Transcript PA-824 - TB Alliance

Issues in testing regimens
containing multiple novel agents
I. Preclinical Testing
Jacques Grosset
Johns Hopkins University School of
Medicine, Baltimore, MD
What is Preclinical Testing?
• In the year 2005, the general objective of
preclinical testing is to determine whether a
drug active in vitro against Mycobacterium
tuberculosis is likely to contribute to improved
treatment of tuberculosis
• Its specific objectives are to assess the toxic,
pharmacokinetic (PK), and pharmacodynamic
(PD) properties of a given drug .
In vitro assessment of properties
of the drug alone
- MIC, lowest drug concentration that prevents
the growth of at least 99% of the inoculated
bacilli (CFU)
- MBC, lowest drug concentration that kills at
least 99% of the inoculated bacilli (CFU)
- EmaxC, Concentration of Maximal Effect, i.e.,
lowest drug concentration beyond which there is
no additional killing
EmaxC
7
6
log10 kill
5
MBC
4
3
MIC
2
1
0
0
0.01 0.03 0.06 0.12 0.25 0.5
1
2
4
concentrations
8
Log10 kill of M. tuberculosis by INH in 7H9 broth
Exp Expt 2, inoculum 5.41 log10cfu; Expt 3, inoculum 6.27 log10
EmaxC
7
Log10 cfu kill
6
5
MBC
4
MIC
3
2
1
0
0
0.12 0.25
0.5
1
2
4
8
16
concentration
32 (mcg/ml)
Log10 kill of M. tuberculosis by RIF in 7H9 broth
Inoculum 6.27 log10 cfu; CFU counts after 2 weeks of culture in 7H9+OADC
In vivo sequential assessment of
properties of the drug alone
1. Toxicity
2. Basic PK assessment: bioavailability (SIT, SBT) at
non-toxic doses
3. Basic PD assessment: dose-ranging activity (MED,
MBD, EmaxD)
4. If basic PK & PD data are favorable (the higher the
ratio of toxic dose/effective dose, the better),
- Dose fractionation studies to establish the PD
parameters most closely correlated with bactericidal
activity (AUC/MIC, Cmax/MIC, Time>MIC )
Example of a dose fractionation study
Jayaram et al, AAC (2003); 47:2118-2124
CFU counts in mouse lung
After 10mg/kg
CFU counts
after 6 days of
treatment
RIF pharmacodynamic parameter
An example
What should not be done….
(to jump to mouse experiments without solid data)
Vehicle Control
CMC
many visible lesions
4wks after infection
INH 25mg/kg
(after 4wks treatment)
no visible lesions
Four unknown compounds
X1 100mg/kg
X2 100mg/kg
X3 100mg/kg
X4 100mg/kg
Even in groups treated with the highest doses, all mice had nodular
lung lesions similar to those observed in controls
Serum Inhibitory Titer of compound “x”
Serum
Control
0
INH po
INH ip
“x1” po
“x1” ip
serum
+, +
Dilutions of serum in 7H9
broth
1/2 1/16 1/32
+
+
+
+
+
+
1/64
1/128
0
0
0
0
+
+
+
+
+
po, single oral dosing; ip, single intraperitoneal dosing
INH, 25mg/kg; Compound “x”, 100mg/kg
+, culture positive; 0, culture negative on day 14
Conclusion: No active serum concentrations of “x”
What should be done ?
Take no short cuts but apply sequential procedures
•
Screen MIC (5, 1.25, and 0.15µg/ml) by
standard validated method
• If, and only if, MIC is favorable (≤1.25
µg/ml), perform serum inhibitory test in the
mouse
• If, and only if, the titer of the serum is
favorable (≥1/4), determine MED, MBD
EmaxD in the mouse.
Dose-ranging activity of PA-824
log CFU in lungs
8
7
MED
6
MBD
5
MED
4
3
2 logs
MBD EmaxD≥200
2
1
IN
H
20
0
PA
10
0
PA
50
PA
25
PA
12
.5
PA
Pr
et
re
at
m
en
t
Un
tre
at
ed
0
After 4 weeks of daily (5/7) treatment in mice aerosol infected with 5x103 CFU
Properties of the drug alone
Secondary in vivo assessment
• Confirmation of “bactericidal” activity:
- select drug- resistant mutants when given alone
- prevent selection of INH-resistant mutants
when combined with INH
• Assessment of “sterilizing” activity: ability of
the compound to kill bacilli that persist after 2
months of daily treatment with RHZ.
Initiation of treatment
Log
10
CFU count
12
Death of untreated controls
10
8
PA-824
6
6.02
5.83
5.53
4
3.94
H
H + PA
RHZ
2
0
Days
-20
0
56
CFU counts in the lungs of mice treated with
PA-824 alone or in combination
Selection of drug-resistant mutants
Proportion of colonies resistant to:
INH
PA-824
Regimen
No treatment1
(0.2 µg/ml)
(2 µg/ml)
1.3 x 10-6
9 x 10-7
INH alone
2.5 x 10-4
--------
PA-824 alone
--------
3.8 x 10-3
INH + PA-824
< 5 x 10-6
5 x 10-6
1Stover
et al, Nature (2000);405:962
Log
10
CFU count
CFU counts in the lungs of mice treated
with PA-824 in the continuation phase
10
9
8
7
6
5
4
3
2
1
0
No treatment
Initial phase
regimen:
RHZ
Continuation
phase regimen:
M
H
PA-824
RH
3.94
2.48
1.92
0.60
0
-3
0
8
16
24
Weeks
Assessment of activity in combination
therapy: Two past examples
1) the impact of rifampin
2) the impact of pyrazinamide
Log10 cfu in lungs
8
7
6
5
INH + SM
INH + RIF
4
3
2
1
months
0
0 1
2 3
6
9
12 18
Comparative bactericidal activity of INH + SM vs. INH + RIF in
mice … as in humans
(Tubercle 1967;48:11-26; Tubercle 1962;43: 201-67; Tubercle 1969; 50 (march suppl):12-21)
Failure and relapse rates after INH+SM and
INH+RIF*
Proportion of Mice and (Humans) with
Positive Cultures:
Regimen
Duration (mo.)
