Untitled

A N T I M I C R O B I A L R E S I S T A N C E Old Drugs, New Purpose: Retooling ExistingDrugs for Optimized Treatment of ResistantTuberculosis Kelly E. Dooley,1 Carole D. Mitnick,2 Mary Ann DeGroote,3 Ekwaro Obuku,4,5 Vera Belitsky,6 Carol D. Hamilton,7Mamodikoe Makhene,8 Sarita Shah,9 James C. M. Brust,9 Nadza Durakovic,6 Eric Nuermberger,1and on behalf of the Efficacy Subgroup, RESIST-TB 1Johns Hopkins University School of Medicine, Baltimore, Maryland, 2Harvard Medical School, Boston, Massachusetts, 3Mycobacterial ResearchLaboratories, Colorado State University, Fort Collins, 4Institute of Human Virology, University of Maryland School of Medicine, Baltimore; 5AIDS ReliefProgramme and Joint Clinical Research Centre, Kampala, Uganda; 6Partners In Health, Boston, Massachusetts, 7Health and Development Sciences, Family Health International, Durham, North Carolina, 8National Institutes of Health, Bethesda, Maryland, and 9Albert Einstein College of Medicine,Bronx, New York Treatment of drug-resistant tuberculosis is hindered by the high toxicity and poor efficacy of second-linedrugs. New compounds must be used together with existing drugs, yet clinical trials to optimize combina- tions of drugs for drug-resistant tuberculosis are lacking. We conducted an extensive review of existingin vitro, animal, and clinical studies involving World Health Organization–defined group 1, 2, and 4 drugsused in drug-resistant tuberculosis regimens to inform clinical trials and identify critical research questions.
Results suggest that optimizing the dosing of pyrazinamide, the injectables, and isoniazid for drug-resistanttuberculosis is a high priority. Additional pharmacokinetic, pharmacodynamic, and toxicodynamic studiesare needed for pyrazinamide and ethionamide. Clinical trials of the comparative efficacy and appropriatetreatment duration of injectables are recommended. For isoniazid, rapid genotypic tests for Mycobacterium tuberculosis mutations should be nested in clinical trials. Further research focusing on optimization of doseand duration of drugs with activity against drug-resistant tuberculosis is paramount.
The World Health Organization (WHO) estimates that optimized background regimens for trials with new com- 650 000 cases of multidrug-resistant (MDR) tuberculosis pounds, the Drug Efficacy Subgroup of RESIST-TB (Re- occurred in 2010. Extensively drug-resistant (XDR) search Excellence to Stop TB Resistance; tuberculosis has been found in every country with the ) reviewed the existing literature on second-line tu- means to test for it ]. To improve treatment of drug- resistant tuberculosis with existing drugs and identify of individual agents to drug-resistant tuberculosis treat-ment. This review summarizes the preclinical and clini-cal evidence and gaps in knowledge for antituberculosisagents classified by the WHO as groups 1, 2, and Received 9 January 2012; accepted 24 April 2012; electronically published 21 4. Groups 3 (fluoroquinolones) and 5 (agents of uncer- 1Additional members of the RESIST-TB Drug Efficacy Subgroup (in alphabetical tain efficacy) are reviewed separately. Priority ranking of order): Jason Andrews, Pepe Caminero, Scott Franzblau, Mark Harrington, Gary Horwith, Kayla Laserson, Sonal Munsiff, Alex Pym, Rajeswari Ramachandran, and research questions—has not previously been undertaken.
Correspondence: Kelly E. Dooley, MD, PhD, Division of Clinical Pharmacology, Johns Hopkins University School of Medicine, 600 N Wolfe St, Osler 527,Baltimore, MD 21287 ([email protected]).
The Author 2012. Published by Oxford University Press on behalf of the InfectiousDiseases Society of America. All rights reserved. For Permissions, please e-mail: In vitro studies were included if they used Mycobac- [email protected].
DOI: 10.1093/cid/cis487 terium tuberculosis laboratory or clinical strains and 572 • CID 2012:55 (15 August) • ANTIMICROBIAL RESISTANCE reported measures of antituberculosis activity. Studies involv- ing animals describing drug efficacy against M. tuberculosis PZA has potent sterilizing activity. Critical areas for future infection were included. Clinical studies were included if research include determining patterns and frequency of PZA they had relevant pharmacokinetic (PK), safety, bacteriologic, resistance among drug-resistant tuberculosis cases, evaluating the clinical significance of PZA resistance, development of A search strategy using “tuberculosis” and the drug being eval- rapid diagnostics to detect resistance, exploration of the risks uated as MeSH terms between January 2008 and September 2011 and benefits of higher doses, including hepatotoxicity with was employed in PubMed and Embase. Articles from 1940–2011 modest dose increases, and establishment of the optimal du- were reviewed if they were in English or French. References at the ration of PZA use for drug-resistant tuberculosis.
end of articles and relevant textbook chapters were searched byhand.
