Of the 58 included patients, 29 (50.0%) patients were considered cured and 4 (6.9%) patients completed their treatment; 25 (43.1%) patients were still on treatment. All but one patient (98.3%), including those still on treatment, were culture and smear-microscopy negative at the time of data collection. Of the individuals with positive sputum smear and cultures at the beginning of the tuberculosis treatment, median (IQR) time to conversion was 47 (21 – 68) days for sputum smear and 47 (22 – 61) days for culture. There was no significant difference between time to microscopy or culture conversion between the two participating hospitals (p=0.062 and p=0.781, respectively). There was a low correlation between AUC/MIC ratio and parameters predictive of treatment outcome, i.e. days to sputum smear-microscopy (r2

linear = 0.364) or to culture conversion (r2

linear = 0.169). The same lack of correlation applies for linezolid Cmin, Cmax, dose in mg/kg/day and days to culture or microscopy conversion with no linear r2

linear higher than 0.138. When cases are categorized based AUC/MIC ratio using 100 as a cutoff,

Chapter 5C

94

Table 3 Hemoglobin (mmol/L) at day 0, day 30, day 60, and day 90 after the first administration of

linezolid

Erythropoietin

Yes n No n p-value

Day 0 8.0 (7.1 – 8.6) 30 8.3 (7.1 – 9.5) 27 0.452

Day 30 7.8 (7.0 – 8.4) 29 7.9 (6.8 – 8.7) 27 0.863

Day 60 7.9 (7.1 – 8.9) 23 7.8 (6.9 – 8.8) 22 0.946

Day 90 7.5 (6.6 – 8.9) 15 7.7 (6.6 – 8.3) 15 0.967

Data presented as median (interquartile range) of hemoglobin (mmol/L). Erythropoietin was administered in a dose of 2000 IE two times a week. Data of day 120, day 150 and day 180 are not presented due to limited number of data. Mann Whitney U test (Wilcoxon rank sum) was used to test whether the distribution of hemoglobin concentrations were the same with or without erythropoietin at every time point.

which is a commonly used target based on other microorganisms, the distribution of days to microscopy (p=0.984) or culture conversion (p=0.241) was the same across both groups.

Adverse events

Linezolid and erythropoietin were co-administered for a median (IQR) of 88 (62 – 142) days. Baseline hemoglobin levels (median; IQR) of patients that did (8.0; 7.1 – 8.6 mmol/L) or did not receive erythropoietin at the start of linezolid therapy (8.3; 7.1 – 9.5 mmol/L) were similar (p=0.452) (Table 3). Furthermore, at all other time points up to 90 days there was no difference between hemoglobin of patients that did or did not receive erythropoietin, nor was there a difference in hemoglobin over time per patient between the two groups (Table 3). There were too few patients with data on days 120 – 180 available to allow for adequate interpretation of these data. None of the patients in Haren received blood transfusions.

Table 4 describes the cumulative dose, days of exposure to linezolid, AUC0–24h, and Cmin

of linezolid for patients with and without linezolid-related adverse events. Peripheral neuropathy was observed in 11 (19%) of the patients. The distribution of linezolid cumulative dosage in mg/kg (p=0.041) and days of exposure to linezolid (p=0.003) differed across patient groups with and without peripheral neuropathy. The cumulative linezolid dose was significantly higher in patients with peripheral neuropathy. They had a median (IQR) of 1,829 (1,414 – 2,255) mg/kg versus those without neuropathy that had a median (IQR) of 1,221 (742 – 1,994) mg/kg (p=0.041). A similar observation was made for the number of linezolid treatment days: median (IQR) 159 (120 – 196) days in patients with and 97

95


5C

Ta

ble

4

Cu

mu

lati

ve

do

se,

da

ys

of

ex

po

sure

to

lin

ez

oli

d,

AU

C0

–2

4h,

an

d C

min

of

lin

ez

oli

d f

or

pa

tie

nts

wit

h a

nd

wit

ho

ut

lin

ez

oli

d-r

ela

ted

ad

ve

rse

ev

en

ts

N (%

)Li

nezo

lid c

umul

ativ

e do

se (

mg/

kg)

p-va

lue

Tota

l lin

ezol

id

expo

sure

(day

s)p-

valu

eLi

nezo

lid A

UC 0–

24h

(mg*

h/L)

p-va

lue

Line

zolid

Cm

in

(mg/

L)p-

valu

e

Ane

mia

Yes

No

9 (1

6%)

30 (5

2%)

918

(550

– 1

964)

1211

(855

– 1

830)

0.48

097

(52

– 15

9)97

(66

– 15

0)0.

768

158

(117

– 2

18)

128

(98

– 17

0)0.

299

3.4

(2.5

– 5

.5)

2.5

(1.5

– 4

.0)

0.16

9

Leuk

open

iaYe

s N

o5

(9%

)40

(69%

)18

98 (1

056

– 26

68)

1046

(739

– 1

813)

0.19

415

0 (1

02 –

352

)86

(62

– 15

0)0.

065

158

(134

– 1

78)

122

(89

– 17

1)0.

314

3.8

(3.0

– 4

.4)

2.6

(1.4

– 4

.2)

0.20

7

Perip

hera

l neu

ropa

thy

Yes

No

11 (1

9%)

25 (4

3%)

1829

(141

4 –

2255

)11

64 (7

55 –

192

2)0.

041

159

(120

– 1

96)

97 (6

6 –

147)

0.00

314

9 (1

04 –

185

)10

7 (9

1 –

156)

0.26

13.

2 (1

.6 –

4.1

)2.

0 (1

.4 –

3.8

)0.

477

Opt

ical

neu

ropa

thy

Yes

No

1 (2

%)

48 (8

3%)

997

1306

(861

– 1

985)

0.73

589 10

5 (6

7 –

169)

0.76

610

012

7 (9

6 –

169)

0.61

22.

52.

6 (1

.5 –

4.0

)0.

98

N: n

umbe

r of p

atie

nts (

perc

enta

ge o

f the

tota

l num

ber o

f pat

ient

s inc

lude

d in

the

stud

y, i.

e. 5

8); A

nem

ia: <

8.7

mm

ol/L

(♂),

<7.5

mm

ol/L

(♀);

Leuk

open

ia: <

4*10

9 /L (♂

,♀).

Line

zolid

AU

C 0–24

h: are

a un

der t

he ti

me

linez

olid

con

cent

ratio

n cu

rve

from

0-2

4 ho

urs;

Lin

ezol

id C

min

: low

est m

easu

red

linez

olid

con

cent

ratio

n.M

ann

Whi

tney

U te

st w

as u

sed

to te

st w

heth

er th

e di

strib

utio

n of

par

amet

ers

was

the

sam

e ac

ross

gro

ups

with

or w

ithou

t ane

mia

, leu

cope

nia,

per

iphe

ral n

euro

path

y,

and

optic

al n

euro

path

y re

spec

tivel

y.

Chapter 5C

96

(66 – 147) days in patients without peripheral neuropathy (p=0.003). Nine percent (5/45; 8 missing data) had leucopenia: no difference was identified in cumulative linezolid dose nor in days of exposure to linezolid (p=0.194 and 0.065, respectively). One patient had optical neuropathy.

The distribution of AUC0–24h was the same for patients with and without anemia (p=0.299), leucopenia (p=0.314), optical neuritis (p=0.612), and peripheral neuropathy (p=0.261). The distribution of Cmin of linezolid was the same across the categories anemia, leucopenia, optical neuritis, and peripheral neuropathy. Only two of 48 patients (2/48; 10 missing data) had diabetes type 2, rendering combined analysis of adverse events and diabetes impossible.

Pharmacokinetic parameters

Patients had a mean (standard deviation, SD) Cmin of 3.1 (2.2) mg/L, a Cmax of 9.4 (3.1) mg/L, and an AUC0–12h of 70.1 (31.9) mg*h/L with all different dosages combined. Pharmacokinetic parameters for the different dose categories are displayed in Table 5. Due to the small number of patients receiving 400 mg, 900 mg, or 1,200 mg linezolid respectively, comparison of pharmacokinetic parameters of these groups was not possible.

There was no correlation between linezolid dose in mg/kg and AUC0–12h with or without excluding P-glycoprotein modulators (r2

linear were 0.012 and 0.001, respectively). Patients that concomitantly received P-gp inhibitors did not display statistically significant different means for any of the pharmacokinetic parameters. Furthermore, ideal body weight did not correlate well with AUC0–12h.

Table 5 Pharmacokinetic parameters of linezolid

Linezolid daily dosage

400 mg (n=5) 600 mg (n=41) 900 mg (n=7) 1,200 mg (n=5) p-value

Cmin (mg/L) 3.7 (1.9) 2.8 (1.8) 3.0 (1.5) 4.6 (5.0) 0.405

Cmax (mg/L) 11.0 (3.8) 9.0 (3.1) 9.3 (2.5) 10.6 (3.1) 0.666

AUC0–12h (mg*h/L) 83.3 (33.9) 65.9 (30.8) 71.3 (19.7) 89.2 (49.3) 0.472

AUC/MIC ratio 499 (375) 357 (252)* 432 (135) 707 (499) 0.193

Data presented as mean (standard deviation). Pharmacokinetic parameters of patients that concomittantly received P-gp modulators are excluded.* n=37 due to 4 missing MICs. P-values calculated using Independent Samples Kruskal-Wallis Test displaying whether the distribution of pharmacokinetic parameters is the same across all linezolid dosage groups, i.e. 400 mg, 600 mg, 900 mg, and 1,200 mg. A non-parametric test was used because of the small number of samples.

97


5C

DISCUSSION

This is the first study to investigate linezolid toxicity in relation to pharmacokinetic linezolid exposure. The efficacy of treatment regimens containing linezolid in this cohort was very good, with a large percentage of MDR-TB patients that was considered cured. The median times observed to microscopy and culture conversion (both 47 days) were similar to those of a recent meta-analysis; in that analysis, microscopy converted to negative after median 43.5 days, and culture after 61 days (10). The proportion of adverse events in the retrospective cohort was lower than previously reported in a large meta-analysis (10). In our cohort, anemia occurred in 16%; peripheral neuropathy in 19%; and optical neuritis in 2% of all patients; anemia was seen in 38%, peripheral neuropathy in 47%, optic neuritis in 13% of patients in the meta-analysis (10). Comparison of dose related effects is difficult due to the multitude of the included studies in the meta-analysis, the non-fixed dose in our retrospective study and small overlap of data (10).

In Haren, 75% of the included patients received erythropoietin in a dose of 2,000 IUs twice a week. Erythropoietine was given in an attempt to prevent anemia which could otherwise lead to cessation of an effective drug in MDR-TB patients. Analysis of the Hb levels revealed no significant differences between patients that did or did not receive preventive erythropoietine. As no evidence for any benefit emerged from this analysis, we would now argue against routine use of erythropoietine in patients receiving linezolid.

Peripheral neuropathy was observed in 19% of the patients. The cumulative linezolid dosage and the number of days patients were exposed to linezolid were observed to be statistically significantly higher in patients who experienced peripheral neuropathy. A possible explanation that cumulative linezolid dosage and days of exposure to linezolid, in contrast to linezolid Cmin or AUC, are correlated to peripheral neuropathy might be that even the lower linezolid Cmin and AUCs are above mitochondrial toxicity threshold (28).

International collaboration in MDR/XDR-TB research is critically important to merge data and reach meaningful cohort sizes to explore important clinical questions. The toxicity data we found, e.g. for anemia and leucopenia in patients with higher Cmin could only be detected by merging data of our two centers. The fact that in the same analysis, Cmin did not predict optical neuritis and peripheral neuropathy could be due to the limited sample size, but moment of pharmacokinetic sampling in relation to the occurrence of adverse events amongst others.

Chapter 5C

98

Several methodological study limitations can be identified, particularly the limited sample size and the retrospective epidemiological nature of our study, and missing data. Unfortunately, 43 patients had to be excluded, because of lack of TDM data or due to recent admission to the hospital. Despite the fact that the inclusion and exclusion process has been transparently documented, this could have led to a selection bias. Furthermore, pharmacokinetic linezolid data were cross-sectionally computed at one or few moments in time. During the course of treatment, linezolid dosages or clinical parameters might have changed. The differences between the two study settings is another relevant limitation. Data were collected at two hospitals by two different researchers. We cannot exclude general differences in standard of care, treatment, and monitoring. The impact of observed differences in treatment regimens is expected to be limited since most differences are seen in drugs with a disputable place in therapy (6). Furthermore, differences were inherent to the individual character of treatment regimens.

Despite the retrospective nature of this study and the above mentioned limitations, our study supports the increasingly important position of linezolid in MDR-TB treatment regimens. Toxicity is more frequent with higher doses, i.e. 600 mg twice daily (15). However, with reduced dosages, adverse events appear manageable and infrequent. Our findings in this retrospective study justify confirmation in a prospective study.

One of the most important questions that remains to be answered is what pharmacokinetic / pharmacodynamic target should be aimed for in the treatment of MDR-TB? There are suggestive in vitro data for M. tuberculosis to support an AUC/MIC ratio of 100 which is adequate for gram-positive organisms (18). In addition, in vivo data showed promising results of linezolid added to other anti tubercular agents (29). Our study showed a weak association between AUC/MIC ratio and days to sputum smear-microscopy or to culture conversion in the presence of multi drug regimen limited by its retrospective nature. However, target AUC/MIC might be obtained in in vitro pharmacodynamic infection models that simulate human pharmacokinetics, such as hollow-fiber models. Next, prospective PK/PD studies should be designed to validate these targets in MDR-TB patients. Perhaps, linezolid dried blood spot and a limited sampling strategy could facilitate TDM in such studies (30). Being able to taper the dose to the lowest effective linezolid dosage with minimal toxicity based on TDM together with affordable, generic linezolid in safe and effective treatment regimen containing new agents, such as PA-824, bedaquiline, and delamanid, could help in the management of tuberculosis worldwide.

99


5C

In conclusion, our retrospective study suggests that the treatment regimens containing linezolid were effective and well tolerated, with high sputum microscopy and culture conversion rates and low frequency of adverse events. Interestingly, peripheral neuropathy seems to be mediated by cumulative dose and days of exposure to linezolid. There was no relevant correlation between pharmacokinetic/pharmacodynamic parameters and patients characteristics.

REFERENCES


2. Ahuja, S. D., D. Ashkin, M. Avendano, R. Banerjee, M. Bauer, J. N. Bayona, M. C. Becerra, A. Benedetti, M. Burgos, R. Centis, E. D. Chan, C. Y. Chiang, H. Cox, L. D’Ambrosio, K. DeRiemer, N. H. Dung, D. Enarson, D. Falzon, K. Flanagan, J. Flood, M. L. Garcia-Garcia, N. Gandhi, R. M. Granich, M. G. Hollm-Delgado, T. H. Holtz, M. D. Iseman, L. G. Jarlsberg, S. Keshavjee, H. R. Kim, W. J. Koh, J. Lancaster, C. Lange, W. C. de Lange, V. Leimane, C. C. Leung, J. Li, D. Menzies, G. B. Migliori, S. P. Mishustin, C. D. Mitnick, M. Narita, P. O’Riordan, M. Pai, D. Palmero, S. K. Park, G. Pasvol, J. Pena, C. Perez-Guzman, M. I. Quelapio, A. Ponce-de-Leon, V. Riekstina, J. Robert, S. Royce, H. S. Schaaf, K. J. Seung, L. Shah, T. S. Shim, S. S. Shin, Y. Shiraishi, J. Sifuentes-Osornio, G. Sotgiu, M. J. Strand, P. Tabarsi, T. E. Tupasi, R. van Altena, M. Van der Walt, T. S. Van der Werf, M. H. Vargas, P. Viiklepp, J. Westenhouse, W. W. Yew, J. J. Yim, and Collaborative Group for Meta-Analysis of Individual Patient Data in MDR-TB. 2012. Multidrug resistant pulmonary tuberculosis treatment regimens and patient outcomes: an individual patient data meta-analysis of 9,153 patients. PLoS Med. 9:e1001300.

3. Falzon, D., N. Gandhi, G. B. Migliori, G. Sotgiu, H. S. Cox, T. H. Holtz, M. G. Hollm-Delgado, S. Keshavjee, K. DeRiemer, R. Centis, L. D’Ambrosio, C. G. Lange, M. Bauer, D. Menzies, and Collaborative Group for Meta-Analysis of Individual Patient Data in MDR-TB. 2013. Resistance to fluoroquinolones and second-line injectable drugs: impact on multidrug-resistant TB outcomes. Eur. Respir. J. 42:156-168.

