Embed Size (px)
Citation preview
Pyrazinamide, an Amazing Old Drug as a Corner Stone in Novel TB Drug Regimens Khisi Mdluli – TB Alliance
5th International Workshop on Clinical Pharmacology of Tuberculosis Drugs 8th, September, 2012 San Francisco, USA
No new drugs for TB in 40 years!
Drug-sensitive TB 4 Drugs, >6 months M(X)DR-TB Few available drugs; >2 years; poorly
tolerated TB/HIV co-infection Drug-drug interactions with
antiretroviral agents (ARVs) Latent TB Infection 9 Month H
Current TB Therapy and Unmet Needs Shorter, simpler therapy
More effective, safer regimens; shorter, simpler therapy
Shorter, Co-administration with ARVs
Shorter , more easily tolerated therapy
Unmet Needs Current Therapy
2
3
• Properties of pyrazinamide (PZA, Z)
• Lessons learned from pre-clinical models
• Contribution of PZA beyond the first 2 months of therapy
• Role of PZA in new combinations
• The mechanism of action – Unknown (pmf, FasI, RpsA, NAD)???
• Resistance to PZA and the need for a PZA DST
• Future studies
Structure of Presentation
Tackling TB Through Technology
4
• Very low molecular weight
• Pro-drug that is activated by a bacterial enzyme, PncA (also pro-drugs: INH, ETH, Nitroimidazoles??)
• Resulting pyrazinoic acid, (POA) is necessary for activity
• Side effects – hepatotoxicity and arthralgia
• Bulkiest portion of the standard regimen – high dose of 1.5g daily
Pyrazinamide – Chemical Properties
Tackling TB Through Technology
5
• Very high in vitro MIC – almost inactive
• Requires low pH
• Better activity against non-replicating organisms
• Synergizes other drugs or drug combinations
• Contributes strongly to treatment shortening
• Suggests that in an infection: – PZA acts on a specific sub-population of persisting bacteria – Resides in an acidic environment – Not as susceptible to other anti-TB drugs – Responsible for most relapse – Enriched in established infections
Pyrazinamide – Biological Properties
Tackling TB Through Technology
Historical Human Treatment Duration Shortening Regimens
Development of Regimens
1946 – First randomized trial : S Monotherapy led to S resistance
1952 – First regimen: S/PAS/H 24 months of therapy
1960s – PAS replaced by E: S/H/E 18 months of therapy
1970s – Addition of R: S/H/R/E 9-12 months of therapy
1980s – S replaced by Z: H/R/Z/E 6-8 months , oral therapy
1940 1950 1960 1970 1980 1990 2000 2010
Lessons Learned from Pre-Clinical Models
1-day incubation period 21-day incubation period
McDermott et al, Am Rev Tuberc 1956: 74:100
PZA is A Lot More Efficacious in an Established Infection
PZA is Not Very Efficacious in an Acute Infection
Rullas et al, AAC 2010; 54:2262
INH
RIF
MXF
PZA
Human-equivalent doses
• 1-day incubation after 10^5 CFU intratracheal instillation • 8 days of treatment • Dotted line is ED99, which is essentially the baseline CFU value
PZA is A Lot More Efficacious in an Established Infection
Pa50
Pa100
PZA J
RIF PNU
LZD
INH MXF
SM
EMB
-2,5
-2
-1,5
-1
-0,5
0C
hang
e in
log
CFU
in lu
ngs
Drug
Log kill between Day 0 and D28
E. Nuermberger PZA Workshop June 2011
PZA is A Lot More Efficacious in an Established Infection
Pa50
Pa100 PZA J
RIF PNU
LZD
INH MXF SM
EMB
-2,5
-2
-1,5
-1
-0,5
0
Cha
nge
in lo
g C
FU in
lung
s Drug
Log kill between Day 0 and D28
1-day incubation period 28-day incubation period
PZA150 = permits bacillary multiplication PZA150 = mean kill of 0.1 log CFU/dose
PZA has Limited Efficacy in Other Mouse Models
Anne Lenaerts, Essentiality of PZA Workshop, Bethesda, 6/1/11
Dose-Ranging Activity of PZA in Acute Infection
0
1
2
3
4
5
6
7
Pre-Rx H3.13 Z150 Z300 Z600 Z900
Lung
log 1
0 CFU
cou
nt
Regimen
Bacteriostasis
1 log kill
Dose-Dependent Activity of PZA Against Chronic TB in Mice and Guinea Pigs
Mice Guinea pigs
• Pyrazinamide given at 1/4 and 1/2 the human-equivalent - minimally active • Human-equivalent doses reduced lung CFUs by ∼1.0 log(10) • Doubling the human-equivalent dose reduced CFUs by 1.7 and 3.0 log(10) in mice and guinea pigs, respectively. • As in humans and mice, pyrazinamide showed significant synergy with rifampin in guinea pigs.
