Measuring Intralesional Drug Exposures in Cavitary TB using Noninvasive In Vivo PET Imaging

Effective treatment of infections depends on achieving adequate antibiotic concentrations at infection sites,
where the pathogen resides. However, with few exceptions, current antibiotic dosing recommendations are
based on achievable plasma concentrations, without specific information on drug concentrations at the site of
infection. However, plasma drug levels do not correlate well with those at infection sites. Cavitary lesions,
which are the hallmark of human tuberculosis (TB), have limited drug penetration and consequently are a risk
factor for treatment failure, recurrence, and the emergence of antibiotic resistance. Direct tissue measurements
are invasive, can be performed in humans only when clinically indicated, and generally provide data at a single
time-point even in animal models. Additionally, given that multiple, pathologically distinct TB lesions coexist
within the same infected-host simultaneously, measurements from one or a few easily accessible lesions are
subject to sampling bias. Finally, current antibiotic treatment strategies are designed for efficacy (e.g. >85%) at
a population level, but ignore the inter- and intra-subject heterogeneity. While shorter treatments could cure
e.g. >70%, tools to identify patients at-risk for treatment failure or requiring longer treatments are needed.
We have developed novel tools to perform noninvasive, simultaneous and unbiased, multi-compartment in
situ measurements of antibiotic concentration-time profiles. First-in-human, whole-body dynamic 11C-rifampin
positron emission tomography (PET) and computed tomography (CT) were performed in newly identified
patients with rifampin-susceptible TB. PET demonstrated spatially compartmentalized rifampin exposures in
the multiple, pathologically distinct TB lesions in the same patient, with low cavitary tissue rifampin exposures.
Repeat PET/CT measurements demonstrated independent temporal evolution of rifampin exposure trajectories
in different lesions within the same patient. Similar findings were re-capitulated by PET/CT in experimentally
infected rabbits with cavitary TB and confirmed using post-mortem analyses. Integrated modeling of the PET-
captured concentration-time profiles in hollow-fiber bacterial kill-curve experiments identified that 35 mg/kg/day
of rifampin is needed to achieve cure in four months for cavitary disease. Optimized antibiotic dosing could
shorten current treatments. Conversely, suboptimal dosing is a major factor for treatment failure and antibiotic
resistance, which the World Health Organization declared as one of the top ten threats to human health.
Our overall goals are to leverage our expertise in novel in vivo imaging tools, animal models of cavitary TB
and hollow-fiber systems to gain mechanistic insights about TB treatments: a) measure the spatial and
temporal distribution of TB drugs active against multi-drug resistant TB (bedaquiline, pretonamid, linezolid
regimen) and optimize cavitary TB treatments; b) identify the key factors contributing to treatment failure, long-
term (relapse-free) cure or able to guide treatments and; c) develop imaging (pathogen-specific or radiography-
based) biomarkers for early identification of subjects at-risk for treatment failure or requiring longer treatments.

Grant Number: 5R01AI153349-05
NIH Institute/Center: NIH
Principal Investigator: WILLIAM BISHAI