4. Falzon, D., E. Jaramillo, H. J. Schunemann, M. Arentz, M. Bauer, J. Bayona, L. Blanc, J. A. Caminero, C. L. Daley, C. Duncombe, C. Fitzpatrick, A. Gebhard, H. Getahun, M. Henkens, T. H. Holtz, J. Keravec, S. Keshavjee, A. J. Khan, R. Kulier, V. Leimane, C. Lienhardt, C. Lu, A. Mariandyshev, G. B. Migliori, F. Mirzayev, C. D. Mitnick, P. Nunn, G. Nwagboniwe, O. Oxlade, D. Palmero, P. Pavlinac, M. I. Quelapio, M. C. Raviglione, M. L. Rich, S. Royce, S. Rusch-Gerdes, A. Salakaia, R. Sarin, D. Sculier, F. Varaine, M. Vitoria, J. L. Walson, F. Wares, K. Weyer, R. A. White, and M. Zignol. 2011. WHO guidelines for the programmatic management of drug-resistant tuberculosis: 2011 update. Eur. Respir. J. 38:516-528.

Chapter 5C

100

5. Migliori, G. B., G. Sotgiu, N. R. Gandhi, D. Falzon, K. DeRiemer, R. Centis, M. G. Hollm-Delgado, D. Palmero, C. Perez-Guzman, M. H. Vargas, L. D’Ambrosio, A. Spanevello, M. Bauer, E. D. Chan, H. S. Schaaf, S. Keshavjee, T. H. Holtz, D. Menzies, and Collaborative Group for Meta-Analysis of Individual Patient Data in MDR-TB. 2013. Drug resistance beyond extensively drug-resistant tuberculosis: individual patient data meta-analysis. Eur. Respir. J. 42:169-179.


7. Lange, C., I. Abubakar, J. W. Alffenaar, G. Bothamley, J. A. Caminero, A. C. Carvalho, K. C. Chang, L. Codecasa, A. Correia, V. Crudu, P. Davies, M. Dedicoat, F. Drobniewski, R. Duarte, C. Ehlers, C. Erkens, D. Goletti, G. Gunther, E. Ibraim, B. Kampmann, L. Kuksa, W. de Lange, F. van Leth, J. van Lunzen, A. Matteelli, D. Menzies, I. Monedero, E. Richter, S. Rusch-Gerdes, A. Sandgren, A. Scardigli, A. Skrahina, E. Tortoli, G. Volchenkov, D. Wagner, M. J. van der Werf, B. Williams, W. W. Yew, J. P. Zellweger, D. M. Cirillo, and for the TBNET. 2014. Management of patients with multidrug-resistant/extensively drug-resistant tuberculosis in Europe: a TBNET consensus statement. Eur. Respir. J. 44:23-63.


9. Cox, H., and N. Ford. 2012. Linezolid for the treatment of complicated drug-resistant tuberculosis: a systematic review and meta-analysis. Int. J. Tuberc. Lung Dis. 16:447-454.



12. Bolhuis, M. S., R. V. Altena, D. V. Soolingen, W. C. Lange, D. R. Uges, T. S. Werf, J. G. Kosterink, and J. W. Alffenaar. 2013. Clarithromycin increases linezolid exposure in multidrug-resistant tuberculosis patients. Eur. Respir. J. 42:1614-1621.

13. Peloquin, C. A. 1997. Using therapeutic drug monitoring to dose the antimycobacterial drugs. Clin. Chest Med. 18:79-87.

14. Srivastava, S., C. A. Peloquin, G. Sotgiu, and G. B. Migliori. 2013. Therapeutic drug management: is it the future of multidrug-resistant tuberculosis treatment? Eur. Respir. J. 42:1449-1453.

101


5C




18. Rodriguez, J. C., L. Cebrian, M. Lopez, M. Ruiz, I. Jimenez, and G. Royo. 2004. Mutant prevention concentration: comparison of fluoroquinolones and linezolid with Mycobacterium tuberculosis. J. Antimicrob. Chemother. 53:441-444.



21. Peng, G. W., R. P. Stryd, S. Murata, M. Igarashi, K. Chiba, H. Aoyama, M. Aoyama, T. Zenki, and N. Ozawa. 1999. Determination of linezolid in plasma by reversed-phase high-performance liquid chromatography. J. Pharm. Biomed. Anal. 20:65-73.

22. Tiberi, S., S. De Lorenzo, R. Centis, P. Viggiani, L. D’Ambrosio, and G. B. Migliori. 2013. Bedaquiline in MDR-/XDR-TB cases: first experience on campassionate use. Eur. Respir. J. 43:289-292.

23. Sotgiu, G., L. D’Ambrosio, R. Centis, G. Bothamley, D. M. Cirillo, S. De Lorenzo, G. Guenther, K. Kliiman, R. Muetterlein, V. Spinu, M. Villar, J. P. Zellweger, A. Sandgren, E. Huitric, C. Lange, D. Manissero, and G. B. Migliori. 2012. Availability of anti-tuberculosis drugs in Europe. Eur. Respir. J. 40:500-503.

24. Pranger, A. D., R. van Altena, R. E. Aarnoutse, D. van Soolingen, D. R. Uges, J. G. Kosterink, T. S. van der Werf, and J. W. Alffenaar. 2011. Evaluation of moxifloxacin for the treatment of tuberculosis: 3 years of experience. Eur. Respir. J. 38:888-894.

25. Bolhuis, M. S., T. van der Laan, J. G. Kosterink, T. S. van der Werf, D. van Soolingen, and J. W. Alffenaar. 2014. In vitro synergy between linezolid and clarithromycin against Mycobacterium tuberculosis. Eur. Respir. J. 44:808-811.

26. Alsaad, N., R. van Altena, A. D. Pranger, D. van Soolingen, W. C. de Lange, T. S. van der Werf, J. G. Kosterink, and J. W. Alffenaar. 2012. Evaluation of co-trimoxazole in treatment of multidrug-resistant tuberculosis. Eur. Respir. J. 42:504-512.

Chapter 5C

102

27. Alsaad, N., B. Wilffert, R. van Altena, W. C. de Lange, T. S. van der Werf, J. G. Kosterink, and J. W. Alffenaar. 2014. Potential antimicrobial agents for the treatment of multidrug-resistant tuberculosis. Eur. Respir. J. 43:884-897.

28. De Vriese, A. S., R. V. Coster, J. Smet, S. Seneca, A. Lovering, L. L. Van Haute, L. J. Vanopdenbosch, J. J. Martin, C. C. Groote, S. Vandecasteele, and J. R. Boelaert. 2006. Linezolid-induced inhibition of mitochondrial protein synthesis. Clin. Infect. Dis. 42:1111-1117.

29. Zhao, W., Z. Guo, M. Zheng, J. Zhang, B. Wang, P. Li, L. Fu, and S. Liu. 2014. Activity of linezolid-containing regimens against multidrug-resistant tuberculosis in mice. Int. J. Antimicrob. Agents. 43:148-153.


In vitro synergy between

linezolid and clarithromycin

against Mycobacterium tuberculosis

M.S. Bolhuis, T. van der Laan, J.G.W. Kosterink, T.S. van der Werf, D. van Soolingen, and J.W.C. Alffenaar

Eur Respir J. 2014 Sep; 44(3): 808–11

Chapter 6

Chapter 6

104

To the Editor:

Approximately 3% of new tuberculosis (TB) cases worldwide represent multidrug-resistant tuberculosis (MDR-TB) (1). In these MDR-TB cases, resistance of Mycobacterium tuberculosis against the otherwise effective rifampicin and isoniazid forces clinicians to diverge to other antimicrobial agents. Such treatment options include World Health Organization (WHO) group 5 drugs linezolid and clarithromycin (1). Linezolid shows excellent efficacy in the treatment of MDR-TB, but its use is often troubled by adverse events (2-4). Linezolid has shown in vitro bacteriostatic activity against M. tuberculosis and is also effective at achieving culture conversion in drug resistant patients (5). In vitro testing revealed that clarithromycin is not very active against M. tuberculosis since the minimal inhibitory concentrations (MIC) are relatively high. Clinical efficacy seems questionable, since MICs, as reported in literature, are significantly higher than achievable serum peak levels in vivo (6). On the contrary, clarithromycin reaches adequate local concentrations in alveolar cells and in epithelial lining fluid, where most mycobacteria reside (7). Besides that lower clarithromycin MICs were observed by the Dutch National Mycobacteria Reference Laboratory (Bilthoven, the Netherlands; unpublished data).

Due to the limited number of new treatment options, optimizing existing treatment regimens is a conceivable option. Exploring synergy between tuberculosis drugs might help in improving treatment regimens. A study that investigated several anti-TB drugs, such as isoniazide, rifampin, and/or ethambutol, but not linezolid, revealed in vitro synergistic activity with clarithromycin against M. tuberculosis (8). In this study, we aim to investigate the possible in vitro synergy between linezolid and clarithromycin in Mycobacterium tuberculosis isolates obtained from multidrug-resistant, resistant, and drug-susceptible TB patients.

We randomly selected a panel of 24 M. tuberculosis isolates from the strain collection of the Tuberculosis Reference Laboratory of the National Institute for Public Health and the Environment (RIVM, Bilthoven, the Netherlands). The selected collection consisted of 13 multidrug-resistant, five drug-sensitive and six mono-resistant M. tuberculosis isolates. Drug susceptibility testing (DST) was performed using two methods: the absolute concentration method (ACM) and a Mycobacteria Growth indicator tube (MGIT) 960 system (9, 10).

For the ACM, we used a sterilized Middlebrook 7H10 agar of pH 6.6 supplemented with oleic-acid-dextrose-catalase (both Becton Dickinson and Company, Sparks, MD) and varying concentrations of drugs (9). The plates were checked for mycobacterial growth after 4, 8, 12, 14, 16, 19, and 21 days. The plates were analyzed when the growth in the control well

105

In vitro synergy between linezolid and clarithromycin

6

without anti-TB drugs was considered sufficient, i.e. when colonies were clearly visible and countable. At this point of time, all wells were checked for growth inhibition. Growth inhibition was defined as less than 90% of the colonies of the control well.

Next to the ACM, the MGIT 960 system with EpiCenter TB eXiST software was used (Becton Dickinson and Company, Sparks, MD) (10). Each tube contained 0.8 ml of Bactec MGIT drug susceptibility supplement and 100 μl of the appropriate drug solution. Growth was monitored hourly. The tubes were analyzed when growth unit (GU) value of the growth control tube, containing a 1:100 dilution of the inocula, reached 400. Growth inhibition was defined as GU value <100.

The checkerboard method was used to study in vitro synergy between linezolid and clarithromycin (both Sigma Aldrich, St.Louis, MO, USA). Linezolid was added in concentrations between 0 – 0.5 μg/mL and clarithromycin with a range of 0 – 8 μg/mL as is shown in Figure 1. We calculated the lowest fractional inhibitory concentration (FIC) to determine synergy as follows: (MIClinezolid, combination / MIClinezolid, alone) + (MICclarithromycin, combination / MICclarithromycin, alone). Synergy was defined as a FIC ≤0.5, indifference as FIC >0.5 to 4, and antagonism as FIC >4 (11).

Of the selected M. tuberculosis isolates (n=24), synergy between clarithromycin and linezolid was determined for 74% by using the MGIT method and in 59% by using the ACM. The median (interquartile range, IQR) FIC was 0.37 (0.31 – 0.47) using the MGIT and 0.50 (0.38 – 0.75) using the ACM. The combination of drugs did not display antagonism in any of the isolates. A median checkerboard composed from all selected M. tuberculosis strains with clarithromycin and linezolid is shown in Figure 1. Synergy was observed in 77% of the MDR-TB isolates (n=13) using MGIT and in 46% in the ACM method. Combining clarithromycin and linezolid resulted in a median (IQR) FIC of 0.37 (0.32 – 0.37) using MGIT and 0.62 (0.375 – 1.0) using ACM in the MDR-TB isolates.

In conclusion, we observed synergy between clarithromycin and linezolid both with the MGIT and with the ACM method. This finding is in line with a previous observation that clarithromycin displayed synergy with isoniazide, rifampin, and/or ethambutol in M. tuberculosis (8).

Although the underlying mechanism is yet to be elucidated, it has been suggested that disorganization or disruption of the outer cell wall layer and the cytoplasmic membrane of the cell envelope by either clarithromycin or linezolid may play a role (8). This disruption

Chapter 6

106

Figure 1 Schematic median checkerboard of Mycobacteria tuberculosis growth inhibition with varying

concentrations of clarithromycin and/or linezolid using A. Mycobacteria Growth Indicator Tubes (MGIT,

n=23) and B. Absolute Concentration Method (ACM, n=23).

Each cell represents the median result, i.e. growth or no-growth, of all tested isolates. Grey cells indicate growth, white cells indicate >90% less growth than controls. $ in order to calculate the fractional inhibitory concentrations (FIC) 4 μg/mL is used as the MICclarithromycin, alone. The lowest FIC cell is marked by ‘FIC’. For example: the FIC of Figure 1B is (MIClinezolid, combination / MIClinezolid, alone) + (MICclarithromycin, combination / MICclarithromycin, alone) = (0.125 / 0.50) + (1.0 / 8.0 ) = 0.375.

2.0 $2 $

) 1.0

g/m

L)

in(μ

go

my

ci

0.5

0

FIC

rith

roC

lar

50

.25

00

.0

A0 0 0 0625 0 125 0 25 0 50

AMGIT

0.0 0.0625 0.125 0.25 0.50

Linezolid (μg/mL)MGIT

Linezolid (μg/mL)

8.0

8

) 4.0

g/m

L) 4

in(μ

go

my

ci

2.0

rith

roC

lar

1.0 FIC

00

.0

B0 0 0 0625 0 125 0 25 0 50

BACM

0.0 0.0625 0.125 0.25 0.50

Linezolid (μg/mL)ACM

Linezolid (μg/mL)

107


6

might allow easier penetration by the other drug, resulting in the observed in vitro synergy. However, this hypothesis assumes that permeability is normally a limiting factor. Further research studying the underlying mechanism is needed and might also explain the fact that we observed in vitro synergy in the majority, but not all of the isolates. Although the majority of the isolates in this research displayed in vitro synergy when clarithromycin and linezolid were combined, it would be more interesting to determine or predict which isolates display synergy before applying these drugs in treatment regimens. Indeed, checkerboard experiments have not been validated or widely accepted for tailoring treatment in individual cases, and therefore cut-off values for FIC to deviate from the theoretical cut-off value of 1.0 have been employed (11). Consequently, the number of isolates displaying synergy might be under- or overestimated.

Previously, we showed that clarithromycin increases linezolid exposure by 44% in MDR-TB patients (12). The implications were summarized as follows: clarithromycin might be used as a booster to increase linezolid exposure, comparable to low-dose ritonavir and protease inhibitors; and the relatively cheap clarithromycin might reduce costs of treatment of the relatively expensive linezolid (13). The in vitro synergy we observed in this study further strenghtens the case for adding clarithromycin as a secondary drug to MDR-TB treatment regimens. The increased drug susceptibility of linezolid and clarithromycin in combination with the increased linezolid exposure might allow for further reduction of linezolid dosage, further reducing costs and adverse events. A prospective evaluation of MDR-TB patients receiving both drugs as a part of their treatment regimen is warranted to investigate efficacy and tolerability in real life. Furthemore, synergy testing should be performed with both other second-line TB drugs, and new TB drugs in the pipeline. Especially since the number of new MDR-TB drugs emerging from the pipeline in the next years is expected to be limited, drug resistance should be avoided at all costs. Optimizing treatment regimens through use of combinations that show synergy could be one strategy to avoid over exended use of new drugs. This is particularly important when considering the new World Health Organization post-2015 Strategy, which is based on the concept of TB elimination (14, 15).

To conclude, clarithromycin and linezolid display in vitro synergy in multidrug-resistant M. tuberculosis isolates. Due to the boosting effect with linezolid, low incidence of adverse effects, low costs, observed higher concentrations in lung tissue, and the in vitro synergy with linezolid and other antimicrobial drugs, the role of clarithromycin might become more important in future MDR-TB treatment.

Chapter 6

108

ACKNOWLEDGMENTS

Reproduced with permission of the European Respiratory Society © Eur Respir J September 2014 44:808-811; published ahead of print May 2, 2014, doi:10.1183/09031936.00041314.

REFERENCES

1. Falzon, D., E. Jaramillo, H. J. Schunemann, M. Arentz, M. Bauer, J. Bayona, L. Blanc, J. A. Caminero, C. L. Daley, C. Duncombe, C. Fitzpatrick, A. Gebhard, H. Getahun, M. Henkens, T. H. Holtz, J. Keravec, S. Keshavjee, A. J. Khan, R. Kulier, V. Leimane, C. Lienhardt, C. Lu, A. Mariandyshev, G. B. Migliori, F. Mirzayev, C. D. Mitnick, P. Nunn, G. Nwagboniwe, O. Oxlade, D. Palmero, P. Pavlinac, M. I. Quelapio, M. C. Raviglione, M. L. Rich, S. Royce, S. Rusch-Gerdes, A. Salakaia, R. Sarin, D. Sculier, F. Varaine, M. Vitoria, J. L. Walson, F. Wares, K. Weyer, R. A. White, and M. Zignol. 2011. WHO guidelines for the programmatic management of drug-resistant tuberculosis: 2011 update. Eur. Respir. J. 38:516-528.