On completion of
treatment
3-6 mo. after
treatment
INH+SM
6
100% (0)
100% (29)
INH+SM
18
35% (0)
75% (~10)
INH +RIF
6
0% (0)
20% (6-7)
INH+RIF
9
0% (0)
0% (1-3)
*From Mitchison; and Grosset & Ji; in Gangadharam & Jenkins, Chapman & Hall, 1998
Conclusion: because the mouse model is a pessimistic model,
results achieved in the mouse are likely achievable in humans
Log10 cfu in lungs
8
7
6
5
INH + SM
INH + RIF
INH + RIF + PZA
4
3
2
1
months
0
0 1 2 3
6
9
12 18
Comparative bactericidal activity of INH + SM, INH + RIF, and
INH + RIF + PZA in mice… as in humans
(Grosset, Tubercle 1978: 59:287; EA/BMRC, Tubercle 1986;67:5)
22
Failure and relapse rates after INH+RIF (HR) and
INH+RIF+PZA (HRZ)
Drug regimen
6HR
2HRZ/4HR
Proportion of Mice and (Humans) with
Positive Cultures:
On completion of
3-6 mo. after
treatment
treatment
0-10% (0)
0%
(0)
40-60% (6-7)
10-30% (1-2)
From Mitchison; and Grosset & Ji; in Gangadharam & Jenkins, Chapman & Hall, 1998
Assessment of activity in
combination regimens
• Activity after incorporation into the first-line
regimen (2RHZ/4RH)
– as supplement
– as substitution
• Activity after incorporation into new regimens
Pre-requisites for combination
experiments in murine model
• Realistic appraisal of doses to be tested
• Assurance of compatible pharmacokinetics
• Selection of infection parameters and
outcomes relevant to human disease
1. Assessment after incorporation in
the first-line regimen (2RHZ/4RH*)
• Activity of moxifloxacin
* R, rifampin; H, isoniazid; Z, pyrazinamide
Log10 CFU in
entire lung
9
8
7
Untreated
6
2RHZ+4RH
5
2RHZM+4RHM
4
2RHM+4RH
3
2RMZ+4RM
2
2MHZ+4MH
1
0
0
1
2
3
4
5
6 Duration of treatment(mos.)
Results of log10 CFU counts from lung homogenates in
mice treated with MXF and standard regimen 2RHZ/4RH.
Conclusions
1. The addition of MXF did not significantly
improve the sterilizing activity of RHZ.
2. The substitution of MXF for R or Z was
detrimental to the activity of RHZ
3. But, the substitution of MXF for H
provided a regimen with substantially
improved sterilizing activity
4. Phase II clinical studies evaluating the
RMZ regimen will soon be underway
2. Assessment after incorporation in
new regimens
• Activity of PA-824 in the RMZ* regimen
* R, rifampin; M, moxifloxacin, Z, pyrazinamide
8
Log(10) CFU count
7
6
5
4
Lung
3
Spleen
2
1
0
Pre-Rx
RMZ
RMZPa PaMZ
Regimens
RPaZ
RMPa
CFU counts after 2 months of treatment
Proportion of mice relapsing after 3
months of therapy
*
*
79%
78%
100%
90%
% relapse
80%
46%
70%
60%
50%
3
40%
2
30%
20%
1
10%
0%
2RMZ + 1RM
2RMZPa +
1RMPa
2PaMZ + 1PaM 2RPaZ + 1RPa
3RMPa
*p<0.05 vs. RMZ
Conclusions
1. The addition of PA-824 to the RMZ regimen did
not improve the sterilizing activity of RMZ.
2. R is more sterilizing than PA-824, and the
substitution of Pa for M or Z was detrimental to the
activity of RMZ
3. But, the substitution of Pa for R provided a
regimen with sterilizing activity approaching that of
RMZ
4. Such a “PaMZ” regimen, without R and H, has
great potential for HIV-TB, MDR-TB, etc.
To conclude
1. To date, the mouse model of TB chemotherapy has
provided results predictive of clinical outcomes.
2. However, it is essential that the model utilizes the:
- appropriate mouse species
- appropriate infection with M. tuberculosis
- equipotent dose of drugs
- appropriate time points to assess cure
3. Preclinical testing of a drug is the best way to determine
whether there is a need for a clinical trial (Nardell &
Rubin, AJRCCM 2005; 172: 1361-62)
Acknowledgements
All of these studies could not have been
performed without the support of:
- TB Alliance
- NIAID (NIAID-DAIDS N01 AI 40007, NIAID K08
AI 58993, NIAID-DAIDS R01 AI 43846).