POSTER PRESENTATIONS

Clinical Pharmacology

P301 Interactions between HIV and HCV therapies: how common and who wins? Torkington, A*; Barnes, J; Cripps, S; Lee, K; Marshall, N; Morgan, H; Pratt, A; Sayers, I; Swabe, J; Swaden, L; Selby, P; Vaghjiani, T; Zimmer, M; Bhagani, S; Collier, J; Macdonald, D; Nelson, M; Prince, M; Wright, M; Ustianowski, A (Manchester, UK)

P302 Efavirenz significantly decreases etonogestrel exposure: results of a bidirectional pharmacokinetic evaluation of efavirenz- and nevirapine-based antiretroviral therapy plus etonogestrel contraceptive implants Chappell, C*; Scarsi, K; Nakalema, S; Chen, B; Riddler, S; Cohn, S; Darin, K; Achilles, S; Lamorde, M (Pittsburgh, USA)

P303 Dolutegravir plasma concentrations according to companion antiretroviral drug: unwanted drug interaction or desirable boosting effect? Cattaneo, D*; Minisci, D; Cozzi, V; Riva, A; Meraviglia, P; Clementi, E; Rizzardini, G; Gervsasoni, C (Milan, Italy)

P304 Determinants of dolutegravir plasma concentrations in the clinical setting Alcantarini, C; Calcagno, A*; Marinaro, L; Ferrara, M; Milesi, M; Trentalange, A; Barco, A; Montrucchio, C; De Nicolò, A; Ariaudo, A; Favata, F; D’Avolio, A; Di Perri, G; Bonora, S (Turin, Italy)

P305 Pharmacokinetics (PK) of darunavir/ritonavir (DRV/RTV) with tenofovir DF/emtricitabine (TDF/FTC) or raltegravir (RAL) in HIV-infected adults enrolled in the NEAT001/ANRS143 study and relationship with virological response Bonora, S; Stohr, W; D’Avolio, A; Cursley, A; Molina, J; Faetkenheuer, G; Vandekerckhove, L; Di Perri, G; Pozniak, A; Richert, L; Raffi, F; Boffito, M* (London, UK)

P306 Evaluation of the drug-drug interaction (DDI) potential between elvitegravir/cobicistat/emtricitabine/tenofovir alafenamide and atorvastatin Custodio, J*; West, S; SenGupta, D; Zari, A; Humeniuk, R; Ling, K; Fordyce, M; Kearney, B (Foster City, USA)

P307 Steady-state pharmacokinetics (PK) of atazanavir/cobicistat and darunavir/cobicistat once daily over 72 hours in healthy volunteers: the importance of PK forgiveness in clinical practice Elliot, E*; Amara, A; Pagani, N; Else, L; Moyle, G; Schoolmeesters, A; Higgs, C; Khoo, S; Boffito, M (London, UK)

P308 Association of SLCO1B1 521T>C (rs4149056) with darunavir/ritonavir (DRV/r) plasma concentrations in HIV-infected individuals enrolled in the NEAT001/ANRS143 study Gurjar, R*; Boffito, M; D’Avolio, A; Schwimmer, C; Yazdanpanah, Y; Doroana, M; Di Perri, G; Bonora, S; Pozniak, A; Richert, L; Raffi, F; Owen, A (Liverpool, UK)

P309 When food can make the difference: the case of elvitegravir-based coformulation Gervsasoni, C*; Baldelli, S; Minisci, D; Meraviglia, P; Clementi, E; Galli, M; Cattaneo, D (Milan, Italy)

P310 Impact of food on the bioavailability of darunavir, cobicistat, emtricitabine and tenofovir alafenamide (DCFTAF), the first protease inhibitor-based complete HIV-1 regimen Crauwels, H; Baugh, B*; Van Landuyt, E; Vanveggel, S; Hijzen, A; Opsomer, M (Raritan, USA)

*Indicates presenting author.

P311 Prevalence of drug-drug interactions (DDI) and its impact on durability among patients receiving elvitegravir/cobicistat/emtricitabine/tenofovir (EVG/C/F/T) and concomitant medication Gomez-Ayerbe, C; Del Campo, S; Vivancos, M; Gaspar, G; Moreno, A; Martin, A; Romero, A; Jimenez, P; Cervero, M; Rodriguez-Sagrado, M; Serrano, R; Muriel, A; Peréz Elías, M* (Madrid, Spain)

P312 Relationships between dolutegravir plasma-trough concentrations, UGT1A1 genetic polymorphisms and side-effects of the central nervous system in Japanese HIV-1-infected patients Yagura, H*; Watanabe, D; Ashida, M; Nakauchi, T; Tomishima, K; Togami, H; Hirano, A; Sako, R; Doi, T; Yoshino, M; Takahashi, M; Yamazaki, K; Uehira, T; Shirasaka, T (Osaka, Japan)

P313 S-protein thiol-omics to assess the redox-modulation effects of antiretroviral drugs Correia, M; Dias, C; Caixas, U*; Grilo, N; Lemos, A; Trigo, D; Pacheco, P; Monteiro, E; Soto, K; Diogo, L; Pereira, S (Lisbon, Portugal)

P314 Utilizing phase 3 clinical trial data to assess adverse event (AE) frequency of a potentially interacting medication (PIM) amlodipine with elvitegravir/cobicistat (EVG/COBI) Podzamczer, D; Tashima, K; Daar, E; McGowan, J; Campbell, T; Slim, J; Thompson, M; Guo, S; Borg, P; Haubrich, R; Das, M; McNicholl, I*; McCallister, S (Foster City, USA)

P315 Real-world antiretroviral plasma levels in HIV-positive patients treated with sofosbuvir-containing DAA for hepatitis C infectionTempestilli, M; Fabbri, G*; Timelli, L; Zaccarelli, M; Bellagamba, R; Cicalini, S; Lupi, F; Gallo, A; Libertone, R; Fazio, S; Antinori, A; Ammassari, A (Rome, Italy)

P316 Detection and analysis of antiretroviral medication errors by a clinical pharmacist in hospitalized HIV patientsPablos Bravo, S; García Muñoz, C; Pulido, F*; Lázaro Cebas, A; Ferrari Piquero, J (Madrid, Spain)

P317 Clinical and genetic factors associated with kidney tubular dysfunction in a real-life single-centre cohort of HIV-positive patientsSalvaggio, S; Giacomelli, A*; Falvella, F; Oreni, L; Meraviglia, P; Atzori, C; Clementi, E; Galli, M; Rusconi, S (Milan, Italy)

P318 Prevalence of drug-drug interactions involving antiretroviral treatment: impact of the integrase inhibitor classMessiaen, P*; Baecke, C; van der Hilst, J (Hasselt, Belgium)

P319 Temporal trend of the plasma level of efavirenz: comparison between CYP2B6-516 GG and GT genotypeWong, C*; Chan, P; Lee, S (Hong Kong, Hong Kong)

POSTER PRESENTATIONS

*Indicates presenting author.

www.pat.nhs.uk

Interactions between HIV and HCV therapies: How common and who wins?

2016

Torkington A1 ,Barnes J2, Cripps S3, Lee K4, Marshall N5, Morgan H6, Muqbil G7, Pratt A8, Sayers I1, Selby P9, Swabe J10, Swaden L5, Vaghjiani T10,

Zimmer M6, Bhagani S5, Collier J3, Macdonald D5, Nelson M6, Prince M4, Wright M9, Ustianowski A1

1. North West ID Unit, Manchester, UK; 2. Heartlands Hospital, Birmingham , UK; 3.Oxford University Hospitals, Oxford, UK; 4. Central Manchester University Hospital,

Manchester, UK. 5. Royal Free London, London, UK; 6.Chelsea and Westminster Hospital, London, UK; 7. St Georges Hospital, London, UK; 8. York Hospital, York,

UK; 9. Addenbrookes Hospital, Cambridge, UK 10. University Hospitals Southampton, UK.

Background: • The current era of HCV direct acting antivirals (DAAs) has allowed

HIV/HCV co-infected patients to achieve similar rates of response to HCV

mono-infected patients1.

• Managing HIV/HCV therapy is complex, often involving drug-drug

interactions (DDIs) between the DAAs, ARVs and other medicines.

• We evaluated the incidence of DDIs in co-infected patients and its impact

on choice of preferred HCV therapy as recommended by NHS England.

Methods: • Retrospective evaluation of all HIV/HCV co-infected patients receiving

DAAs seen across 10 UK centres from June 2015 till May 2016.

• Data were collected on demographics, HCV genotype, choice of DAA and

ARVs and any changes made to these or additional monitoring required.

• The Liverpool hep-druginteractions.org website2 was used to evaluate

presence and severity of potential drug interactions.

Conclusions: • Managing HIV/HCV co-infected patients is clearly complex requiring review

and modification of both HIV and HCV therapy with additional monitoring.

• The renal monitoring associated with the tenofovir/ledipasvir DDI needs

standardising as patients are being monitored when it is not necessary.

• The role of the specialist pharmacy team is key to managing this cohort.

Co-medicines and monitoring: • 728 co-medicines were identified in 153/198 (77%) patients (median

3.5/patient).

• 186/728 (25%) amber DDIs (close monitor/dosage adjustment required)

were identified in 147/198 (74%) of patients, with 24/728 (3%) red (do not

co-administer) observed for 20/198 (10%) of patients.

• The need for additional monitoring were reported for 75/198 (38%) of

patients due to potential DDIs with the DAA chosen. Renal monitoring for

tenofovir/ ledipasvir co-administration was reported in 22/198 (11%) of

patients. The monitoring was only required in 9/22 (40%) of those patients.

Impact of DDIs on ART and HCV DAA choice: • 36/198 (18%) required alteration to their HIV regimen prior to DAA therapy,

24/36 (66%) of which received Abbvie 2D/3D (ritonavir) based DAA.

• 6/198 (3%) required adaptation of HCV regimen due to current ART

regimen.

ARV Change N (%) of patients Omit ritonavir on Abbvie 2D/3D 13 (7) NNRTI to unboosted Integrase 13 (7) PI to unboosted Integrase 3 (1) PI changed 2 (1)

Boosted Integrase converted to unboosted Integrase 1 (1)

Other 4 (2) Nil change made/possible 162 (82)

Limitations: • Retrospective evaluation.

• DDIs with recreational drugs, including chems may be underrepresented.

0% 5% 10% 15% 20% 25% 30% 35% 40%

INI + 2NRTI

PI + 2NRTI

NNRTI + 2NRTI

Other PI based regimen

Other ART

Off ART

39% (78/198)

29% (58/198)

21% (42/198)

7% (13/198)

3% (5/198)

1% (2/198)

0% 10% 20% 30% 40% 50% 60%

Harvoni +/- ribavirin

Abbvie 3D +/- ribavirn

Sofosbuvir/PegIFN/Ribavirin

Sofosbuvir/Daclatasvir +/- ribavirin

Abbvie 2D +/- ribavirin

Other

55% (109/198)

17% (33/198)

17% (33/198)

6% (11/198)

5% (9/198)

1% (2/198)

Figure 1: ART regimen co-prescribed with HCV therapy

Figure 2: HCV regimens

Characteristics Outcome Results N=198

Male sex N (%) 177 (89)

Age, years Median 49 years

HCV genotype

1 2 3 4

157 (81) 2 (1) 17 (8)

21 (10)

Cirrhotic N (%) 84 (42)

Prior HCV non-responder N (%) 95 (48)

Co-medicines: Total observed

Number per patient

N (%)

Median

728 3.5

Table 1: Baseline characteristics

Table 2: Changes to HIV ARVs to accommodate HCV therapy

1. Arends et al. Natural history and treatment of HCV/HIV coinfection: Is it time to change paradigms? J Hepatol. 2015 Nov;63(5):1254-62

2. http://www.hep-druginteractions.org/

Background:

• Women in Sub-Saharan Africa are increasingly using the etonogestrel (ENG) implant for contraception.

• Commonly used antiretrovirals, efavirenz (EFV) and nevirapine (NVP) are cytochrome P450 (CYP) inducers; ENG is metabolized by CYP3A4.

• We characterized ENG pharmacokinetics (PK) in HIV-positive Ugandan women receiving ENG implants with EFV- or NVP-based ART, or no ART.

• Also, we compared EFV and NVP concentrations before and after ENG implant insertion.

Efavirenz significantly decreases etonogestrel exposure: results of a bidirectional pharmacokinetic

evaluation of efavirenz- and nevirapine-based antiretroviral therapy plus etonogestrel contraceptive implant

Acknowledgements: This work was funded by the Society for Family Planning Research. C. Chappell is supported by NIH/NICHD K12HD043441 Building Interdisciplinary Research Careers in Women’s Health. We appreciate all the participants for volunteering their time. Corresponding author (C. Chappell) contact: chappellca@upmc.edu

Background and Methods

Results

Conclusions

Table 2. ENG Plasma Concentrations Over 24 Weeks (pg/mL)1

Table 1. Demographic Characteristics by Study Group1

Poster # P302

1Department of Obstetrics, Gynecology and Reproductive Sciences, University of Pittsburgh, 2Magee-Womens Research Institute, 3Department of Pharmacy Practice, College of Pharmacy, University of Nebraska Medical Center, 4Infectious Diseases Institute, Makerere University College of

Health Sciences, 5Division of Infectious Diseases, Department of Medicine, University of Pittsburgh, 6Division of Infectious Diseases, Feinberg School of Medicine, Northwestern University

• Over 24 weeks of combined use, ENG exposure was 84% lower in women using EFV-based ART compared to ART-naïve

women. No significant influence of NVP-based ART on ENG exposure was observed.

• Similar to other studies, these findings highlight significant concern for reduced contraceptive effectiveness of the ENG

implant in HIV-infected women on EFV-based ART.

• Although statistically lower EFV concentrations were observed after ENG insertion, all participants in the EFV group had

therapeutic concentrations at week 24.

Catherine A. Chappell1,2, Kimberly K. Scarsi3, Shadia Nakalema4, Beatrice A. Chen1,2, Sharon A. Riddler5, Susan E. Cohn6, Kristin M. Darin6, Sharon L. Achilles1,2, Mohammed Lamorde4

Methods:

• Non-randomized, parallel-group, 3-arm study:

• ART-naïve, EFV-, or NVP-based ART (on stable ART with

undetectable HIV-RNA at enrollment)

• The women on EFV-based ART had an IUD

in place at enrollment.

• Sparse PK sampling (ENG, EFV, NVP) was collected per Figure 1.

• Plasma was collected 12-14 h post-EFV or 11-13 h post-NVP dose.

• ART and ENG concentrations were quantitated using validated

HPLC and HPLC-MS/MS methods, respectively.

64 participants screened

ART naïve: 20 women

NVP-based ART: 22 women

EFV-based ART: 22 women

20 women enrolled

20 women completed

primary endpoint

20 women completed

primary endpoint

20 women completed

primary endpoint

20 women enrolled

20 women enrolled

-1 HIV RNA >400 -1 concern for pregnancy

-2 HIV RNA >400

Fig. 2: Flow Diagram

• Screening occurred between September 2014 and July 2015

Table 3. Geometric mean (GM) of trough NVP and EFV serum concentrations before and after ENG implant insertion Time from implant insertion

NVP (mg/L) GM (90%CI)

EFV (mg/L) GM (90%CI)

Pre-insertion 6.9 (6.9-7.8) 4.7 (3.6-5.8)

Week 4 6.7 (5.6-7.8) 4.0 (2.6-5.5)

Week 12 6.1 (4.9-7.3) 3.7 (2.8-4.7)

Week 24 5.7 (4.7-6.8) 3.6 (2.6-4.5)

GM Ratio1 0.83 (0.78-0.88) 0.75 (0.71-0.79)

P-Value2 0.227 0.009 1GM ratio of pre-insertion:week 24 2Wilcoxon signed-rank test comparing pre-insertion to week 24

Study week

ART-Naïve (N=20)2

NVP (N=19)2

EFV (N=19)2

NVP:ART-Naïve3

EFV:ART-Naïve3

1 831

(738, 924) 647

(548, 746) 137

(118, 156) 0.78

(0.74-0.81) 0.16

(0.16-0.17)

4 507

(445, 568) 457

(394, 521) 87

(74, 99) 0.90

(0.88-0.91) 0.17

(0.17-0.17)

12 441

(382,500) 453

(398, 509) 63

(55, 71) 1.03

(1.02-1.04) 0.14

(0.14-0.14)

24 362

(308, 415) 341

(310, 373) 66

(60, 72) 0.94

(0.90-1.01) 0.18

(0.17-0.20)

AUC4 11,120 10,470 1,800 0.94

(0.94-0.94) 0.16

(0.16-0.16)

1Compared by Wilcoxon signed-rank test 2Presented as geometric mean (GM) with 90% confidence intervals 3Presented as geometric mean ratio (GMR) with 90% confidence intervals 4GM area under the curve (AUC) from 0 to 24 weeks (pg*wk/mL)

19 evaluable participants

19 evaluable participants

20 evaluable participants

-1 processing error

-1 Nonadherence

Characteristic ART-Naïve (N=20)

NVP-Based ART

(N=19)

EFV-Based ART

(N=19) Age (year) 27.5 (25.0-30.0) 32.0 (28.0-34.5) 29.0 (25.0-34.0)

Weight (kg) 65.5 (55.3-71.5)* 56.0 (52.5-69.5) 57.0 (48.5-59.5)†

BMI (kg/m2) 25.1 (21.5-28.8) 22.1 (21.3-26.5) 22.3 (20.7-23.6)

Married, n (%) 8 (40%)* 14 (73.7%)† 14 (73.7%)†

Prior Live Births 2.0 (1.75-3.0) 3.0 (2.0-4.0) 3.0 (2.0-4.0)

CD4 Count (cells/mL)

884 (689.5-1124.8)*

552 (438.5-718.5)†

549.0 (378.0-990.0)†

Duration on ART (Months) ----- 32.1 (27.0-53.6)* 23.3 (17.8-25.5)†

0

5

10

15

20

0

5

10

15

20

Fig.3. Trough NVP and EFV serum concentrations over study (mg/L)

NVP EFV

1Presented as either n(%) or median (interquartile range), as appropriate. * p<0.05 when compared by Student’s T-test or Chi-Square tests to the group noted by †

Dolutegravir plasma concentrations according to companion antiretroviral drug: unwanted drug interaction or desirable boosting effect?(P303)

Dario Cattaneo1, Davide Minisci2, Valeria Cozzi1, Agostino Riva2 Paola Meraviglia2, Emilio Clementi1,3, Massimo Galli2,3 and Cristina Gervasoni2

1Unit of Clinical Pharmacology and 2Department of Infectious Diseases and, ASST Fatebenefratelli Sacco, Milan, Italy. 3Università degli Studi di Milano, Milan, Italy

Background and Objectives

Main pharmacologic advantages of dolutegravir compared with other integrase inhibitors are its suitability for once-daily administration, no need for pharmacokinetic boosting, a high barrier to resistance and its modest drug-to-drug interaction profile.