7. Honeybourne, D., F. Kees, J. M. Andrews, D. Baldwin, and R. Wise. 1994. The levels of clarithromycin and its 14-hydroxy metabolite in the lung. Eur. Respir. J. 7:1275-1280.

109


6




11. Odds, F. C. 2003. Synergy, antagonism, and what the chequerboard puts between them. J. Antimicrob. Chemother. 52:1.

12. Bolhuis, M. S., R. V. Altena, D. V. Soolingen, W. C. Lange, D. R. Uges, T. S. Werf, J. G. Kosterink, and J. W. Alffenaar. 2013. Clarithromycin increases linezolid exposure in multidrug-resistant tuberculosis patients. Eur. Respir. J. 42:1614-1621.

13. Srivastava, S., C. A. Peloquin, G. Sotgiu, and G. B. Migliori. 2013. Therapeutic drug management: is it the future of multidrug-resistant tuberculosis treatment? Eur. Respir. J. 42:1449-1453.

14. Diel, R., R. Loddenkemper, J. P. Zellweger, G. Sotgiu, L. D’Ambrosio, R. Centis, M. J. van der Werf, M. Dara, A. Detjen, P. Gondrie, L. Reichman, F. Blasi, G. B. Migliori, and European Forum for TB Innovation. 2013. Old ideas to innovate tuberculosis control: preventive treatment to achieve elimination. Eur. Respir. J. 42:785-801.

15. D’Ambrosio, L., M. Dara, M. Tadolini, R. Centis, G. Sotgiu, M. J. van der Werf, M. Gaga, D. Cirillo, A. Spanevello, M. Raviglione, F. Blasi, G. B. Migliori, and on behalf of the European national programme representatives. 2014. Tuberculosis elimination: theory and practice in Europe. Eur. Respir. J. 43: 1410-1420.

General discussion and

future perspectives

Chapter 7

Chapter 7

112

In this thesis, we aimed to describe the clinical pharmacology of linezolid, with special focus on therapeutic drug monitoring (TDM) to optimize treatment of patients suffering from multidrug-resistant tuberculosis (MDR-TB). In order to attain our objectives, we studied the two main areas of pharmacology, i.e. the pharmacokinetics (PK) and pharmacodynamics (PD) of linezolid in MDR-TB.

Pharmacokinetic drug-drug interactions

In Chapter 2, we performed a review of literature on drug-drug interactions of linezolid and other drugs from the oxazolidinone group. The multitude of PK drug-drug interactions that are presented in the review of literature (chapter 2) underline that drug-interactions are an important factor to take into account when designing a treatment regimen. In these PK drug-drug interactions, the perpetrator changes PK parameters of the victim, possibly resulting in a clinically relevant increased or decreased exposure of the victim drug. Examples include the interaction between the perpetrator clarithromycin and the victim linezolid – studied in this thesis – resulting in a significant increase of linezolid exposure (1, 2). This interaction could potentially lead to an increased exposure of linezolid, a drug that is already infamous for its toxicity, forcing clinicians to temporarily cease treatment with linezolid. However, there are also examples of victim anti-TB drugs with a decreased exposure as a result of a PK drug-drug interaction. Lowered exposure to subtherapeutic levels could result in acquired resistance, limiting the number of effective drugs even further. Subtherapeutic exposure could also lead to treatment failure – possibly even resulting in death.

From the point of view of tuberculosis treatment, PK drug-drug interactions with the anti-TB drug as a victim might be perceived as being of most interest. However, drug interactions with anti-TB drugs as a perpetrator are potentially just as important. Suboptimal treatment of co-morbidities might delay or render adequate tuberculosis treatment impossible. Since delay of treatment does not only worsen outcomes of treatment of the individual patient, but also has a negative impact on transmission within a community (3), delay in the initiation of the intensive phase of treatment due to drug-drug interactions might also be unfavourable.

Besides PK drug-drug interactions that are mentioned in summaries of product characteristics (SmPCs) or are known from literature, physicians must always be vigilant for new drug-interactions. For instance, the PK drug-drug interaction between linezolid and clarithromycin (2) is not yet described in the SmPC of linezolid (4). Moreover, the product

113

General discussion and future perspectives

7

information explicitly states linezolid is not metabolised by humane CYP-isoforms (4). This underlines the importance of pharmacovigilance during the post-marketing phase. This is particularly important for drugs that are often used for the treatment of MDR-TB. These anti-MDR-TB drugs are used off-label, such as linezolid and clarithromycin, or on-label but with an accelerated registration program followed by a confirmatory trial, such as bedaquiline (5). Perhaps, an intensive monitoring programme of new anti-TB drugs and new combinations could help to facilitate the gathering of real world clinical data.

Designing treatment regimens

Clinicians often face a difficult challenge when designing treatment regimens consisting of at least four likely effective drugs for the treatment of MDR-TB, as recommended by the World Health Organisation guideline (6). In some low-resources countries, the challenge might be even greater due to lack of available drug susceptibility testing (DST) (7). In these cases, population-based data on DST and surveillance data determine the selection of likely effective drugs to design treatment regimens. However, besides drug susceptibility, there are other factors to take into account when designing a treatment regimen such as pattern of adverse events, drug-drug interactions, route of administration (e.g. intravenously versus orally), availability of the drug in the treatment setting, and even costs.

Of these items that should be taken into account when designing a treatment regimen, adverse events deserve special attention. In cases where in vitro DST reveals sensitivity, the susceptible drug might not be suitable to add to treatment regimens due to adverse events that have occurred. Most clinical studies on diagnosis or treatment of tuberculosis present data on the in vitro drug susceptibility of the included patients. However, the number of feasible treatment options left might even be of greater relevance for clinicians than the number of resistant drugs. In order to allow comparison between clinical studies, it would be helpful to standardize the range of anti-tuberculosis drugs for which DST is performed together with the methods used. A recent study underlined the importance of DST. Perhaps, an addendum on the 2005 manuscript of Laserson et al. “Speaking the same language: treatment outcome definitions for MDR-TB” could help to propagate speaking the same language on the subject of DST as well (8).

Chapter 7

114

Targets for Therapeutic Drug Monitoring

Although PK / PD parameters are critically important, unfortunately, there are no clear PK / PD targets yet that can be used for TDM for linezolid. However, several steps were made in defining these targets. For instance, in vitro studies revealed maximum killing rates at concentrations twice the MIC (9, 10). Linezolid also showed excellent activity at an AUC (area under the time concentration curve) / MPC90 (mutant prevention concentration of 90% of the strains) ratio of 116 in isolates of drug-resistant M. tuberculosis (11). At the moment, a linezolid AUC / MIC ratio of at least 100 is strived for in MDR-TB. However, it should be noted that this target ratio is not based on in vitro analysis of M. tuberculosis isolates but on other microorganisms, such as Streptococcus pneumoniae and Streptococcus aureus (12). Although there are several arguments suggesting the PK/PD target of an AUC/MIC ratio of 100 might be correct, an in vitro PD study is warranted, followed by a clinical validation in a prospective study.

In our retrospective study, in which almost all patients had an AUC/MIC ratio >100, we found no correlation between AUC/MIC ratio and sputum or culture conversion. This might suggest the AUC/MIC ratios were above target at a plateau on which further increasing the ratio does not have a clear effect on efficacy. Perhaps, doses could be lowered even further without loss of efficacy, whilst reducing toxicity.

Obviously, designing clinical PD studies by simply withholding treatment in one cohort is not ethically justified. Unfortunately, PD studies in animals are often not possible due to toxicity in animals at drug exposures that are relevant in humans (13). Because of these limitations, hollow fiber PD infection models of tuberculosis might provide a solution (13). These models take half-lives of drugs and clinically applied dosing schedules into account, mimicking in vitro exposure of M. tuberculosis in patients to anti-TB drugs as closely as possible. Previously, this method was used to determine the rifampicin PK/PD target in tuberculosis strains, i.e. a daily AUC0–24h/MIC of 24 (14). We propose to conduct a similar study to determine an appropriate PK/PD target for linezolid in MDR-TB strains. Once the PK/PD target is more clearly established for linezolid in MDR-TB, the target should be clinically validated.

The identification of the linezolid AUC/MIC ratio for treatment of MDR-TB will possibly reveal a plethora of subjects to study, but more importantly it might improve MDR-TB treatment regimens containing an effective anti-tuberculosis drug. One of the positive effects on the research of the applicability of linezolid in the treatment of MDR-TB could be that

115


7

the identification of the linezolid PK/PD target might result in a more homogeneous dosing of linezolid in clinical studies. Results from different studies will be more easily comparable since the attained targets will be similar. This could rapidly increase the number of eligible patients in cohorts of whom TDM data are available, thereby revealing information that would had not have been discovered with small numbers of patients or when too high dosages are applied, a practice that is probably common today (chapter 5).

TDM using standard or alternative sampling techniques, such as DBS sampling or oral fluid sampling, could contribute to limiting adverse effects by tapering the linezolid dose to the lowest possible effective dose. This tapered linezolid dose could particularly be helpful in optimizing treatment regimens of extra vulnerable groups of patients, such a patients with multi-morbidity, pediatric, geriatric, and pregnant or lactating patients. In these patients, pharmacokinetics might be altered. Furthermore, these groups are often not included in studies because of safety reasons, resulting in a lack of and delay on information. Nevertheless, MDR-TB obviously also affects children, the elderly and pregnant women and thus these patients might need to be treated with linezolid as well. In all of the above examples, development of PK models for specific populations is warranted. These PK models for specific patient groups could enable precise dose adjustments based on TDM, resulting in fast attainment of PK/PD targets. These data could also contribute to determining adequate standard dosages for these groups of patients.

As for adding linezolid to treatment regimens of pregnant or lactating patients with MDR-TB, the lack of knowledge on in utero effects of linezolid on the pregnant patient and her (unborn) child, the possible efficacy of the drug, the risk of adverse effects and alternatives have to be deliberated by the physician. The fact that untreated MDR-TB is lethal to both the mother and the unborn child and the fact that multidrug-resistance severely limits the number of treatment options that are still feasible, will possibly force clinicians to prescribe linezolid. In these cases, tapering linezolid to the lowest still effective dose could possibly limit the chances of negative effects on the pregnant patients and their (unborn) child. Development of guidelines and increasing the number of peer-reviewed publications on the treatment of pregnant MDR-TB patients including short- and long-term follow-up of the child, would be very helpful for clinicians. This will allow pregnant MDR-TB patients to make better-informed decisions on the risks of treatment.

Chapter 7

116

Linezolid-clarithromycin interaction

Another subject that warrants future study, is the PK drug-drug interaction between linezolid and clarithromycin (chapter 3). After confirming the results in a new cohort, the next step should be to elucidate the underlying mechanism of the interaction. Although several studies refer to an interaction with linezolid as a victim and other drugs as a perpetrator, as being modified by P-gp (15, 16), this is merely based on one in vitro study combined with the summary of product characteristics (4, 17). Both the summary of product characteristics and Wynalda et al. state that linezolid is not metabolized through cytochrome P-450 iso-enzymes (4, 17). Since the perpetrator drugs were P-glycoprotein and cytochrome P-450 3A4 modulators, the hypothesis was postulated that the interaction with linezolid might be P-pg mediated (15, 16). However, there is one article – written by an employee of Pfizer, the manufacturer of linezolid – stating that the linezolid drug-interaction might not be P-gp mediated (18). This raises the question whether Pfizer might have unpublished data on file, contributing to the knowledge on the mechanism of the drug-drug interactions with linezolid as a victim. It would be interesting to elucidate the drug-drug interaction, preferably by publishing these data followed by a replication of the study to confirm the findings.

Once the exact mechanism is elucidated, the effect of other perpetrator drugs using the same mechanism should be studied. If, for instance, the mechanism turns out to be CYP3A4 mediated, it would be interesting to quantify the effect of other CYP3A4 modulators such as grapefruit juice, carbamazepine or ketoconazole.

Another alluring sequel to the performed prospective PK drug-drug interaction study that we performed would be to quantify the effect of clarithromycin in a higher daily dose than the previously administered 500 mg of clarithromycin, such as a dose of 1000 mg extended release oral formulation once daily. This dose of 1000 mg clarithromycin daily is often administered for several indications without important adverse effects and this dosage is generally well tolerated. The concentration-enhancing effect on linezolid would possibly be more pronounced after administration of 1000 mg clarithromycin than what we observed after administration of 500 mg clarithromycin (2).

The in vitro synergy between linezolid and clarithromycin that we described in chapter 6 of this thesis should be followed by studies of other combinations of second line anti-tuberculosis drugs. Investigating in vitro synergy could help in optimizing existing treatment regimens. Since the number of anti-TB drugs that emerges from the pipeline

117


7

is limited, it is critically important to protect these new drugs from acquiring resistance. Strategies to design and optimize new treatment regimens containing these drugs are warranted.

Elucidating the exact mechanism underlying the observed synergy would contribute to our knowledge of the disease and the mechanism of action of the drugs used. Furthermore, this would possibly answer the question whether the observed synergy is relevant in humans in clinical practice or is only limited to an in vitro setting. Detailed knowledge on synergy, but also on inhibition, could alter the way treatment regimens are designed.

Generic linezolid: cheap linezolid for everyone?

In May 2015, the patent of linezolid will expire. As a result, cheaper generic versions of linezolid will become available within the following years. As a matter of fact, in India – where the patents of linezolid are not recognized by the authorities – cheaper versions of linezolid have been available for years (19). The lowered price of linezolid will make the drug affordable for a larger group of patients, including patients in low- and middle-income countries. The affordability will possibly lead to an increased incorporation of linezolid in treatment regimens. This, in turn, might result in increased antimycobacterial pressure and the likelihood of emerging resistance increases (20). Clinicians should ensure addition of linezolid to adequate treatment regimens to protect this effective anti-tuberculosis drug. Attention should be paid to adherence, preferably administering linezolid under directly observed therapy (DOT) (21). Therapeutic drug monitoring should be applied to identify subtherapeutic drug exposure of linezolid, but also to identify subtherapeutic exposure of other drugs of the treatment regimen leaving linezolid as monotherapy.

The long treatment duration is one of the great obstacles in treating MDR-TB. The impact on patients of being on treatment for at least 18 months is not to be underestimated. Possibly, the duration of treatment might also be one of the reasons of non-adherence or cessation of treatment. Shortening the treatment duration would have an enormous impact on the treatment of MDR-TB. Shortened treatment duration might not only increase the chances of treatment completion in an adequate manner, but it might also save the tuberculosis programs costs per treated patient. Recently, the shortened 9-months Bangladesh treatment regimen showed promising outcome in over 500 MDR-TB patients (22). In contrast with clofazimin (23) and pretomanid, previously known as PA-284 (24),

Chapter 7

118

there is no clear evidence suggesting linezolid is effective in killing slowly replicating persistent microorganisms (25). However, the effective linezolid might play a future role in a comparable shortened treatment regimen: not as a killer of persisters but as an effective anti-MDR-TB drug in the intensive phase. When the saved resources would be used for research in the field of developing new anti-TB drugs, shortening the treatment duration might indirectly even help in reaching the WHO goal of eradicating TB by the year 2050.

However, perhaps in the following years the role of structure analogues of linezolid in treating MDR-TB will become greater than the role of linezolid. Preliminary results in a murine model of one of these analogues, suggest that sutezolid (PNU-100480) may have the potential to shorten TB regimens in both normal sensitive and MDR-TB (26). Another advantage of linezolid structure analogues might be that they display less toxicity than linezolid (10, 27). It will be interesting to study to what extent structure analogues will display less toxicity than linezolid in clinical practice and if findings from this thesis on linezolid, such as the PK drug-drug interaction and in vitro synergy with clarithromycin, will also be observed in the newer oxazolidinones.

Conclusion

In the General Introduction of this thesis, we referred to famous African musicians that fell victim to TB. Hopefully, the research in this thesis combined with the studies proposed in the future perspectives part of the General Discussion, will further contribute to the quality of treatment of MDR-TB. Perhaps, the research will contribute to improved treatment in both low- and high-income countries. It might even prevent a modern day African musician from falling victim to TB.

119


7

REFERENCES


2. Bolhuis, M. S., R. V. Altena, D. V. Soolingen, W. C. Lange, D. R. Uges, T. S. Werf, J. G. Kosterink, and J. W. Alffenaar. 2013. Clarithromycin increases linezolid exposure in multidrug-resistant tuberculosis patients. Eur. Respir. J. 42: 1614-1621.