Ahmad et al AAC 55(4):1527-1532, 2011
PZA Effect in Whole Blood Culture
M. bovis BCG M. tuberculosis H37Rv
- -0.135 - +0.220
Vit D -0.151
Vit D+U -0.372 Z effect: U -0.311 Z effect:
Vit D+U+Z -0.410 -.038 U+Z -0.411 -.100
Wallis et al, AAC 2011; 55:567 & PLoS ONE 2012; 7:e30479
16
• MTB replicating slowly at pH 5.8 was exposed to PZA at human encountered concentration-time profiles
• Daily pyrazinamide dosing for 28 days accurately achieved: – the pyrazinamide pharmacokinetic parameters
– the lack of early bactericidal activity
– a sterilizing effect rate of 0.10 log(10) CFU/ml per day starting on day 6 of therapy
– a time to the emergence of resistance of the from 2 to 3 weeks of monotherapy as in patients
• The sterilizing effect was linked to the PZA AUC/MIC
• Resistance suppression was associated with Time above MIC
• Monte Carlo simulations of patients demonstrated: – PZA concentrates in epithelial lining fluid 18x
– target serum AUC is ~83 is achieved in 80-90% of virtual patients
– PZA does not concentrate in alveolar macrophagess
– target serum AUC is ~1750 (achieved <0.1% of virtual patients)
– Therefore the bacilli susceptible to PZA in humans are extracellular
Pyrazinamide PK/PD in in vitro HFS
Tackling TB Through Technology
Gumbo T, Dona CS, Meek C, Leff R. AAC. 2009. 53(8):3197-204.
Correlation of PD Parameters and Effect in Mice
AUC/MIC
T>MIC Cmax/MIC
Ahmad et al, ICAAC 2010
Correlation of PD Parameters and Effect in Mice
EC90 = AUC0-24h of 123 µg-h/ml associated w/ 0.106 log10 CFU/d reduction in mice (like extended EBA in man)
By comparison, in the in vitro HFS, the AUC associated with a 0.11 log CFU/ml/d reduction was 1500 µg-h/ml, or 12x higher
Serum exposures produced by Z doses recommended for humans (eg, AUC0-24= 200-550 µg-h/ml) are sufficient for maximal bactericidal activity vs. intracellular bacilli in mice.
The discrepancy between the predictions based on the in vitro system and the results in mice, may be explained if bacilli in activated macrophages of chronically infected mice are more susceptible to Z than the “young” extracellular bacilli at pH 5.8 in vitro.
Ahmad et al, ICAAC 2010
Conclusions
• The efficacy of Z is demonstrable in a variety of models. • Factors determining efficacy include:
– stage of infection and – pH – Others??
• These factors must be considered in using these models to optimize dosing, and select the best combinations.