For drug-sensitive tuberculosis, the role of ethambutol (EMB)is to protect companion drugs against resistance. However, EMB resistance among patients with drug-resistant tuberculo- sis reaches 50%–60%. Mutations in the embB gene confer a 2–8-fold increase in MIC [Genotypic testing for EMB resis-tance is 57% sensitive and 92% specific [ Pyrazinamide (PZA) is a prodrug activated by pncA whose mechanism of action is still being elucidated (Table PZA In mice, the minimal effective dose is 100 mg/kg/d, which pro- requires acidic conditions to exert antituberculosis activity. Re- duces an AUC equivalent to 15 mg/kg/d in humans. Higher sistance to PZA is conferred by mutations in pncA or rpsA [ doses are required for bactericidal activity. Although 100 mg/kg The absence of dominant mutations in the pncA gene repre- prevents emergence of resistance to companion drugs, protection sents a substantial limitation for rapid molecular testing.
PZA activity is correlated with the ratio of the area under the Doses <15 mg/kg cannot prevent emergence of resistance to time-concentration curve (AUC) to the minimum inhibitory companion drugs, so 15 mg/kg/d probably represents the clin- concentration (MIC) in preclinical models []. Doubling the ical minimal effective dose []. In combination with INH, human-equivalent PZA dose increases bactericidal and steril- EMB produces superior 6-month culture conversion rates at izing effects in mice and guinea pigs ]. Importantly, PZA 25 vs 12.5 mg/kg. EMB at a daily dosage of 25 mg/kg plus a has synergistic effects when given together with investigational good sterilizing agent is highly active ]. However, EMB- related optic neuritis is dose and duration dependent and maybe irreversible. Incidence is 5% with the 25 mg/kg dose and <1% with the 15 mg/kg/d dose but may be decreased by PZA has treatment-shortening effects in regimens contain- ing isoniazid (INH) with or without rifampin []. Earlybactericidal activity (EBA) trials with PZA demonstrated minimal EBA from days 0–2 (EBA0-2, marker of early bac- The principal role of EMB is to prevent resistance to compan- terial killing) and modest EBA from days 2–14 (EBA2-14 is ion drugs. This property should extend to treatment of drug- a proposed surrogate of sterilizing activity), but enhanced resistant tuberculosis caused by EMB–susceptible isolates.
EBA0-14 of several investigational drugs [Although the Higher daily doses (eg, 25 mg/kg) are probably more active contribution of PZA to rifampin-containing regimens is but increase the risk of ocular toxicity. Research priorities limited to the first 2 months of treatment, its optimal include determination of frequency of EMB resistance among duration in other regimens has not been evaluated. Though patients with drug-resistant tuberculosis and the safety and ef- PZA is likely to improve drug-resistant tuberculosis treat- ficacy of intermittent high-dose EMB.
ment outcomes against susceptible strains, use of PZA fordrug-resistant tuberculosis is complicated by (1) the high incidence of PZA resistance among drug-resistant tubercu- Isoniazid (INH) is a prodrug activated by M. tuberculosis’ losis strains, (2) challenges in performing drug susceptibil- catalase-peroxidase enzyme KatG. The activated drug binds ity testing, and (3) a poor understanding of clinical InhA, an enoyl-acyl carrier protein reductase enzyme, inhib- iting fatty acid synthesis. Partial loss of KatG function ANTIMICROBIAL RESISTANCE • CID 2012:55 (15 August) • 573 Summary Information About World Health Organization Class 1, 2, and 4 Drugs for Tuberculosis After activation by PncA, pyrazinoic acid translation, It may also inhibit ATPsynthesis Inhibits transcription by binding to RpoB, Cmax/MIC is associated withresistance suppression, as forrifampin pharmacokinetic studies with AMK inmice suggest that daily doses of20–45 mg/kg produce more human-equivalent exposures presumably inhibits protein synthesisby binding to the 16S ribosomal RNAencoded by rrs 0.3–1.2 µg/mL in 7H12broth, 2.5–10 µg/mL in7H10 broth alanine:D-alanine ligase, thus blockingcell wall peptidoglycan synthesis Abbreviations: AMK, amikacin; ATP, adenosine triphosphate; AUC, area under the time-concentration curve; Cmax, maximum concentration; CS, cycloserine; KM, kanamycin; MIC, minimum inhibitory concentration;PAS, para-aminosalicylic acid; TZ, terizidone.