3. Storla, D. G., S. Yimer, and G. A. Bjune. 2008. A systematic review of delay in the diagnosis and treatment of tuberculosis. BMC Public Health. 8:15-2458-8-15.


5. Cox, E., and K. Laessig. 2014. FDA approval of bedaquiline--the benefit-risk balance for drug-resistant tuberculosis. N. Engl. J. Med. 371:689-691.


7. Cox, E., and K. Laessig. 2014. FDA approval of bedaquiline--the benefit-risk balance for drug-resistant tuberculosis. N. Engl. J. Med. 371:689-691.

8. Laserson, K. F., L. E. Thorpe, V. Leimane, K. Weyer, C. D. Mitnick, V. Riekstina, E. Zarovska, M. L. Rich, H. S. Fraser, E. Alarcon, J. P. Cegielski, M. Grzemska, R. Gupta, and M. Espinal. 2005. Speaking the same language: treatment outcome definitions for multidrug-resistant tuberculosis. Int. J. Tuberc. Lung Dis. 9:640-645.

9. Wallis, R. S., W. Jakubiec, V. Kumar, G. Bedarida, A. Silvia, D. Paige, T. Zhu, M. Mitton-Fry, L. Ladutko, S. Campbell, and P. F. Miller. 2011. Biomarker-assisted dose selection for safety and efficacy in early development of PNU-100480 for tuberculosis. Antimicrob. Agents Chemother. 55:567-574.

10. Wallis, R. S., R. Dawson, S. O. Friedrich, A. Venter, D. Paige, T. Zhu, A. Silvia, J. Gobey, C. Ellery, Y. Zhang, K. Eisenach, P. Miller, and A. H. Diacon. 2014. Mycobactericidal activity of sutezolid (PNU-100480) in sputum (EBA) and blood (WBA) of patients with pulmonary tuberculosis. PLoS One. 9:e94462.

11. Rodriguez, J. C., L. Cebrian, M. Lopez, M. Ruiz, I. Jimenez, and G. Royo. 2004. Mutant prevention concentration: comparison of fluoroquinolones and linezolid with Mycobacterium tuberculosis. J. Antimicrob. Chemother. 53:441-444.

12. Andes, D., M. L. van Ogtrop, J. Peng, and W. A. Craig. 2002. In vivo pharmacodynamics of a new oxazolidinone (linezolid). Antimicrob. Agents Chemother. 46:3484-3489.

13. Gumbo, T., A. Louie, M. R. Deziel, L. M. Parsons, M. Salfinger, and G. L. Drusano. 2004. Selection of a moxifloxacin dose that suppresses drug resistance in Mycobacterium tuberculosis, by use of an in vitro pharmacodynamic infection model and mathematical modeling. J. Infect. Dis. 190:1642-1651.

Chapter 7

120

14. Gumbo, T., A. Louie, M. R. Deziel, W. Liu, L. M. Parsons, M. Salfinger, and G. L. Drusano. 2007. Concentration-dependent Mycobacterium tuberculosis killing and prevention of resistance by rifampin. Antimicrob. Agents Chemother. 51:3781-3788.




18. Gandelman, K., T. Zhu, O. A. Fahmi, P. Glue, K. Lian, R. S. Obach, and B. Damle. 2011. Unexpected effect of rifampin on the pharmacokinetics of linezolid: in silico and in vitro approaches to explain its mechanism. J. Clin. Pharmacol. 51:229-236.


20. Osorio, N. S., F. Rodrigues, S. Gagneux, J. Pedrosa, M. Pinto-Carbo, A. G. Castro, D. Young, I. Comas, and M. Saraiva. 2013. Evidence for diversifying selection in a set of Mycobacterium tuberculosis genes in response to antibiotic- and nonantibiotic-related pressure. Mol. Biol. Evol. 30:1326-1336.

21. Weis, S. E., P. C. Slocum, F. X. Blais, B. King, M. Nunn, G. B. Matney, E. Gomez, and B. H. Foresman. 1994. The effect of directly observed therapy on the rates of drug resistance and relapse in tuberculosis. N. Engl. J. Med. 330:1179-1184.

22. Aung, K. J., A. Van Deun, E. Declercq, M. R. Sarker, P. K. Das, M. A. Hossain, and H. L. Rieder. 2014. Successful ‚9-month Bangladesh regimen‘ for multidrug-resistant tuberculosis among over 500 consecutive patients. Int. J. Tuberc. Lung Dis. 18:1180-1187.

23. Xu, J., Y. Lu, L. Fu, H. Zhu, B. Wang, K. Mdluli, A. M. Upton, H. Jin, M. Zheng, W. Zhao, and P. Li. 2012. In vitro and in vivo activity of clofazimine against Mycobacterium tuberculosis persisters. Int. J. Tuberc. Lung Dis. 16:1119-1125.

24. Tyagi, S., E. Nuermberger, T. Yoshimatsu, K. Williams, I. Rosenthal, N. Lounis, W. Bishai, and J. Grosset. 2005. Bactericidal activity of the nitroimidazopyran PA-824 in a murine model of tuberculosis. Antimicrob. Agents Chemother. 49:2289-2293.

25. Dietze, R., D. J. Hadad, B. McGee, L. P. Molino, E. L. Maciel, C. A. Peloquin, D. F. Johnson, S. M. Debanne, K. Eisenach, W. H. Boom, M. Palaci, and J. L. Johnson. 2008. Early and extended early bactericidal activity of linezolid in pulmonary tuberculosis. Am. J. Respir. Crit. Care Med. 178:1180-1185.

121


7

26. Williams, K. N., C. K. Stover, T. Zhu, R. Tasneen, S. Tyagi, J. H. Grosset, and E. Nuermberger. 2009. Promising antituberculosis activity of the oxazolidinone PNU-100480 relative to that of linezolid in a murine model. Antimicrob. Agents Chemother. 53:1314-1319.


Summary

Summary

124

In this thesis, the general aim was to describe the clinical pharmacology of linezolid in multidrug-resistant tuberculosis (MDR-TB). In order to do so, several aims were formulated to shed light on different pharmacokinetic (PK) and pharmacodynamic (PD) aspects of linezolid in MDR-TB. Since treatment regimens have to be composed of at least four anti-TB drugs that are likely to be effective and since the number of new-drugs that emerge from the pipeline is limited, clinicians are forced to use drugs with limited knowledge about safety and efficacy to treat these patients. One of these drugs, the oxazolidinone linezolid, is used off-label in the treatment of MDR-TB. Linezolid holds promise to be highly effective in treatment regimens of MDR-TB, however linezolid is infamous for its adverse events, such as peripheral and optical neuropathy, anaemia, and myelosuppression. Because of these adverse events clinicians often feel forced to withdraw linezolid from MDR-TB treatment regimens. This is particularly important in patients that are infected with Mycobacterium tuberculosis, a microorganism that is only susceptible for few selected drugs.

To reduce toxicity, dose reduction was studied and obviously, this helped to reduce adverse events. However, dose reduction should never compromise treatment efficacy, since reduced efficacy can result in failing treatment regimens, acquired resistance and even death. To help clinicians manoeuvre through the field of toxicity while retaining efficacy, we aim to contribute to the knowledge of PK and PD of linezolid in MDR-TB with special focus on TDM of linezolid to optimize treatment of patients suffering from MDR-TB.

Pharmacokinetics of linezolid

In order to attain our objectives, in Chapter 2, we performed a literature search on drug-drug and drug-food interactions of anti-microbial drugs, including those used in multidrug-resistant tuberculosis treatment regimens, such as clarithromycin and linezolid. The pharmacokinetic interactions, as described in Chapter 2, are a modified version of a published manuscript with a broader scope, studying other antimicrobial drugs as well. For each drug that was studied, including linezolid (oxazolidionones) and clarithromycin (macrolides), literature was searched using the Medical Subject Headings (MeSH) term ‘drug interaction’ combined with the drug name and group name. The search was confined by different limitations, with the goal of improving the quality of the search results. For oxazolidinones, the search also aimed to reveal drug-interactions of other, not yet commercially available drugs of the oxazolidione group, e.g. sutezolid (PNU-100480),

125

Summary

*

posizolid (AZD2563), radezolid (RX-1741), and torezolid (TR-701). Due to the recent discovery of some of these drugs and due to the limited post-marketing knowledge of linezolid, the number of drug-interactions studies was very limited on linezolid, and even absent on the other oxazolidinones.

The results revealed, that a fat meal decreased linezolid absorption and antacids had no clinical significant effect on absorption of linezolid. The metabolism of linezolid appeared not to be influence by CYP modulators, however literature suggests that linezolid might be a P-glycoprotein (P-gp) substrate.

From this literature study, we concluded that there are drug-drug and drug-food PK interactions with linezolid, with variable relevance. The change in PK parameters of linezolid might influence therapy through increased toxicity or decreased efficacy. Finally, the sheer amount of pharmacokinetic drug interactions, with linezolid, but also other anti-mycobacterial drugs, require multidisciplinary vigilance of physicians, pharmacists, and other health care professionals.

Chapter 3 focuses on one of these pharmacokinetic drug-drug interactions. As a part of routine therapeutic drug monitoring (TDM) at the Tuberculosis Center Beatrixoord (Haren, the Netherlands), we discovered a drug-drug interaction between linezolid and clarithromycin in a 42-year-old male TB-patient. In this patient, the area under the time concentration curve (AUC0–12h) of linezolid was increased from 29 mg*h/L to 108 mg*h/L. Based on co-administrated medication, lack of significant changes in his liver or renal function, and coincidence with the start of co-administration of clarithromycin 1000 mg once daily, we hypothesised that there might be an interaction between linezolid and the potent P-gp pump inhibitor clarithromycin.

Based on this observation, a prospective, single-arm, fixed-order study was designed to quantify the interaction. Male and female patients aged ≥18 years who were diagnosed with MDR-TB were eligible for inclusion in our study. Pregnant or lactating patients, subjects that had previously shown hypersensitivity to linezolid or macrolide antibiotics, patients that had hypokalaemia or concomitantly received P-glycoprotein modulators were excluded. All patients received linezolid 300 mg twice a day. The AUC was determined at three time points in the study: at baseline, after two weeks of co-administration of linezolid with 250 mg clarithromycin once daily, and finally after two weeks of co-administration of linezolid with 500 mg clarithromycin.

Summary

126

In concordance with the calculated sample size, five of the seven included patients were suitable for evaluation. The median (interquartile range, IQR) of linezolid AUC was 36.3 (33.2 – 46.3) mg*h/L without co-administration of clarithromycin. After co-administration with 250 mg clarithromycin once daily, the median (IQR) linezolid AUC increased to 61.0 (34.6 – 63.9) mg*h/L (p=0.686) and after 500 mg clarithromycin to 67.2 (66.9 – 76.0) mg*h/L (p=0.043). In other words, linezolid exposure increased by a median (IQR) of 44% (23 – 102%) after co-administration with 500 mg clarithromycin compared to baseline, whereas 250 mg clarithromycin had no statistically significant effect. As for safety and tolerability, co-administration was well tolerated by most patients; none experienced severe adverse effects, one patient reported Common Toxicity Criteria grade 2 gastrointestinal adverse events. Based on this study, we concluded that further research in a larger cohort is needed to provide insight into the observed inter-patient variation, perhaps caused by genetic variability polymorphism. Furthermore, we concluded that TDM of linezolid is advisable when linezolid is co-administrated with clarithromycin until effect size of the PK interaction becomes predictable.

Despite the advisable nature of TDM when administering linezolid with or without clarithromycin, in many parts of the world conventional sampling might be hindered by logistical problems. In order to facilitate easy analysis of the anti-TB drugs linezolid, we developed and clinically validated two new sampling methods using dried blood spot (DBS) and oral fluid analysis (Chapter 4).

In order to clinically validate the DBS sampling, we recruited patients that were admitted to Tuberculosis Center Beatrixoord (Haren, the Netherlands) that were receiving linezolid as part of their treatment regimen for MDR-TB for whom routine TDM was scheduled. Sampling was performed at steady-state. Venous blood samples were obtained before drug intake, and at 1, 2, 3, 4, and 8 hours after drug intake. The venous blood samples were used as a control, but also to generate venous dried blood spots (vDBS) by dropping 50 μL onto Whatman 31 ET CHR paper. DBS samples were obtained at 2 and 8 hours after dosing by dropping blood from a finger prick, directly on DBS paper. Both the vDBS and DBS samples were dried at room temperature and stored in sealed plastic bags with desiccant sachets at -20°C until analysis. Just before analysis, an 8-mm-diameter disc was punched out of each (v)DBS after which extraction of these discs was performed. After further sample preparation, the samples were analysed using a validated liquid chromatography-tandem mass spectrometric (LC-MS/MS) method. The DBS analytical method was validated in accordance with the U.S. Food and Drug Administration guidance. Besides these common

127

Summary

*

validation tests, special tests were performed to evaluate the impact of haematocrit value (Hct) at 20, 25, 30, 35, 40, 45, and 50%, and blood spot volume set at 30, 50, 70, and 90 μL. Blood spot volume and Hct values were set at 50 μL and 35% during the validation of the analysis to represent tuberculosis patients.

The DBS assay showed linearity over the analytical concentration range (0.05 – 30 mg/L linezolid) with a pooled correlation coefficient, r2, of 0.9947. Reproducibility, presented as bias and coefficient of variation (CV) were less than 17.2% for the lower limit of quantification (0.05 mg/L linezolid), and less than 7.8% for other levels (0.25 – 30 mg/L linezolid). The method showed a high recovery of approximately 95% and a low matrix effect less than 8.7%. DBS samples proved to be stable for 2 months at 37°C. Bland-Altman analysis revealed that the ratio of the linezolid concentration in DBS to that in plasma was 1.2 (95% confidence interval [CI], 1.12 – 1.27).

From this study, we concluded that DBS analysis of linezolid might be a useful tool for TDM of linezolid in the treatment of MDR-TB patients. The easy sampling procedure and high sample stability may be useful, especially in underdeveloped countries with limited resources and where conventional plasma sampling is not feasible.

In Chapter 4, we also developed and clinically validated an oral fluid sampling method. Patients that took part in the prospective study described in chapter 3 were included. All patients received 300 mg linezolid twice daily and 250 mg clarithromycin once daily when the steady state blood and oral fluid samples were obtained. The samples were collected simultaneously before and at 1, 2, 3, 4, 8, and 12 h after drug intake. Oral fluid samples were collected using a small cotton roll (Salivette; Sarstedt, Leicester, United Kingdom). Patients chewed on the Salivette for approximately two minutes after which the samples were centrifuged and then stored at -20°C until analysis. Linezolid and clarithromycin concentrations were analysed using validated LC-MS/MS analysis methods.

Passing-Bablok regressions and Bland-Altman analysis revealed oral fluid analysis of linezolid to be suitable for TDM in MDR-TB patients as an alternative for conventional serum analysis. Since the ratio of the linezolid concentration in serum to that in oral fluid was calculated to be 0.97 (95% CI, 0.92 to 1.02) using Bland-Altman analysis, no correction factor is needed for the interpretation of linezolid oral fluid concentrations. However, the Bland-Altman analysis of clarithromycin concentration serum / oral fluid revealed a ratio of 3.07 (95% CI, 2.45 to 3.69). We observed a non-linear relationship between the concentrations of the metabolite of clarithromycin, hydroxyclarithromycin, in serum and in oral fluid. Based on

Summary

128

this study, we concluded that the analysis of linezolid and clarithromycin in oral fluid might be an easy, non-invasive alternative in MDR-TB patients when conventional sampling might not be possible or desirable.

Previous chapters, but also studies on inter- and intra-patient variability of linezolid pharmacokinetics performed by other research groups, indicate that there might be an important role for TDM. Nevertheless, several recent studies did not incorporate TDM data in their research. Therefore, we advocated the use of TDM combined with drug susceptibility testing, among other things, in two letters to the editor (Chapter 5). Also, we pointed out the importance of directly observed therapy (DOT) in clinical trials, except perhaps in cases where compliance is highly probable. Furthermore, we suggested that a randomised controlled trial of linezolid versus a placebo should be designed and planned, in addition to an adequate background regimen that is designed using DST and TDM.

The suggestions that we made based on these recent studies, encouraged us to retrospectively study linezolid efficacy, safety, and tolerability in relation to linezolid pharmacokinetics in our Tuberculosis Center Beatrixoord (Haren, the Netherlands). In order to increase the sample size, an international collaboration was initiated allowing for inclusion of multidrug-resistant tuberculosis patients from the E. Morelli Hospital (Sondalo, Italy) as well. We selected MDR-TB patients that received linezolid as a part of their treatment regimen, excluding patients younger than 18 years, patients lacking data due to recent admission to either reference hospital, and patients of whom no TDM data were available. Two researchers retrieved anonymous retrospective data, including patient characteristics, treatment information, pharmacokinetic data, and treatment outcome.