Most of pharmacologic information of dolutegravir derive

from studies conducted in healthy volunteers; data in real life settings are, therefore, lacking. Here we carried out a pharmacokinetic survey in HIV-infected patients given dolutegravir as part of their antiretroviral therapy.

Methods

In this retrospective analysis we included consecutive HIV-infected patients from our local database treated with dolutegravir for at least one month with a request for therapeutic drug monitoring of dolutegravir plasma trough concentrations.

Collected blood samples had to be taken 24 hours after

the last drug intake (a time window of 20 min was considered acceptable), immediately before drug administration, to ensure that these samples can be considered as trough concentrations.

Drug concentrations were assessed by high performance liquid chromatography method with UV-detection previously developed in our lab.

The main demographic and clinical data were recorded for

each patient. Comparisons between dolutegravir trough concentrations clustered according to the companion antiretroviral drug were performed by ANOVA.

Results 1: demographic characteristics

Patients (n) 80 Males (%) 63% Caucasians (%) 93% Age (years) 52 [45-57]* Naïve to antiretroviral treatment (%) 16% CD4 cell count (cells/mL) 573 [450-795]* Patients with viral load >37 copies/mL (n) 2/80 Body weight (Kg) 72 [64-79]* Serum alanine aminotrasferase (IU/mL) 31 [21-41]* Serum creatinine (mg/dL) 0.9 [0.8-1.1]* *Data were given as median [interquartile range]

Results 2: concomitant therapy

Conclusions

Atazanavir coadministration greatly inhibited dolutegravir metabolism, ultimately resulting in a 2- to 4-fold increase in drug disposition.

This drug-to-drug interaction – related to the atazanavir-

mediated inhibition of UDP-glucuronisyltransferase 1A1, the main enzyme involved in the metabolism of dolutegravir – might become potentially relevant in all clinical conditions which require higher than conventional dolutegravir exposure.

At variance with other NNRTIs, rilpivirine coadministration

resulted in dolutegravir concentrations comparable to those measured in patients given darunavir, or abacavir/emtricitabine suggesting negligible inducing effect of this drug on dolutegravir metabolism.

Comedications Patients

(n) Dolutegravir

levels (ng/mL)*

Antiretrovirals Abacavir/emtricitabine 12 1045 [856-1115] Atazanavir (85% at 400 mg qd) 26 2399 [1929-4070] Darunavir (800/100 mg qd) 26 756 [556-1048] Efavirenz 1 58 Etravirine 3 25, 182, 931 Rilpivirine 12 603 [432-1373] Qd: once daily; *Data were given as median [interquartile range]

Dolutegravir plasma trough concentrations clustered according to the companion antiretroviral drug

0

1600

3200

4800

6400

[Dol

uteg

ravi

r], n

g/m

L

Rilpivirine Darunavir Atazanavir

**

Abacavir/ emtricitabine

**p<0.001 versus other groups. Dashed lines depict the protein-adjusted 90% inhibitory concentration for wild-type and resistant virues (64 and 640 ng/mL, respectively)..

Determinants of Dolutegravir Plasma Concentrations in the Clinical Setting

Alcantarini C, Calcagno A, Marinaro L, Ferrara M, Milesi M, Trentalange A, Barco A, Montrucchio C, De Nicolò A,

Ariaudo A, Favata F, D’Avolio A, Di Perri G, Bonora S.

Unit of Infectious Diseases, Department of Medical Sciences, University of Torino, Torino, Italy

P 304

Background

Dolutegravir (DTG) is the latest available integrase strand transfer

inhibitor.[1]

It is primarily metabolized via UGT1A1 (with CYP 3A4 as a minor pathway) and it is substrate of p-glycoprotein.[2]

Few drug-drug interactions have been observed but data on DTG pharmacokinetics in the clinical setting are limited.[3,4]

Results

363 samples were available from 149 patients (median 1, range 1-19 per patient).

Median age and body mass index were 49.3 years (46.4-54.5) and 24.2 Kg/m2 (20.8-27.7); 102 patients (68.4%) were male and 50 (33.5%) were HCV-positive.

Samples were withdrawn 22.5 hours (10.8-24.2) after drug intake [198 (54.5%) were trough values] and DTG median concentrations were 1107 ng/mL (399-2549).

324 (91.3%) and 31 (8.7%) samples were from patients on once-daily or twice-daily DTG: trough values were 660 ng/mL (255-1237) and 2674 ng/mL (1000-3474), respectively.

Inter-patient variability was high (102%) and lower in patient on twice-daily DTG (56.9% vs. 108%); intra-patient variability (calculated in 10 patients with >4 trough samples, all on once-daily DTG) was 64.7%.

A significant correlation was observed between DTG concentrations and age (rho 0.21 and p<0.001, rho 0.58 and p<0.001 considering Cmax) (figure 1).

Higher concentrations were observed in patients on atazanavir (2321 vs. 922 ng/mL, p<0.001), no significant differences in darunavir/r intakers as compared to the others (figure 2), while border-line lower in those on valproic acid (n=7, 829 vs. 1132 ng/mL, p=0.08) (figure 3).

At multivariate linear regression analysis age, post-dose time, atazanavir use (p<0.001) and, border-line, valproic acid use (p=0.06) were independent predictors of DTG concentrations explaining approximately 46% of its variability.

Conclusions DTG PK showed significant variability; in the majority of patients plasma concentrations were in the expected range of efficacy;

A significant DTG higher exposure was observed with concomitant atazanavir and in older patients while darunavir administration had no impact on DTG exposure; a population pharmacokinetic study, with the inclusion of pharmacogenetic variants, is ongoing;

The unexpected trend towards lower DTG plasma concentrations in valproic acid intakers warrant confirmation in formal interaction studies.

The Torino Therapeutic Drug Monitoring (TDM) registry was used and patients on DTG, with fully available data (demographic, time after dose, concomitant medications) were included; patients on rifampin were excluded.

Data are described as medians (interquartile ranges) and analysed through non-parametric tests.

A multivariate linear regression analysis was performed including variables with p-values below 0.10 at univariate tests.

Figure 3 Dolutegravir concentrations with valproic acid Figure 1 Association between dolutegravir concentrations and age

Materials and Methods

Figure 2 Dolutegravir concentrations with Protease Inhibitors

r = 0.21 p<0.001

Correspondence to: Dr. Chiara Alcantarini Amedeo di Savoia Hospital Tel: +390114393980 chiara.alcantarini@gmail.com

References 1. Rathbun RC et al. Dolutegravir, a second-generation integrase inhibitor for the treatment of HIV-1 infection. Ann Pharmacother. 2014 Mar;48(3):395-403. 2. ViiV Healthcare. Tivicay prescribing information available at: http://www.accessdata.fda.gov/drugsatfda_docs/label/2013/204790lbl.pdf 3. Dailly E et al. Influence of nevirapine administration on the pharmacokinetics of dolutegravir in HIV-1 infected patients. 16th International Workshop on Clinical Pharmacology of HIV & Hepatitis Therapy May 2015, Washington DC, USA [abstr. 64] 4. Le MP et al. Dolutegravir drug interaction with DRV/r or ATV/r: impact on its pharmacokinetic? 16th International Workshop on Clinical Pharmacology of HIV & Hepatitis Therapy, May 2015, Washington DC, USA [abstr. 75]

• No associations were observed between DRV concentration and sex, age, ethnicity or weight. We did not see lower DRV concentrations in the subgroups with baseline CD4 <200 cells/μL or VL ≥100,000 copies/mL. Higher DRV concentrations were associated with both higher RTV and (in RAL arm) higher RAL concentration (p<0.001).

• No association was found between DRV concentration (modelled Ctrough) at W4 and virological failure at or after W32 in analyses of both arms together or separately for each arm.

Table 1: Darunavir (DRV) model parameter estimates and relative standard errors. Table 2: Secondary darunavir (DRV) pharmacokinetic (PK) parameters stratified by randomisation arm [geometric mean (95% confidence intervals, CI)]. Differences in DRV PK parameters in the raltegravir (RAL) arm vs. tenofovir disoproxil fumarate/emtricitabine (TDF/FTC) arm are expressed as geometric mean ratio (GMR) (90% CI). Figure 1: Concentration-time data of (a) darunavir (DRV) and (b) ritonavir (RTV) at week 4 and 24 following treatment initiation (log-linear scale; n=722 patients, n=1319 concentrations). A) B) Conclusions • DRV exposure was not affected by age, gender, weight or ethnicity but showed a significant association

with RTV and RAL concentrations.

• DRV concentrations decreased by 10% in presence of RAL as compared to TDF/FTC: although statistically significant this slight difference is unlikely to have clinical significance.

• There was no evidence of an association between DRV concentrations and virological failure at or after

W32.

• Further analysis is on-going to confirm the lack of a relationship between RAL exposure and virological failure and to determine DRV and RAL MEC.

References: 1. Raffi et al. Lancet. 2014 Nov 29;384(9958):1942-51; 2. Molto et al. Clin Pharmacokinet 2013; 52 (7): 543-53; 3. Arab-Alameddine et al. J Antimicrob Chemother 2014; 69 (9): 2489-98

Background • Limited prospective pharmacokinetic (PK) data are available on DRV/RTV once-daily (OD) and raltegravir

(RAL) twice-daily (BD) in antiretroviral (ARV)-naïve HIV-infected individuals.

• The NEAT 001/ANRS 143 trial is a phase III, open-label, randomized, non-inferiority trial comparing the efficacy and safety of two first-line regimens in HIV-infected antiretroviral-naïve subjects over at least 96 weeks: DRV/RTV+ tenofovir/emtricitabine (TDF/FTC) versus DRV/RTV+ RAL [1].

• NEAT 001/ANRS143 96 weeks (W)-randomised study demonstrated non-inferiority of first-line ARV therapy with DRV/RTV (800/100mg OD) plus RAL (400mg BD) compared with DRV/RTV plus TDF/FTC (245/200mg OD) [1].

• However, higher failure rates in the RAL arm were seen in those with low CD4 counts and high viral load (VL) at baseline.

• We here present the population PK analysis performed in the NEAT001/ANRS143 study. Objectives To investigate the relationship between plasma concentrations of ARV drugs and: 1. Concomitant drug: DRV/RTV in the presence of TDF/FTC versus RAL (presented here) 2. Patient’s characteristics (age, sex, ethnicity, weight/height/BMI) (presented here) 3. Virological failure (presented here) 4. CD4 change from baseline (on-going) 5. Adverse events (on-going) 6. Adherence (on-going) 7. Drug disposition genes (poster P308) 8. To define the DRV and RAL minimum effective concentrations (MEC) or other threshold parameter for

prediction of virological failure (on-going) Methods • Virological failure as defined as the first occurrence of any of the following three components: i) change of

any component of the initial randomised regimen before week 32 because of documented insufficient virologic response, HIV RNA reduction <1 log10 copies/mL by week 18, HIV RNA ≥400 copies/mL at week 24 (confirmed by a subsequent measurement); ii) failure to achieve virologic response by week 32 (defined as HIV RNA ≥50 copies/ml at week 32, confirmed by a subsequent measurement); iii) HIV RNA ≥50 copies/ml (confirmed by a subsequent measurement) at any time after week 32.

• Blood for PK analysis of study drugs (TFV, FTC, DRV, RTV, and RAL) was collected 4 and 24 W after ARV initiation.

• Nonlinear mixed effects modelling was applied to W4 and W24 (single random samples) DRV

concentration-time data simultaneously (NONMEM v.7.3) to estimate PK parameters for each patient at each sampling occasion (AUCτ, Cmax, Cτ). Samples were excluded if time post-dose was missing or > 30 hours and if DRV or RTV concentrations were below the lower limits of the bioanalytical assay.

• The impact of age, weight, sex, ethnicity, randomisation arm (RAL vs. TDF/FTC as the reference arm), and time matched RTV concentration on DRV apparent oral clearance (CL/F) was also investigated.

• We also examined if DRV concentration was lower in patients with CD4<200cells/μL or VL≥100,000copies/mL at baseline. Cox regression was used to associate W4 drug levels with virological failure at or after W32 (defined as confirmed VL ≥50copies/mL), adjusting for baseline VL.

Results Study population • Seven hundred and twenty-two patients (89% male, 83% Caucasian) were included in the analysis

contributing 1319 DRV and RTV concentrations (n=671 week 4, n=648 week 24).

• Fifty percent of patients were randomised to both the RAL arm and TDF/FTC arm.

• Median (range) age and weight were 37 years (18-76) and 72 kg (41-135), respectively.

• The majority of patients [704/722 (98%)] received DRV/RTV 800/100 mg OD. Alternative doses included 400/100 mg (n=8), 600/100 mg n=1), 1200/100 mg (n=1), 1600/100 mg (n=8) all OD and 400/100 mg BD (n=1).

PK modelling • DRV and RTV concentrations over time for W4 and W24 are shown in Figure 1.

• DRV PK was described by a 1-compartment oral model with absorption rate constant (ka) fixed to a

literature value of 1.04 h-1 [2,3].

• Following multivariate analysis time-matched RTV concentration and randomisation arm were significantly associated with DRV CL/F (Table 1).

• DRV CL/F was increased by 10% in the presence of RAL compared to TDF/FTC. Model parameters are summarised in Table 1.

• Secondary PK parameters (AUCτ, Cmax and Cτ) stratified by randomisation arm are presented in Table 2.

• AUCτ, Cmax and Cτ were 11, 6 and 14% lower in the RAL arm (Table 2).

Pharmacokinetics (PK) of darunavir/ritonavir (DRV/RTV) with tenofovirDF/emtricitabine (TDF/FTC) or raltegravir (RAL) in HIV-infected adults

enrolled in the NEAT001/ANRS143 study and relationship with virologicalresponse.

Stefano Bonora1; Laura Dickinson2; Wolfgang Stöhr3; Antonio D'Avolio1; Adam Cursley3; Jean-Michel Molina4; Gerd Faetkenheuer5; Linos Vandekerckhove6; Giovanni Di Perri1; Anton Pozniak7; Laura Richert8; Francois Raffi9; Marta Boffito7,10

1University of Turin Unit of Infectious Diseases Turin Italy; 2University of Liverpool, UK; 3University College London Medical Research Council London United Kingdom; 4University of Paris Diderot Infectious Diseases Paris France; 5University Köln Unit of Internal

Medicine Koln Germany; 6Ghent University and Ghent University Hospital HIV Translational Research Unit Ghent Belgium; 7Chelsea and Westminster NHS Trust St Stephens Centre London United Kingdom; 8University of Bordeaux INSERIM ISPED Bordeaux France;9Nantes University Hospital Infectious and Tropical Diseases Nantes France; 10Imperial College, London

Poster N 305 - HIV Glasgow, 23-26 October 2016, Glasgow, UK

0.01

0.1

1

10

100

0 5 10 15 20 25 30 35

DR

V m

g/L

Time (h)

DRV WK4 DRV WK24

0.001

0.01

0.1

1

10

100

0 5 10 15 20 25 30 35

RTV

mg/

L

Time (h)

RTV WK4 RTV WK24

Evaluation of the Drug Interaction Potential between Elvitegravir/Cobicistat/Emtricitabine/Tenofovir Alafenamide

and AtorvastatinJoseph M. Custodio, Steve K. West, Devi SenGupta, Arian Zari, Rita Humeniuk,

Kah Hiing J. Ling, Marshall Fordyce, Brian P. KearneyGilead Sciences, Inc., Foster City, CA 94404 USA

• An increase in AVA exposures was observed following coadministration with E/C/F/TAF• AVA (10 mg single dose) did not result in changes in the PK of the components of E/C/F/TAF• ThesestudyfindingsprovideguidanceontheconcomitantuseofE/C/F/TAFandAVAinpatientsinitiatingor

currently on AVA therapy• All study treatments were well tolerated

� Elvitegravir/cobicistat/emtricitabine/tenofovir alafenamide (EVG/COBI/FTC/TAF (150/150/200/10 mg; E/C/F/TAF; Genvoya™) is a single tablet regimen (STR) approved for treatment of HIV infection1

– COBI, a pharmacokinetic (PK) enhancer in E/C/F/TAF, is an inhibitor of the metabolizing enzyme CYP3A, and the transporters P-gp, BCRP, and OATP1B1/1B31

� Atorvastatin (AVA), a 3-hydroxy-3-methylglutaryl-coenzyme A (HMG-CoA) reductase inhibitor, is a generically available, commonly prescribed statin for lipid-lowering in HIV-infected individuals2

– AVA is a substrate of the enzyme CYP3A and is extensively metabolized• The CYP3A-mediated metabolites 2-hydroxy-AVA (2-OH-

AVA; orthohydroxylated) and 4-hydroxy-AVA (4-OH-AVA; parahydroxylated) are both active

– AVA is a substrate of the transporters P-gp, BCRP, and OATP1B1/1B32

– AVA has limited liability as a perpetrator of drug drug interactions (DDI)2,3

� The risk of skeletal muscle adverse events (e.g., myopathy) is increased during concomitant use of statins with strong CYP3A inhibitors2,4

� Accordingly, the prescribing information of AVA contains cautionary dosing recommendations regarding various concomitant medications that are strong CYP3A inhibitors such as HIV protease inhibitors

– Darunavir/ritonavir: Do not exceed AVA 20 mg/day5

– Atazanavir/ritonavir: Titrate AVA dose carefully and use the lowest dose6

– Lopinavir/ritonavir: Use AVA with caution and at the lowest necessary dose7

� All treatments were well tolerated and all subjects completed the study

� No clinically relevant laboratory abnormalities related to study treatments were observed

� The majority of AEs were mild (Grade 1) in severity and no Grade 3 or 4 AEs or serious AEs were observed

� Coadministration of E/C/F/TAF + AVA resulted in a 2.6-fold and 2.3-fold increase in AVA AUC and Cmax, respectively – This observed DDI was of lesser magnitude than lopinavir/ritonavir + AVA (5.9-fold and 4.7-fold increase in AVA AUC and Cmax,

respectively2) � The increase in AVA exposures following coadministration with E/C/F/TAF were consistent with the inhibition of intestinal Pgp and/or

BCRP and inhibition of the CYP3A-mediated metabolism of AVA by COBI � The steady state exposures of EVG, COBI, TAF, and TFV were consistent with historical data following administration of E/C/F/TAF

� The primary objective of this study was to evaluate the effect of administration of E/C/F/TAF on the PK of AVA and quantify the magnitude of the DDI

� The secondary objective was to evaluate the safety and tolerability of administration of E/C/F/TAF and AVA, as single agents and in combination

� Compared with administration of AVA alone, E/C/F/TAF + AVA led to increases in the exposures of AVA

� Higher AVA exposures following E/C/F/TAF + AVA were accompanied by a marked decrease in 2-OH-AVA, a CYP3A-mediated metabolite of AVA

– 2-OH-AVA concentrations were undetectable in the majority of subjects following coadministration of E/C/F/TAF + AVA