• Tolerability and resistance status of infecting organism
PZA Contribution Beyond the First 2 Months
EBA of PZA in Combinations
Jindani et al, AJRCCM 2003; 167:1348 Diacon et al, AAC 2010; 54:3402 Diacon et al, AAC 2012; 56:3027 Diacon et al, Lancet 2012; epub July 23
EBA0-2 EBA2-14
Companion Drugs
No Z Z Diff No Z Z Diff
Nil -.017 .054 .061 .024 .114 .090
H .722 .485 -.237 .112 .096 -.016
S .071 .118 .047 .130 .177 .047
SH .379 .797 .418 .071 .151 .080
SHR .320 .694 .374 .143 .161 .018
J -.022 .079 .101 .076 .143 .067
Pa .134 .170 .036 .105 .148 .043
PZA Adds Sterilizing Activity to RIF-INH, but Only in the Initial Phase in Mice
Total mice n (%) relapsing 6 mo after treatment
completion
6RH 52 19 (37%)
2RHZ / 4RH 47 5 (11%)
Total mice n (%) relapsing 6 mo after treatment
completion
6RH 52 31 (60%)
3RH / 3RHZ 55 32 (58%)
p= 0.0042
Pathol Biol 1982;30:444
Contribution of PZA Beyond the 1st 2 Months in the SHZ Regimen
0102030405060708090
100
0 2 6 9 12
% c
ultu
re-p
ositi
ve m
ice
Months
2SHZ / 10H2SHZ / 10HZ
Only 1 CFU per mouse
Bull Int Union Tuberc 1978;53:5
Contribution of Z Beyond the 2nd Month in 2nd-Line Regimens
012345678
0 1 2 3 4 5 6
Lung
log 10
CFU
cou
nt
Months
2RHZ / 4RH2MEZA / 4ME2MEZA / 4MEZ2LEZA / 4LE2LEZA / 4LEZ
M = moxifloxacin L = levofloxacin E = ethionamide A = amikacin
Regimen Proportion (%) relapsing after treatment for:
5 months 6 months 7 months
2 RHZ / 4 RH 7/30 (23%) 0/30 (0%) ND
2 MEZA / 5 MEZ 28/29 (97%) 17/29 (59%) 6/30 (20%)
2 LEZA / 5 LEZ 26/26 (100%) 23/29 (79%) 11/29 (38%)
Nuermberger et al., Demystifying PZA. 2012
Conclusions
• PZA may contribute sterilizing activity beyond the first 2 months in the murine model, but only in regimens without INH and RIF.
• Reasons for the difference may include: – overlapping killing of PZA-susceptible persisters by RIF – antagonism of PZA’s sterilizing effect by INH – more rapid resolution of inflammation with more potent RHZ combo
Role of PZA in New Combinations
Efficacy in M. tuberculosis ‘Active’ Mouse Infection Model: Enhanced Contribution from PZA in Combination Regimens
• BALB/c mice, high dose aerosol (3.5 log10 CFU from late log phase culture M. tuberculosis H37Rv) • Establish infection for 14d (lung burden > 7 log10 CFU at d14 post-infection) prior to treatment • PO drug treatment (1x/d, 5d/wk, x multiple months) • Harvest lungs: gross path, histology and plate homogenate to Middlebrook 7H11 agar to enumerate tissue CFU
E. Nuermberger PZA Workshop June 2011
3.17
6.34
2.73
5.24
2 months treatment 1 month treatment
Tasneen et al 2008 AAC 52(10) :3664
>3 log10 CFU >3.5 log10
CFU
Treatment group
Proportion (%) with positive M.tb cultures 3 mo after completing treatment for:
2 mo 3 mo 4 mo 5 mo 6 mo
2HZR/3HR 50% (7/14)
14% (2/14)
0% (0/14)
JZ 0% (0/15)
0% (0/15)
JZ + (R, M, L or Pa)
0-7% (0-1/15)
0% (0/15)
JZ + (P or C)
0% (0/15)
0-7% (0-1/15)
0% (0/15)
Tasneen et al, AAC (2011);55:5485
JZ-containing combinations, including JZPa, accomplish in 2-3 months what takes the standard regimen 5-6 months
The Efficacy of JZ-Based Combinations
Lung CFU cts % (proportion) relapsing D-17 D0 W4 (+12) W6 (+12) W8 (+12) W10 (+12)
Untreated 4.41 ± 0.08 8.32 ± 0.26 PZM 47%
(7/15) 13%
(2/15) JZ 93%
(14/15) 67%
(10/15) 53%
(8/15)* JZC 7%1,3
(1/15) 0%1,4 (0/15)
JZU 53%2 (8/15)
40% (6/15)
1p≤ 0.005 vs. JZ; 2p= 0.0528 vs. JZ 3p< 0.05 vs. JZU; 4p= 0.0507 vs. JZU
CFZ and PNU-100480 significantly improve the sterilizing activity of the JZ building block
Williams et al, AAC (2012)
The Effects of Clofazimine and PNU on JZ
Efficacy of Pyrazinamide-Containing vs Pyrazinamide-Sparing Regimens
Antimicrob. Agents Chemother. 2011, 55(12):5485.