In both centers, a combined number of 58 MDR/XDR-TB patients that had received linezolid and underwent TDM were included. However, of 54 eligible patients in Sondalo, 36 had to be excluded, of whom 27 due to absence of pharmacokinetic data of linezolid. In the combined cohort, DST revealed a median (IQR) MIC for linezolid of 0.5 (0.25 – 0.5) mg/L. Of the 58 included patients, 29/58 (50.0%) patients were considered cured; 4/58 (6.9%) patients completed their treatment; and 25/58 (43.1%) patients were still on treatment at the moment of data collection. All but one patient (98.3%), including those still on treatment, were culture and smear-microscopy negative when data were collected. In this combined cohort, no correlation was observed between microscopy or culture conversion and the pharmacokinetic/pharmacodynamic parameters, the AUC/MIC ratio, which were above the proposed target in almost all cases.

129

Summary

*

Retrospective analysis pointed out that patients with peripheral neuropathy had received a higher median (IQR) cumulative dose (1,829 (1,414 – 2,255) mg/kg versus 1,164 (755 – 1,922) mg/kg (p=0.041)) or received linezolid for a longer median (IQR) period of time (159 (120 – 196) versus 97 (66 – 147) days (p=0.003)) compared to patients without peripheral neuropathy. Another interesting finding from this study was that co-administration of erythropoietin with linezolid had no effect on preventing anaemia. At 0, 30, 60, and 90 days after starting linezolid, there was no difference between haemoglobin (Hb) of patients that did or did not receive erythropoietin, nor was there a difference in Hb over time per patient between the two groups.

From our retrospective study, we concluded that the treatment regimens containing linezolid were effective and well tolerated. High sputum microscopy and culture conversion rates and low frequency of adverse events were observed in our cohort. Furthermore, we concluded that peripheral neuropathy seemed to be mediated by cumulative dose and days of exposure to linezolid. We were not able to observe a relevant correlation between pharmacokinetic/pharmacodynamic parameters, such as AUC/MIC ratio, and efficacy parameters. Perhaps this is due to the fact that in our cohort almost every patient had an AUC/MIC ratio of >100.

Pharmacodynamics of linezolid

Although in this thesis most emphasis is placed on pharmacokinetic of linezolid, in Chapter 6 we focused on a pharmacodynamic interaction between linezolid and clarithromycin M. tuberculosis isolates. In collaboration with the Dutch National Mycobacteria Reference Laboratory (Bilthoven, the Netherlands), we performed an in vitro synergy test using the checkerboard method. A panel of 24 M. tuberculosis isolates were randomly selected from their collection. We used two previously described methods, i.e. the absolute concentration method (ACM) and the MGIT 960 system, with varying concentrations of drugs. Linezolid was added in concentrations between 0 – 0.5 mg/L and clarithromycin with a range of 0 – 8 mg/L in a checkerboard fashion. After growth inhibition was determined for all combinations of isolates and concentrations, we calculated the lowest fractional inhibitory concentration (FIC) to determine synergy. The FIC was calculated as: (MIC of linezolid in combination / MIC of linezolid alone) + (MIC of clarithromycin in combination / MIC of clarithromycin alone), and defined synergy as an FIC ≤0.5, indifference as 0.5<FIC≤4, and antagonism as FIC >4.

Summary

130

Of the 24 selected M. tuberculosis isolates, synergy between clarithromycin and linezolid was determined for 74% by using the MGIT method and in 59% by using the ACM. No antagonism was observed. Since the combination is most likely to be co-administered in MDR-TB patients, we also focused on MDR-TB isolates. In these MDR-TB isolates (n=13), synergy was observed in 77% using MGIT and in 46% using the ACM method with a median (IQR) FIC of 0.37 (0.32 – 0.37) and 0.62 (0.375 – 1.0) respectively. Based on these data, we concluded that clarithromycin and linezolid display in vitro synergy in multidrug-resistant M. tuberculosis isolates. The in vitro synergy that was observed in this study might play a role in designing future MDR-TB treatment.

In Chapter 7 we discussed whether the objectives of this thesis, as presented in the general introduction, were met. Furthermore, we placed the most important findings from this thesis in perspective of published literature and clinical practice. For instance, we explained the importance of pharmacokinetic drug-drug interaction studies of anti-TB drug as a victim, but also as a perpetrator. Moreover, we explained the importance of individualized treatment regimens and defined targets for therapeutic drug monitoring. Finally, we discussed future perspectives following from the results of this thesis. These future perspectives include the possible impact of the patent of linezolid expiring in the following years and subjects that warrant future study based on the studies that were performed for this thesis.

Nederlandse samenvatting


132

Tuberculose

Wereldwijd overlijden jaarlijks meer dan één miljoen mensen aan de infectieziekte tubercu-lose. Tuberculose wordt veroorzaakt door tuberkelbacillen die officieel worden aangeduid als Mycobacterium tuberculosis. Tuberculose wordt meestal overgedragen van mens tot mens via het inademen van tuberkelbacillen, die in wolken van fijne druppels – aërosolen – via de lucht worden verspreid door mensen met besmettelijke tuberculose. Door hoesten en niezen kunnen anderen besmet worden, vooral wanneer deze mensen langdurig met de besmettelijke tuberculosepatiënt in dezelfde ruimte verblijven, en als die ruimte weinig geventileerd wordt. Na het inademen van deze druppels met bacteriën zorgt het afweersysteem meestal dat de bacteriën opgeruimd worden. Bij ongeveer 10% van de mensen met een normale afweer is het afweersysteem niet in staat de bacterie voldoende te bestrijden. In deze gevallen nestelt de bacterie zich in het lichaam. Meestal ontstaat er vervolgens een latente, oftewel slapende, tuberculose-infectie waarbij de patiënt geen klachten en ziekteverschijnselen ontwikkelt. Als het afweersysteem te zwak is kan tuberculose ontstaan doordat de tuberkelbacil ‘wakker’ wordt en weer gaat delen. In de meeste gevallen gebeurt dit binnen enkele maanden of jaren na de besmetting, soms later. De besmette persoon krijgt dan klachten, meestal in de vorm van hoesten, nachtzweten en ongewild gewichtsverlies. De tuberkelbacil komt vrijwel altijd via de luchtwegen het lichaam binnen en veroorzaakt daar de meeste ziekteverschijnselen. De bacterie kan ook op andere plaatsen in het lichaam opduiken en daar bijna alle organen aantasten. Voorbeelden hiervan zijn de longvliezen, de lymfeklieren, de darmen, het buikvlies, de nieren, de geslachtsorganen, de lever, de hersenvliezen, de hersenen en de botten, vooral die van de wervelkolom.

Multiresistente tuberculose

Bij iedere celdeling van de tuberkelbacil kan door toeval een fout, oftewel een mutatie, ontstaan. Af en toe ontstaat hierbij een mutant, oftewel een gemuteerde bacterie, die ongevoelig is voor een bepaald tuberculosemedicament. Het aantal bacteriën bij patiënten met tuberculose is enorm groot. Tussen al die bacteriën zijn er altijd enkele mutanten die tegen een of ander antibioticum resistent zijn. Deze mutanten zouden overleven wanneer de behandeling alleen met dit ene antibioticum, waartegen ze resistent zijn, zou plaatsvinden. Bij opeenvolgende fouten in de behandeling waarbij telkens maar één antibioticum wordt gebruikt, kan een patiënt uiteindelijk een infectie hebben met bacteriën die resistent zijn tegen meerdere middelen tegelijk. Om dit te voorkomen moet bij de behandeling van tuberculose altijd een combinatie van verschillende antibiotica gebruikt worden.

133


*

Er kan een situatie ontstaan waarbij de bacterie niet meer gevoelig is voor de standaard-behandeling met isoniazide, rifampicine, pyrazinamide en ethambutol, waarmee wereldwijd veel ervaring is opgedaan. Als de bacterie ongevoelig is voor de twee meest gebruikte en krachtigste middelen, namelijk rifampicine en isoniazide, spreekt men van multiresistente tuberculose (MDR-TB). Helaas komt MDR-TB vaak voor. Op basis van gegevens van de Wereldgezondheidsorganisatie (WHO) denken we dat 3,6 – 20% van de nieuwe tuberculosepatiënten besmet is met MDR-TB.

Wanneer een patiënt besmet is met MDR-TB is de behandeling veel complexer dan die bij normaal gevoelige tuberculose, aangezien in het geval van MDR-TB de meest efficiënte middelen per definitie niet kunnen worden ingezet als onderdeel van de behandeling. Omdat deze middelen, rifampicine en isoniazide, niet gebruikt kunnen worden voor de behandeling van de tuberculose dient er uitgeweken te worden naar reserve-antibiotica. Volgens de behandelrichtlijn van de WHO behoort de medicamenteuze behandeling van MDR-TB te bestaan uit een combinatie van tenminste vier antibiotica waarvoor de bacterie hoogstwaarschijnlijk nog gevoelig is. Artsen hebben hierbij beschikking over vijf verschillende groepen reserve-antibiotica op basis van een indeling van de WHO. In de vijfde groep zitten de antibiotica linezolid en claritromycine. Op dit moment is er echter relatief weinig ervaring met de behandeling van MDR-TB met deze middelen en is er nog weinig bekend over de effectiviteit bij tuberculose. Hierdoor dient de patiënt geen zes maanden, maar soms tot 24 maanden te worden behandeld en is het onzeker of de patiënt überhaupt kan genezen. Wanneer de kennis over de effectiviteit, bijwerkingen en veiligheid van linezolid en claritromycine bij de behandeling van MDR-TB zou toenemen, zouden deze middelen mogelijk een grotere rol kunnen krijgen bij de behandeling.

Farmacologie

Om geneesmiddelen effectief in te kunnen zetten bij een medicamenteuze behandeling is kennis van de farmacologie van de toegepaste geneesmiddelen nodig. Farmacologie is de wetenschap die zich bezighoudt met de wisselwerking van het werkzame bestanddeel van een geneesmiddel met het aangrijpingspunt waar het middel moet werken. Bij infectieziekten zoals tuberculose dient het geneesmiddel aan te grijpen op levende tuberkelbacillen. De farmacologie bestaat uit twee onderdelen: farmacokinetiek en farmacodynamie. Farmacokinetiek beschrijft ruwweg wat het lichaam met het geneesmiddel doet en farmacodynamie hoe het geneesmiddel aangrijpt op het ziekteproces in het lichaam. Bij infectieziekten is dat vooral hoe het middel de ziektekiem aanpakt.


134

Het begrip farmacokinetiek beschrijft de reis van een geneesmiddel door het lichaam. Die reis begint bij opname of absorptie uit het maag-darmkanaal. Na inslikken moet het middel uit het darmkanaal opgenomen worden in het bloed. Vervolgens wordt het middel over het lichaam verdeeld (distributie), omgezet (metabolisme) en uitgescheiden (eliminatie). In combinatie met de dosering van het geneesmiddel hebben deze processen een concentratie van het geneesmiddel in het lichaam als gevolg. In algemene zin kan gezegd worden dat een te hoge concentratie in relatie tot farmacodynamische parameters giftige bijwerkingen kan veroorzaken. Daartegenover staat dat bij een te lage concentratie het geneesmiddel onwerkzaam is. Bij antibiotica kan een te lage concentratie de bacteriën met verminderde gevoeligheid selecteren waardoor uiteindelijk zelfs resistentie kan ontstaan.

Onderzoek naar farmacokinetiek richt zich vaak op het meten van de concentratie van een geneesmiddel in het lichaam. Het liefst zouden we de concentratie meten op de plaats waar het geneesmiddel moet werken, maar dat lukt eigenlijk alleen als we een weefselmonster van een aangedaan lichaamsdeel zouden verwijderen voor onderzoek. We gaan er van uit dat het geneesmiddel via de bloedbaan bij de infectiehaard moet komen en dat in ontstoken weefsel de bloedvoorziening voldoende is, zodat de bloedconcentratie overeenkomt met de weefselconcentratie van het geneesmiddel. In de farmacokinetiek worden meestal bloedmonsters gemeten. Naast bloedmonsters kunnen echter ook urine- en speekselmonsters gebruikt worden of een gedroogde bloeddruppel uit een vingerprik op een papier, de zogeheten dried blood spot (DBS). Farmacodynamisch onderzoek bij tuberculose bestaat veelal uit laboratoriumonderzoek naar de gevoeligheid van bacteriën voor specifieke geneesmiddelen.

Beide bovenstaande begrippen kunnen gecombineerd worden: farmacokinetiek / farmaco-dynamie (PK/PD). PK/PD relateert de gemeten concentratie van het geneesmiddel aan de benodigde concentratie, op basis van de gevoeligheid van de bacterie, op de plek van de infectiehaard. Het doseren op basis van PK/PD noemt men therapeutic drug monitoring (TDM). Met behulp van TDM zou onderzocht kunnen worden in hoeverre de concentratie van of blootstelling aan linezolid aansluit bij de gevoeligheid van de bacterie. Als de concentratie of blootstelling te hoog is, kunnen vermijdbare bijwerkingen of vergiftigingsverschijnselen optreden. De dosering van het geneesmiddel moet dan verlaagd worden. Wanneer de concentratie of blootstelling te laag is, zou de dosering van het geneesmiddel juist moeten worden verhoogd om onderbehandeling en resistentie te voorkomen.

135


*

Linezolid

Linezolid is een oxazolidinone-antibioticum dat geregistreerd is voor de behandeling van infecties van de huid en de longen met resistente stafylokokken. Linezolid wordt ook ingezet tegen MDR-TB, alhoewel het daarvoor niet geregistreerd is. Er zijn aanwijzingen dat linezolid een erg effectief geneesmiddel is in de behandeling van MDR-TB. Een nadeel is echter dat linezolid bekend staat om zijn bijwerkingen. De belangrijkste bijwerkingen zijn beschadiging van de oogzenuw, doofheid en prikkeling in de benen, voeten en handen door zenuwbeschadiging (perifere neuropathie), bloedarmoede en een onderdrukte bloedaanmaak in het beenmerg. Door deze bijwerkingen zien artsen zich vaak gedwongen te stoppen met toediening van linezolid als onderdeel van de tuberculosebehandeling. Voor de patiënt kan het moeten stoppen met linezolid zeer nadelige gevolgen hebben aangezien de multiresistente bacterie vaak slechts voor een paar geneesmiddelen gevoelig is.

Uit onderzoek is gebleken dat het verlagen van de dosering van linezolid het aantal bijwer-kingen kan verminderen. Een gevaar van het verlagen van de linezoliddosering is echter dat dit de effectiviteit van linezolid in het gedrang kan brengen. Dit zou het falen van de behandeling, ontwikkelen van ongevoeligheid van de bacterie voor linezolid en zelfs overlijden van de patiënt tot gevolg kunnen hebben.

Het doel van dit proefschrift is om zorgverleners te ondersteunen bij het manoeuvreren tussen enerzijds voldoende werkzaamheid en anderzijds bijwerkingen van linezolid door de klinische farmacologie van linezolid te onderzoeken wanneer het wordt ingezet bij de behandeling van MDR-TB. Hierbij wordt specifiek aandacht besteed aan de mogelijke rol van TDM bij het optimaliseren van de behandeling met linezolid.

Een ander belangrijk probleem is dat linezolid nooit alleen, maar altijd in combinatie met andere geneesmiddelen moet worden gegeven, terwijl de patiënt misschien ook nog andere geneesmiddelen krijgt. Al die geneesmiddelen kunnen met elkaar botsen: het ene middel kan de concentratie van het andere beïnvloeden, bijvoorbeeld doordat beide middelen op dezelfde manier in de lever worden omgezet of doordat de uitscheiding wordt beïnvloed. In dit proefschrift is daarom een literatuuroverzicht van de verschillende farmacokinetische interacties tussen geneesmiddelen opgenomen. Een interactie tussen linezolid en claritromycine, die beide worden ingezet bij de behandeling van MDR-TB, hebben we daarom speciaal bestudeerd. Hiernaast hebben we een tweetal nieuwe methodes opgezet om linezolid te meten in speeksel en DBS. We hebben de meetmethode van speekselmonsters en de toepasbaarheid hiervan in de praktijk onderzocht. Ook hebben we de linezolidblootstelling


136

en bloedconcentratie onderzocht in eerder afgenomen monsters, waarbij we vooral aandacht hebben besteed aan de effectiviteit, veiligheid en verdraagbaarheid bij MDR-TB patiënten. Tenslotte is een van de doelen om in het laboratorium te onderzoeken in hoeverre er een farmacodynamische beïnvloeding is tussen linezolid en claritromycine in MDR-TB bacteriën. Door middel van bovenstaande doelen beogen we meer inzicht te krijgen in de farmacologie van linezolid bij de behandeling van MDR-TB, met bijzondere aandacht voor TDM.