Downloaded from http://cid.oxfordjournals.org/ Summary of Research Questions to Inform Optimization of Drug-Resistant Tuberculosis Treatment Regimens Translation of Preclinical Findings to Clinical Settings PK/PD studies in animals to define the PD parameter most closely correlated withactivity Use of PK/PD Relationships to Optimize Dosing PK/PD studies in in vitro and animal models Can target attainment be further optimized? Dose-ranging human efficacy trials with PK/PD Can toxicity be reduced through changes in Human efficacy and tolerability studies using Can toxicity be reduced through changes in tolerability studies of new formulations,studies of novel delivery systems for existingdrugs Enhanced Use of Genotypic and Phenotypic Testing to Inform Drug Choices Can rapid resistance tests identify patients Correlation of rapid genotypic resistance test likely to benefit from a drug that ordinarily Human genotyping to identify mutations likely to affect drug absorption or clearance in What is the correlation between phenotypic Randomized trial of drug in question in patients whose isolates have documented phenotypic resistance, or collection of phenotypic and genotypic data of the drug in question amongpatients enrolled in trials of other compounds Designing an Optimized Background Regimen of Existing Drugs What is the activity of this drug in humans? Randomized trial of an optimized background regimen with substitution of injectableagents What is the optimal duration of treatment Randomized trial comparing different drug Abbreviations: AMK, amikacin; CM, capreomycin; CS/TZ, cycloserine and terizidone; EBA, early bactericidal activity; EMB, ethambutol; ETA, ethionamide; INH,isoniazid; KM, kanamycin; PAS, para-aminosalicylic acid; PD, pharmacodynamic; PK, pharmacokinetic; PZA, pyrazinamide; RBT, rifabutin.
typically results in INH MICs of 2–8 µg/mL, while complete promoter mutant, and 25 mg/kg/d had bactericidal activity loss results in MICs ≥ 16 µg/mL. Mutations in inhA and its against a katG mutant (AUC/MIC of approximately 40 and 15, promoter cause low-level INH resistance and cross-resistance with ethionamide []. Data on the distribution of INHMICs of clinical drug-resistant tuberculosis isolates are sparse, but up to 43%–75% of drug-resistant tuberculosis strains In humans, INH exposures are variable and determined by may remain susceptible to INH at clinically achievable N-acetyltransferase (NAT2) genotype Against INH- susceptible strains, INH AUC >10.5 μg × h/mL (AUC/MIC >100) occurs in slow acetylators receiving 3 mg/kg and rapid acetylators receiving 6 mg/kg/d and is associated with near- Maximal INH activity in vitro and in mice is achieved at an maximal EBA. However, considerable bactericidal activity is AUC/MIC ratio value of 100–200. Still, an INH dose of 10 achieved with doses as low as 1.25 mg/kg []. To achieve mg/kg/d exhibited bactericidal activity against an inhA such bactericidal activity against strains with low-level ANTIMICROBIAL RESISTANCE • CID 2012:55 (15 August) • 575 Priority Ranking of Drugs for Additional Research on Optimization for Drug-Resistant Tuberculosis Treatment Barriers to Optimization for Drug-Resistant Sterilizing activity in first-line regimens, so may Resistance may be common in multidrug-resistant shorten drug-resistant tuberculosis treatment Phenotypic resistance testing problematic Synergistic effects with new drugs in clinical Multiple different mutations can confer resistance, impeding development of rapid genotypicresistance test Cheap, well tolerated, and widely available MIC distribution among resistant strains unknown Low-level resistance may be overcome with Correlation between genotypic resistance test result Rapid genotypic resistance test may predict Interpatient variability in PK and acetylation which patients with drug-resistant tuberculosis Susceptibility to injectables confers better Amikacin not widely available and expensive Poor early bactericidal activity prevents using this Relative efficacy of injectable drugs is unknown Optimal treatment duration of injectable use is Large sample sizes needed to study comparative Parenteral use requirement makes this group a target for replacement as new oral drugs are Only second-line oral drug with potential Use of isoniazid in initial tuberculosis treatment may select for isoniazid and ethionamide cross- Relationship between drug exposure and GI Ability of rapid genotypic tests to predict susceptibility Better tolerated than many second-line drugs Resistance may be common in drug-resistant Relationship between drug exposure and ocular tuberculosis strains given the use of ethambutol in Animal models suggest that neuroprotective Concerns over ocular toxicity may limit use doses that agents may prevent optic neuritis, allowing for Poor GI tolerability and risk of hypersensitivity Lower doses may have similar activity, better tolerability than currently recommended dose Not amenable to animal studies given marked Lower doses may have similar activity, better Serious central nervous system side effects tolerability than currently recommended dose Most drug-resistant tuberculosis strains are rifabutin Some drug-resistant tuberculosis strains may Even drug-resistant tuberculosis strains considered Rapid genotypic resistance tests may predict rifabutin susceptible have reduced rifabutin susceptibility compared with wild-type strains Current genotypic resistance tests may not identify discordant susceptibility to rifampin and rifabutin Abbreviations: GI, gastrointestinal; MIC, minimum inhibitory concentration; PK, pharmacokinetic.