� Coadministration of E/C/F/TAF + AVA, did not affect the PK of EVG

� Mean exposures of the 2-OH-AVA metabolite were ~90% lower following E/C/F/TAF + AVA vs AVA alone

– Only 2-OH-AVA is presented because 4-OH-AVA was undetectable in the majority of subjects following AVA alone and all subjects following coadministration E/C/F/TAF + AVA

� Coadministration of E/C/F/TAF + AVA, did not affect the PK of TAF, or its metabolite TFV

� PhaseI,open-label,fixed-sequence,threeperiod,singlecenterstudy in healthy subjects

� Study treatments were administered in the morning with 240 mL water, within 5 minutes of completion of a standardized meal, following a 10 hour overnight fast

� Intensive PK assessments performed on Days 1, 13 and 14 – Plasma concentrations were determined using validated liquid

chromatography-tandem mass spectrometry (LC/MS/MS) assays � PK parameters estimated using noncompartmental methods and

WinNonlin 6.4 (Pharsight Corp., Sunnyvale, California, USA) � Statistical comparisons for EVG, COBI, TAF, TFV, AVA and

its metabolites (2-OH-AVA and 4-OH-AVA) were made using geometric least-squares mean ratios (GMRs), associated 90% confidenceintervals(CIs),andano-effectboundaryof70–143%(EVG, COBI, TAF, TFV AUC, Cmax and Ctau; AVA AUC) and 50-200% (AVA Cmax),

� Safety (adverse events [AEs]; laboratory tests) assessed throughout dosing and follow-up period (7 ± 2 days after last treatment)

Study Design

Demographics

© 2016 Gilead Sciences, Inc. All rights reserved.HIV Glasgow, 23-26 October 2016, Glasgow, UK

Gilead Sciences, Inc.333 Lakeside Drive

Foster City, CA 94404800-445-3252

Introduction

Safety

Results

Methods

Objectives

Results

Conclusions

Discussion

References1. Genvoya USPI and SmPC2. Lipitor USPI and SmPC3. Lennernäs, H. Clin Pharmacokinet. 2003.4. FDA Drug Safety Communication (Issued March 1, 2012): Interactions between certain HIV or hepatitis C drugs and

cholesterol-lowering statin drugs can increase the risk of muscle injury

Poster number

P306

Passcode: P306

Period 1 Period 2 Period 3Day 1 Days 2-3 Days 4-13 Day 14

AVA (10mg) Washout E/C/F/TAF E/C/F/TAF+AVA (10mg)

Subject DispositionEnrolled/Completed, n 16/16Mean age, y (range) 34 (20-44)Sex, n (%) Male 14 (88) Female 2 (12)Weight , kg (range) 82 (71-97)Race/ethnicity, n (%) Black 9 (56) White 7 (44) Hispanic or Latino 6 (38)

AVA PK Parameter Mean (%CV)

E/C/F/TAF+AVAN=16

AVA N=16 GMR (90% CI)

AUCinf (h*ng/mL) 38.4 (32) 14.6 (27) 260 (231, 293)

AUClast (h*ng/mL) 34.9 (33) 12.0 (30) 291 (263, 323)

Cmax (ng/mL) 3.0 (51) 1.3 (56) 232 (191, 282)

TAF PK Parameter Mean (%CV)

E/C/F/TAF + AVAN=16

E/C/F/TAF N=16 GMR (90% CI)

AUClast (h*ng/mL) 226 (34) 224 (33) 99.9 (83.7, 119)

Cmax (ng/mL) 184 (58) 183 (40) 96.6 (72.8, 128)

TFV PK ParameterMean (%CV)

E/C/F/TAF + AVAN=16

E/C/F/TAFN=16 GMR (90% CI)

AUCtau (h*ng/mL) 265 (12) 269 (12) 98.6 (96.6, 101)

Cmax (ng/mL) 15.8 (14) 16.7 (25) 96.0 (88.5, 104)

Ctau (ng/mL) 9.6 (14) 9.5 (14) 101 (99.1, 104)

EVG PK Parameter Mean (%CV)

E/C/F/TAF + AVAN=16

E/C/F/TAF N=16 GMR (90% CI)

AUCtau (h*ng/mL) 21300 (24) 23300 (31) 92.5 (87.1, 98.2)

Cmax (ng/mL) 1770 (18) 1970 (29) 91.1 (84.8, 98.0)

Ctau (ng/mL) 387 (48) 442 (55) 88.2 (80.8, 96.3)

2-OH-AVA PK ParameterMean (%CV)

E/C/F/TAF +AVAN=16

AVA N=16 GMR (90% CI)

AUCinf (h*ng/mL) NC 17.1 (30) ‒

AUClast (h*ng/mL) 0.9 (170) 13.4 (33) 9.84 (6.33, 15.3)

Cmax (ng/mL) 0.1 (113) 1.1 (32) 12.2 (9.45, 15.7)

5. Prezista USPI and SmPC6. Reyataz USPI and SmPC7. Kaletra USPI and SmPC

AVA

EVG

2-OH-AVA

TAF

TFV

Pharmacokinetics

� Coadministration of E/C/F/TAF + AVA, did not affect the PK of COBI

COBI PK Parameter Mean (%CV)

E/C/F/TAF + AVAN=16

E/C/F/TAFN=16 GMR (90% CI)

AUCtau (h*ng/mL) 9700 (32) 9710 (34) 100 (96.3, 104)

Cmax (ng/mL) 1430 (26) 1380 (23) 103 (98.2, 107)

Ctau (ng/mL) 39.7 (85) 34.3 (92) 110 (98.5, 123)

COBI

Time (h)

Con

cent

ratio

n(n

g/m

l)

0.1

1

10

E/C/F/TAF+AVAAVA

Time (h)

Con

cent

ratio

n(n

g/m

l)

0.001

0.01

0.1

1

10AVA

Time (h)

Con

cent

ratio

n(n

g/m

l)

100

1000

10000

E/C/F/TAF+AVAE/C/F/TAF

Time (h)

Con

cent

ratio

n(n

g/m

l)

1

10

100

1000

10000

E/C/F/TAF+AVAE/C/F/TAF

Time (h)

Con

cent

ratio

n(n

g/m

l)

1

10

100

E/C/F/TAF+AVAE/C/F/TAF

Time (h)

Con

cent

ratio

n(n

g/m

l)

0.1

1

10

100

1000E/C/F/TAF+AVAE/C/F/TAF

NC: Not calculable because the terminal phase of 2-OH-AVA could not be estimated following coadministration of E/C/F/TAF+AVA

Figure 1: Geometric mean steady state plasma concentrations of darunavir andatazanavir when boosted by 150 mg of cobicistat over 72 hours.

Figure 2: Geometric mean steady state plasma concentrations of darunavir andatazanavir when boosted by 150 mg of cobicistat over 72 hours AND by 100 mg ofritonavir [4].

Figure 3: Cobicistat (COBI), (a) atazanavir (ATV) and (b) darunavir (DRV) urineconcentration decay between 24 and 72 hours post-dose expressed as geometricmean and 95% confidence intervals (95%CI; dashed lines).

(a)

(b)

Conclusions

• This study investigated the PK forgiveness of two cobicistat-PI/b in plasma.

• Different concentration decay patterns and relationships with cut-offs used to define therapeutic concentrations were seen for darunavir and atazanavir, which may be partially explained by cobicistat the t1/2 (shorter with darunavir than atazanavir).

• For the first time, we also measured drug PK forgiveness in urine, which can be an easier marker of adherence.

• Further clinical data are warranted to inform the differential impact of delayed or missed doses with cobicistat boosted PIs with different PK tails.

Background

Ritonavir-boosted protease inhibitors (PI/b) such as atazanavirand darunavir are an instrumental option as third agents in the management of HIV [1].

Advantages of pharmacological boosting include increased drug exposure and a prolonged half-life thereby reducing pill burden, allowing once daily dosing, and in the case of PIs, achieving a high genetic barrier to resistance [2].

The use of ritonavir as a boosting agent, however, presents a number of disadvantages, and cobicistat, a structural analogue without antiviral activity, is now available as an alternative PKenhancer and has been co-formulated with PIs [3].

Although cobicistat inhibits CYP3A4 with a potency similar to that of ritonavir, it is characterized by the lack of enzyme-inducing activity [3].

At the 150 mg dose, cobicistat provides bioequivalent exposures of atazanavir (300 mg once daily) and darunavir (800 mg once daily) compared with those observed with 100 mg of ritonavir once daily [3].

PK forgiveness is important in clinical practice in order to understand the management of late and missed doses, particularly with PIs as their use is increasingly targeted to complex cases of viral resistance, poor adherence and extensive ART experience.

Drug persistence in plasma is dependent on its t1/2 (which itself depends on CL and V). As such, antiretroviral agents with longer t1/2 may be more forgiving and allow for forgotten doses, especially if drug concentrations remain therapeutic until the patient reinitiates drug intake.

Objectives

We previously presented data on the PK forgiveness of once daily ritonavir-boosted darunavir and atazanavir, showing, in particular, a favorable atazanavir PK tail and a slight increase in decline rate for both PIs as ritonavir concentrations decrease [4].

The objectives of this study were therefore to investigate the steady-state PK of darunavir/cobicistat and atazanavir/cobicistatonce daily dosing over 72 hours following drug intake cessation in plasma and in urine.

Methods

Healthy volunteers consented to taking part to the study in writingand when confirmed eligible, they received a fixed-dosecombination of atazanavir 300 mg + cobicistat 150 mg once dailyfor 10 days, followed by a 10 day washout period and then afixed-dose combination of darunavir 800 mg + cobicistat 150 mgonce daily for 10 days.

Full PK profiles were assessed for each phase for the 72 hoursfollowing day 10 (steady-state).

Drug concentrations were assessed by a validated LC/MSmethod at the University of Liverpool, UK. The limit ofquantification LLQ (plasma & urine) for all analytes was 10 ng/ml.Concentrations <LLQ were expressed as half LLQ values.

PK parameters were determined over 24 hours in plasma and tothe last measurable concentration (Clast) in plasma and urine (24-72 hours post-dose) by non-compartmental methods(WinNonlin).

Results

Study population

Sixteen volunteers completed all phases of the study.

Of the 16 subjects, median (range) age was 38 (20 to 54) years,and the median body mass index (BMI) was 25 (22 to 31) kg/m2.Nine were female, nine self-identified as Caucasians, six asblack, and two as Asian.

Treatment was generally well tolerated, and no serious adverseevents occurred during the study. No clinically relevant changesin laboratory parameters were reported.

Pharmacokinetics

Geometric mean (GM) plasma concentration versus time curvesfor atazanavir and darunavir when combined with cobicistat andwith ritonavir are shown in Figure 1 and 2 and PK parameters aresummarized in Table 1. Table 2 shows the drug concentrationsmeasured 24, 30, 36, 48 hours post dose and the number ofsubjects below target per time-point.

The suggested minimum effective concentration for HIV isolatesfor atazanavir is 150 ng/mL (equivalent to 10 times the proteinbinding corrected minimum inhibitory concentration [IC50] forwild-type virus).

The suggested minimum effective concentration for protease-resistant HIV isolates for darunavir is 550 ng/mL (equivalent to10 times the protein binding corrected minimum IC50 for wild-typevirus).

Cobicistat PK parameters are illustrated in Table 3.

Although urine concentrations of atazanavir and darunavir arelower than those measured in plasma, urine PK profiles showedthe same concentration decay trends of plasma, suggesting thaturine concentrations at the end or later than the end of thedosing interval may be a promising marker of adherence/PKforgiveness (Figure 3).

Table 1: Darunavir and atazanavir steady state pharmacokinetic(PK) parameters, expressed as geometric mean and 95%confidence intervals, range and coefficient of variation, measuredover 24 and 72 hours.

For comparison, when boosted with ritonavir, the published GMterminal elimination t1/2 to 72 hours of atazanavir was 6.74 hoursand the t1/2 measured over the dosing interval of 24 hours was13.72 hours.

When boosted with ritonavir, the published GM terminalelimination the t1/2 to 72 hours of darunavir was 6.48 hours andthe t1/2 measured over the dosing interval of 24 hours was 10.70hours.

Table 2: Plasma concentrations of darunavir and atazanavirmeasured at 24, 30, 36, 48 hours post dose, expressed asgeometric mean (GM) and range, and number of subjects belowtarget per time-point.

Table 3: Cobicistat steady state plasma PK parameters,expressed as GM and 95% CI, range and coefficient of variation,measured over 24 and 72 hours with darunavir and atazanavir.

Steady-state pharmacokinetics (PK) of atazanavir/cobicistat and darunavir/cobicistat once daily over 72 hours in healthy volunteers: the

importance of PK forgiveness in clinical practice

Emilie Elliot1,2, Alieu Amara2, Nicole Pagani1, Laura Else2, Graeme Moyle1, Alex Schoolmeesters1, Chris Higgs1, Saye Khoo2, Marta Boffito1,3

1St Stephen’s Centre, Chelsea and Westminster Hospital, London, UK; 2University of Liverpool, Liverpool, UK; 3Imperial College, London, UK

Poster P307 - HIV Glasgow, 23-26 October 2016, Glasgow, UK

PK parameters ATV 300mg OD

t1/2 (0-24 h) t1/2 (0-Clast h) Cmax (ng/ml) C24 (ng/ml) Clast (ng/ml)AUC0-24

(ng.h/ml)AUC0-Clast(ng.h/ml)

Geomean 9.69 6.77 3718.85 759.20 6.36 37713.15 46128.91

low 95% 9.24 6.22 3308.00 612.57 1.29 32661.47 38592.12

up 95% 12.83 7.54 4940.55 1290.07 19.00 51555.93 67844.20

Min 6.32 5.42 844.97 256.10 5.00 11413.66 14057.77

Max 19.26 9.96 7282.82 2666.54 77.28 83763.28 128322.91

CV (%) 33 20 40 73 178 46 56

PK parameters DRV 800mg OD

t1/2 (0-24 h) t1/2 (0-Clast) Cmax (ng/ml) C24 (ng/ml) Clast (ng/ml)AUC0-24

(ng.h/ml)AUC0-Clast(ng.h/ml)

Geomean 10.41 6.35 5515.02 1032.56 8.80 58099.81 66710.08

low 95% 9.18 5.88 4949.07 837.92 6.01 51464.12 58145.46

up 95% 12.94 7.03 6566.03 1625.74 14.44 70391.27 83214.29

Min 5.23 4.25 2855.55 372.96 7.50 26404.49 29317.22

Max 19.15 8.48 8365.97 3359.34 41.13 111312.19 141982.09

CV 35 18 29 65 84 32 36

Hours post dose 24 30 36 48

Darunavir (ng/mL) GM (range) 1033 (373-3359) 381 (97-257) 109 (7.5-594) 45 (7.5-149)

N of subjects below target (550 ng/mL) 3/16 11/16 15/16 16/16

Atazanavir (ng/mL) GM (range) 759 (249-2667) 407 (148-1679) 201 (65-1093) 66 (14-949)

N of subjects below target (150 ng/mL) 0/16 2/16 5/16 13/16

PK parameters cobicistat 150 mg with darunavir

T1/2 (0-

24 h)

T1/2 (0-

Clast)

Tmax (h)

Tlast (h)

Cmax (ng/ml)

C24 (ng/ml)

Clast (ng/ml)

AUC0-24 (ng.h/ml)

AUC0-Clast (ng.h/ml)

Pre-dose (ng/ml)

24 Hr CL/F (L/h)

Clast Hr CL/F (L/h)

Geomean 3.81 3.62 2.54 32.25 1250.25 27.56 5.00 9532.06 9681.21 28.07 15.74 15.49

Low 95% 3.49 3.34 2.26 30.00 1149.77 22.29 5.00 8677.55 8790.87 21.41 13.30 14.05

Up 95% 4.29 3.98 3.12 35.25 1392.73 51.37 5.00 10857.17 11078.72 45.67 17.96 17.76

Min 2.59 2.59 1.00 24.00 932.46 5.00 5.00 6167.33 6254.42 10.88 10.40 10.04

Max 5.60 5.55 4.00 48.00 1867.32 120.90 5.00 14425.77 14933.25 74.28 24.32 23.98

CV (%) 21 18 32 16 20 81 0 23 23 61 23 24

PK parameters cobicistat 150 mg with atazananvir

T1/2 (0-

24 h)

T1/2 (0-

Clast)

Tmax (h)

Tlast (h)

Cmax (ng/ml)

C24 (ng/ml)

Clast (ng/ml)

AUC0-24 (ng.h/ml)

AUC0-Clast (ng.h/ml)

Pre-dose (ng/ml)

24 Hr CL/F (L/h)

Clast Hr CL/F (L/h)

Geomean 4.43 4.21 2.35 36.30 1408.02 49.59 5.00 10553.97 10923.56 41.64 14.21 13.73

Low 95% 3.95 3.87 2.10 33.74 1293.37 42.07 5.00 9589.47 9904.56 31.42 12.96 12.50

Up 95% 5.19 4.69 2.90 39.76 1577.76 79.63 5.00 12058.87 12535.27 74.21 16.18 15.70

Median 4.32 4.17 3.00 36.00 1381.49 56.35 5.00 10569.53 10735.04 37.28 14.41 14.22

Min 3.14 3.21 1.00 30.00 929.72 14.15 5.00 7825.70 8145.20 11.11 10.22 9.95

Max 8.39 6.13 4.00 48.00 1986.37 156.24 5.00 14680.79 15068.30 172.94 19.17 18.42

CV (%) 28 19 33 17 20 63 0 23 24 80 23 23

ritonavir [4].