Efficacy of Pyrazinamide-Containing vs Pyrazinamide-Sparing Regimens
% (Proportion) Relapsing After M1 After M2 After M4
Untreated
2RHZ/3RH 15/15 (100%)*
2JPZ 15/15 (100%) 0/15 (0%)
4JMZ 15/15 (100%) 5/15 (33%) 0/15 (0%)
2PMZ 15/15 (100%) 15/15 (100%)
4PaMZ 15/15 (100%) 10/15 (67%)
2PaPZ 15/15 (100%) 15/15 (100%)
4PaPM 15/15 (100%) 13/15 (87%)
4JPM 15/15 (100%) 7/14 (50%)
4JPaM 15/15 (100%) 7/14 (50%)
Antimicrob. Agents Chemother. 2011, 55(12):5485.
The Sterilizing Power of PZA
After M1 After M2 After M4 Untreated
2RHZ/3RH 15/15 (100%)*
JZP 15/15 (100%) 0/15 (0%)
JZM 15/15 (100%) 5/15 (33%) 0/15 (0%)
JPM 15/15 (100%) 7/14 (50%)
JPaM 15/15 (100%) 7/14 (50%)
Antimicrob. Agents Chemother. 2011, 55(12):5485.
PZA is critical to achieve dramatic treatment shortening, but the novel JPaM regimen is still superior to RHZ
*RHZ still had 1.65 +/- 0.23 log CFU in lungs at M4 time point
p < 0.05 vs. RHZ
Efficacy of RHZ-Sparing Regimens Lung CFU Counts % (Proportion) Relapsing
D-13 D0 M1 M2 M2 (+3) M3 (+3) M4 (+3)
Untreated 3.54 7.25 2RHZ/4RH 4.73 + 0.29 3.04 + 0.27 100%
(15/15) 64%
(9/14)
JUCPa 3.48 + 0.57 0.37 + 0.75 93% (14/15) 13% (2/15)1,2
7% (1/15)
JUC 3.37 + 0.74 0 87% (13/15)
27% (4/15)1,2
7% (1/15)
JUPa 3.99 + 0.89 0.97 + 1.18 100% (15/15)
43% (6/14)1,2
0% (0/15)
JCPa 4.39 ± 0.51 1.55 ± 1.14 100% (15/15)
60% (9 /15)
33% (5/15)
UCPa 4.47 + 0.39 0.82 + 1.64 100% (15/15)
100% (15/15)
100% (15/15)
Williams et al, AAC (2012)
Bedaquiline ± PZA – Mouse vs Humans
• TB Alliance NC-001 trial
Andries et al, Science (2005); 307:223 Ibrahim et al, AAC (2007); 51:1011 Lounis et al, AAC (2008); 52:3568
Adapted from Diacon et al, ICAAC (2011) Lounis et al, AAC (2008); 52:3568
PaMZ is a Synergistic Combination in Mice
0
1
2
3
4
5
6
7
8
9
0 4 8
Untreated
PaMZ
PaM
PaZ
MZ
Weeks
Comparison of NC-001 and Mouse EBA Results -2
.5-2
-1.5
-1-.5
0.5
logC
FU c
hang
e fro
m b
asel
ine
0 2 4 6 8 10 12 14Day
TMC207 TMC207 & PyrazinamideTMC207 & PA-824 PA-824 & Pyrazynamide PA-824 & Pyr & Moxifloxacin Rifafour e275
Bi-linear Regression: logCFU change from baseline
Nuermberger et al, AAC (2008); 52:1522 Diacon et al, ICAAC (2011)
Resistance to PZA and the Need for a PZA DST
Resistance to PZA • Estimated that ~50% of MDR isolates globally are PZAR strains
– demonstration of RIFR is sufficient for classification as MDR-TB (ie., INHR) • Nearly one-third of retreated tuberculosis patients in Peru have PZAR strains (Vazquez-Campos et al, 2004 Int. J. Tuberc Lung Dis 8:465-472) • Estimated that ~5% of M. tuberculosis isolates globally are PZAR strains • Conclusion from a 1997 mss that “all genuine PZAR strains … were found to have pncA mutations” (Scorpio et al 1997 AAC 41(3):540)
• subject to interpretation as to what criteria are used to claim PZAR
• several literature examples that cite PZAR strains w/o any pncA mutations
Consensus in the field (i) the vast majority of Mtb PZAR strains possess mutations in pncA (ii) only a small number of Mtb PZAR isolates have wild type pncA nucleotide sequence (iii) additional mechanisms for resistance to PZA (e.g., mutations at rpsA) may still yet
be uncovered
Given this, is it a sound strategy to invest in the development of a pncA-centric molecular-based diagnostic for PZA DST? – YES, it seems that way!!