resistance (MIC ≤ 0.5 μg/mL), a dose of ≥600 mg is proba- bly needed. Higher INH doses may yield bactericidal activity High-dose INH may be useful in the treatment of drug- against strains with MICs of 1 or 2 µg/mL, depending on resistant tuberculosis, but efficacy will depend on INH dose, acetylator status. Addition of high-dose INH (16–18 mg/kg) patient acetylator status, and degree of INH resistance. In but not standard-dose INH (5 mg/kg) to drug-resistant patients infected with M. tuberculosis with isolated inhA tuberculosis treatment increased sputum culture conversion mutations, high-dose INH should be more effective than rates in a recent trial ]. Peripheral neuropathy was more ETA. Additional studies are needed to determine the INH common with high-dose INH, but pyridoxine was not MIC distribution among drug-resistant tuberculosis isolates and the ability of rapid genotypic resistance testing to 576 • CID 2012:55 (15 August) • ANTIMICROBIAL RESISTANCE predict INH MIC. EBA studies can confirm the relationship ≥6 months of drug-resistant tuberculosis treatment. KM and between INH AUC/MIC and effect and enable optimal AMK inhibit ribosomal protein synthesis. Both have poor ac- dosing recommendations based on genotypic resistance tivity against slowly multiplying mycobacteria. Mutations in assays. A rapid means of determining acetylator status rrs confer high-level resistance to both agents, although some would be helpful to customize dose selection. The safety of KM-resistant strains retain susceptibility to AMK. The Hain® high-dose INH requires investigation.
GenoType MTBDRsl assay is sensitive and specific for detect-ing KM/AMK resistance.
Rifabutin (RBT) is a rifamycin antibiotic with more potent in vitro activity than rifampin. Although 12%–36% of rifampin- AMK is more potent than KM in vitro and in mice. Whereas resistant clinical isolates reportedly remain susceptible to RBT, weak bactericidal activity is observed with human-equivalent the breakpoint defining susceptibility is not evidence based. In doses of AMK, similar KM doses are bacteriostatic [ fact, RBT MICs against rifampin-resistant strains are virtuallyalways higher than the wild-type distribution, indicating that RBT is unlikely to be fully active against rifampin- As monotherapy, AMK at doses of 5–15 mg/kg/d has minimal resistant isolates. Although new line probe assays can identify EBA and no dose-response effect . No clinical trial has eval- specific rpoB mutations that do not shift the RBT MIC as uated the contribution of KM or AMK to drug-resistant tuber- much as the rifampin MIC, it is unclear whether such assays culosis regimens. However, findings in cohort studies suggest that patients with multidrug-resistant tuberculosis and resistanceto injectables have lower treatment success than patients with preserved susceptibility to injectables [, KM and AMK In mice, RBT has dose-dependent bactericidal activity. At cause nephrotoxicity, vestibulotoxicity, and ototoxicity; the latter 10 mg/kg/d, RBT monotherapy reduces lung and spleen 2 are related to cumulative dose and commonly irreversible [ CFU counts by 3–5 log in mice infected with drug-sensitiveM. tuberculosis, but the exposures observed with this dose may not be tolerable in humans, and efficacy against rifam- KM is the most often-used injectable agent for drug-resistant pin-resistant strains has not been assessed [].
tuberculosis. AMK is more potent but more expensive, and the mouse-to-human PK/PD correlates are unknown. Both RBT is widely distributed in tissues, has poor bioavailability, and drugs have significant, potentially irreversible, toxicities related autoinduces its metabolism. At 600 mg/d (twice the usual clini- to cumulative dose. Despite its modest bactericidal activity in mice, AMK exhibits little to no EBA in patients. However, im- Trials comparing rifampin 600 and RBT 300 mg/d (together proved outcomes in patients with drug-resistant tuberculosis with standard tuberculosis drugs) against drug-sensitive tuber- with preserved susceptibility to injectables compel their use, culosis, found similar efficacy and tolerability []. Owing to despite medical and logistical disadvantages. Understanding cross-resistance, RBT is rarely used for drug-resistant tuberculo- the comparative efficacy of KM and AMK, their specific con- sis, but RBT has been part of successful regimens to treat XDR tribution to multidrug therapy, and the optimal duration of treatment will require preclinical and clinical trials.
RBT may retain some activity against a small proportion of Capreomycin (CM) is a polypeptide antibiotic that inhibits rifampin-resistant drug-resistant tuberculosis strains. However, protein synthesis. Mutations in the mycobacterial tlyA gene more rigorous study of the impact of specific rpoB mutations confer CM resistance, and mutations in its rrs gene may on RBT susceptibility in light of achievable RBT exposures confer cross-resistance to aminoglycosides and CM.
is needed to gauge the accuracy of rapid rpoB genotyping andidentify the minority of patients who may benefit from RBT.