PK parameters ATV 300mg ODPK parameters ATV 300mg OD

t1/2 1/2 1/2 (0(0-24 h)24 h) t1/2 1/2 1/2 (0(0-Clastlastlast h)h) Cmax max max (ng/ml)(ng/ml) C242424 (ng/ml)(ng/ml) Clastlastlast (ng/ml)(ng/ml)AUC0-24

(ng.h/ml)(ng.h/ml)AUC0-Clast((ng.hng.h/ml)/ml)

Geomean 9.69 6.77 3718.85 759.20 6.36 37713.15 46128.91

low 95% 9.24 6.22 3308.00 612.57 1.29 32661.47 38592.12

up 95% 12.83 7.54 4940.55 1290.07 19.00 51555.93 67844.20

Min 6.32 5.42 844.97 256.10 5.00 11413.66 14057.77

Max 19.26 9.96 7282.82 2666.54 77.28 83763.28 128322.91

CV (%)CV (%) 33 20 40 73 178 46 56

PK parameters DRV 800mg ODPK parameters DRV 800mg OD

t1/2 1/2 1/2 (0(0-24 h)24 h) t1/2 1/2 1/2 (0(0-Clastlastlast)) Cmaxmaxmax ((ngng/ml)/ml) C242424 (ng/ml)(ng/ml) Clastlastlast (ng/ml)(ng/ml)AUC0-24

(ng.h/ml)(ng.h/ml)AUC0-Clast(ng.h/ml)(ng.h/ml)

Geomean 10.41 6.35 5515.02 1032.56 8.80 58099.81 66710.08

low 95% 9.18 5.88 4949.07 837.92 6.01 51464.12 58145.46

up 95% 12.94 7.03 6566.03 1625.74 14.44 70391.27 83214.29

Min 5.23 4.25 2855.55 372.96 7.50 26404.49 29317.22

Max 19.15 8.48 8365.97 3359.34 41.13 111312.19 141982.09

CV 35 18 29 65 84 32 36

Hours post dose24 30 36 48

Darunavir (ng/mL) GM (range) 1033 (373-3359) 381 (97-257) 109 (7.5-594) 45 (7.5-149)

N of subjects below target (550 ng/mL) 3/16 11/16 15/16 16/16

Atazanavir (ng/mL) GM (range) 759 (249-2667) 407 (148-1679) 201 (65-1093) 66 (14-949)

N of subjects below target (150 ng/mL) 0/16 2/16 5/16 13/16

PK parameters cobicistat 150 mg with darunavir

T1/2(0-

24 h)

T1/2(0-

Clast)

Tmax(h)

Tlast(h)

Cmax(ng/ml)

C24(ng/ml)

Clast (ng/ml)

AUC0-24(ng.h/ml)

AUC0-Clast(ng.h/ml)

Pre-dose (ng/ml)

24 HrCL/F(L/h)

Clast HrCL/F (L/h)

Geomean 3.81 3.62 2.54 32.25 1250.25 27.56 5.00 9532.06 9681.21 28.07 15.74 15.49

Low 95% 3.49 3.34 2.26 30.00 1149.77 22.29 5.00 8677.55 8790.87 21.41 13.30 14.05

Up 95% 4.29 3.98 3.12 35.25 1392.73 51.37 5.00 10857.17 11078.72 45.67 17.96 17.76

Min 2.59 2.59 1.00 24.00 932.46 5.00 5.00 6167.33 6254.42 10.88 10.40 10.04

Max 5.60 5.55 4.00 48.00 1867.32 120.90 5.00 14425.77 14933.25 74.28 24.32 23.98

CV (%) 21 18 32 16 20 81 0 23 23 61 23 24

PK parameters cobicistat 150 mg with atazananvir

T1/2(0-

24 h)

T1/2(0-

Clast)

Tmax(h)

Tlast(h)

Cmax(ng/ml)

C24(ng/ml)

Clast (ng/ml)

AUC0-24(ng.h/ml)

AUC0-Clast(ng.h/ml)

Pre-dose (ng/ml)

24 HrCL/F(L/h)

Clast HrCL/F (L/h)

Geomean 4.43 4.21 2.35 36.30 1408.02 49.59 5.00 10553.97 10923.56 41.64 14.21 13.73

Low 95% 3.95 3.87 2.10 33.74 1293.37 42.07 5.00 9589.47 9904.56 31.42 12.96 12.50

Up 95% 5.19 4.69 2.90 39.76 1577.76 79.63 5.00 12058.87 12535.27 74.21 16.18 15.70

Median 4.32 4.17 3.00 36.00 1381.49 56.35 5.00 10569.53 10735.04 37.28 14.41 14.22

Min 3.14 3.21 1.00 30.00 929.72 14.15 5.00 7825.70 8145.20 11.11 10.22 9.95

Max 8.39 6.13 4.00 48.00 1986.37 156.24 5.00 14680.79 15068.30 172.94 19.17 18.42

CV (%) 28 19 33 17 20 63 0 23 24 80 23 23

0

1

10

100

1000

10000

24 36 48 60 72

[Dru

g] (n

g/m

l)

Time (h)

[DRV] Urine n = 16

[COBI] Urine n = 16

0

1000

24 36 48 60 72

[Dru

g] (n

g/m

l)

Time(h)

[ATV] Urine n = 16

[COBI] Urine n = 16

This study was funded by Bristol-Myers SquibbReferences: 1. Churchill et al. HIV Med. 2016;17 Suppl 4:s2-s104; 2. Moyle et al. HIV Med. 2001;2:105-13; 3. Marzolini et al. J Antimicrob Chemother. 2016 Jul;71(7):1755-8; 4. Boffito et al. Antimicrob Agents Chemother. 2011 Sep;55(9):4218-23.

BACKGROUNDDarunavir (DRV) is metabolised by cytochrome P450 (CYP) 3A4 (CYP3A4) and CYP3A5 and is a substrate for the hepatic influx transporter OATP1B1 (encoded by SLCO1B1). The pregnane X receptor (PXR; NR1I2) and constitutive androstane receptor (CAR; NR1I3) play important roles in transcriptional regulation of enzymes and transporters. Here, the effect of polymorphisms within these genes on the pharmacokinetics of DRV was investigated in participants from the NEAT001 / ANRS 143 study.

Association ofAssociation of SLCO1B1SLCO1B1 521T>C521T>C (rs4149056)(rs4149056) withAssociation ofdarunavirAssociation ofAssociation of SLCO1B1SLCO1B1SLCO1B1Association ofdarunavirdarunavir/ritonavir (DRV/r)

521T>C521T>C (rs4149056)(rs4149056)(rs4149056) withwith521T>C521T>C/ritonavir (DRV/r)/ritonavir (DRV/r)/ritonavir (DRV/r) plasma concentrations

withplasma concentrationsplasma concentrations in darunavir

HIVdarunavirHIVHIV-darunavirdarunavir/ritonavir (DRV/r)/ritonavir (DRV/r)/ritonavir (DRV/r) plasma concentrationsplasma concentrationsplasma concentrations in in darunavirdarunavirHIVHIV--infected individuals enrolled in the NEAT001/ANRS143infected individuals enrolled in the NEAT001/ANRS143 study.Rohan Gurjar, Marta Boffito, Antonio D'Avolio, Christine Schwimmer, Yazdan Yazdanpanah, Manuela Doroana, Giovanni Di Perri, Stefano Bonora, Anton Pozniak, Laura Richert, Francois Raffi, Andrew Owen, and the NEAT001/ANRS143 Trial Study Group.

Results

Linear Regression Multivariate Regression

DarunavirWeek 4 Week 24 Week 4 Week 24

P P P P

Age 0.637 0.546

Sex 0.586 0.236

Weight week 4 0.923

Weight week 24 0.291

Time of Post-dose plasma collection <0.001 <0.001 <0.001 <0.001

Ritonavir concentration <0.001 <0.001 <0.001 <0.001

Trial Randomisation (RAL Vs TDF/FTC) 0.134 0.003 0.013 <0.001

CYP3A4 (rs35599367) 0.435 0.652

SLCO1B1 (rs4149056) 0.043 0.713 0.048CYP3A5 (rs776746) 0.264 0.148 0.049NR1I2 (rs2472677) 0.596 0.871

NR1I3 (rs2307424) 0.628 0.197

Ritonavir

Age 0.979 0.966

Sex 0.315 0.621

Weight week 4 0.965

Weight week 24 0.844

Time of Post-dose plasma collection <0.001 <0.001 <0.001 <0.001

Darunavir concentration <0.001 <0.001 <0.001 <0.001

Trial Randomisation (RAL Vs TDF/FTC) 0.443 0.062

CYP3A4 (rs35599367) 0.569 0.452

SLCO1B1 (rs4149056) 0.95 0.241

CYP3A5 (rs776746) 0.817 0.186

NR1I2 (rs2472677) 0.689 0.019 0.027NR1I3 (rs2307424) 0.107 0.355

Table 1. Summary of stepwise multivariate linear regression.

A total of 1163 plasma concentrations were available from 588 participants. Darunavir plasma concentrations were significantly higher at week 4 (P =0.013) and week 24 (P = < 0.001) in the tenofovir/emtricitabine (TDF/FTC) arm (2790 ± 2110 ng/mL and 2648 ± 2061ng/mL) compared to theraltegravir (RAL) arm (2698 ± 2377 ng/mL and 2216 ± 1698 ng/mL). (Figure 1 A and B).

In week 4 pooled analysis, SLCO1B1 rs4149056 was associated with DRV plasma concentrations (P = 0.048). Plasma DRV concentrations at week 4were 2719 ± 2348 ng/mL, 2842 ± 1996 ng/mL and 2506 ± 1459 ng/mL for TT,TC and CC genotype groups, respectively (Figure 2).

At Week 24 DRV plasma concentrations were associated (P = 0.049) with CYP3A5 (rs776746). Plasma DRV concentrations were 2490 ± 2012 ng/mL,2300 ± 1597 ng/mL and 2429 ± 1606 ng/mL for TT, TC and CC genotype groups respectively. (Figure 3).

NR1I2 (rs2472677) was significantly associated with ritonavir plasma concentrations (P = 0.027) at week 24. Plasma concentrations of ritonavir at week24 were 149.02 ± 246.47 ng/mL, 213.4 ± 471.84 ng/mL and 260.4 ± 701.93 ng/mL for TT, TC and CC genotype groups, respectively (Figure 4).

ConclusionsMethodsNEAT 001 / ANRS 143 was a randomised study that demonstrated non-inferiority of first-line ART with DRV/ritonavir (800/100mg once daily) plus raltegravir (RAL; 400mg twice daily) compared with DRV/r plus tenofovir/emtricitabine (TDF/FTC, 245/200mg once daily). Blood samples were collected at week 4 and week 24 post therapy initiation at any time <24hr post-dose. DNA was extracted from whole blood and genotyping for CYP3A4 (rs35599367), CYP3A5 (rs776746), SLCO1B1 (rs4149056; 521T>C), NR1I3 (rs2307424) and NR1I2 (rs2472677; 63396C>T) polymorphisms was conducted using real-time PCR based allelic discrimination. Demographic data, time of sample collection, randomisation and plasma drug concentrations were assessed using step-wise multivariate regression (Table 1).

Lower DRV plasma concentrations at week 4 and 24 were observed in patients receiving RAL as compared to those receiving TDF/FTC suggesting a possible interaction between DRV and RAL.. This may influence the emergence of resistance associated mutations (RAM) leading to virological failure, which was seen in patients receiving RAL, but needs to be further evaluated1. At week 4, lower DRV plasma concentrations were seen C homozygotes for SLCO1B1 521T>C. This association is different to that previously reported for lopinavir and statins, where concentrations were higher in C homozygotes2. In previous studies CYP3A5 (rs776746) has shown to lower plasma concentrations of lopinavir.3 The effect of PXR; NR1I2 (rs2472677; 63396C>T) has been previously reported other PIs such as atazanavir.4

References –1. Lambert-Niclot S, George EC, Raffi F et al. Antiretroviral resistance at virological failure in the NEAT 001/ANRS 143 trial: raltegravir plus darunavir/ritonavir or tenofovir/emtricitabine plus darunavir/ritonavir as first-line ART. J Antimicrob Chemother 2016 Apr;71(4):1056-62.

doi: 10.1093/jac/dkv427. Epub 2015 Dec 24.2. Kohlrausch FB, de Cassia Estrela R, Barroso PF, Saurez-Kurtz G. The impact of SLCO1B1 polymorphisms on the plasma concentration of lopinavir and ritonavir in HIV-infected men. Br J C in Pharmacol 2010 Jan;69(1):95-8. doi: 10.1111/j.1365-2125.2009.03551.x3. Berno, Giulia, Zaccarelli, Mauro, D’Arrigo, Roberta. Potential implications of CYP3A4, CYP3A5 and MDR-1 genetic variants on the efficacy of Lopinavir/Ritonavir (LPV/r) monotherapy in HIV-1 patients. Berno G et al. Journal of the International AIDS Society 2014, 17(Suppl

3):195894. Alessandro Schipani,Marco Siccardi, Andrew Owen. Population Pharmacokinetic Modeling of the Association between 63396C3T Pregnane X Receptor Polymorphism and Unboosted Atazanavir Clearance. ANTIMICROBIAL AGENTS AND CHEMOTHERAPY, Dec. 2010, p. 5242–

5250

Figure 1. Impact of trial randomisation (RAL Vs TDF/FTC) on C0-24 of darunavir at (A)week 4 (B)week 24

Figure 2. Influence of SLCO1B1 521T>C (rs4149056)on darunavir C0-24 (A) TT (B) TC (C) CC at week 4

Figure 3. Influence of CYP3A5 (rs776746) on darunavir C0-24 (A)CC (B) TC (C) TT at week 24

Figure 4. Influence of NR1I2 (rs2472677) on ritonavir C0-24 (A) TT (B) TC (C) CC at week 24

Results 1: demographic characteristics

Patients (n) 75 Males (%) 80% Caucasians (%) 75% Duration of Stribild therapy (days) 173 [56-356]*

Concomitant HAART (n) darunavir (n=35); atazanavir (n=2)

CD4 cell count (cells/mL) 487 [354-697]* Patients with viral load >37 copies/mL (n) 14/75 Body weight (Kg) 70 [65-77]* Serum alanine aminotrasferase (IU/mL) 31 [21-41]* Serum creatinine (mg/dL) 0.9 [0.8-1.1]* *Data were given as median [interquartile range]

Conclusions

The role of food in improving Stribild® bioavailability is underestimated in real life settings. Since this condition may increase the risk for patients to experience suboptimal drug exposure, it is important that all health professionals more convincingly advise their patients to take Stribild® in fed conditions. On the other hand, the key role of patient education and patient responsibility to be fully adherent with the instructions provided by healthcare professionals on how to correctly take the therapy should not be underestimated.

When food can make the difference: the case of elvitegravir-based co-formulation (P309)

Cristina Gervasoni1, Sara Baldelli2, Davide Minisci1, Paola Meraviglia1, Emilio Clementi2,3, Massimo Galli1,3 and Dario Cattaneo2

1Department of Infectious Diseases and 2Unit of Clinical Pharmacology, ASST Fatebenefratelli Sacco, Milan, Italy. 3Università degli Studi di Milano, Milan, Italy

Background and Objectives

In the product monograph of the fixed-dose combination containing elvitegravir, cobicistat, tenofovir and emtricitabine (Stribild®) it is recommended that the formulation should be administered under fed conditions to optimize drugs exposure.

Here we assessed to what extent this advice is applied in the real life scenario by measuring elvitegravir, tenofovir and cobicistat plasma trough concentrations in HIV-infected patients given Stribild® alone or as part of antiretroviral therapy as per their daily routine practice.

Methods

Results 2: timing of Stribild® intake

23%

17%

36%

9%

15% breakfast (fed) lunch (fed) supper (fed) mid-morning (fasting) late-evening (fasting)

24% of patients assumed Stribild® in fasting conditions

Results 3: distribution of drug concentrations

In this analysis we included consecutive HIV-infected patients fulfilling the following conditions:

a) treated with Stribild® for at least one month b) with at least one request for therapeutic drug monitoring

of elvitegravir and tenofovir plasma trough concentrations c) no clinical evidence of gastrointestinal impairment d) not given drugs known to affect elvitegravir or tenofovir

pharmacokinetics. Collected blood samples had to be taken 24 hours after

the last drug intake (a time window of 20 min was considered acceptable), immediately before drug administration (trough concentrations).

The main demographic, biochemical and clinical data were

recorded for each patient, together with detailed information on the time of blood drawing and the time of last drug intake

Drug trough concentrations were assessed by LC-MS/MS.

The lower limit of quantifications (LOQ) were: - elvitegravir 25 ng/mL - cobicistat 5 ng/mL - tenofovir 10 ng/mL

0

500

1000

1500

2000

2500

Elvi

tegr

avir

(ng/

mL)

0

200

400

600

800

1000

Teno

fovi

r (ng

/mL)

0

60

120

180

240

300

Cob

icis

tat (

ng/m

L)

Fasting Fed Fasting Fed Fasting Fed

0

300

600

900

1200

1500

Elvi

tegr

avir

(ng/

mL)

Teno

fovi

r (ng

/mL)

Cob

icis

tat (

ng/m

L)

Fasting Fed 0

80

160

240

320

400

Fasting Fed 0

40

80

120

160

200

Fasting Fed

Each symbol identifies an individual patient

Drug concentrations measured in 6 out of the 12 patients with elvitegravir levels <LOQ given Stribild® in the fasting vs. fed settings

• A food effect was observed for DRV following administration of DCFTAF – Cmax, AUClast and AUCinf for DRV decreased following fasted conditions compared with fed conditions – this is consistent with what has been previously reported for DRV, when co-administered with

ritonavir (a 32% decrease in DRV AUClast was seen in fasted versus fed conditions) or COBI (39% to 56% decrease in DRV PK parameters was observed in fasted versus fed conditions)11,12

– the food effect for DRV was previously shown to be similar for different types of food.12

• Differences in exposure to COBI, FTC and TAF in fed versus fasted conditions are not considered to be clinically relevant.

• Administration of DCFTAF was generally well tolerated under both fed and fasted conditions, and no new safety issues were identified.

• Consistent with prescribing recommendations for other DRV formulations,13 it is recommended that DCFTAF be taken with food, which is also the recommendation in the ongoing Phase III AMBER and EMERALD trials in HIV-1-infected adults.

Impact of food on the bioavailability of darunavir, cobicistat, emtricitabine and tenofovir alafenamide (DCFTAF), the first protease inhibitor-based, single-tablet complete HIV-1 regimenHerta Crauwels,1 Bryan Baugh,2 Erika Van Landuyt,1 Simon Vanveggel,1 Anja Hijzen,1 Magda Opsomer1

1Janssen Pharmaceutica NV, Beerse, Belgium; 2Janssen Research & Development LLC, Raritan, NJ, USA

P310

References1. Dejesus E, et al. J Acquir Immune Defic Syndr 2009;51:163–74.

2. Hodder SL, et al. AIDS Patient Care STDS 2010;24:87–96.

3. Willig JH, et al. AIDS 2008;22:1951–60.

4. Clay PG, et al. Medicine (Baltimore) 2015;94:e1677.

5. Mills A, et al. J Acquir Immune Defic Syndr 2015;69:439–45.

6. DHHS. Guidelines for the use of antiretroviral agents in HIV-1-infected adults and adolescents. July 14, 2016 [accessed 12 September 2016]. Available at: https://aidsinfo.nih.gov/contentfiles/lvguidelines/adultandadolescentgl.pdf.

7. European AIDS Clinical Society Guidelines, Version 8.0. October 2015 [accessed 12 September 2016]. Available at: http://www.eacsociety.org/files/guidelines_8.0-english-revised_20160610.pdf

8. Ruane PJ, et al. J Acquir Immune Syndr 2013;63:449–55.

9. Custodio JM, et al. ASCPT 2015. Abstract PI-052.

10. De Clercq E. Biochem Pharmacol; 2016 pii:S0006-2952(16)30069-7

11. Kakuda TN, et al. Antivir Ther 2014;19:597–606.

12. Sekar V, et al. J Clin Pharmacol 2007;47:479–84.

13. PREZISTA® (darunavir). Full Prescribing Information. June 2016 [accessed 12 September 2016]. Available at: http://www.prezista.com/sites/default/files/pdf/us_package_insert.pdf#zoom=100

Acknowledgements and disclosures • We would like to thank the study participants and site staff for their participation and support during the study; the Janssen team members from the

bioanalysis department, in particular Vera Hillewaert; Jiri Letal who was the study statistician; and Janssen Research & Development team members, in particular Kimberley Brown, Julia Sugumar and Eric Y. Wong for their input into this poster.