Active Site Residues Metal Chelation Residues Role in Catalysis Implicated from Modeling
Distribution of MtbPncA Missense Mutations in PZAR Strains of M. tuberculosis
Zimic et al, Infection Genetics and Evolution (2010) 10:346 Chang et al, AAC epub 18-July-2011
M1T/I A3P/E, L4S/W, I5N/S, I6T, V7A/F/D/G/I D8E/N/Y/G/H, V9A/G, Q10K/P/R, D12A/G/N, F13S, C14R/Y/W/G, G17D/S, L19P, G23V, G24D A25E, A26G, L27P/R, A28D/E Y34D/S, L35P/R, A36V, Y41H H43P, V45G, A46V/E, T47A/P/S, K48E/T, D49A/V/G/N, H51N/Y/P/Q/R, D53A/N, P54T/L/Q/R/S, H57D/P/R, F58L, S59P, T61P, P62T/R/H/L, D63A/G/H, Y64D, S66P, S67P, W68R/G/L/S, P69L/R, H71Y/D/R/E/P/T, C72R/W, T76P/I, G78C/D/V, A79V, H82D/R/L, L85P/R, T87M I90L/S, V93L, F94P/S/L K96E/Q/T/N, G97S/C/D, Y99D, A102T/V, Y103H/C/S, S104R/C, G105D, G108D, K111Q, T114P, L116R/P N118T, W119R/L, L120P/R, R121P R123L/P, V125F/G/D V128G, V130G G132V/A/D/S, I133N/S/T, A134V/F, T135P, D136H/N/G/Y, H137R/P, C138R/S/Y, V139L/M/G/A R140S, Q141P, T142M/K/P/A, A146T/V, R148S L151S T153N, R154G, V155G/A, L156Q, V157G L159P/R/V, T160P, A161P, G162D, V163R, S164P T167I, T168N, A171T/E/V/P, L172A/P, M175V/T V180A/F L182S
Loop1 - β1 -
Loop2 - α1 -
Loop3 - β2 -
Loop4 -
β3 - Loop5 -
α2 – Loop6 –
β4 - Loop7 –
α3 - Loop8 -
β5 - Loop9 -
α4 - Loop10 -
β6 –
Parallel beta sheet w/alpha helices
NH2 terminus
CO2H terminus
Mutations in Genes Involved in Resistance to Pyrazinamide
• Analyzed whole genome sequences of 215 Mycobacterium tuberculosis strains representing all lineages described.
• Lineages coded by color as follows: Lineage 1 (L1), L2, L3, L4, L5, L6.
• Strains analyzed per lineage: L1(46), L2(37), L3(35), L4(64), L5(16), L6(15), two animals strains and BCG.
• Identified non-synonymous substitutions leading to a change of amino acid.
• Identify small (<40bp insertions or deletions) • Methods
• Identify strains (and lineages) involved in mutations • Asses the impact of mutations by predicting whether an amino acid substitution
affects protein function. SIFT prediction is based on the degree of conservation of amino acid residues in sequence alignments derived from closely related sequences, collected through PSI-BLAST.
PncA Polymorphism in Global Strains
Conclusions
• PZA is likely to be a key player in any regimen used against susceptible isolates, with optimal dose and duration likely dependent on the susceptibility of the isolate and the sterilizing activity of companion drugs
• Therefore, improved PZA-DST has to be developed both for clinical trials and for point of care
• Greater understanding of exposure-response relationships in a variety of pre-clinical models and in the clinical setting is necessary to optimize dosing
• Careful study of PZA-resistant mutants selected on therapy or isolated from non-responders may also help understand MOA and clinical significance of specific mutations
• Animal models can be useful in determining the clinical significance of specific PncA mutations
Acknowledgements
• Eric Nuermberger • Charles Peloquin • Koen Andries • Bob Wallis • Tawanda Gumbo • Anne Lenaerts • Paul Liberator