Resultaten

In hoofdstuk 2 beschrijven we een literatuuronderzoek naar interacties van enkele antibiotica met andere geneesmiddelen of voedsel. In het literatuuronderzoek is niet alleen gezocht naar interacties met het antibioticum linezolid, maar ook naar vele andere verschillende antibiotica. Voor een betere leesbaarheid hebben we ervoor gekozen om in hoofdstuk 2 alleen de resultaten van het onderzoek naar de oxazolidinone-antibiotica, inclusief linezolid, weer te geven. Hoofdstuk 2 is daarom een voor het onderwerp van dit proefschrift aangepaste versie van het oorspronkelijke onderzoek.

We hebben de literatuur op een systematische manier doorzocht op alle geneesmiddelen die binnen het bereik van het onderzoek vielen. We gebruikten de zoekterm ‘drug interaction’ (geneesmiddelinteractie) uit het gestructureerde trefwoordensysteem, de Medical Subject Headings (MeSH) database. Deze zoekterm combineerden we met de generieke geneesmid-delnaam en de naam van de geneesmiddelgroep. Om het aantal resultaten te beperken en de kwaliteit van de resultaten te verhogen, hebben we vervolgens enkele filters toegepast. Zo werden bijvoorbeeld alleen Engelstalige wetenschappelijke artikelen geselecteerd voor het literatuuronderzoek. Uit de geneesmiddelgroep oxazolidinonen hebben we op de generieke geneesmiddelnaam linezolid en op de nog niet commercieel verkrijgbare geneesmiddelen uit de zelfde groep: sutezolid (PNU-100480), posizolid (AZD2563), radezolid (RX-1741) en torezolid (TR-701) gezocht.

Het aantal onderzoeken dat geneesmiddelinteracties beschreef met linezolid of een van de andere oxazolidinone-antibiotica was beperkt. Mogelijk speelt het feit dat enkele van deze geneesmiddelen pas recent ontdekt zijn en dat er slechts beperkte ervaringsgegevens bij patiënten bekend zijn hierbij een rol. Een vette maaltijd bleek de opname van linezolid vanuit de darm te verminderen. De opname van linezolid vanuit de darm werd niet of nauwelijks verstoord door het gebruik van maagzuurremmers.

137


*

Er zijn interacties van linezolid met enkele geneesmiddelen bekend die de concentratie van linezolid in het lichaam kunnen beïnvloeden. Er worden verschillende mechanismen geopperd om die interacties te verklaren. Zo zouden de geneesmiddelen de omzetting van linezolid beïnvloeden door verandering van het belangrijke ontgiftingssysteem in de lever, het zogenaamde cytochroom P450 enzymsysteem. Andere auteurs denken dat de waargenomen geneesmiddelinteracties veroorzaakt zouden kunnen worden door de beïnvloeding van de effluxpomp P-glycoproteïne.

Geconcludeerd kan dus worden dat er enkele interacties zijn tussen het antibioticum linezolid en andere geneesmiddelen, maar ook met voedsel. In de praktijk zijn deze interacties niet allemaal even belangrijk. De veranderde farmacokinetische parameters zouden de behan-deling van MDR-TB direct kunnen beïnvloeden door middel van toxiciteit of verminderde effectiviteit van het antibioticum, maar ook indirect door beïnvloeding van middelen voor de behandeling van andere kwalen dan tuberculose. Tenslotte laat het grote aantal geneesmiddelinteracties nog eens zien aan artsen, apothekers en andere zorgverleners hoe belangrijk die geneesmiddelinteracties zijn voor de behandeling van patiënten.

In hoofdstuk 3 beschrijven we een nieuwe geneesmiddelinteractie van linezolid. Het meten van geneesmiddelconcentraties ten behoeve van TDM is een standaard onderdeel van de zorg die we verlenen aan MDR-TB patiënten binnen het Universitair Medisch Centrum Groningen, Centrum voor Revalidatie / Tuberculose Centrum Beatrixoord (Haren, Nederland; hierna: Tuberculose Centrum Beatrixoord). Tijdens een dergelijke routine-analyse bij een 42-jarige mannelijke tuberculosepatiënt ontdekten we bij toeval een genees-middelinteractie tussen linezolid en claritromycine. Bij deze patiënt bleek de blootstelling aan linezolid, weergegeven als de oppervlakte onder de concentratie-tijd curve (AUC) per 12 uur, verhoogd van 29 mg*h/L naar 108 mg*h/L. We konden het verschil niet verklaren met behulp van bekende geneesmiddelinteracties. Ook boden leverproeven en nierfunctietesten geen aanknopingspunt om de verhoogde blootstelling te kunnen verklaren. Het ontstaan van de verhoogde blootstelling aan linezolid viel samen met het toevoegen van claritromycine in een dosering van eenmaal daags 1000 mg aan de behandeling. Op basis van het bovenstaande ontwikkelden we de hypothese dat er mogelijk een interactie zou kunnen zijn tussen linezolid en claritromycine.

Dit heeft ons doen besluiten een open-label, prospectief, één arm, vaste volgorde onderzoek uit te voeren. Het doel van dit onderzoek was om een eventuele farmacokinetische interactie tussen linezolid en claritromycine in maat en getal weer te geven. Zowel volwassen mannen


138

als vrouwen, bij wie MDR-TB was vastgesteld, kwamen in aanmerking voor deelname aan dit onderzoek. Zwangeren, patiëntes die borstvoeding gaven, patiënten met eerder aangetoonde overgevoeligheid voor linezolid of een van de geneesmiddelen uit de claritromycine geneesmiddelgroep, patiënten met een kaliumtekort en patiënten die tegelijkertijd een geneesmiddel toegediend kregen dat de werking van het P-glycoproteïne beïnvloedde, kwamen niet in aanmerking voor deelname aan het onderzoek. Alle deelnemende patiënten kregen gedurende het gehele onderzoek tweemaal daags 300 mg linezolid toegediend. Bij alle patiënten werd de blootstelling gemeten op drie momenten gedurende het onderzoek: 1. wanneer de patiënt alleen linezolid kreeg zonder claritromycine, 2. nadat dezelfde patiënt gedurende twee weken linezolid had ontvangen met eenmaal daags 250 mg claritromycine en 3. nadat deze gedurende twee weken linezolid had ontvangen met eenmaal daags 500 mg claritromycine.

Er hebben zeven patiënten deelgenomen aan het onderzoek en van vijf van deze patiënten konden we de gegevens gebruiken voor de uiteindelijke analyse. Van tevoren hadden we berekend dat dit aantal patiënten voldoende groot was om tot een duidelijk antwoord te komen.

De mediane blootstelling aan linezolid was 36,3 (33,2 – 46,3) mg*h/L zonder gelijktijdige toediening met claritromycine. Na gelijktijdige toediening met eenmaal daags 250 mg claritromycine bleek de mediane (interkwartielbereik, IQR) blootstelling aan linezolid verhoogd tot 61,0 (34,6 – 63,9) mg*h/L (p=0,686) en na gelijktijdige toediening met 500 mg claritromycine tot 67,2 (66,9 – 76,0) mg*h/L (p=0,043). Met andere woorden: na toevoegen van eenmaal daags 500 mg claritromycine is de blootstelling aan linezolid met een mediaan (IQR) van 44% (23 – 102%) verhoogd ten opzichte van toediening zonder claritromycine. Toediening van 250 mg claritromycine had geen statistisch significant effect op de blootstelling van linezolid.

Het gelijktijdig toedienen van claritromycine en linezolid kon door de meeste patiënten goed verdragen worden. Tijdens het onderzoek zijn geen ernstige bijwerkingen opgetreden. Bij één van de patiënten is een milde tot matig-ernstige (Groep 2, Common Toxicity Criteria, CTC) maag-darmbijwerking opgetreden. Op grond van de resultaten uit dit onderzoek concluderen we dat verder onderzoek in een grotere groep van patiënten wenselijk is. Er is vooral behoefte aan inzicht in de oorzaak van het verschil in effectmaat van de waargenomen farmacokinetische interactie tussen de individuele patiënten. Mogelijk was een van de oorzaken verschil in erfelijke aanleg. Tenslotte concluderen we dat TDM van linezolid aan

139


*

te raden is indien linezolid tegelijkertijd wordt toegediend met claritromycine totdat het effect van de geobserveerde interactie beter te voorspellen is.

Ondanks het feit dat TDM van linezolid in sommige gevallen aangeraden wordt, is het op klassieke wijze afnemen van verschillende bloedmonsters in de tijd in grote delen van de wereld niet goed uitvoerbaar. Om het afnemen van monsters voor TDM van linezolid te vergemakkelijken, hebben we twee nieuwe methoden ontwikkeld om monsters af te nemen en er vervolgens de waarde van vastgesteld (gevalideerd) in hoofdstuk 4. We hebben methoden ontwikkeld om de linezolidconcentratie te kunnen meten in gedroogde druppels bloed op papier (Dried Blood Spot, DBS) en in speeksel (oral fluid).

Om het meten van linezolid in DBS te valideren, hebben we patiënten uit het Tuberculose Centrum Beatrixoord die linezolid kregen als onderdeel van de MDR-TB behandeling opgenomen in het onderzoek. We hebben bloedmonsters verzameld in de plateaufase (steady state) vlak voor inname en 1, 2, 3, 4 en 8 uur na inname van linezolid. Met behulp van deze bloedmonsters hebben we ‘veneuze DBS’ (vDBS) monsters gemaakt door 50 μL aderlijk bloed te druppelen op geschikt Whatman 31 ET CHR DBS-papier. Ook hebben we DBS-monsters van patiënten verzameld op 2 en 8 uur na het toedienen van linezolid door een druppel bloed uit een vingerprik direct op DBS-papier te laten vallen. In beide gevallen hebben we de bloeddruppels laten drogen bij kamertemperatuur en opgeslagen in een afsluitbare plastic zak met droogpoeder. Hierna zijn de monsters in de vriezer bij -20°C bewaard totdat de monsters geanalyseerd werden. Vlak voor het analyseren hebben we een schijf met een diameter van 8 mm uit het papier met de gedroogde bloeddruppel geboord. Vervolgens hebben we de monsters onderzocht met behulp van een gevalideerde vloeistofchromatografie tandem massaspectrometrische (LC-MS/MS) methode. We hebben de nieuwe DBS-methode gevalideerd in overeenstemming met regelgeving van de Amerikaanse Food and Drug Administration (FDA). Naast het validatie-onderzoek op basis van deze regelgeving, zijn enkele aanvullende onderzoeken uitgevoerd specifiek voor DBS-analyse.

Bij DBS wordt een ingedroogd bloedmonster in een schijfje filterpapier opnieuw in oplossing gebracht zodat een geneesmiddelconcentratie wordt gemeten. De hoeveelheid vloeistof wordt via een standaardmethode toegevoegd, waarbij wordt aangenomen dat bij iedereen de concentratie bloedcellen en hemoglobine (Hb) het zelfde is. Sommige mensen hebben echter bloedarmoede, anderen hebben juist een hoog Hb. De hoogte van het Hb, die recht evenredig is met de hematocrietwaarde, heeft invloed op de geneesmiddelconcentratie. Ook de grootte


140

van de bloeddruppels heeft invloed op de met DBS gemeten concentratie. Beide parameters zouden de resultaten van de DBS-analyse kunnen beïnvloeden. We hebben daarom de invloed van een hematocriet van 20, 25, 30, 35, 40, 45 en 50% en bloeddruppelvolumes van 30, 50 70 en 90 μL onderzocht. Voor de validatie hebben we de hematocrietwaarde op 35% en het bloeddruppelvolume op 50 μL ingesteld.

De DBS-analyseresultaten waren recht evenredig met gemeten bloedwaarden over het bereik van de gehele analytische concentratie van 0.05 – 30 mg/L linezolid met een gepoolde correlatiecoëfficiënt van r2 = 0,9947. De reproduceerbaarheid was goed met een systematische fout (bias) en variatiecoëfficiënt (CV) kleiner dan 17,2% voor de laagst gevalideerde concentratie (0,05 mg/L linezolid). Voor de andere concentraties (0,25 – 30 mg/L linezolid) waren de waardes van de bias en CV kleiner dan 7,8%. Het effect van verschillen in hematocriet en bloeddruppelvolume was acceptabel. De methode heeft een goede recovery van ongeveer 95%. Er is een klein matrixeffect van minder dan 8,7% gemeten. De DBS-monsters bleken stabiel te zijn nadat deze gedurende twee maanden bewaard waren bij 37°C. Een Bland-Altman-analyse liet zien dat de ratio van de linezolidconcentratie in DBS / plasma 1,2 was (95% betrouwbaarheidsinterval, 1,12 – 1,27). Deze correctiefactor moet gebruikt worden om de concentratie gemeten in DBS om te rekenen naar de plasmaconcentratie van linezolid.

We concluderen dat de DBS-analyse van linezolid een gemakkelijke, bruikbare methode is om TDM van linezolid bij MDR-TB uit te voeren. De relatief simpele monsternameprocedure en de grote chemische stabiliteit van de monsters maken de analyse zeer gebruiksvriendelijk. Dit maakt de methode onder andere zeer geschikt voor ontwikkelingslanden.

Tevens hebben we een speekselmonsteranalyse ontwikkeld en gevalideerd (hoofdstuk 4). Hiertoe hebben we gegevens van de patiënten gebruikt die deelnamen aan het onderzoek in hoofdstuk 3. Zoals in hoofdstuk 3 beschreven, kregen alle patiënten gedurende twee weken tweemaal daags 300 mg linezolid en eenmaal daags 250 mg claritromycine toegediend. Bloed- en speekselmonsters werden tegelijk afgenomen in de plateaufase. Monsters werden vlak voor toediening van de medicatie verzameld en op 1, 2, 3, 4, 8 en 12 uur na het toedienen van de medicatie. Speeksel werd verzameld met behulp van Salivettes (Sarstedt, Leicester, Engeland). Patiënten kauwden gedurende twee minuten op het katoenen watje van de Salivette. Vervolgens werden de monsters gecentrifugeerd en opgeslagen bij -20°C tot het moment van analyse. Linezolid en claritromycine concentraties zijn vervolgens gemeten met behulp van de door ons gevalideerde LC-MS/MS methode.

141


*

Passing-Bablok regressie en Bland-Altman analyse hebben aangetoond dat analyse van linezolid in speeksel een geschikt alternatief is voor conventionele bloedmonsters van patiënten met MDR-TB. Er is geen correctiefactor nodig voor de interpretatie van speekselmonsters, aangezien de ratio tussen de linezolidconcentratie in serum en de concentratie in speeksel met behulp van Bland-Altman analyse berekend is op 0,97 (95% betrouwbaarheidsinterval: 0,92 – 1,02). De Bland-Altman analyse van de claritromycineconcentratie liet echter een ratio tussen serum en speeksel zien van 3,07 (95% betrouwbaarheidsinterval: 2,45 – 3,69). We hebben een niet recht evenredige relatie waargenomen tussen de concentraties van het metaboliet hydroxyclaritromycine (in de lever gevormd omzettingsproduct van claritromycine) in serum en speeksel. We concluderen dat de analyse van linezolid en claritromycine in speeksel een makkelijke, non-invasieve methode is om monsters te verzamelen bij MDR-TB patiënten. Dit alternatief zou vooral kunnen worden overwogen wanneer het gebruikelijke verzamelen van bloedmonsters niet mogelijk of wenselijk is.

De mogelijke rol van TDM van linezolid bij MDR-TB bespraken wij in de eerdere hoofd-stukken. Anderen hebben gewezen op verschillen in farmacokinetiek van linezolid tussen verschillende patiënten en in één patiënt in loop van de tijd. Het afnemen en analyseren van bloedmonsters ten behoeve van TDM wordt helaas niet altijd bij geneesmiddelonderzoeken uitgevoerd. Om het belang te onderstrepen dat we hieraan hechten, hebben we een tweetal brieven ingestuurd naar wetenschappelijke tijdschriften in reactie op een tweetal publicaties waarbij geen TDM was uitgevoerd (hoofdstuk 5). Naast het belang van TDM hebben we ook getracht aandacht te vestigen op het belang van het innemen van tuberculosemedicatie onder toezicht van zorgprofessionals (directly observed therapy, DOT). Alleen bij hoge uitzondering, wanneer therapietrouw hoogstwaarschijnlijk is, zou overwogen kunnen worden om van dit principe af te wijken. Tenslotte hebben we aangegeven dat er grote behoefte is aan het ontwerpen en uitvoeren van een dubbelblind gerandomiseerd onderzoek – een onderzoek waarbij noch de patiënt noch de dokter weet of de patiënt linezolid of een placebo krijgt – bovenop een standaard behandeling. Hierbij is het van belang om ook farmacokinetische gegevens en gegevens over de gevoeligheid van de betrokken bacterie voor de gekozen antibiotica te verzamelen.