• This study was sponsored by Janssen Pharmaceuticals.

• All authors are full-time employees of Janssen and potential stockholders of Johnson & Johnson.

• Medical writing support was provided by Ian Woolveridge from Zoetic Science, Macclesfield, UK, an Ashfield Company. Support for medical writing assistance was provided by Janssen Pharmaceuticals.

• These data have been presented previously: Crauwels et al. 21st IAC 2016. Abstract and poster THPEB064.

Plas

ma

conc

entr

atio

n of

DRV

(ng/

mL)

9000

8000

7000

6000

5000

4000

3000

2000

1000

0

0 4 8 12 16 20 24Time (hours)

DCFTAF 800/150/200/10mg under fasted conditions (N=23)DCFTAF 800/150/200/10mg under fed conditions (N=24)

Figure 1. Mean ± SD DRV plasma concentration-time curves following administration of a single oral dose of DCFTAF under fed (standardised high-fat breakfast) or fasted conditions.

Plas

ma

conc

entr

atio

n of

CO

BI (n

g/m

L)

1000

900

800

700

600

500

400

300

200

100

0

0 4 8 12 16 20 24Time (hours)

DCFTAF 800/150/200/10mg under fasted conditions (N=23)DCFTAF 800/150/200/10mg under fed conditions (N=24)

Figure 2. Mean ± SD COBI plasma concentration-time curves following administration of a single oral dose of DCFTAF under fed (standardised high-fat breakfast) or fasted conditions.

Plas

ma

conc

entr

atio

n of

FTC

(ng/

mL)

3000

2500

2000

1500

1000

500

0

0 4 8 12 16 20 24Time (hours)

DCFTAF 800/150/200/10mg under fasted conditions (N=24)DCFTAF 800/150/200/10mg under fed conditions (N=24)

Figure 3. Mean ± SD FTC plasma concentration-time curves following administration of a single oral dose of DCFTAF under fed (standardised high-fat breakfast) or fasted conditions.

Plas

ma

conc

entr

atio

n of

TA

F (n

g/m

L)

240

200

160

120

80

40

0

0 2 4 6 8 10 12Time (hours)

DCFTAF 800/150/200/10mg under fasted conditions (N=24)DCFTAF 800/150/200/10mg under fed conditions (N=24)

Figure 4. Mean ± SD TAF plasma concentration-time curves following administration of a single oral dose of DCFTAF under fed (standardised high-fat breakfast) or fasted conditions.

Introduction• Once-daily, single-tablet, complete HIV-1 regimens have been shown to improve treatment adherence,

patient satisfaction, and virologic outcomes compared with multi-tablet regimens.1–4

• DCFTAF is the first protease inhibitor (PI)-based, single-tablet, complete HIV-1 regimen, for once-daily use – This regimen combines darunavir (DRV, D 800mg) with the pharmacoenhancer, cobicistat (COBI, C 150mg),

and with emtricitabine (FTC, F 200mg), and the prodrug of tenofovir, tenofovir alafenamide (TAF, 10mg)5

• Current treatment guidelines recommend DRV, boosted with ritonavir (preferred option) or COBI (alternative option), combined with other antiretroviral drugs, for HIV-1-infected treatment-naïve patients6,7

• TAF provides comparable efficacy to tenofovir disoproxyl fumarate (TDF) at a much lower dose due to better stability and more directed distribution,8 resulting in lower systemic exposure to tenofovir and fewer adverse effects, particularly a lower risk of renal toxicity and bone density changes.5,9,10

• In a Phase II trial of 153 treatment-naïve, HIV-1-infected adults, DCFTAF had similar virologic efficacy to DRV + COBI + FTC/TDF, with significantly improved renal and bone safety and no treatment-emergent resistance to PIs or any components of the regimen.5

• The efficacy and safety of DCFTAF is under investigation in two, large, Phase III studies – AMBER (NCT02431247) is evaluating the efficacy and safety of DCFTAF versus a regimen comprising

a DRV/COBI fixed-dose combination with FTC/TDF in approximately 670 treatment-naïve, HIV-1-infected patients

– EMERALD (NCT02269917) is evaluating the efficacy and safety of switching to DCFTAF as compared to remaining on a PI boosted with low-dose ritonavir or COBI, in combination with FTC/TDF in approximately 1100 virologically suppressed, HIV-1-infected patients.

• Given the impact of food on the systemic exposure to DRV,11–13 the recommended intake for DCFTAF is once-daily with food. The present study was undertaken to assess the impact of food on the single-dose pharmacokinetics of the components of DCFTAF in healthy volunteers.

MethodsStudy design and treatments• TMC114FD2HTX1002 (NCT02475135) was a Phase I, open-label, randomised, two-period, single-centre,

crossover study in HIV-negative, healthy adult participants.

• In two treatment sessions, participants received a single oral dose of DCFTAF under fasted conditions or 30 minutes after a standardised high-fat breakfast with a washout period of at least 7 days in between each treatment session.

• The standardised high-fat breakfast (928kCal; 56g fat) consisted of two eggs fried in butter, two strips of bacon, two slices of white bread with butter, one croissant with one slice of cheese, and 240mL (8oz) of whole milk.

• The study protocol and amendments were reviewed and approved by an Independent Ethics Committee. The trial was conducted in accordance with the Declaration of Helsinki, Good Clinical Practice guidelines and applicable regulatory requirements. All participants provided written informed consent.

Pharmacokinetic (PK) and safety evaluations• Blood samples were taken predose and 0.25, 0.5, 0.75, 1, 1.5, 2, 3, 4, 5, 6, 8, 12, 18, 24, 36, 48 and

72 hours after DCFTAF intake.

• PK profiles of the component drugs were determined up to 72 hours for DRV and COBI, 48 hours for FTC and 12 hours for TAF.

• Plasma concentrations of DRV, COBI, FTC and TAF were determined using validated high-performance liquid chromatography-mass spectrometry/mass spectrometry assays

– The lower limit of quantification was 5.0ng/mL for DRV, COBI and FTC, and 1.0ng/mL for TAF.

• PK parameters were determined using non-compartmental analysis (WinNonlin version 6.2.1, Pharsight, Mountain View, CA, USA)

– The PK parameters determined for each of the component drugs included: time to maximum plasma concentration (tmax), terminal elimination half-life (t1/2term), maximum plasma concentration (Cmax), area under the plasma concentration-time curve (AUC, calculated by linear-linear trapezoidal summation) from time of administration up to the last timepoint with a measurable concentration post-dose (AUClast) and

extrapolated to infinity (AUCinf).

• The least square (LS) means of natural logarithm-transformed Cmax, AUClast and AUCinf for each component were estimated using a linear mixed-effects model, controlling for treatment, sequence, and period as fixed effects, and subject as a random effect (SAS version 9.3, SAS Institute, Cary, NC, USA).

• The food effect was evaluated using geometric mean ratios of PK parameters for each component drug under fed (reference) or fasted (test) conditions (test/reference)

– A 90% confidence interval (CI) was constructed around the difference between the LS-means of test and reference, back-transformed using the exponential function, and compared with the 80% to 125% boundaries of no effect.

• Safety and tolerability were assessed at regular intervals throughout the study. The follow-up period was 7 to 10 days after the last intake of study medication or after study discontinuation.

ResultsParticipant disposition and baseline characteristics • Twenty-four participants were randomised (12 per dosing sequence), all of whom completed the study and

were included in the PK and safety analyses.

• Twelve of the volunteers (50%) were female and all 24 (100%) were white. Median (range) age and body mass index were 35 (18 to 54) years and 23.7 (19.8 to 29.5) kg/m2, respectively.

Effect of food on the bioavailability of the components of DCFTAF • PK parameters for DRV, COBI, FTC and TAF following single-dose administration of DCFTAF under fed

(high-fat breakfast) and fasted conditions are summarised in Table 1.

• Analysis of the geometric mean ratios for Cmax, AUClast and AUCinf indicated that the exposure to DRV, when administered as DCFTAF, was 30% to 45% lower in fasted (test) compared with fed (reference) conditions (Table 1)

– Figure 1 shows mean ± SD DRV plasma concentrations over time following administration of a single-dose of DCFTAF, under fed and fasted conditions.

• Similarly, exposure to COBI was 16% to 30% lower under fasted versus fed conditions (Table 1) – Mean ± SD COBI plasma concentrations over time following intake of a single dose of DCFTAF under fed

and fasted conditions are shown in Figure 2.

• For FTC, the Cmax was 26% higher in fasted compared with fed conditions, while AUClast was comparable under both conditions (Table 1)

– Figure 3 shows mean ± SD FTC plasma concentrations over time following administration of a single-dose of DCFTAF under fed and fasted conditions.

• For TAF, the Cmax following DCFTAF administration was 82% higher in fasted conditions, while the AUCinf was 20% lower in fasted compared with fed conditions (Table 1). The AUClast of TAF was comparable for test and reference, with the 90% CI of the LS mean ratio within the 80% to 125% boundaries of no effect

– Mean ± SD TAF plasma concentrations over time following intake of a single-dose of DCFTAF under fed and fasted conditions are shown in Figure 4.

Safety and tolerability• In total, 9/24 (38%) and 10/24 (42%) volunteers experienced an adverse event (AE) under fasted and fed

conditions, respectively, following single-dose administration of DCFTAF.

• All AEs were grade 1 or 2 in intensity, with no grade 3 or 4 events reported – the most common AEs (observed in >2 volunteers during any treatment phase) were headache

(3/24 [13%] fasted vs 5/24 [21%] fed) and nausea (4/24 [17%] fasted vs 2/24 [8%] fed) – three volunteers reported AEs that were grade 2 in intensity: irritable bowel syndrome (under fasted

conditions), and nausea and headache (under fed conditions) – incidence of AEs by preferred term was generally comparable under fasted and fed conditions.

• No deaths or serious AEs occurred during the study, and no volunteers discontinued the study because of an AE.

• Laboratory abnormalities were mostly grade 1 or 2. No consistent or clinically relevant changes in blood chemistry or haematology were observed.

Table 1. DRV, COBI, FTC and TAF PK parameters and statistical analyses following administration of a single dose of DCFTAF under fed (standardised high-fat breakfast) and fasted conditions.

DRV COBI FTC TAF

PK parameter, mean (SD)†

Fasted (test) n=23‡

Fed (high fat) (reference)

n=24‡

Fasted (test)n=23§

Fed (high fat) (reference)

n=24

Fasted (test) n=24¶

Fed (high fat) (reference)

n=24††

Fasted (test)

n=24‡‡

Fed (high fat) (reference)

n=24¶

Cmax, ng/mL 4089 (1846) 6629 (1543) 704 (368) 711 (164) 2247 (573) 1785 (486) 180 (90.6) 107 (65.2)

tmax, hours 3.00 (1.00–8.02) 5.00 (1.50–8.00) 3.00 (1.00–6.00) 5.00 (2.00–6.10) 1.00 (0.50–2.00) 2.00 (0.75–5.00) 0.50 (0.25–0.75) 0.88 (0.25–5.00)

AUClast, ng•h/mL 67,504 (35,642) 93,541 (39,730) 5771 (3206) 6168 (2260) 11,593 (2573) 11,499 (2055) 106 (44.7) 117 (51.5)

AUCinf, ng•h/mL 72,147 (36,009) 94,686 (40,882) 6136 (3064) 6258 (2268) 12,286 (2729) 10,029 (1079)§§ 109 (47.7) 125 (57.3)

t1/2term, hours 7.0 (2.3) 7.8 (3.5) 4.1 (0.9) 3.9 (0.6) 10.8 (1.2) 10.7 (1.2)§§ 0.3 (0.2) 0.5 (0.1)

Geometric mean ratio, % (90% CI)

n¶¶ 23 vs 24 23 vs 24 24 vs 24 24 vs 24

Cmax 54.99 (46.73–64.71) 76.96 (55.70–106.33) 125.99 (112.85–140.65) 182.29 (140.50–236.50)

AUClast 65.65 (56.76–75.92) 70.90 (51.13–98.30) 100.12 (96.29–104.10) 89.54 (81.20–98.72)

AUCinf 70.25††† (59.49–82.95) 84.39‡‡‡ (68.52–103.95) – 80.38§§§ (73.04–88.45)

†Except tmax = median (range); ‡n=20, §n=22, ¶n=16, ††n=7, ‡‡n=21 for AUCinf, t1/2term§§Accurate determination not possible for more than 50% of participants; interpret with caution; ¶¶test vs reference†††n=20 for test and reference; ‡‡‡n=22 for test; §§§n=21 for test and n=16 for referenceSD = standard deviation

Presented at HIV Drug Therapy, Glasgow, UK, 23–26 October 2016

This poster is available on request from Dr Herta Crauwels: HCRAUWEL@its.jnj.com

Conclusion

•Cristina Gomez-Ayerbe1; Santos Del campo1; Maria Jesus Vivancos1, Gabriel Gaspar2, Ana Moreno1, Andres Martin3, Alberto Romero4, Patricia Jimenez4, Miguel Angel Cervero5, Miguel Angel Rodriguez-Sagrado1, Regino Serrano6, Alfonso Muriel1, María Jesús Peréz Elías1.

• 1Hospital Ramón y Cajal, IRYCIS Madrid Spain; 2Hospital de Getafe , Getafe Spain; 3Hospital Universitario Puerta del Mar Cádiz Spain; 4Hospital Universitario Puerto Real, Cádiz Spain;5Hospital Severo Ochoa, Leganés Spain; 6Hospital del Henares, Coslada Spain.

Prevalence of drug-drug interactions and its impact on durability among patients receiving elvitegravir/cobicistat/emtricitabine/tenofovir (EVG/C/F/T) and concomitant medication

P311

BACKGROUND and OBJECTIVE

RESULTS

METHODS

• In clinical trials patients with different degree potential interactions were excluded, so few experience is available in polimedicated patients.

Cobicistat, a component of single treatment regimen EVG/C/F/T, is a potent Cytocrome P450 inhibitor, so many drug-drug interactions (DDI) are expected1.

Sometimes you need to use drugs classified in the category of potential interaction / or use with caution2

From the real practice perspective, we evaluated the clinical impact of DDI associated with the use of EVG/C/F/T and concomitant medication (CM).

Time and reasons to change EVG/C/F/T were compared according to CM status (No DDI, if patient has no CM or no interactions were expected and, DDI present, if CM has been identified by LHDI tool with potential interactions or contraindicated.

Time and reasons to change EVG/C/F/T were compared according to CM status (No DDI, if patient has no CM or no interactions were expected and, DDI present, if CM has been identified by LHDI tool with potential interactions or contraindicated.

CM was recorded since EVG/C/F/T was started, and categorized, according to the Liverpool HIV Drug interactions (LHDI) tool, www.hiv-druginteractions.org/. Highest degree of DDI found was applied

CM was recorded since EVG/C/F/T was started, and categorized, according to the Liverpool HIV Drug interactions (LHDI) tool, www.hiv-druginteractions.org/. Highest degree of DDI found was applied

No interactions expected Potential interactions Use with caution and formally contraindicated

From July 2014 to January 2016, we retrospectively reviewed all patients starting a new EVG/C/F/T regimen, both in naive and switching scenarios, in 6 hospitals of Spain

From July 2014 to January 2016, we retrospectively reviewed all patients starting a new EVG/C/F/T regimen, both in naive and switching scenarios, in 6 hospitals of Spain

Baseline CD4 median (IQR) and HIVBaseline CD4 median (IQR) and HIV-RNA (% 1.57)

521 cel/mcL (328-718) 75%

Naive= 43 (18%); Switch= 199 (82%)

AIDS Stage 28%

Route of HIV infection IDU 35% MSM 34% HTX 26%

Mean age: 47 years (18-68)

Sex Women: 61 (25%)

Median Follow-up before Median Follow-up before EVG/C/FTC/TDF

13.7 years (IQR 38-51)

Baseline Characteristics of 242 patients included Concomitant

Medication

No concomitant Medication

89 (37%)

Concomitant Medication 153 (63%)

Potential Interaction 111

(46%)

Contraindicated 3 (1.2%)

DDI group 114 (47%)

No Drug Interaction

39 (16%)

Any Drug Interaction 114 (47%)

46

15

4 5 5 7

18

0

5

10

15

20

25

30

35

40

45

50 Reasons to start

EVG/C/FTC/TDF (%)

After a median exposure of 278 days to EVG/C/F/T, 13% of patients changed its therapy due to • Toxicity 6.2% • Other reasons 3.3% • DDI 2.1% • Simplification 0.8%

No differences were observed in median time to change EVG/C/F/T according to DDI status • neither in the univariate log Rank test P (0.69) • nor in the Cox regression analysis P (0.64) after

adjusting for sex, aids, CD4 cell count, and HIV-RNA .

No DDI group 128 (53%)

the probability of remaining in E/C/F/T therapy after 10 months was 86% for Non DDI vs.88% for DDI Patients, P (0.69)

Conclusions In clinical practice, despite a large number of drug-drug interactions are predicted for EVG/C/F/T and concomitant medication, apparently these DDI not influenced a shorter durability. . 1. Xu L, Liu H, Murray BP, Callebaut C, Lee MS, Hong A, et al. Cobicistat (GS-9350): a potent and selective inhibitor of human CYP3A as a novel pharmacoenhancer. ACS Med Chem Lett. 2010;1:209–13. 2.University of Liverpool. HIV drug interaction checker. Available from: www.hiv-druginteractions.org.

Relationships between dolutegravir plasma-trough concentrations, UGT1A1 genetic polymorphisms, and side-effects of central nervous system in Japanese HIV-1-infected patients

Demographics and genotypes of participants(Table 1)

Dolutegravir (DTG) is a second-generation integrase inhibitor used to treat HIV-1-infected patients. DTG has shown anti-HIV effects non-inferior to those of other drugs in phase-3 trials and can be conveniently taken once daily1). The characteristic side-effects of DTG include central nervous system side-effects (CNSSEs) leading to drug discontinuation in some cases2). Furthermore, DTG is primarily metabolized by UGT1A1, and a weak correlation exists between DTG plasma-trough concentrations and UGT1A1 genetic polymorphisms3). The principal aim of the study was to explore DTG plasma-trough concentrations association with CNSSEs. Moreover, we considered whether UGT1A1 genetic polymorphisms could predict DTG CNSSEs.

DTG plasma-trough concentrations were measured in 101 Japanese HIV-1 patients who were taking DTG at Osaka National Hospital from June 2014 to March 2016, and UGT1A1 genetic screening (*6 and *28) was performed. DTG trough levels were measured by liquid chromatography-mass spectrometry4). UGT1A1 was genotyped using the sequence method.

