Transcript Slide 1

PK/PD: New Microbial Diseases
and Model Systems
Tawanda Gumbo, MD
Associate Professor of Medicine,
Division of Infectious Diseases, University of
Texas Southwestern Medical Center,
Dallas, Texas
Mitchison’s three population model hypothesis
• Log-phase growth bacilli (LPG):
Killed rapidly by INH during early bactericidal
activity. Termed BACTERICIDAL effect.
• Slowly replicating bacilli under acidic conditions
(SRB): Killed by PZA during ~8 weeks of therapy.
Believed mostly in macrophages
• Non-replicating persistent bacilli (NRP):
Killed by rifamycins during short bursts of metabolism
but needs up to 6 months to be eradicated. Size
SRB>>NRP
Mitchison DA. The basic mechanisms of chemotherapy. Chest 1979; 76: 771-81
Models for PK/PD work in TB
• Require that one of each of the three Mtb
populations be emulated
• Require that free drug pharmacokinetics of anti-TB
at site of infection be correctly recapitulated
• Mouse models traditionally used. However, it is
becoming clear that the bacillary populations in
mice are different from cavitary TB
M. tuberculosis in the hollow fiber system
Gumbo T, et al. J Infect Dis 2006;195:194-201
GROWTH OF MYCOBACTERIUM TUBERCULOSIS IN THE
HOLLOW FIBER SYSTEM: log-phase growth population
• 5 week experiment
8
• 104 initial inoculum
Log10 CFU/mL
6
• Log-phase growth
pattern during days
1 through 8
4
2
0
0
10
20
Tim e (Days)
30
40
• Bacterial burden
relatively stable
thereafter
ISONIAZID
Background
• Isoniazid has the greatest bactericidal activity,
but ceases to kill after 2-3 days, believed due
to depleting M. tuberculosis in log phase
growth
ISONIAZID
Questions
Is the cessation of isoniazid microbial kill due
to depletion of M. tuberculosis in exponential
growth phase?
ISONIAZID AGAINST M. TUBERCULOSIS
Experimental Design
• Slow and fast acetylators simulated
• Doses studied for each acetylation phenotype:
– 0, 25, 100, and 300 mg a day
• In addition, a dose of 600 mg a day for fast
acetylators
ISONIAZID AGAINST M. TUBERCULOSIS
Resistance, Not Extinction Responsible For Effect Cessation
M. tuberculosis density (log10 CFU/ml)
6
5
4
3
2
Simulated isoniazid-susceptible population
1
Simulated isoniazid-resistant population
Simulated total population
0
0
20
40
Observed isoniazid-resistant population
Observed total population
60
80
Time in hours
100
120
Gumbo T, Louie A, Liu W, Bhavnani S, Ambrose P,Brown D, Drusano GL. (2006).
Journal of Infectious Diseases 195:194-201
PZA dose-effect & dose scheduling studies
• All media acidified to pH 5.8 for entire study, and Mtb
grew slowly in the HFS, to simulate SRB under acidic
conditions
• Half-life =11hr
• Doses mimicked: 0, 7.5, 15, 30, 60, 90, 120 mg/kg daily for
28 days (dose effect)
• EC20, EC40, EC60 then examined in dose-scheduling
studies
• Checked microbial kill and actual PK parameters
achieved
Relationship between PZA AUC/MIC and kill
• The PK/PD index most closely associated with
pyrazinamide sterilizing effect was AUC/MIC
•
E=6.52-(2.89 x
AUC/MIC1.21
34.22 1.2 +AUC/MIC1.21
r2 =0.93; p<0.01
BIG QUESTION
• Where in pulmonary lesions is this large
population that PZA kills?
– The majority, starting with Mackeness & McDermott in
the 1950s and 1960s believe it is within macrophages
• Few voices here and there: extra-cellular
Rates of sterilizing effect with
standard dosing
• Standard dose of 2grams PZA per day in
patients produces 0.1 log10CFU/ml/day in 90%
of patients
• From our inhibitory sigmoid Emax relationship,
the AUC0-24/MIC associated with such a kill
rate of 0.1log10CFU/ml/day is 120
Population PK data
• 100,000 patients simulated using ADAPT II & ADAPT 5
• Pop PK: Wilkins J et al. Eur. J Clin. Pharmacol. 2006; 62:727-735.
– Serum clearance= 3.4 L/h, Volume=30 L.
• PZA penetration in ELF and alveolar macrophages
Conte et al. Antimicrob. Agents Chemother. 1999; 43, 1329-1333
Alveolar macrophage/serum ratio=0.52 to 1.14
Epithelial lining fluid/serum ratio=13.9-21.7
• MICs at pH 5.8:
.
Salfinger M & Heifets LB. Antimicrob Agents Chemother
1988;32:1002-4
Monte Carlo simulation assuming
intracellular location
• MCS targeting alveolar macrophage PZA concentration associated
with AUC0-24/MIC of 120.
•
•
•
•
Processing….
Processing….
Processing…
Processing….
• …..in 0.07% of patients!
• But we know that 2G of pyrazinamide has good sterilizing
effect rates in patients.
Monte Carlo simulation assuming
extracellular location
• MCS targeting ELF associated with AUC0-24/MIC of 120.
•
•
•
•
Processing….
Processing….
Processing…
Processing….
• …..in 80.0-90.0% of patients!
• Match!
PZA conclusions
• Based on these simulations, seems that the
majority opinion is in this case likely incorrect
• In fact, in reviewing Robert Koch’s autopsy
reports that he published, he commented
that most Mtb in pulmonary cavities was
extracellular
Wasana Siyambalapitiyage, Crystal Norton, Sandirai Musuka,
Shashikant Srivasatava, Carleton Sherman