Preclinical EvaluationsIn one in vitro experiment, CM was the only drug tested with significant activity against hypoxic, nonreplicating M. tubercu-losis. However, its activity against persistent M. tuberculosis in animals has not been evaluated [In mice, CM has bacter- Kanamycin (KM), its semisynthetic derivative amikacin iostatic activity and a narrow therapeutic margin. Daily doses (AMK), or capreomycin (see below) is given by injection for of 120–150 mg/kg have weak growth inhibitory effects [].
ANTIMICROBIAL RESISTANCE • CID 2012:55 (15 August) • 577 CM prevents emergence of resistance to INH at 300 mg/kg/d, Strategies to improve tolerability by dividing doses may nega- but such doses produce renal tubular necrosis in mice.
tively affect efficacy by preventing serum concentrations fromexceeding MIC [Hypothyroidism is a significant clinical Current dose and treatment duration for CM are largely basedon case series from the 1960s Among previously treated patients, CM at a dosage of 1 g/d plus 2 additional drugs was Among group 4 agents, only ETA has bactericidal activity successful in 50%–85%. CM for 120 days was better than treat- against M. tuberculosis. However, attainment of bactericidal ment for 60 days when CM was given daily with PAS; drug exposures in patients is limited by poor tolerability. In- treatment beyond 6 months, though, provided no further formation regarding the drug exposures needed to produce benefit. CM use is limited by reversible, dose-dependent renal bactericidal effects and the potential for achieving such expo- toxicity (1%–2% of patients) and potentially irreversible oto- sures in patients with drug-resistant tuberculosis remains toxicity, which is related to cumulative exposure (2%–3%).
scarce. PK/PD studies in in vitro and animal models couldhelp determine PD targets and establish the human-equivalent dose in mice. In addition, better understanding of human In mice, CM is bacteriostatic and less effective than aminogly- population PK and toxicodynamics could identify the cosides. Despite the intriguing observation of bactericidal ac- minimal effective dose of ETA. The utility of rapid genotypic tivity against nonreplicating M. tuberculosis in vitro, this resistance testing to identify patients who are unlikely to finding has not been demonstrated in vivo. The most benefit from ETA should be studied.
common mycobacterial resistance mutations occurring withCM use do not confer cross-resistance to KM or AMK, but the converse may not be true. Whether use of CM as the Highly specific for M. tuberculosis, para-aminosalicylic acid initial injectable agent affords a treatment advantage by (PAS) is a bacteriostatic agent. Resistance to PAS is associated keeping other options open is unclear, because CM is less with mutations of mycobacterial thyA, but this mechanism ac- potent than KM or AMK in preclinical studies, and there are counts for only 6% of phenotypic resistance no comparative clinical trials. The optimal dose and durationof treatment are unknown.
Preclinical EvaluationsEarly guinea pig studies demonstrated that daily treatment was more effective than intermittent therapy, but the human- equivalent dose in animals and the PK/PD correlates of PASactivity are unknown [].
Ethionamide (ETA) and prothionamide are prodrugs requir- ing enzymatic activation by M. tuberculosis EthA to inhibit New delayed-release formulations produce higher PAS expo- InhA, a target shared with INH. Mutations in mycobacterial sures, overcoming the rapid metabolism and clearance of early inhA or its promoter confer reduced susceptibility to INH and formulations ]. Although PAS has not been evaluated in ETA; mutations in ethA cause ETA and prothionamide mono- EBA studies, clinical trials showed that monotherapy at a dosage of 10 g/d for 3 months produced clinical improvementand had efficacy similar to that of streptomycin [Case reports describe effective drug-resistant tuberculosis treatment In mice, doses ≥25 mg/kg are bactericidal. Doses as low as with PAS-containing regimens. The use of PAS is limited by its 12.5 mg/kg prevent selection of drug-resistant mutants by toxicities—gastrointestinal irritation, myxedema, hypokalemia, INH There are no data confirming pharmacodynamic life-threatening hypersensitivity reactions, and B12 deficiency.
(PD) targets or human-equivalent mouse doses.
PAS is a poorly tolerated, bacteriostatic agent most useful for ETA was originally evaluated as an agent to prevent resistance preventing emergence of resistance to companion drugs. As to INH or treat INH-resistant disease ]. Interest in ETA with other oral second-line agents, little is known about PK/ waned when better-tolerated alternatives became available.
PD correlates of PAS activity. In vitro and animal model Gastrointestinal intolerance is the Achilles’ heel of ETA and studies to define the lowest exposure necessary for maximal mandates gradual dose increases to the highest tolerable dose.