DTG plasma-trough concentrations were compared in patients in whom CNSSEs did and did not develop, and then the frequency of CNSSEs was examined among three groups:.A) patients with homozygous mutations in UGT1A1*6/*28 or compound

heterozygous mutations in*6/*28;B) patients with heterozygous mutations in *6/*28; andC) wild-type

1) Raffi F, et al. Lancet Infect Dis, 2013;11:927-35.2) Van den Berk G, et al. Conference on Retroviruses and Opportunistic Infections (CROI), 2016, Abstract 948. 3) Yagura H, et al. World STI & HIV Congress, 2015, P17.29.4) Takahashi M, et al. The 28th Annual Meeting of the Japanese Society for AIDS Research, 2014, P-025

Yagura H1, Watanabe D2, Ashida M2, Nakauchi T1, Tomishima K1, Togami H3, Hirano A3, Sako R1, Doi T1, Yoshino M4, Takahashi M5, Yamazaki K1, Uehira T2 and Shirasaka T21 Department of Pharmacy, Osaka National Hospital, Osaka, Japan2 AIDS Medical Center, Osaka National Hospital, Osaka, Japan3 Department of Pharmacy, Nagoya Medical Center, Aichi, Japan4 Department of Pharmacy, Osaka-minami medical center, Osaka, Japan5 Department of Pharmacy, Suzuka Hospital, Mie, Japan

Grouping of genotypeA

(*6/*6, *28/*28,*6/*28)

B(*6/-,*28/-)

C(-/-)

p

Participants (n, %) 13 (13%) 37 (37%) 51 (50%)

Age (years)43

[39–49]

42

[42–56]

45

[41–51]0.8604

Males (n, %) 13 (100%) 32 (86%) 47 (92%) 0.3152

Body weight (kg)64

[58–68]

67

[59–74]

65

[58–71]0.6604

CD4 cell count (cells/μL)490

[425–597]

409

[334–546]

501

[315–624]0.2597

Participants with

HIV-1-RNA level <50 at

time of sampling (n, %)

13

(100%)

36

(97%)

50

(98%)0.8343

Use of antiretroviral agents

(n, %)

Tenofovir 9 (69%) 19 (51%) 28 (55%) 0.7417

Abacavir 3 (23%) 18 (49%) 21 (41%) 0.4861

Protease inhibitor 0 (0%) 0 (0%) 0 (0%) 1.0000

NNRTI 1 (8%) 0 (0%) 2 (4%) 0.2230

Duration of DTG treatment

(days), median [IQR]

68

[39–88]

80

[63–168]

126

[63–181]0.1570

HBV infection (n, %) 1 (8%) 1 (3%) 4 (8%) 0.5781

HCV infection (n, %) 0 (0%) 0 (0%) 2 (4%) 0.3678

Frequencies of CNSSEs in the three groups were: A, 23%; B, 25%; and C, 18%. No significant differences in frequency or symptoms of CNSSEs were evident in terms of UGT1A1 genetic polymorphisms.

ConclusionsAlthough UGT1A1 genetic polymorphisms are not predictive of

DTG CNSSEs, the data suggest that a relationship may exist between DTG plasma-trough concentrations and CNSSEs.

Fig. Correlation between UGT1A1 polymorphisms and DTG plasma-trough concentrations. Horizontal straight line indicates median value. *p < 0.05

Median [IQR]

Materials & Methods

IQR, interquartile range; NNRTI, non-nucleoside reverse transcriptase inhibitor.

References

Osaka National Hospital2-1-14, Hoenzaka, Chuo-ku, Osaka, Japan

Phone: +81-6-6942-1331http://www.onh.go.jp/mokuji/mokuji.html

Comparison of DTG plasma-trough concentrations and CNSSEs (Figure)

Median DTG plasma-trough concentration was significantly higher in patients with CNSSEs (1.34 μg/mL) than in those without CNSSEs (1.06 μg/mL) .

Background Results

Correlation between UGT1A1 polymorphismsand CNSSEs (Table 2)

Genotype grouping A B C p

Frequencies of CNSSEs

(n,%)

3/13(23%)

9/37(25%)

9/51(18%) 0.7306

CNSSEs symptomsHeadache 2 4 2 0.2815Insomnia 1 3 2 0.4697Irritability 0 1 2 0.7712

Light-headedness 0 0 2 0.3678Depression 0 0 1 0.6095Dizziness 0 1 0 0.4175

without CNSSEs with CNSSEs

DTG

con

cent

ratio

n (μ

g/m

L)

*

P312

• Redox-imbalance plays a role on viral, inflammatory and immunological response of

HIV-infection1.

• S-thiolation is a reversible oxidative modification that occurs when a low molecular

weight thiol – cysteine (Cys), glutathione (GSH), homocysteine (Hcy), cysteinylglycine

(CysGly) establishes a disulfide bond between its –SH group and a –SH group of a

free Cys residue in a protein2.

• This process, that protects protein from irreversible oxidation, is a relevant redox-

buffer in blood and has regulatory functions2.

• Several antiretrovirals , as efavirenz (EFV) and nevirapine (NVP), were implicated with

oxidative/electrophilic stress generation that might have implications in drug

efficacy/toxicity3.

• The plasma proteins S-thiolation profile (RSSP) might represent a pharmacodynamic

biomarker for assessing the redox-modulating effects of drugs.

AIM:

1) To evaluate the relation between RSSP and the immunological

status in HIV-infected patients without cART (naïve).

2) To investigate and compare the RSSP profile between naïve, NVP-

and EFV-containing cART patients.

• Study protocol received prior approval from Hospitals Ethics Committee and all

patients signed an informed consent.

• Patients were stratified according to cART use: naïve, NVP-cART and EFV-cART

• Anthropometric and clinical data recorded (Table 1).

• Exclusion criteria: kidney and hepatic dysfunction; detectable viral load in treated

patients

• Total and free thiols quantified by an HPLC with fluorescence detection method

(HPLC-FD) (Figure 1). The RSSP fraction was determined by subtraction of the free

fraction to the total concentration of the each thiol.

• Immunological status of patients is positively related to their RSSP profile, which is influenced by ageand cART.

• GSSP levels are the most influenced by cART, decreasing with NVP and increasing with EFV.

• This data supports that EFV and NVP have different redox-modulating effects. While sharing similarmechanisms of action, these differences might be reflected in its toxicity profile: neurotoxicity (EFV)

and hepatotoxicity (NVP), that will be evaluated in further studies.

1. Background

2. Patients & Methods

3. Results

4. Conclusions

Figure 1. Quantification of total and free thiols in serum.

1. Aquaro S, et al. Future HIV Ther. 2008; 2:327-38; 2. Dalle-Donne I, et al. Antioxid Redox Signal 2008; 10:445-73.;

3. Caixas U, et al.Toxicology. 2012; 301:33-9.

5. References

FCT: EXPL/DTP-FTO/1792/2013; PD/BD/105892/2014(CGD);

SFRH/BD/86791/2012(NMG).

iNOVA4Health: 201601-02-021

Financial support

: sofia.pereira@nms.unl.pt; mjoao.correia@nms.unl.ptCorrespondence

Variable Naïve NVP EFV p-value

N 22 30 83 -

Age (years) 42 ± 10 48 ± 10 43 ± 11 ns

CD4 cell count (cell/µL) 4789 ± 192 600 ± 301 642 ± 239 0.025a

Viral load (cps/mL) 43096 ± 55670 BQL BQL -

Table 1. Anthropometric and clinical data

Data are presented as mean ± SD. BQL, bellow quantification limit; NA, not available;ns, not significant; a One-Way ANOVA

0 200 400 600 800 1000100

150

200

250r= 0.539p= 0.014

CD4+ (cell/µL)

Cys

SSP

( µM

)

0 200 400 600 800 10000

1

2

3

4r= 0.493p= 0.044

CD4+ (cell/µL)

GSS

P ( µ

M)

0 200 400 600 800 10000

10

20

30

40r=0.451p=0.046

CD4+ (cell/µL)

Cys

Gly

SSP

( µM

)

A. B. C.

Figure 2. Association between CD4+ T cell count in naïve patients (n=22) and A. S-cysteinylated (CysSSP),

B. S-glutathionylated (GSSP) and C. S-cysteinylglycinylated (CysGlySSP) protein profile.

Multivariable analysis

• GSSP was influenced by:• Age B: -0.04, 95% CI (-0.06; -0.02), p=0.001• cART B: 1.7, 95% CI (1.1; 2.2), p<0.001

• CysSSP was influenced by:• Age B: 1.3, 95% CI (0.8;1.9), p<0.001

S-protein thiolomics to assess the redox-modulation effects of antiretroviral drugs

M. João Correia1, Clara G. Dias1, Umbelina Caixas1,2, Nádia M. Grilo1, Ana R. Lemos1, Diva Trigo3, Patrícia Pacheco3, Emília C. Monteiro1, Karina Soto1,3,

Lucília N. Diogo1, Sofia A. Pereira1

1 CEDOC, Chronic Disease Research Centre, NOVA Medical School|Faculdade de Ciências Médicas, Universidade Nova de Lisboa, Lisboa, Portugal; 2

Centro Hospitalar de Lisboa Central, EPE, Lisboa, Portugal; 3Hospital Prof. Dr. Fernando Fonseca, EPE, Lisboa, Portugal

Figure 3. Comparison between naïve (n=22), nevirapine (n=30) and efavirenz (n=83) groups of A. S-homocysteinylated (HcysSSP), B. S-cysteinylated (CysSSP), C. S-glutathionylated (GSSP) and D. S-cysteinylglycinylated (CysGlySSP) proteins profile. One-Way ANOVA for B. and D. and Kruskal-Wallis test for A.and C.; * p<0.05, ** p<0.01, ***p<0.001.

Naïve NVP EFV0

50

100

150

200

250***

*

CysSSP

( µM)

Naïve NVP EFV0

5

10

15

20*

HcysSSP

( µM)

Naïve NVP EFV0

1

2

3

4

5***

**

GSSP

( µM)

Naïve NVP EFV0

10

20

30

40**

***

CysGlySSP

( µM)

A. B.

C. D.

1

iNOVA4Health

1. Title (per)Sulfidomics: benchmarking mechanisms underlying drug toxicity and drug resistance in precision medicine 2. Team

3. Objectives - Major goals: 1. To evaluate the impact of (per)sulfide-based metabolism dysfunction on drug resistance and drug-induced kidney injury.

Specifically, this project is aimed at evaluating in vitro - the relative contribution of metabolic pathways involved in the synthesis and breakdown of sulfide/thiol-containing metabolites; - the role of persulfides (cysteine persulfide, CysSSH, and glutathione persulfide, GSSH) in drug conjugation, detoxification/resistance - the role of N-acetyltransferase 8 (NAT8) in drug detoxification/resistance

2. To translate the role of common targets and players of (per)sulfide-based metabolism and drug toxicity/resistance found in vitro into clinical studies

4. Background - A new matching theory: Human (per)sulfidome & drug toxicity/resistance interplay --- Metabolic pathways of sulfidome: GSH and H2S homeostasis --- Sulfur metabolism is of paramount relevance in human physiology, particularly through the synthesis of two major physiological regulators: glutathione (GSH), a key determinant of the cellular and systemic redox status, and hydrogen sulfide (H2S), recently recognized as the third ‘gasotransmitter’ (like NO and CO), modulating multiple physiological processes [1]. Presently, there are three recognized endogenous sources of H2S: cystathionine β-synthase (CBS), cystathionine γ-lyase (CGL), and 3-mercaptopyruvate sulfurtransferase (MST) [2, 3]. The canonical end-product of CBS and CGL activity is cysteine, which is further converted to GSH via the sequential action of glutamate-cysteine ligase (GCL) and glutathione synthetase (GS). Both CBS and CGL can also catalyze the conversion of cystine into cysteine persulfide, CysSSH [4]. H2S is catabolized in mitochondria by a sulfide oxidizing unit, involving a sulfide:quinone oxidoreductase (SQR), a persulfide dioxygenase (ETHE1), rhodanese (Rhod) and sulfite oxidase (SOx) [5]. The major sulfide oxidation products are thiosulfate and sulfate. In the course of sulfide oxidation, glutathione persulfide (GSSH) is formed by SQR and it is the preferred substrate of both Rhod and ETEH1. --- Emergence of persulfides as “super-nucleophiles” --- Despite the classic views of GSH and H2S as key elements in sulfur biochemistry and signaling, persulfides and polysulfides (particularly glutathione persulfide, GSSH, and cysteine persulfide, CySSH) have been recently proposed as the major reactive players with a high impact on cell signaling [4, 6]. Interestingly, their increased reactivity and the stronger nucleophilic character of the persulfide moiety suggest a higher propensity to form adducts with electrophiles as drugs, toxic metabolites and endogenous compounds [6]. However, the physiological

Utilizing Phase 3 Clinical Trial Data to Assess Adverse Event (AE) Frequency of a Potentially Interacting Medication (PIM)

Amlodipine with Elvitegravir/Cobicistat (EVG/COBI)D Podzamczer1, K Tashima2, E Daar3, J McGowan4, T Campbell5, J Slim6, M Thompson7,

S Guo8, P Borg9, M Das8, R Haubrich8, I McNicholl8, S McCallister8

1Hospital Universitari de Bellvitge, Barcelona, Spain; 2The Miriam Hospital, Providence, RI, USA; 3Los Angeles Biomedical Research Institute at Harbor-UCLA Medical Center, Torrance, CA, USA;4North Shore University Hospital, Manhasset, NY, USA; 5University of Colorado, Denver, CO, USA; 6Saint Michael’s Medical Center, Newark, NJ, USA;

7AIDS Research Consortium of Atlanta, Atlanta, GA, USA; 8Gilead Sciences, Foster City, CA, USA; 9Gilead Sciences, Stockley Park, UK

� We used five methods to define drug-related adverse events with the use of amlodipine which can potentially interact with cobicistat: – treatment-emergent AEs (unspecified) grade 1-4 by organ class

– treatment-emergent AEs (unspecified) grades 2-4

– amlodipine-specific grade 2-4 treatment-emergent AEs

– common treatment-emergent AEs (defined as those AE occurring in ≥10% of subjects)

– treatment-emergent AEs leading to premature E/C/F/TAF or E/C/F/TDF discontinuation

� Overall or common AEs were similar in participants that did or did not use concomitant amlodipine

� Treatment-emergent AEs grades 2-4 and amlodipine-specific AEs grades 2-4 were higher for participants using amlodipine, but only one participant discontinued E/C/F/TDF or E/C/F/TAF potentially due to an amlodipine AE (local swelling)

� AEs that occurred more often in amlodipine users were palpitations, ataxia, nervous system disorders and peripheral edema

� Because E/C/F/TDF or E/C/F/TAF may increase the level of amlodipine when co-administered, the lowest clinically effective amlodipine dose should be used when combined with E/C/F/TAF

� Overall, patients may use amlodipine if the prescriber determines that the benefits may exceed the potential risks and that appropriate monitoring is conducted

� TAF, a novel oral prodrug of tenofovir (TFV), is more stable in plasma than TDF, allowing for a 10-fold lowering of dose, which results in a substantial reduction (90%) in circulating TFV, while achieving a 4-fold increase in intracellular levels of TFV diphosphate (TFV-DP) and comparably high efficacy1,2

� EVG/COBI (E/C) has shown high rates of efficacy, and when combined in a single tablet regimen (STR) with emtricitabine/tenofovir alafenamide (F/TAF), improved bone and renal safety in treatment-naïve and experienced participants compared to F/tenofovir disoproxil fumarate (F/TDF)

� Previous research suggested that vertigo, nervous system disorders and edema were more common in E/C when combined with amlodipine compared to E/C alone3

� We evaluated the clinical consequences of use of the potentially interacting medication amlodipine in 9 large phase 3 clinical trials. Clinical monitoring is advised when amlodipine is administered with E/C

� To determine if the frequency of adverse events (AE) is increased in subjects who use amlodipine when combined with E/C compared to subjects not using amlodipine

� There were 4,667 participants available from the 9 trials � Amlodipine users tended to be older with a higher percentage of

female and Black subjects (Table 1)

� We retrospectively pooled data from 5 treatment-naïve studies and 4 treatment-experienced studies to assess AEs associated with concomitant use of amlodipine

� Subjects were included in the analysis if they received an STR of E/C/F with either TDF or TAF

Naïve studies � Five Phase 3 randomized, double-blind, double-dummy, active-

controlled studies (treatment-naïve)2,4-6

– GS-US-292-0104; E/C/F/TDF vs E/C/F/TAF – GS-US-292-0111; E/C/F/TDF vs E/C/F/TAF – GS-US-236-0102; E/C/F/TDF vs efavirenz/F/TDF – GS-US-236-0103; E/C/F/TDF vs ritonavir-boosted

atazanavir+F/TDF – GS-US-236-0128; E/C/F/TDF vs ritonavir-boosted

atazanavir+F/TDF in women

Experienced study � Four Phase 3 randomized, active-controlled, open-label study

(treatment experienced)7-10

– GS-US-292-0109; E/C/F/TDF vs E/C/F/TAF – GS-US-292-0112; E/C/F/TAF – GS-US-236-0115; E/C/F/TDF vs ritonavir-boosted protease

inhibitor+F/TDF – GS-US-236-0121; E/C/F/TDF vs a non-nucleoside reverse

transcriptase inhibitors+ F/TDF

Definitions � Treatment-emergent AEs are:

– Any AEs with onset date on or after the study drug start date and no later than 30 days after the study drug stop date, or any AEs leading to study drug discontinuation

� A list of drug-specific AEs was generated for amlodipine based on AEs reported in Micromedex, Lexi-Comp, the package insert and the AHFS Drug Information 201511-14

Endpoints � We evaluated the following treatment emergent endpoints:

– treatment-emergent AEs (unspecified) grade 1-4 by organ class (Table 2)

– treatment-emergent AEs (unspecified) grades 2-4 (Table 3) – amlodipine-specific grade 2-4 treatment-emergent AEs by

grade (Table 4) – common treatment-emergent AEs (defined as those AE

occurring in ≥10% of subjects) (Table 5) – treatment-emergent AEs leading to premature E/C/F/TAF or

E/C/F/TDF discontinuation (Table 6) � Statistical comparisons between users and non-users were

conducted using two-sided Fisher exact tests at the significance level of 0.05 with no correction for multiple comparisons

Clinical Observations � There were no cases of AV node heart block, hypotension and

no clinically significant ECG changes between amlodipine users and non-users

� Some of the AEs reported may be due to the indication for using amlodipine. For example, hypertension (a potential bias by indication)

� The frequency of AEs observed are within the reported amlodipine frequencies from the Micromedex and Norvasc package insert11,13

– palpitations up to 4.5% – peripheral oedema up to 10.8% – ataxia <1% – dizziness 3.4%

� The AEs that occurred in greater than 10% of subjects included the following: diarrhea, nausea, upper respiratory tract infection, nasopharyngitis, headache

� There was no statistical difference in occurrence of common treatment-emergent AEs between users and non-users of amlodipine

� There was a statistically significant difference in discontinuations between subjects using amlodipine versus not (p=0.031)

� The overall number of discontinuations was small (N=8) � In the amlodipine group, discontinuations were due to 1) eye

irritation, eye pain, eye pruritis; 2) local swelling, disturbance in attention; 3) liver injury; 4) increased serum creatinine; 5) lung cancer; 6) hepatic encephalopathy; 7) renal failure (2)

� Of the subjects that discontinued E/C-containing regimens, the only event consistent with an amlodipine-related AE was the subject who discontinued with local swelling

� Dose of amlodipine was not recorded � Small number of actual individual adverse events � No correction for false discovery due to large number of

statistical tests � A multi-variate model to rule out confounders by indication is

pending � Lack of an amlodipine without E/C arm confounds the ability to

determine the baseline frequency of amlodipine AEs

Table 1: Baseline Characteristics- Subjects using Amlodipine or not from 9 Studies

Table 2: Treatment-Emergent AEs (unspecified) by Organ Class†

Table 3: Treatment-Emergent AEs (unspecified) grade 2-4†

Table 4: Amlodipine-Specific Grade 2-4 AEs by Grade

Table 5: Common Treatment-Emergent AEs-occurrence of any event (events that occurred in ≥10% of subjects)

Table 6: AEs Leading to Premature E/C/F/TAF or E/C/F/TDF Discontinuation

© 2016 Gilead Sciences, Inc. All rights reserved.HIV Glasgow, 23-26 October 2016, Glasgow, UK

Ian McNicholl, PharmDGilead Sciences, Inc.