De aanbevelingen die we deden in de twee ingestuurde brieven, hebben ons aangespoord om zelf ook het nut van TDM te bestuderen bij onze eigen patiënten die eerder linezolid hadden gebruikt. We hebben de veiligheid en verdraagbaarheid onderzocht in samenhang met de farmacokinetische en farmacodynamische gegevens de we terug hebben gezocht. Om de steekproef te vergroten hebben we niet alleen patiënten geïncludeerd uit het Tuberculose


142

Centrum Beatrixoord, maar ook uit het E. Morelli Ziekenhuis (Sondalo, Italië). We hebben patiënten geselecteerd die linezolid toegediend kregen als onderdeel van behandeling van MDR-TB. Patiënten jonger dan 18 jaar oud, patiënten waarvan onvoldoende gegevens beschikbaar waren door recente opname in een van beide ziekenhuizen en patiënten waarbij geen bloedmonsters zijn afgenomen voor TDM van linezolid, kwamen niet in aanmerking voor het onderzoek. Twee onderzoekers, één in Italië en één in Nederland, hebben van de geselecteerde patiënten alle belangrijke gegevens verzameld, met in elk geval ook de farmacokinetische gegevens en de behandelresultaten, waarbij we ervoor gezorgd hebben dat de privacy van patiënten niet werd geschonden.

In totaal konden we gegevens van 58 MDR-TB patiënten uit beide centra gebruiken. Van 36 van de 54 patiënten uit Italië waren de gegevens onvolledig. Zo waren er bij 27 van de 36 geen bloedmonsters afgenomen ten behoeve van TDM van linezolid. De mediane (IQR) minimale groeiremmende concentratie (MIC) van de bacteriën voor linezolid was 0,5 (0,25 – 0,5) mg/L. Terugkijkend bleek dat 50% (29/58) van de patiënten genezen was. Op het moment van het verzamelen van de gegevens had 6,9% (4/58) van de patiënten de behandeling afgerond, maar 43,1% (25/58) was nog bezig met de behandeling. Op één patiënt na waren alle patiënten kweeknegatief (‘geen groei’) op het moment van verzamelen van de data. We hebben in dit cohort geen samenhang kunnen ontdekken tussen omslag in de sputumkweek en PK/PD parameters. Mogelijk speelt het feit dat de AUC/MIC ratio in bijna alle patiënten boven het beoogde doel was een rol.

Terugkijkend zagen we dat patiënten met de bijwerking/vergiftiging perifere neuropathie een hogere mediane (IQR) totale linezoliddosering hadden gekregen. Zo kregen patiënten met perifere neuropathie 1829 (1414 – 2255) mg/kg ten opzichte van 1164 (755 – 1922) mg/kg bij de patiënten zonder perifere neuropathie (p=0,041). Ook kregen de patiënten met perifere neuropathie gedurende een langere mediane (IQR) periode linezolid toegediend (159 (120 – 196) dagen) ten opzichte van patiënten zonder perifere neuropathie (97 (66 – 147) dagen) (p=0,003). Een andere interessante bevinding uit dit onderzoek was dat het toevoegen van erythropoiëtine (“epo” voor het aanmaken van rode bloedcellen) aan behandelschema’s met linezolid om bloedarmoede voorkomen, geen toegevoegde waarde bleek te hebben bij het voorkomen daarvan. We hebben daarvoor het Hb op dag 0, 30, 60 en 90 na starten van linezolid onderzocht. Er was geen verschil tussen het Hb van patiënten die wel of geen erythropoiëtine toegediend kregen. Er was er ook geen verschil waar te nemen tussen het Hb in de tijd tussen deze twee groepen.

143


*

Terugkijkend leken de behandelregimes met linezolid effectief en waren er niet veel problemen met bijwerkingen. We zagen dat de sputumkweken van de meeste patiënten negatief werden en dat de bijwerkingen heel beperkt waren.

Farmacodynamiek van linezolid

In hoofdstuk 6 beschrijven we een farmacodynamische interactie tussen linezolid en claritromycine in tuberculosebacteriën. Bij dit onderzoek hebben we samengewerkt met het Rijksinstituut voor Volksgezondheid en Milieu (RIVM, Bilthoven, Nederland) bij het uitvoeren van een laboratoriumtest. We hebben de zogeheten schaakbord methode gebruikt om eventuele synergie (versterkend effect: 2 plus 2 = 5) of juist antagonisme (verzwakkend effect: 2 plus 2 = 3) tussen linezolid en claritromycine te onderzoeken. We hebben hiertoe willekeurig 24 tuberkelbacilstammen geselecteerd uit de biobank van het RIVM bestaande uit bij patiënten geïsoleerde stammen. We gebruikten een tweetal al ontwikkelde methoden om groei/groeiremming vast te stellen, namelijk de absolute concentratie methode (ACM) en de Mycobacterium Growth Inhibition Tube (MGIT) 960 methode. Hierbij hebben we groei/groeiremming onderzocht na toevoeging van verschillende concentraties van beide geneesmiddelen op een schaakbord-achtige wijze. Linezolid hebben we toegevoegd in concentraties van 0 – 0,5 mg/L en claritromycine in concentraties van 0 – 8 mg/L. Nadat groei/groeiremming geanalyseerd was bij alle mogelijke combinaties van concentraties van beide geneesmiddelen met de twee methoden hebben we de laagste fractionele groeiremmende concentratie (FIC) berekend. De FIC werd als volgt berekend: MIC van linezolid in combinatie met claritromycine gedeeld door de MIC van linezolid zonder claritromycine plus de MIC van claritromycine in combinatie met linezolid gedeeld door de MIC van claritromycine zonder linezolid. Synergie was gedefinieerd als een FIC≤0,5, geen invloed (indifference) als een 0,5<FIC≤4 en antagonisme als een FIC>4.

In 74% van de 24 willekeurig geselecteerde tuberculosebacteriën bleek er synergie te zijn tussen claritromycine en linezolid wanneer de MGIT methode gebruikt werd. Wanneer de ACM methode gebruikt werd was dit percentage 59%. Er werd in geen van de geselecteerde bacteriën antagonisme gezien. Aangezien de combinatie van linezolid en claritromycine in de praktijk vooral bij MDR-TB patiënten wordt toegepast, hebben we extra aandacht besteed aan de selecteerde bacteriën van MDR-TB patiënten. In deze MDR-TB bacteriën (n=13) bleek synergie te zijn in 77% van de gevallen gemeten met de MGIT methode en in 46% van de gevallen gemeten met de ACM methode. Ook hier werd in geen van de gevallen antagonisme


144

waargenomen. De mediane (IQR) FIC die berekend werd bij de MDR-TB patiënten was 0,37 (0,32 – 0,37) en 0,62 (0,375 – 1,0). We concluderen dan ook dat claritromycine en linezolid synergetisch werken in een deel van de laboratoriumtesten met MDR-TB bacteriën. Dit kan misschien helpen bij het ontwerpen van toekomstige MDR-TB behandelschema’s.

In hoofdstuk 7 bespreken we de belangrijkste bevindingen van dit proefschrift: wat voegt het toe aan wat we al wisten, en hoe kunnen we de resultaten gebruiken voor patiënten met MDR-TB? We bespreken het belang van farmacokinetische geneesmiddelinteracties met antibiotica als slachtoffer, maar ook als veroorzaker van de interactie. Hiernaast besteden we aandacht aan het belang van geïndividualiseerde behandelschema’s en de noodzaak om PK/PD streefwaarden voor TDM te onderzoeken. Tenslotte gaan we op het toekomstperspectief in: 1. Waar verwachten we dat de behandeling van MDR-TB naar toe zal gaan in de komende jaren; 2. Welke rol zou linezolid kunnen spelen bij de behandeling van (multiresistente) tuberculose; 3. Wat zou in de komende jaren de impact kunnen zijn van het verlopen van het patent van linezolid en 4. Wat zou (vervolg)onderzoek kunnen zijn waaraan op dit moment het meeste behoefte is.

Co-authors of manuscripts

presented in this thesis

Co-authors of manuscripts presented in this thesis

146

Affiliations during the conductance of the research

J.W.C. Alffenaar University of Groningen, University Medical Center Groningen, Department of Hospital and Clinical Pharmacy (before 2014), Department of Clinical Pharmacy and Pharmacology (since 2014), Groningen, the Netherlands.

R. van Altena University of Groningen, University Medical Center Groningen, Tuberculosis Center Beatrixoord, Haren, the Netherlands.

J.R.B.J. Brouwers University of Groningen, Department of Pharmacotherapy and Pharmaceutical Care, Groningen, the Netherlands.

S. De Lorenzo E. Morelli Hospital AOVV, Reference center for MDR-TB and HIV-TB, Sondalo, Italy.

B. Greijdanus University of Groningen, University Medical Center Groningen, Department of Hospital and Clinical Pharmacy, Groningen, the Netherlands

K. van Hateren University of Groningen, University Medical Center Groningen, Department of Hospital and Clinical Pharmacy, Groningen, the Netherlands

R.A. Koster University of Groningen, University Medical Center Groningen, Department of Hospital and Clinical Pharmacy, Groningen, the Netherlands.

J.G.W. Kosterink University of Groningen, University Medical Center Groningen, Department of Hospital and Clinical Pharmacy (before 2014), Department of Clinical Pharmacy and Pharmacology (since 2014), Groningen, The Netherlands and University of Groningen, Department of Pharmacy, Section Pharmacotherapy and Pharmaceutical Care, Groningen, the Netherlands.

T. van der Laan National Tuberculosis Reference Laboratory, National Institute for Public Health and the Environment (RIVM), Bilthoven, the Netherlands.

W.C.M. de Lange University of Groningen, University Medical Center Groningen, Tuberculosis Center Beatrixoord, Haren, the Netherlands.

G.B. Migliori WHO Collaborating Centre for Tuberculosis and Lung Diseases, Fondazione S Maugeri, Care and Research Institute, Tradate, Italy.

P.N. Panday University of Groningen, University Medical Center Groningen, Department of Hospital and Clinical Pharmacy, Groningen, the Netherlands.

147

Co-authors of manuscripts presented in this thesis

*

A.D. Pranger University of Groningen, University Medical Center Groningen, Department of Hospital and Clinical Pharmacy, Groningen, the Netherlands.

D. van Soolingen National Tuberculosis Reference Laboratory, National Institute for Public Health and the Environment (RIVM), Bilthoven, the Netherlands and Radboud University Nijmegen Medical Center, Departments of Clinical Infection Diseases and Pulmonary Diseases and Medical Microbiology, Nijmegen, the Netherlands.

G. Sotgiu Clinical Epidemiology and Medical Statistics Unit, Department of Biomedical Sciences, University of Sassari, Research, Medical Education and Professional Development Unit, AOU Sassari, Italy.

S. Tiberi Imperial College Healthcare NHS Trust, London, UK.

D.R.A. Uges University of Groningen, University Medical Center Groningen, Department of Hospital and Clinical Pharmacy, Groningen, the Netherlands.

D.H. Vu University of Groningen, Department of Pharmacotherapy and Pharmaceutical Care, Groningen, the Netherlands; University of Groningen, University Medical Center Groningen, Department of Hospital and Clinical Pharmacy, Groningen, the Netherlands and Hanoi University of Pharmacy, Hanoi, Vietnam.

T.S. van der Werf University of Groningen, University Medical Center Groningen, Departments of Internal Medicine, and Pulmonary Diseases & Tuberculosis, Groningen, The Netherlands and University of Groningen, University Medical Center Groningen, Tuberculosis Center Beatrixoord, Haren, the Netherlands.

Dankwoord

Dankwoord

150

Het slotakkoord! Naast het feit dat onderzoek doen vooral leuk is, is het ook erg fijn om iets – al is het maar een klein beetje – bij te kunnen dragen aan de behandeling van de tuberculose. Zonder hulp van velen had ik dit punt nooit bereikt. Allereerst wil ik de patiënten, die zonder eigenbelang deelnamen aan het onderzoek, hartelijk bedanken.

Ik ben vooral mijn promotores en copromotor ontzettend dankbaar.

Prof. dr. T.S. van der Werf, beste Tjip. Ik heb ontzettend veel bewondering voor jou als arts, promotor en onderzoeker. Hartelijke dank voor je kritische blik op alle geschreven stukken. Vooral bij de eerste manuscripten had je veel op te merken (‘de METc maakte meer spelfouten dan jij’ en ‘kleine suggesties mijnerzijds’). Zonder je tekstuele, maar vooral ook inhoudelijke input zou het proefschrift nooit deze staat bereikt hebben. In het begin moest ik soms wat wennen aan je directe manier van communiceren. Maar ik kan me herinneren dat we elkaar tegenkwamen in een gang in het UMCG en dat je, zonder iets te zeggen, je hand opstak waarna we elkaar zwijgend een high five gaven. Toen wist ik dat de samenwerking goed zou komen. Ook heb ik erg veel gehad aan het samen peer-reviewen. Dit heeft geholpen bij het gestructureerd inhoudelijk en methodologisch kijken naar onderzoeken en kritisch kijken naar de nieuwswaarde. Tenslotte wil ik je graag hartelijk bedanken voor de benodigde uitleg en bovenal humor tijdens de TB-visites.

Prof. dr. J.G.W. Kosterink, beste Jos. Jij was niet alleen promotor, maar ook nog eens opleider tijdens een deel van dit promotietraject. Ik heb altijd ontzettend veel bewondering gehad voor het feit dat je je drukke baan als hoofd van de afdeling Klinische Farmacie en Apotheek – sinds januari 2014 Klinische Farmacie en Farmacologie – combineert met het opleiden van ziekenhuisapothekers, klinisch farmacologen, farmaciestudenten en promovendi. Toch was je altijd beschikbaar voor promotiebesprekingen, inhoudelijk overleg over op te zetten onderzoek, het reviseren van manuscripten, maar ook voor ad hoc overleg wanneer ik dat nodig had. Hartelijk dank daarvoor! Tenslotte wil ik je – ook hier nog eens – hartelijk danken voor het opleiden tot ziekenhuisapotheker. Ik heb ontzettend veel geleerd.

Prof. dr. D.R.A. Uges, beste Donald. Zowel voor als tijdens je emeritaat speelde je een grote rol bij het onderzoek. Geweldig dat je nu nog steeds tijd vrij maakt voor het onderzoek. Niet alleen op afstand door het reviseren van stukken tekst, maar ook nu zit je altijd trouw bij de promotiebesprekingen. Dit geeft wel aan dat je een promotor in hart en nieren bent. In de eerste fase van het promotieonderzoek zat ik deels op het laboratorium als ziekenhuisapotheker in opleiding. Zonder een goed draaiend laboratorium, waar je het onderzoek altijd een belangrijke plaats gaf, had dit proefschrift nooit zijn huidige staat

151

Dankwoord

*

bereikt. Ook denk ik met veel dank terug aan de eerste periode als projectapotheker en later als ziekenhuisapotheker op het laboratorium waar je me altijd bij alle interessante cases betrok. Hier is misschien wel de basis gelegd voor mijn interesse in PK/PD en überhaupt in onderzoek.

Dr. J.W.C. Alffenaar, beste Jan-Willem. Ik heb er ontzettend veel bewondering voor hoe je binnen korte tijd na het afronden van je eigen promotie als co-promotor verschillende promovendi met succes begeleidt, een goed (inter)nationaal netwerk op hebt gebouwd en overal mogelijkheden ziet. Je ziet overal kansen en mogelijkheden. Een mooie eigenschap bij het opzetten van onderzoek. Ik wil je bedanken voor alle inhoudelijke gesprekken over het onderzoek en het mij betrekken bij verschillende ideeën en vooral voor het vertrouwen dat je in me had. Na het opschrijven van het case-report, ontstond bijna vanzelfsprekend de mogelijkheid om hier een promotieonderzoek van te maken, wat ik met beide handen aangreep.

Graag zou ik de beoordelingscommissie, bestaande uit prof. dr. D.M. Burger, prof. dr. F.G.J. Cobelens en prof. dr. G. M. M. Groothuis, willen bedanken voor het beoordelen van het manuscript.

Er zijn veel mensen die een grote rol hebben gespeeld bij het totstandkomen van dit proefschrift. Ik zou dan ook allen willen bedanken die een directe bijdrage hebben geleverd aan de inhoud van dit proefschrift.