( presumably bacteriostatic) effect, in concert with human PK 578 • CID 2012:55 (15 August) • ANTIMICROBIAL RESISTANCE studies, could identify lower, more tolerable PAS doses that prioritizes research opportunities to optimize the use of WHO-defined group 1, 2, and 4 drugs in the treatment ofdrug-resistant tuberculosis (Table Important gaps in un- derstanding the pharmacokinetics and pharmacodynamics of Cycloserine (CS) and terizidone have broad-spectrum antimi- these agents prevent confident recommendations regarding crobial activity. As an N-methyl-D-aspartate receptor partial dosing and duration needed to maximize efficacy with accept- agonist, CS competes with γ-aminobutyric acid in the brain, able safety and tolerability. Focused research using preclinical causing central nervous system (CNS) side effects. Overexpres- models and small proof-of-concept trials may be sufficient to sion of mycobacterial alr, which encodes the primary target, guide dose optimization. Questions regarding appropriate du- confers CS resistance. Phenotypic susceptibility testing is no- ration require longer studies with efficient adaptive designs.
toriously difficult to perform, and little correlation exists Better understanding of the clinical implications of resistance between treatment outcomes and in vitro findings testing ( phenotypic and genotypic) is also critical.
Such research efforts are essential to complement observa- tional studies, eg, of the “Bangladesh” regimen, [] which are It is difficult to demonstrate the efficacy of CS in animal models.
not designed to evaluate individual contributions of drugs.
This is partly due to PK differences. The half-life of CS is 12 The 3 high-priority drug groups reviewed were used in the hours in humans and 23 minutes in mice []. Furthermore, Bangladesh regimen and are part of short-course regimens greater circulating concentrations of D-alanine in mice may an- tagonize CS activity. Therefore, the human-equivalent dose of New drug candidates in clinical development call into CS remains a mystery. In mice and guinea pigs, daily doses of question the need for research to optimize use of existing 150–300 mg/kg have little or no therapeutic effect second-line drugs. Existing drugs, however, will be used in combination with new agents for the foreseeable future, CS was studied in the 1950s in complicated or resistant tuber- making it imperative to optimize their use to protect new culosis cases. CS monotherapy (1–1.5 g in 4 divided doses) drugs (Table In addition, evolution of resistance will inevi- produced rapid clinical improvement, and in one study cultures tably follow the introduction of new drugs, making it unlikely were converted to negative in one-third of patients with new or that existing agents will be removed from clinical use. Finally, chronic tuberculosis. When CS was combined with INH, clini- studies that result in regulatory approval of new agents may cal improvements were seen, but INH resistance emerged not inform their best use. They may not be studied in combi- CS and ETA promoted culture conversion in a majority of pa- nation with other new agents, with truly optimized back- tients in whom other regimens failed []. To date, the individ- ground regimens, or for adequate durations. Additional trials ual contribution of CS in drug-resistant tuberculosis regimens will be necessary to define their optimal use. The same lessons has not been evaluated. CNS side effects are common, serious, highlighted in this review need to be applied to new drugs; and dose related, severely limiting the use of CS.
such trials will provide opportunities to address key researchpriorities established here. Improved quality of evidence, through dedicated preclinical and clinical research, is essential CS is a bacteriostatic agent with clinically proven antitubercu- to achieve the goal of shorter, more effective, and less toxic losis efficacy. Its use is hindered by serious CNS toxicity.
regimens for drug-resistant tuberculosis.
Although animal models are not informative, in vitro modelscould be used to define the PK/PD correlates of activity and optimize drug exposures necessary for bacteriostatic effects.
We acknowledge the assistance of Raju Ravikiran who helped with the literature review for capreomycin.
Each author conducted the initial in-depth review of the literature related to ≥1 study drug, including selection, review, and critical interpre-tation of published studies. All authors participated in the ultimate selec- Programmatic management of drug-resistant tuberculosis is tion of included articles, the synthesis and interpretation of data, and the complex. Increased toxicity and reduced potency or accessibil- This work was supported by the National Insti- ity of second-line drugs result in poor outcomes. The evidence tutes of Health (grant K23AI080842 to K. E. D. and K23AI083088 to base to guide doses and combinations of these drugs for treat- J. C. M. B.) and the Doris Duke Charitable Foundation (Clinical Scientist ment is limited and of low quality [Heeding a call for re- Development Award 2007071 to S. S.).
search into “the most effective use of existing second-line Potential conflicts of interest. All authors: No reported conflicts.
All authors have submitted the ICMJE Form for Disclosure of Potential antituberculosis therapies and other antimicrobials available to Conflicts of Interest. Conflicts that the editors consider relevant to the treat drug-resistant TB” [], this review identifies and content of the manuscript have been disclosed.
ANTIMICROBIAL RESISTANCE • CID 2012:55 (15 August) • 579 21. Davies G, Cerri S, Richeldi L. Rifabutin for treating pulmonary tuber- culosis. Cochrane Database Syst Rev 2007: CD005159.