333 Lakeside DriveFoster City, CA 94404

800-445-3252Email: Ian.McNicholl@gilead.com

Introduction Results

Materials & Methods

Objective

Conclusions

Limitations

References

Acknowledgments

1. Ruane P, et al. J Acquir Immune Defic Syndr 2013; 63:449-5.2. Sax P, et al. Lancet 2015; 385:2606-15.3. Tashima K, et al. Intl Workshop on Co-morbidities and Adverse Drug Reactions 2016. Poster 20.4. Sax P, et al. Lancet 2012; 379: 2439-48.5. DeJesus E, et al. Lancet 2012; 379: 2429-38.6. Squires K, et al. Lancet HIV 2016; 3: e410-e420.7. Mills A, et al. Lancet Infect Dis 2016; 16: 43-52.8. Pozniak A, et al. J Acquir Immune Defic Syndr 2016; 71: 530-37.9. Arribas JR, et al. Lancet Infect Dis 2014; 14: 581-89.10. Pozniak A, et al. Lancet Infect Dis 2014; 14: 590-99.11. Micromedex® Solutions. Truven Health Analytics 2012-2016.12. Lexi-Comp Online® Hudson, Ohio: Lexi-Comp, Inc.; October 15, 2015.13. Norvasc® package insert. Pfizer, Inc., New York, NY. March 2015.14. AHFS Drug Information 2015. American Society of Health-System Pharmacists, Bethesda, MD.

We extend our thanks to the patients and their families. This study was funded by Gilead Sciences, Inc.

Poster number

P314

Passcode: P314

Amlodipine Use (N=153)

No Amlodipine Use (N=4514)

Median age, y 50 38Sex Female, % 25 18Race, % White 46 64 Black 50 23 Hispanic/Latino ethnicity 10 19Median HIV-1 RNA, log10 c/mL naïve subjects 4.52 4.64Median CD4 count, cells/mm3

treatment-naïve subjects 398 (N=86) 386 (N=2643) treatment-experienced subjects 606 637Median BMI (kg/m2) 28.7 24.9

Adverse Event Amlodipine Use (N=153)

No Amlodipine Use (N=4514)

Any event 94% (143)* 91% (4101) Infections and Infestations 69% (105) 66% (2991) Gastrointestinal 48% (73) 47% (2107) Musculoskeletal 42% (65) 34% (1523) Nervous system 36% (55) 27% (1200) Metabolism and Nutrition 31% (47) 14% (611) General 31% (48) 21% (939) Respiratory 28% (43) 23% (1047) Skin 25% (38) 27% (1208) Injury and Poisoning 19% (29) 19% (852) Psychiatric 19% (29) 24% (1100) Vascular 16% (24) 6% (290) Investigations 12% (19) 11% (493)

Adverse Event Amlodipine Use (N=153)

No Amlodipine Use (N=4514)

Any event 67% (103)* 56% (2529)Hypertension 5% (8) 1% (55)Peripheral oedema 5% (7) 0.3% (16)Pain in extremity 4% (6) 2% (71)Headache 4% (6) 3% (146)Diarrhea 3% (5) 4% (173)Nausea 3% (4) 2% (96)Muscle spasms 3% (4) 0.5% (24)Paraesthesia 3% (4) 0.2% (9)Asthma 3% (5) 0.7% (33)Anxiety 3% (5) 2% (95)Insomnia 3% (5) 2% (86)Cough 3% (4) 2% (75)

Adverse Event Amlodipine Use (N=153)

No Amlodipine Use (N=4514)

Any event 49% (75) 47% (2113)

Adverse Event Amlodipine Use (N=153)

No Amlodipine Use (N=4514)

Any event 5% (8) 2% (105)

Adverse Event Amlodipine Use (N=153)

No Amlodipine Use (N=4514) P-value

Any event 14% (22) 5% (237) <0.001* grade 2 12% (19) 4% (194) <0.001 grade 3 2% (3) 1% (43) NSpalpitations 0.7% (1) <0.1% (4) NS grade 2 0 <0.1% (4) NS grade 3 0.7% (1) 0 0.033peripheral oedema 5% (7) 0.4% (16) <0.001 grade 2 5% (7) 0.4% (16) <0.001ataxia 0.7% (1) <0.1% (1) NS grade 2 0.7% (1) 0 0.033 grade 3 0 <0.1% (1) NSnervous system disorders† 3% (4) 0.8% (34) 0.035 grade 2 2% (3) 0.7% (32) NS grade 3 0.7% (1) <0.1% (2) NS

†Listed events that occurred in >10% of subjects in the amlodipine arm*P for overall comparison >0.05

†Listing events occurring in >2% in the amlodipine arm*P=0.006 for comparison of any AE between amlodipine use and non-use

*comparisons done for grade 2-4 events between users and non-users of amlodipine†includes dizziness (2), convulsion (1), ataxia (1), sedation, lethargy, somnolenceNS=not significant

Massimo, Tempestilli; Gabriele, Fabbri; Laura, Timelli; Mauro, Zaccarelli; Rita, Bellagamba; Stefania, Cicalini; Federico, Lupi; Anna L, Gallo; Raffaella, Libertone;

Simone, Fazio; Andrea, Antinori; Adriana, Ammassari

National Institute for infectious Disease «L.Spallanzani» IRCCS, Rome, Italy

massimo.tempestilli@inmi.it

* VLLV (very low-level viremia) = HIV RNA detectable <40 cp/ml§ VS (virological suppression) = HIV RNA not detectable <40 cp/ml

General characteristics Patients (n=66)

Gender, no. (%)- male

- female 55 (83.3)11 (16.7)

Age, mean (±SD), y 54.0 (5.3)Weight, mean (±SD), kg 73.2 (12.0)BMI, mean (±SD), kg/m2 24.3 (3.4)

Creatinine, mean (±SD), mg/dl 0.9 (0.2)

Co-morbidity, no. (%) 36 (54.5)Co-medications, no. (%) 34 (51.5)

HCV characteristicsHCV genotype- 1 a, no. (%)- 1 b, no. (%)

- 2 a/b, no. (%)- 3 a/b, no. (%)

- 4, no. (%)

26 (39.4)7 (10.6)1 (1.5)

22 (33.3)10 (15.2)

Liver stiffness at transient elastography - F1, no. (%)- F3, no. (%)- F4, no. (%)

1 (1.5)13 (19.7)50 (75.8)

Ribavirin, no. (%) 28 (42.4)HIV characteristics

Days spent on current cART, median (IQR) 498 (151-1223)

HIV RNA >=40HIV RNA VLLV*, no. (%)

HIV-RNA VS§, no. (%)

4 (6.1)10 (15.2)51 (77.3)

CD4 cells/mmc, median (IQR) 522 (341-693)

Before-DAA During-DAA

Real-world antiretroviral plasma levels in HIV-positive patients treated with sofosbuvir-containing DAA for hepatitis C infection

Background: Drug-drug interactions (DDI) between hepatitis C direct-acting antiviral agents (DAA) and HIV antiretrovirals(ARV) are frequent. To date, most information has been obtained from phase 1 drug–drug interaction studies in healthy volunteers and drug combinations permitted in phase 2 and 3 HIV/HCV co-infection trials [1-2]. Aim of this study was to

investigate antiretroviral (ARV) plasma trough levels before and during Sofosbuvir (SOF)–based treatment in HCV/HIV co-infected patients treated in the real world setting.

Material and Methods: This study is a monocentre, prospective, open-label, observational cohort study. HIV/HCV co-infected persons undergoing HCV treatment with standard dose of DAA and antiretrovirals are enrolled. Patients also need to receive the

same ARVs for at least two weeks before starting DAA treatment and to have HIV RNA <40 cp/ml at baseline. Antiretroviral regimen is prescribed by clinical care providers based on antiretroviral treatment history, previous HIV genotypic resistance

testing, tolerability, and recommendations for management of HCV/HIV co-infected persons in need of HCV treatment. The Ctrough of ARVs is measured using a validated high-performance liquid chromatography (HPLC). Blood samples are collected before and after two months of DAA treatment. For the purpose of this analysis, estimated change of Ctrough form before to

during DAA was obtained by using a random effect linear regression. A minimum of 7 Ctrough coupled values for each ARV were required for final statistical analysis.

Results: To date, 66 out of 91 enrolled patients were analyzed: 27 received SOF+LDV (40.9%), 16 SOF+dactatasvir (24.4%), 6 SOF+simperevir (9.1%) and 17 SOF+ribavirin (25.8%). Concurrent ARVs included atazanavir (n=8), darunavir (n=19), raltegravir(n=19), efavirenz (n=7), etravirine (n=8) and rilpivirine (n=8). No statistically significant difference in Ctrough of considered ARVs

was found in samples obtained before and during SOF-based treatment (figure 1). In 2/66 patients (3.0%) at least one HIV RNA detectable >40 copies/ml during SOF-based treatment was observed. Consequences of loss of virological suppression, such as

resistance development or treatment change, are still under observation.

Tab.1 Characteristics of the 66 ARV-treated HIV/HCV co-infected

patients receiving Sofosbuvir-based

treatment for Hepatitis C infection

Figure 2. Median Ctrough plasma

concentration of atazanavir, darunavir

and raltegravirbefore and during

SOF-based treatment.

Conclusions: In this large HIV/HCV co-infected patient population observed in the real-world setting, no significant modifications in ARV concentrations during SOF-based DAA treatment were observed for the most commonly used antiretrovirals. Nonetheless,

loss of virological suppression does occur during DAA treatment and thus monitoring of plasma drug levels and viral load are advisable.

References: 1. Molina JM, Orkin C, Iser DM, Zamora FX, Nelson M, Stephan C, et al. Sofosbuvir plus ribavirin for treatment of hepatitis C virus in patients co-infected with HIV (PHOTON-2): a multicentre, open-label, non-randomised, phase 3 study. Lancet. 2015;385:1098–106.

2. Saeed S, Strumpf EC, Walmsley SL, Rollet-Kurhajec K, Pick N,Martel-Laferrière V, et al. How Generalizable Are the Results From Trials of Direct Antiviral Agents to People Coinfected WithHIV/HCV in the Real World? Clin Infect Dis. 2016;62:919-26.

A clinical pharmacist reviewed prescriptions of all hospitalized HIV patients with ART on a daily basis. Medication reconciliation was made comparing outpatient medication records with the treatment prescribed to the patient at admission. The pharmaceutical intervention was carried out through a text message associated with the electronic prescription and a phone call to the physician in charge of the patient. The degree of acceptance of interventions was evaluated.

105 patients 124 admissions = 73,4%

Mean (SD) age: 49 (±8.48)

Regarding the published data, the overall medication error rate in HIV patients admitted to a hospital varies between 5.8% and 86%(1), depending on the methodology and study duration. Admission of an HIV-infected patient by a physician not specialized in Infectious Diseases could be a risk factor for drug-related problems. Most of the described errors are prescribing errors(2,3), highlighting the need for a detailed and accurate medication reconciliation on admission.

Error rate was as high as in other studies. Medication reconciliation on admission by a pharmacist helps to correct these errors. Collaboration between hospital pharmacists and HIV Unit with physicians not specialized in Infectious Diseases, and the development of strategies are needed to prevent medication errors in HIV patients at admission.

41 errors

32 patients had at least

one error (30.5%)

32 patients had at least 32 patients

one error had at least

(30.5%)one error (30.5%)

one error

Descriptive observational study. HIV infected patients with any antiretroviral treatment (ART) admitted to a hospital ward were included. The study lasted 5 months (March-July 2015).

Primary outcome

ART error rate

Secondary outcomes

Type of ART error: omission of treatment wrong schedule wrong dose wrong drug pharmacological interaction (according to the University of Liverpool classification of interactions).

Error rate in each type of ward (medical or surgical).

Number of times error reach to patient.

Time until correction of medication errors.

P316

Pablos Bravo, S1; García Muñoz, C1; Pulido F2; Lázaro Cebas, A1; Ferrari Piquero, JM1. 1Pharmacy Department. Hospital Universitario 12 de Octubre. Madrid.

2HIV Unit. Internal Medicine Department. Hospital Universitario 12 de Octubre. Madrid.

34 29

22 15

TYPE OF ERRORS(%)

WRONG DRUG Prescription of TDF+FTC instead of ABC+3TC (4/6) OMISSION Omission of all ART (4/9) WRONG SCHEDULE ETR 200 mg/24h instead of twice daily (5/12) WRONG DOSE 3TC 300 mg/24h in CrC<50ml/min (3/14)

EXAMPLE OF DRUGS AFFECTED (ratio)

Error rate: 34% surgical and 33% medical wards 29%: Errors reached to patients.

54 hours: Median time of exposition to errors 46.3%: Pharmaceutical interventions accepted.

300 mg/24h in <50ml/min (3/14)

INTERACTIONS: 13 forbidden co-administration/ 505 potential interactions Pharmacist intervention (N=2): protease inhibitor + statins The rest was controlled by HIV physicians

Legends: ART: antiretroviral treatment; TDF: tenofovir; FTC: emtricitabine; ABC: abacavir; 3TC: lamivudine; ETR: etravirine; CrC: creatinine clearance. References: 1Li EH, Foisy MM. Antiretroviral and Medication Errors in Hospitalized HIV-Positive Patients. Ann Pharmacother. 2014;48:998-1010. 2Liedtke MD, Tomlin CR, Skrepnek GH, Farmer KC, Johnson PN, Rathbun RC. HIV Pharmacist's Impact on Inpatient Antiretroviral Errors. HIV Med. 2016. 3Zucker J, Mittal J, Jen SP, Cheng L, Cennimo D. Impact of Stewardship Interventions on Antiretroviral Medication Errors in an Urban Medical Center: A 3-Year, Multiphase Study.

Pharmacotherapy. 2016;36:245-51.

P317

Clinical and genetic factors associated with kidney tubular dysfunction in a real‐life single‐centre cohort of HIV‐positive patients

Salvaggio SE1, Giacomelli A1, Falvella FS2, Oreni ML1,  Meraviglia P3, Atzori C3, Clementi E2, Galli M1, Rusconi S11  Infectious Diseases Unit, DIBIC‐University of Milan, Italy; 2 ASST‐Fatebenefratelli‐Sacco, Clinical Pharmacology Dept., Milan ,Italy; 

3 ASST Fatebenefratelli‐Sacco I Division of Infectious Diseases, Milan ItalyBackgroundTenofovir (TDF) is one of the most widely used antiretroviral drug. Despite the high degree of tolerability a small percentage of patients experienced alteration in tubular functionduring TDF use1. Intracellular TDF disposition is regulated by ATP‐binding cassette (ABC) drug efflux transporters and, a reduced transport activity may be implicated in accumulationof TDF into the cells2‐5. The aim of our study was to assess the major determinant of TDF associated tubular dysfunction in a real‐life setting including the usefulness of single‐nucleotide polymorphisms (SNPs) mapping into ABCC2, ABCC4 and ABCC10 genes

Materials and MethodsWe retrospectively analyzed all HIV positive patients who were followed at the Infectious Diseases Unit, DIBIC Luigi Sacco, University of Milan from April 2013 to June 2015. Allpatients treated with TDF during their antiretroviral history and who underwent a genotypization for the functional variants mapping in ABCC2 (‐24 C>T/rs717620 and 4544G>A/rs8187710), ABCC4 (3463 A>G/rs1751034) and ABCC10 (rs2125739 T>C) were evaluated. We defined Kidney tubular disfunction (KTD) in our cohort as the presence ofabnormal proteinuria and/or phosphaturia at 24 h urine collected during the period of treatment with TDF. The aim of our study was to assess the distribution of the differentabnormal proteinuria and/or phosphaturia at 24 h urine collected during the period of treatment with TDF. The aim of our study was to assess the distribution of the differentgenotype of ABCC2, ABCC4 and ABCC10 in patients treated with TDF with or without KTD. Baseline characteristics were compared between patients with or without KTD by using adescriptive statistical analysis (χ2 for categorical variables and Mann‐Whitney test for continuous variables). Associations between genotypes and KTD were tested by univariate andmultivariate logistic regression analyses. The impact of all variables was estimated with univariate analysis, and those with P < 0.20 were incorporated into multivariate analysis.Statistical significance was defined at 2‐sided P value< 0.05.