Allereerst de artsen van de TB-afdeling: drs. O.W. Akkerman, drs. R. van Altena en drs. W.C.M. de Lange. Zonder jullie hulp bij het opzetten van het onderzoek, bij het verzamelen van monsters en bij de inclusie van patiënten, was dit onderzoek nooit tot stand gekomen. We zeggen soms gekscherend dat alle tuberculosepatiënten op de locatie Beatrixoord op enig moment in een onderzoek terecht komen. Hoewel dat natuurlijk niet helemaal waar is, geeft dit wat mij betreft wel jullie onderzoeks-mindedness aan. Verder wil ik jullie, maar ook zeker dr. Y (Ymkje) Stienstra en Tjip, hartelijk danken voor de TB-visites. Dank voor het feit dat jullie mij als simpele apotheker meegenomen hebben in de wereld van de patiënt en de behandelaar, maar ook dank voor de humor tijdens deze besprekingen. Beste Onno, geweldig om met jou op OK monsters te verzamelen bij een wervel-TB operatie. Succes met het afronden van jouw boekje! Beste Richard, hartelijke dank, succes met je proefschrift en geniet van Myanmar. Wie weet ben ik ooit eens in de buurt. Beste Wiel, dank voor het mij op sleeptouw nemen bij de GGD en de mooie verhalen tijdens de visites.

Dankwoord

152

Ook zijn er enkele apothekers die een directe, inhoudelijk bijdrage hebben geleverd aan dit proefschrift: drs. P.V. Nannan Panday, drs. A.D. Pranger en dr. D.H. Vu. Beste Prashant, hartelijk dank voor alle supervisie tijdens mijn opleiding tot ziekenhuisapotheker. Je hebt als consulent infectieziekten ontzettend veel kennis en expertise opgebouwd over medicamenteuze behandeling van infecties welke zeer van pas kwamen bij het opstellen van onze review naar geneesmiddelinteracties van antibiotica. Hiernaast waardeer ik je ook vooral als directe collega. Beste Arianna, vooral bij het prospectieve onderzoek heb je, met de ervaring die jij reeds had, me veel geholpen. Hiernaast denk ik met veel plezier terug aan de baseballwedstrijd die we samen keken: Let’s go Giants! Dear Hoa, I really enjoyed our cooperation on the analytical studies such as the linezolid in DBS study.

Hiernaast zijn er twee laboratoria geweest die dit onderzoek mogelijk gemaakt hebben. Beide spelen niet alleen bij het onderzoek, maar ook bij de routinezorg een zeer grote rol voor de tuberculose. Allereerst dank aan alle analisten van het laboratorium van de ziekenhuisapotheek van het UMCG. Iedereen is wel op enig moment betrokken geweest bij het onderzoek. B. Greijdanus, beste Ben; R.A. Koster (binnenkort dr. Koster!), beste Remco; K. van Hateren, beste Kay. Hartelijke dank voor het bijdragen van jullie analytische expertise aan enkele van de artikelen in dit proefschrift. Hiernaast wil ik ook bioanalyse-analisten, Albert-Jan, Erwin, Gerben, Hiltjo, Jan, en Mireille bedanken voor hun inzet: merci!

Het laboratorium van het RIVM speelde een belangrijke rol bij het uitvoeren van drug susceptibility testen, maar ook bij enkele onderzoeken. Prof. dr. D. van Soolingen, beste Dick en T. van der Laan, beste Tridia. Bedankt voor de prettige samenwerking.

I would like to thank the co-authors that contributed to the work in this thesis: Prof. dr. J.R.B.J. Brouwers (dank!), Simon Tiberi, Giovanni Sotgiu, Saverio De Lorenzo, Giovanni Battista Migliori. Mille Grazie! I hope we’re able to continue our international collaboration. Working together is the way forward.

Ook zou ik graag de verpleging van het Tuberculose Centrum Beatrixoord willen bedanken voor de rol die ze hebben gespeeld in dit onderzoek. Voor dit type onderzoek is het toedienen van het juiste geneesmiddel op het juiste moment van groot belang. Het is dan ook erg fijn dat jullie dit met zijn allen inzien en hier zo goed mee om gaan. Ook denk ik met plezier terug aan de scholingsmomenten gerelateerd aan het onderzoek, waarbij jullie zelfs bij elkaar DBS monsters afnamen.

153

Dankwoord

*

Hiernaast ben ik het secretariaat van de apotheek erg dankbaar voor de hulp die ze altijd bieden. Specifiek voor het onderzoek door middel van het versturen van brieven naar de METc en andere instanties en het inplannen van afspraken met mede-onderzoekers, maar ook in het algemeen in deze laatste fase van het onderzoek en tijdens het alledaagse werk in het UMCG. Annemiek, Jessica en Wianda: dank jullie wel!

Verder ben ik alle apothekersassistenten erg dankbaar voor de ondersteuning bij het onder-zoek. Vooral dank aan de collega’s van de KGO, de balie, het magazijn, de voorraadbereidingen en satellietapotheek Beatrixoord. Jullie hulp maakte dat dit onderzoek te combineren was met andere taken binnen de apotheek.

Tenslotte ben ik de staf, (ziekenhuis)apothekers en oud-collegae van de voormalige afdeling Klinische Farmacie en Apotheek erg dankbaar voor de feedback en de discussies: Anet, Annelies, Barbara, Bart, Bob, Daan, Derk, Eli, Esther, Eva, Frank-Jan, Gea, Hèlen, Hendrikus, Hilma, Iemke, Jan, Joke, Kim, Lisanne, Marian, Marieke, Marina, Marijn, Marjolijn, Marjolijn, Matthijs, Minke, Nour, Reinout Susan, Sylvia en Trea. Dank voor alles! Ik ben blij met jullie als collega’s. Ook hartelijk dank aan alle nieuwe collega’s van de voormalige Klinische Farmacologie van prof. Dick de Zeeuw. Ik verheug me op een intensieve samenwerking in de komende jaren.

Hiernaast zijn er natuurlijk ook velen die indirect hebben bijgedragen aan dit onderzoek. Ik wil daarom al mijn familie en vrienden bedanken voor de mooie tijden!

Ik heb tijdens mijn studietijd met erg veel plezier wedstrijd geroeid bij de A.G.S.R. Gyas. De combinatie van feesten, sociaal leven, met zijn allen trainen voor een doel, nooit opgeven (‘pijn is tijdelijk, opgeven is voor eeuwig’) heeft bijgedragen aan hoe ik nu ben. Dank aan al mijn Gyas-vrienden en -kennissen, maar vooral aan mijn ploeggenoten, coaches en stuurtjes: Bram, Niels, Thijs, Janneke, Karijn, Alma, Maurice, Jan, Juliette, Nanne, Irma, Koen, Carolien en Sybolt.

Dr. E.A.M. Festen, beste Noortje. Ik bewonder je gedrevenheid, je zelfvertrouwen, je eigenzinnigheid. Je hebt een grote rol gespeeld bij het feit dat ik promotieonderzoek ben gaan doen.

Beste Djoelan, dank voor alle gezelligheid en theetjes op het terras ‘Chez Mathieu’ aan de Vismarkt. Je bent een van de meest sociale mensen die ik ken.

Dankwoord

154

Ook zou ik graag iedereen bedanken met wie ik schaats of fiets. Zonder regelmatig sporten zou ik gek worden. Is er iets lekkerder na een dag werken aan je onderzoek dan lekker schaatsen of een ‘rondje Zoutkamp’ doen? Rik en Anton, ik hoop dat we ooit nog een keer de echte Elfstedentocht mogen doen.

Beste Bas, Njord, Robbert, Peter en Tim. Ik ben ontzettend blij dat we elkaar hebben leren kennen via of op het Praedinius. We hebben de afgelopen jaren veel mooie dingen gedaan. Om maar een paar hoogtepunten te noemen: Lowlands, weekendje Boekarest, Pitch festival, maar ook op stand dineren bij ‘In de Molen’. Het plannen is soms lastig met alle volle agenda’s en de verschillende steden, landen, continenten waar we wonen. Het resultaat is echter altijd de moeite waard.

Ik ben blij met iedereen die ik via Mijntje heb leren kennen. Haar hele familie, maar vooral Wil, Ben, Margriet, Jeroen, Maartje en Phillip, en al haar vrienden. Fijn dat Mijntje zulke leuke vriendjes en vriendinnetjes heeft.

Ook zou ik mijn hele familie – alle ooms, tantes, neven, nichten en andere familie – willen bedanken voor de steun en interesse. Lieve opa’s en oma, lieve Anna, Hein en Thies. mede dankzij jullie heb ik zulke mooie jeugdherinneringen. Ik hoop dat ik nog jaren van jullie mag genieten in goede gezondheid.

Ook zou ik mijn paranimfen, Jasper en Wouter, willen bedanken. Ik heb erg veel aan jullie beiden gehad tijdens mijn promotie. Jasper, geweldig dat je enige tijd geleden besloten hebt naar Groningen terug te komen. Hoewel het samen sporten er niet vaak genoeg van komt, hebben de avondjes in de stad, lekker eten met kookclub ‘Mathieu en Jasper’ en het samen rondhangen gezorgd voor de nodig ontspanning. Wie weet komt die marathon er ooit nog van. Wouter, mijn derde ‘broertje’, ik denk met veel plezier terug aan de avondjes Eurosonic, onze gezamelijke playlist, de BBQ’s bij jullie in de tuin, de echte koffie en onze werklunch bij een welbekend Schots etablissement. Geweldig om met jou te ouwehoeren in de apotheek. Jasper en Wouter, dank voor jullie praktische tips over het promoveren. Het wordt een mooie dag!

Lieve Jaap en Marielou. Bedankt voor alles. Dankzij jullie opvoeding ben ik zo geworden als ik nu ben. Op geen enkel vlak kan ik me een beter voorbeeld dan jullie wensen. Dank voor jullie steun. Lieve Oskar en Jules. Ik ben ontzettend blij met jullie als broers! Dank voor de mooie tripjes, leuke weekenden en de gezelligheid. Jules, dank voor de vakantie naar een van mijn droombestemmingen: Soedan. Het was onvergetelijk! Oskar en Tessa, geweldig dat jullie gaan trouwen! Ik wens jullie het allerbeste.

155

Dankwoord

*

Allerliefste Niamey, geachte Prop-Joe. Bedankt.

Liefste Mijntje. Wat kan een simpele treinreis toch veel veranderen. Ik ben ontzettend dankbaar dat ik toevallig tegenover jou kwam te zitten. Dank voor je steun en begrip voor het onderzoek. Maar bovenal dank dat de afgelopen jaren met jou zo geweldig waren: mooie feestjes, geweldige reizen (ooit nog Nigeria?), familie opzoeken in Madrid, serietjes kijken op de bank, stedentripjes, lekker koken, dansen door de woonkamer, theater en concertbezoek, samen krantjes lezen en nog veel meer. Ik kan niet wachten op wat de toekomst met jou in petto heeft.

About the author

About the author

158

Curriculum vitae

Mathieu Sander Bolhuis was born on May 11th 1981 in Groningen, the Netherlands. In 1999, he graduated from the Praedinius Gymnasium, a secondary school in Groningen, and started to study Pharmacy at the University of Groningen. In 2005, he obtained his master’s degree and in 2007 he graduated as a pharmacist.

From 2004 until 2006, Mathieu was part of a Men’s Heavyweight Crew of the student Rowing Club A.G.S.R. Gyas (Groningen, the Netherlands). In 2006, he stayed in Ghana as a part of a Pharmacy internship.

After graduation, Mathieu started working as a pharmacist at the Diaconessenhuis (Leiden, the Netherlands). In 2008, he returned to Groningen to work at the department of Hospital and Clinical Pharmacy at the University Medical Center Groningen (UMCG). Here he worked at the laboratory and as a Clinical Trial Pharmacist.

In December 2008, he started his specialization to become a hospital pharmacist, under supervision of prof.dr. Jos Kosterink. As a part of this specialization, a prospective study was carried out, which turned out to be the beginning of his PhD project. In December 2012, he was registered as a hospital pharmacist and he continued working at the UMCG.

At the moment, Mathieu works as a hospital pharmacist at the department of Clinical Pharmacy and Pharmacology of the UMCG where he is responsible for the pharmaceutical care of the Center for Rehabilitation / Tuberculosis Center Beatrixoord (Haren, the Netherlands). In addition, he works on a project aiming to implement a new electronic patient database for the UMCG and started with a training to become a clinical pharmacologist.

Mathieu lives together with his girlfriend Mijntje ten Brummelaar and their two cats, Niamey (2011) and Prop-Joe (2011). He enjoys travelling, cycling and West-African music from the seventies. One day he hopes to ride and finish the Elfstedentocht.

159

About the author

*

List of publications related to this thesis

Bolhuis, M. S., T. van der Laan, J. G. Kosterink, T. S. van der Werf, D. van Soolingen, and J. W. Alffenaar. 2014. In vitro synergy between linezolid and clarithromycin against Mycobacterium tuberculosis. Eur. Respir. J. 44:808-811.

Bolhuis, M. S., R. van Altena, D. van Soolingen, W. C. Lange, D. R. Uges, T. S. van der Werf, J. G. Kosterink, and J. W. Alffenaar. 2013. Clarithromycin increases linezolid exposure in multidrug-resistant tuberculosis patients. Eur. Respir. J. 42:1614-1621.

Bolhuis, M. S., R. van Altena, K. van Hateren, W. C. de Lange, B. Greijdanus, D. R. Uges, J. G. Kosterink, T. S. van der Werf, and J. W. Alffenaar. 2013. Clinical validation of the analysis of linezolid and clarithromycin in oral fluid of patients with multidrug-resistant tuberculosis. Antimicrob. Agents Chemother. 57:3676-3680.

Bolhuis, M. S., A. D. Pranger, and J. W. Alffenaar. 2012. Linezolid: safety and efficacy monitoring. Eur. Respir. J. 39:1275-1276; author reply 1276-1277.

Bolhuis, M. S., R. van Altena, and J. W. Alffenaar. 2012. Comment on: daily 300 mg dose of linezolid for multidrug-resistant and extensively drug-resistant tuberculosis: updated analysis of 51 patients. J. Antimicrob. Chemother. 67:2055-6; author reply 2056-2057.

Vu, D. H., M. S. Bolhuis, R. A. Koster, B. Greijdanus, W. C. de Lange, R. van Altena, J. R. Brouwers, D. R. Uges, and J. W. Alffenaar. 2012. Dried blood spot analysis for therapeutic drug monitoring of linezolid in patients with multidrug-resistant tuberculosis. Antimicrob. Agents Chemother. 56:5758-5763.

Bolhuis, M. S., P. N. Panday, A. D. Pranger, J. G. Kosterink, and J. W. Alffenaar. 2011. Pharmacokinetic drug interactions of antimicrobial drugs: a systematic review on oxazolidinones, rifamycines, macrolides, fluoroquinolones, and Beta-lactams. Pharmaceutics. 3:865-913.

Bolhuis, M. S., R. van Altena, D. R. Uges, T. S. van der Werf, J. G. Kosterink, and J. W. Alffenaar. 2010. Clarithromycin significantly increases linezolid serum concentrations. Antimicrob. Agents Chemother. 54:5418-5419.

About the author

160

Other publications

Sturkenboom, M. G., O. W. Akkerman, M. S. Bolhuis, W. C. de Lange, T. S. van der Werf, and J. W. Alffenaar. 2014. Adequate design of pharmacokinetic - pharmacodynamic studies will help optimize TB treatment for the future. Antimicrob. Agents Chemother. Accepted (AAC05173-14).

Hofman, S., M.S. Bolhuis, R.A. Koster, O.W. Akkerman, C. Stove, and J.W. Alffenaar. 2014. Dried blood spots to support treatment of pulmonary infections. Bioanalysis. Accepted.

Vu, D. H., R. A. Koster, M. S. Bolhuis, B. Greijdanus, R. van Altena, D. H. Nguyen, J. R. Brouwers, D. R. Uges, and J. W. Alffenaar. 2014. Simultaneous determination of rifampicin, clarithromycin and their metabolites in dried blood spots using LC-MS/MS. Talanta. 121:9-17.

Akkerman, O. W., R. van Altena, M. S. Bolhuis, T. S. van der Werf, and J. W. Alffenaar. 2014. Strategy to limit sampling of antituberculosis drugs instead of determining concentrations at two hours postingestion in relation to treatment response. Antimicrob. Agents Chemother. 58:628-13.

De Lorenzo, S., J. W. Alffenaar, G. Sotgiu, R. Centis, L. D’Ambrosio, S. Tiberi, M. S. Bolhuis, R. van Altena, P. Viggiani, A. Piana, A. Spanevello, and G. B. Migliori. 2012. Efficacy and safety of meropenem/clavunate added to linezolid containing regimens in the treatment of M/XDR-TB. Eur. Respir. J. 42:1386-1392.

Labberton, L., R. E. Herder, M. S. Bolhuis, R. C. Schellekens, and D. R. Uges. 2011. Formulation of a tacrolimus suspension with extended expiry date using raw materials. EJHP Practice. 17:36-40.

Documents

University of Groningen Linezolid in multidrug-resistant ... · Chapter 2 Pharmacokinetic drug interactions of antimicrobial drugs: a systematic ... tuberculosis (TB) is a bigger