1. World Health Organization. Global tuberculosis control: WHO report 22. Sanders WE Jr, Hartwig C, Schneider N, Cacciatore R, Valdez H. Ac- 2011, WHO Press, World Health Organization, Geneva, Switzerland; tivity of amikacin against mycobacteria in vitro and in murine tuber- 2. Shi W, Zhang X, Jiang X, et al. Pyrazinamide inhibits trans-translation 23. Donald PR, Sirgel FA, Venter A, et al. The early bactericidal activity in Mycobacterium tuberculosis. Science 2011; 333:1630–2.
of amikacin in pulmonary tuberculosis. Int J Tuberc Lung Dis 2001; 3. Ahmad Z, Fraig MM, Bisson GP, Nuermberger EL, Grosset JH, Kara- kousis PC. Dose-dependent activity of pyrazinamide in animal models 24. Chan ED, Strand MJ, Iseman MD. Multidrug-resistant tuberculosis of intracellular and extracellular tuberculosis. Antimicrob Agents Che- (TB) resistant to fluoroquinolones and streptomycin but susceptible to second-line injection therapy has a better prognosis than extensively 4. Fox W, Ellard GA, Mitchison DA. Studies on the treatment of tuber- drug-resistant TB. Clin Infect Dis 2009; 48:e50–2.
culosis undertaken by the British Medical Research Council tuberculo- 25. Kim DH, Kim HJ, Park SK, et al. Treatment outcomes and survival sis units, 1946–1986, with relevant subsequent publications. Int based on drug resistance patterns in multidrug-resistant tuberculosis.
J Tuberc Lung Dis 1999; 3:S231–79.
Am J Respir Crit Care Med 2010; 182:113–9.
5. Jindani A, Dore CJ, Mitchison DA. Bactericidal and sterilizing activi- 26. Peloquin CA, Berning SE, Nitta AT, et al. Aminoglycoside toxicity: ties of antituberculosis drugs during the first 14 days. Am J Respir daily versus thrice-weekly dosing for treatment of mycobacterial dis- eases. Clin Infect Dis 2004; 38:1538–44.
6. Mphahlele M, Syre H, Valvatne H, et al. Pyrazinamide resistance 27. Heifets L, Simon J, Pham V. Capreomycin is active against non- among South African multidrug-resistant Mycobacterium tuberculosis replicating M. tuberculosis. Ann Clin Microbiol Antimicrob 2005; 4:6.
isolates. J Clin Microbiol 2008; 46:3459–64.
28. Klemens SP, DeStefano MS, Cynamon MH. Therapy of multidrug- 7. Starks AM, Gumusboga A, Plikaytis BB, Shinnick TM, Posey JE. Mu- resistant tuberculosis: lessons from studies with mice. Antimicrob tations at embB codon 306 are an important molecular indicator of ethambutol resistance in Mycobacterium tuberculosis. Antimicrob 29. Popplewell AG, Miller JD, Greene ME, Landwehr A. Capreomycin in original treatment cases of pulmonary tuberculosis. Ann N Y Acad Sci 8. Hillemann D, Rusch-Gerdes S, Richter E. Feasibility of the GenoType MTBDRsl assay for fluoroquinolone, amikacin-capreomycin, and eth- 30. Grumbach F. [The treatment of experimental tuberculosis in mice by ambutol resistance testing of Mycobacterium tuberculosis strains and the association of isoniazid and ethionamide in different doses: appli- clinical specimens. J Clin Microbiol 2009; 47:1767–72.
cation of the results to clinical posology]. Rev Tuberc Pneumol (Paris) 9. Grumbach F. Activite antituberculeuse chez la souris, de l’ethambutol en association avec l’isoniazide ou l’etionamide. Ann Inst Pasteur 31. Angel JH, Bhatia AL, Devadatta S, et al. A controlled comparison of cycloserine plus ethionamide with cycloserine plus thiacetazone in pa- 10. Radenbach KL. Minimum daily efficient dose of ethambutol: general tients with active pulmonary tuberculosis despite prolonged previous review. Bull Int Union Tuberc 1973; 48:106–11.
chemotherapy. Tubercle 1963; 44:215–24.
11. Pyle MM. Ethambutol in the retreatment and primary treatment of 32. Zhu M, Namdar R, Stambaugh JJ, et al. Population pharmacokinetics tuberculosis: a four-year clinical investigation. Ann N Y Acad Sci of ethionamide in patients with tuberculosis. Tuberculosis (Edinb) 12. Griffith DE, Brown-Elliott BA, Shepherd S, McLarty J, Griffith L, 33. Mathys V, Wintjens R, Lefevre P, et al. Molecular genetics of para- Wallace RJ Jr. Ethambutol ocular toxicity in treatment regimens for aminosalicylic acid resistance in clinical isolates and spontaneous Mycobacterium avium complex lung disease. Am J Respir Crit Care mutants of Mycobacterium tuberculosis. Antimicrob Agents Chemo- 13. Schaaf HS, Victor TC, Venter A, et al. Ethionamide cross- and co- 34. Karlson AG, Gainer JH, Feldman WH. The effect of dihydrostrepto- resistance in children with isoniazid-resistant tuberculosis. Int mycin para-aminosalicylate on experimental tuberculosis in guinea J Tuberc Lung Dis 2009; 13:1355–9.
pigs. Am Rev Tuberc 1950; 62:149–55.