Results

.100

s

Distribution of different genotypes at rs1751034 of ABCC4 in patients with or without KTD

p = 0 01*

Overall populationN=158

Patientswith KTDn = 42

Patientswithout KTD

n = 116p

N % N % N %ABCC2   ‐24 C>T,  rs717620 

C/C 104 65.38 25 59.5 79 68.10 503C/T 52 32 9 16 38 1 36 31 0

0

20

40

60

80

P ti t ith KTD P ti t ith t KTD

Percen

tage of p

atients

GG

AG

AA

p = 0,01 0.503C/T 52 32.9 16 38.1 36 31.0T/T 2 1.3 1 2.4 1 0.9

ABCC2 4544 G>A, rs8187710  G/G 125 79.1 32 76.2 93 80.2

0.296G/A 29 18.4 10 23.8 19 16.4A/A 4 2.5 0 0 4 3.4

ABCC4  3463 A>G,  rs1751034 A/A 92 58.2 27 64.3 65 56.0

0.010A/G 56 35.5 9 21.4 47 40.5G/G 10 6.3 6 14.3 4 3.5

ABCC1 T>C,  rs2125739

The percentage of patients with KTD was higher among those with "GG" genotype at rs1751034 of ABCC4compared to patients with normal tubular function [6 (14.3%) vs 4 (3.5%), p=0.01]. No statistical significantdifferences were observed regarding the distribution of ABCC2 and ABCC10 SNPs. *χ2

Patients with KTD Patients without KTD T/T 80 50.6 19 45.2 61 52.60.394T/C 66 41.8 21 50.0 45 38.8

C/C 12 7.6 2 4.8 10 8.6Baseline characteristics

OverallpopulationN=158

Patients withKTDn = 42

Patients withoutKTD

n = 116p*

Female [n (%)] 34 (21.5) 8 (19.0) 26 (22.4) 0.649Age (yrs) [median (IQR)] 42 (35‐48) 42 (37‐45) 42 (34‐49) 0.591Non‐Caucasian [n (%)] 5 (3.2) 2 (4.8) 3 (2.6) 0.490Previous therapy duration (yrs)  [median (IQR)] 0.0 (0.0‐6.1) 0.6 (0.0‐8.2) 0.0 (0.0‐3.8) 0.034CD4+ (cells/µL) [median (IQR)] 373 (228‐599) 430 (251‐635) 371 (224‐580) 0.638

Logistic regression of factors involved in thedevelopment of KTD

Bivariate analysis Multivariate analysis[median (IQR)] 373 (228 599) 430 (251 635) 371 (224 580) 0.638HIV‐RNA (log cps/mL)[median (IQR)] 3.8 (0.0‐4.9) 0.0 (0.0‐5.0) 3.9 (0.0‐4.9) 0.081Creatinine (mg/dL)  [median (IQR)] 0.84 (0.72‐0.94) 0.84

(0.70‐0.99) 0.83 (0.73‐0.92) 0.534

GFR (MDRD equation)  [median (IQR)] 103 (89‐117) 106 (89‐118) 101 (89‐117) 0.586HCV coinfection [n (%)] 18 (11.4) 6 (14.3) 12 (10.3) 0.491TDF duration (yrs) [median (IQR)] 5.5 (2.0‐8.5) 6.3 (2.5‐9.2) 4.9 (1.9‐7.9) 0.090Use of protease inhibitors [n (%)] 103 (65.2) 26 (61.9) 77 (66.4) 0.602Diabetes [n (%)] 11 (7 0) 5 (11 9) 6 (5 2) 0 142

y yOR  (95% CI)  p OR  (95% CI)  p

Female vsmale 0,814 0,336 ‐ 1,974 0,650Caucasian vs Non‐Caucasian 1,883 0,304 ‐ 11,684 0,497Baseline age (per year) 0,989 0,956 ‐ 1,023 0,524Previous therapy duration (per year) 1,057 0,988 ‐ 1,131 0,110 0,948 0,854 ‐ 1,051 0,309Baseline CD4 cell count (per cells/mL) 1,000 0,999 ‐ 1,001 0,907Baseline Plasma HIV RNA level ( x log cp/mL) 0,824 0,695 ‐ 0,977 0,026 0,809 0,640 ‐ 1,021 0,075Hepatitis C virus coinfection 1,444 0,505 ‐ 4,131 0,493Baseline creatinine (per mg/dL) 1,999 0,138 ‐ 28,861 0,611Duration of treatment with TDF (per year) 1,076 0,971 ‐ 1,193 0,162 1,032 0,901 ‐ 1,182 0,647Use of protease inhibitors 0,823 0,396 ‐ 1,712 0,602Hypertension 2 039 0 769 5 405 0 152 0 915 0 256 3 266 0 891

During the period of observation 158 patients were selected and genotyped. No statisticalsignificant differences were observed among these two groups of patients regarding age,gender and ethnicity, non‐Caucasian patients were 2 (4.8%) and 3 (2.6%), respectively. Nodifferences were also observed in the distribution of hypertension and diabetes amonggroups. Patients who experienced KTD had a higher prevalence of bone disease [23 (54.8%)vs 32 (27.6%); p=0.002]. * P‐values are for χ2 or Fisher's exact test and Mann‐Whitney test

Carriers of "G" allele in homozygous status at rs1751034 of ABCC4 showed a significant association with KTD (OddsRatio 4.67, 95%CI 1.25‐17.46, p=0.02) in bivariate analysis, but this association was not confirmed by multivariableanalysis.

Conclusions

Diabetes [n (%)] 11 (7.0) 5 (11.9) 6 (5.2) 0.142Hypertension [n (%)] 20 (12.7) 8 (19.0) 12 (10.3) 0.146Bone disease [n (%)] 55 (34.8) 23 (54.8) 32 (27.6) 0.002

Hypertension 2,039 0,769 ‐ 5,405 0,152 0,915 0,256 ‐ 3,266 0,891Diabetes 2,477 0,714 ‐ 8,595 0,153 3,187 0,728 ‐ 13,948 0,124Bone disease 3,178 1,529 ‐ 6,603 0,002 3,412 1,396 ‐ 8,338 0,007Genotype CC at ABCC2      ‐24 (vs CT and TT) 0,689 0,332 ‐ 1,428 0,316Genotype GG at ABCC2   4544 (vs GA and AA) 0,791 0,340 ‐ 1,841 0,587Genotype GG at ABCC4   3463 (vs AG and AA) 4,667 1,247 ‐ 17,464 0,022 2,128 0,440 ‐ 10,292 0,348Genotype TT at ABCC10  (vs TC and CC) 0,745 0,367 ‐ 1,513 0,415

Conclusions• In our real‐life cohort, 26% of patients treated with TDF manifested signs related to kidney tubular dysfunction.• A trend for a KTD protective role for Baseline plasma HIV‐RNA level was evidenced by our analysis.• According to our results ABCC4 rs1751034 appeared to be a genetic determinant of KTD; however, further validation studies are needed for therapy personalization.

Bibliography1. Hamzah L, Booth JW, Jose S, et al; HIVCKD Study. Renal tubular disease in the era of combination antiretroviral therapy. AIDS. 2015 Sep 10;29(14):1831‐6.2. Ray AS, Cihlar T, Robinson KL, et al. Mechanism of active renal tubular efflux of tenofovir. Antimicrob Agents Chemother 2006; 50:3297–304. 10.3. Rungtivasuwan K, Avihingsanon A, Thammajaruk N, et al. Influence of ABCC2 and ABCC4 polymorphisms on tenofovir plasma concentrations in Thai HIV‐infected patients. Antimicrob Agents Chemother. 2015;59(6):3240‐5.4. Rodríguez‐Nóvoa S, Labarga P, Soriano V, et al. Predictors of kidney tubular dysfunction in HIV‐infected patients treated with tenofovir: a pharmacogenetic study. Clin Infect Dis. 2009 Jun 1;48(11):e108‐16.5. Nishijima T, Komatsu H, Higasa K, et al. Single nucleotide polymorphisms in ABCC2 associate with tenofovir‐induced kidney tubular dysfunction in Japanese patients with HIV‐1 infection: a pharmacogenetic study. Clin Infect Dis. 2012

Dec;55(11):1558‐67.

Conclusions Integrase inhibitor use did not result in lower or higher risk for drug-drug inter-actions in our patient cohort. However, dolutegravir-based treatment showed a significantly lower risk for DDI, which was not the case for elvitegravir.

peter.messiaen@jessazh.be tel +32 (0)11309485

Materials and Methods The study population comprised all adult HIV-1 positive patients starting antiretro-viral treatment (ART) in our center from 2009 till april 2016. All prescribed comedications since start of ART were recorded retrospectively from the medical files. The complete treatment was screened for DDIs using the most recent version of the University of Liverpool HIV drug interaction database (www.hiv-druginteractions.org). DDIs were scored as absent, potential, contra-indicated prescription or not assessable due to lack of data. Statistical analysis was performed with SPSS 19. Student’s t-test was used to deter-mine differences in continuous variables between subgroups. The differences bet-ween other parameters were evaluated with Fisher’s exact test.

Introduction Antiretroviral (ARV) agents pose a high risk for drug-drug interactions (DDI) with other ARV and non-ARV drugs. Induction or inhibition of different cytochrome P450 enzymes by protease inhibitors (PI) or non-nucleoside reverse transcriptase inhibitors (NNRTI) is one of the major but not exclusive metabolic pathways poten-tially leading to an increased risk of toxicity or loss of efficacy of ARV and non-ARV drugs. Partly metabolized by other pathways (Table 1), the integrase inhibitor (INI) class might show a more favorable profile. The aim of this study was to investigate the prevalence of potential DDI in patients who recently started ARV and to evalu-ate the effect of INI introduction in clinical practice.

Peter Messiaen1,2, Charlotte Baecke1, Jeroen van der Hilst1,2

Prevalence of drug-drug interactions involving antiretroviral treatment: impact of the integrase inhibitor class

Results Of the 145 patients included, 28% (n=41) were treated on a NNRTI-based regimen,

30% (n=44) on a PI-based regimen and 42% (n=61) on a INI-based regimen (Fig 1). 78% (n=113) of the patients took comedication. Polypharmacy (≥ 5 comedications)

was seen in 26% of patients, significantly correlated with age (p=0.024). Potential DDI was seen in 63% (n=71) of the patients with comedication and in

32% (160/503) of all non-ARV prescriptions. These involved mainly antimicrobial drugs (33%), cardiovascular drugs (19%) and central nervous system drugs (19%). Contra-indicated prescriptions were detected in 1% (n=6), disproportionally more involving gastro-intestinal drugs (66%).

cART treated patients with comedication and DDI trended towards older age, took significantly more comedications and had a significantly lower CD4 T-cell nadir at start of ART compared to those without DDI (Table 2). There was no difference in treatment outcome between the two groups.

1 Department of Infectious Diseases and Immunity, Jessa Hospital, Hasselt, Belgium 2 BIOMED Research institute, Hasselt University, Hasselt, Belgium

Generic Name Adult Dosing Recommendations Route of Metabolisation

Efavirenz (EFV) 600mg qd Predominantly cytochrome P450 (CYP3A4) and CYP2B6 iso-enzymes meta-bolized

Rilpivirine (RPV) 25mg qd Predominantly cytochrome P450 (CYP3A family) metabolized

Nevirapine (NVP) Start 200mg qd for 14 days, then 200mg bd or 400mg qd “extended release” Predominantly cytochrome P450 (CYP3A family) metabolized

Atazanavir (ATZ) 300 mg qd + booster (100 mg ritonavir) Predominantly cytochrome CYP3A family metabolized. Ritonavir is a po-tent CYP3A-inhibitor, enzyme inducer and CYP3A4 substrate

Darunavir (DRV) 600mg bd + booster 100mg ritonavir Conditional alternative dosing: 800mg qd + booster (100mg ritonavir or cobicistat 150mg)

Almost exclusivley cytochrome P450 (CYP3A4) metabolized. Ritonavir is a potent CYP3A-inhibitor, enzyme inducer and CYP3A4 substrate. Cobicistat is a more selective CYP inhibitor and to a small extent metabolized by CYP3A4 and in minor pathway by CYP2D6

Lopinavir (LPV) 400mg bd + booster ritonavir 100mg Predominantly cytochrome CYP3A family metabolized. Ritonavir is a po-tent CYP3A-inhibitor, enzyme inducer and CYP3A4 substrate

Dolutegravir (DTG) 50 mg qd in INI- naive patients, 50 mg bd in INI-experienced patients

Predominantly UGT1A1- mediated glucuronidation, cytochrome P450 (CYP3A4) metabolisation as minor pathway

Elvitegravir (EVG) 150 mg qd + booster (100 mg ritonavir or cobi-cistat 150mg)

Predominantly cytochrome P450 (CYP3A4) metabolized, minor pathways via UGT1A1/3 glucuronidation and oxidative metabolisation. Cobicistat is a more selective CYP inhibitor and to a small extent metabolized by CYP3A4 and in minor pathway by CYP2D6

Raltegravir (RAL) 400 mg bd UGT1A1- mediated glucuronidation

Characteristics cART treated patients with

comedicaton and DDIa (n=71)

cART treated patients with comedication without DDIa

(n=42) P valueb

Median age in years (IQR) 44 (37-51) 39 (33-50) 0.123 Male gender, n (%) 51 (71.8) 34 (81.0) 0.368 Median CD4+ T-cell nadir, cells/µl (IQR) 185 (55-282) 350 (150-420) 0.034 Viral suppression <50 copies/ml, n (%) 70 (98.6) 40 (95.2) 0.554 HBV coinfection (HBsAg-positive), n (%) 1 (1.4) 0 (0.0) 1.000 HCV coinfection, n (%) 5 (7.0) 3 (7.1) 1.000 Median number of non ART comedication, n (IQR) 4 (2-8) 2 (1-4) <0.000 Polypharmacy, n (%) 31 (43.7) 7 (16.7) 0.004 Backbone

TDF/FTC, n (%) 56 (78.9) 23 (54.8) 0.010 ABC/3TC, n (%) 12 (16.9) 16 (38.1) 0.014 AZT/3TC, n (%) 3 (4.2) 2 (4.8) 1.000

Third agent NNRTI, n (%) 17 (23.9) 12 (28.6) 0.658 PI, n (%) 28 (39.4) 9 (21.4) 0.030 INI, n (%) 26 (36.6) 21 (50) 0.174 EVG, n (%) 11 (15.5) 3 (7.1) 0.246 DTG, n (%) 15 (21.1) 17 (40.9) 0.046 RAL, n (%) 0 (0.0) 1 (2.4) 1.000

Table 1: Overview of cART third agents in this study, adult dosing recommendations and route of metabolisation

Fig 1: Chart of the distribution of cART third agents used in the study population (n=145). Co-lors are applied per cART drug class: NNRTI (green); PI (blue) and INI (brown).

P318

Fig 2: Plot of the odds ratios (OR) of cART thrid agents as risk factor for potential or contra-indicated drug-drug interaction (DDI). The black vertical line indicates OR=1, signifying no increased (right) or decreased (left) risk. 95% confidence intervals (CI) are indicated.

Table 2: Patient characteristics of cART treated patients with comedication (n=113), categorized in those with and without DDI

IQR: interquartile range; (a) DDI comprises potential DDI + contra-indicated prescriptions; (b) cART treated patients with comedication and DDI vs those without DDI

PI-based ART was an independent risk factor for potential or contra-indicated DDI (odds ratio (OR) 2.36; 95% confidence interval (CI) 1.14-4.90; p=0.030) (Fig 2). There was no higher risk associated with NNRTI-based ART (OR 0.66; 95% CI 0.32-1.36) as well as for INI-based regimens (OR 0.64; 95% CI 0.33-1.25). A significantly lower risk for drug-drug interaction was seen with dolutegravir-based treatment (OR 0.47; 95% CI 0.22-0.98; p=0.046), though not for elvitegravir-based ART (OR 1.76; 95% CI 0.64-4.82).

The effect of the cART backbone itself on the individual risk for DDI was minimal: careful analysis showed that in only 3 patients DDI was solely due to the backbo-ne: one patient treated with TDF/FTC + EFV and 2 patients with TDF/FTC + DTG.

qd: once-daily; bd: twice-daily

Bonnie Wong1, Denise Pui Chung Chan2, Shui Shan Lee2 1Integrated Treatment Centre, Centre for Health Prevention, Department of Health, Hong Kong 2Stanley Ho Centre for Emerging Infectious Diseases, The Chinese University of Hong Kong

Background

Methods

Results

Conclusion

Therapeutic drug monitoring (TDM) is available as a supplemental clinical service to HIV patients receiving highly active antiretroviral therapy (HAART) in Hong Kong. A high performance liquid chromatography (HPLC) system is in place used to determine plasma drug level and RT-PCR is used to determine CYP2B6-516 genotypes of patients taking EFV. Mid-dose plasma level of EFV of patients started on an EFV-based treatment regimen at Year One (defined as more than 2 months and less than 2 years after treatment initiation) were evaluated. As a sub-study, EFV-treated patients of Integrated Treatment Centre, the largest HIV clinic in Hong Kong, with blood tests performed at 2 or more time points were analyzed.

Efavirenz (EFV) is a commonly used NNRTI command used in HAART regimen for HIV-1 patients. It is metabolized by hepatic cytochrome P450 (CYP) 2B6, which is genetically polymorphic. Genotype 516TT is associated with decreased plasma clearance of EFV and a higher incidence of neurological complications. The pharmacokinetic difference between 516GG and 516GT after long term use of EFV has, however, received less attention.

Fig 1. Comparison of EFV level between GG & GT genotype in the 1st year of HAART

Fig 2. Temporal change of plasma EFV levels – GG vs GT in the sub-study

Year 1 Year 2 Year 3 Year 4/5 GG

(n=23) GT

(n=18) GG

(n=25) GT

(n=11) GG

(n=23) GT

(n=9) GG

(n=30) GT

(n=7) Mean 2.65 3.33 3.03 3.88 2.75 3.04 2.38 2.24 SD 1.27 1.40 1.31 1.54 1.15 1.61 0.97 0.71 p value 0.11 0.10 0.57 0.73

Plasma EFV level of 110 patients were examined in the first part of the study. Their mean age at diagnosis was 42.2 years (SD=11.7), of which 108 (98.2%) were male. Their CYP2B6-516 genotypes were: GG 56 (50.9%), GT 44 (40.0%) and TT 10 (9.1%), the distribution of which was in Hardy-Weinberg equilibrium. At Year One, the mean EFV level of GT was 3.74±1.21mg/L, which was significantly higher than that of GG (3.11±1.17mg/L, p=0.009 in t-test). No significant difference in EFV level between GG and GT could be observed over time when exploring data from 67 patients in the sub-study.

EFV level in patients with CYP2B6-516GT is genotype generally higher than those with GG in the first 2 years after initiation of EFV regimen. Nevertheless, the difference of EFV level between two genotypes is not significant when temporal changes were evaluated. Differential pattern of auto-induction may explain the results elicited in this study. Extrapolation of the results is however cautioned in view of the small number of patient samples tested, which may also be compounded by the high inter- and intra-individual variation of plasma EFV levels.

International Congress on Drug Therapy in HIV Infection, 23-26 October 2016, Glasgow, UK

Acknowledgement: The study was supported by the Council of the AIDS Trust Fund (MSS191R), the Government of Hong Kong SAR. Technical support was provided by Li Ka Shing Institute of Health Science, The Chinese University of Hong Kong.

Table 2. Temporal change of plasma EFV levels – GG vs GT in the sub-study

EFV-treated patients n=110 % Age of Dx (mean ± SD) 42.2 ± 11.7 Male gender 108 98.2 GG genotype 56 50.9 GT genotype 44 40.0 TT genotype 10 9.1 Patients of sub-study n=67 % Age of Dx (mean ± SD) 42.3 ± 12.9 Male gender 65 97.0 Ethnic Chinese 56 83.6 GG genotype 46 68.7 GT genotype 21 31.3

Table 1. Characteristics of patients receiving EFV-based regimen

P319