14. Cynamon MH, Zhang Y, Harpster T, Cheng S, DeStefano MS. High- 35. Peloquin CA, Henshaw TL, Huitt GA, Berning SE, Nitta AT, James dose isoniazid therapy for isoniazid-resistant murine Mycobacterium GT. Pharmacokinetic evaluation of para-aminosalicylic acid granules.
tuberculosis infection. Antimicrob Agents Chemother 1999; 43:2922–4.
15. Parkin DP, Vandenplas S, Botha FJ, et al. Trimodality of isoniazid 36. PARA-AMINOSALICYLIC acid treatment in pulmonary tuberculosis.
elimination: phenotype and genotype in patients with tuberculosis.
Am J Respir Crit Care Med 1997; 155:1717–22.
37. Kam KM, Sloutsky A, Yip CW, et al. Determination of critical 16. Donald PR, Sirgel FA, Botha FJ, et al. The early bactericidal activity of concentrations of second-line anti-tuberculosis drugs with clini- isoniazid related to its dose size in pulmonary tuberculosis. Am cal and microbiological relevance. Int J Tuberc Lung Dis 2010; 14: J Respir Crit Care Med 1997; 156:895–900.
17. Katiyar SK, Bihari S, Prakash S, Mamtani M, Kulkarni H. A randomised 38. Conzelman GM Jr, Jones RK. On the physiologic disposition of controlled trial of high-dose isoniazid adjuvant therapy for multidrug- cycloserine in experimental animals. Am Rev Tuberc 1956; 74:802–6.
resistant tuberculosis. Int J Tuberc Lung Dis 2008; 12:139–45.
39. Wolinsky E, Steenken W Jr. Cycloserine alone and in combination 18. Yoshida S, Suzuki K, Iwamoto T, et al. Comparison of rifabutin sus- with other drugs in experimental guinea pig tuberculosis. Am Rev ceptibility and rpoB mutations in multi-drug-resistant Mycobacterium tuberculosis strains by DNA sequencing and the line probe assay.
40. Epstein IG, Nair KG, Boyd LJ, Auspitz P. Cycloserine-isoniazid com- J Infect Chemother 2010; 16:360–3.
bination therapy in virgin cases of pulmonary tuberculosis. Dis Chest 19. Kelly BP, Furney SK, Jessen MT, Orme IM. Low-dose aerosol infection model for testing drugs for efficacy against Mycobacterium tuberculo- 41. Falzon D, Jaramillo E, Schunemann HJ, et al. WHO guidelines for the sis. Antimicrob Agents Chemother 1996; 40:2809–12.
programmatic management of drug-resistant tuberculosis: 2011 20. Chan SL, Yew WW, Ma WK, et al. The early bactericidal activity of update. Eur Respir J 2011; 38:516–28.
rifabutin measured by sputum viable counts in Hong Kong patients 42. Fauci AS. Multidrug-resistant and extensively drug-resistant tubercu- with pulmonary tuberculosis. Tuber Lung Dis 1992; 73:33–8.
losis: the National Institute of Allergy and Infectious Diseases 580 • CID 2012:55 (15 August) • ANTIMICROBIAL RESISTANCE Research agenda and recommendations for priority research. J Infect 45. Van Deun A, Maug AK, Salim MA, et al. Short, highly effective, and inexpensive standardized treatment of multidrug-resistant tuberculo- 43. Mitnick CD, Castro KG, Harrington M, Sacks LV, Burman W. Ran- sis. Am J Respir Crit Care Med 2010; 182:684–92.
domized trials to optimize treatment of multidrug-resistant tuberculo-sis. PLoS Med 2007; 4:e292.
44. Center for Global Health Policy. A project of the Infectious Diseases *For a full list of relevant references for the drugs reviewed in this Society of America and the HIV Medicine Association. Published by manuscript, the reader is referred to the RESIST-TB website, www.resisttb.
Infectious Diseases Society of America, Arlington, VA, 2010.
ANTIMICROBIAL RESISTANCE • CID 2012:55 (15 August) • 581

Source: http://www.resisttb.com/wp-content/uploads/2013/08/Old-Drugs-New-Purpose-Retooling-Existing-Drugs-for-Optimized-Treatment-of-Resistant-Tuberculosis.pdf