Pulmonary Tuberculosis: Developing Drugs for Treatment

Administrative guidance, not enacted legislation. This captured publisher edition preserves the listed document and attachments; later developments are not independently consolidated.

Pulmonary Tuberculosis:
Developing Drugs for
Treatment
Guidance for Industry
U.S. Department of Health and Human Services
Food and Drug Administration
Center for Drug Evaluation and Research (CDER)
May 2026
Clinical/Antimicrobial

Pulmonary Tuberculosis:
Developing Drugs for
Treatment
Guidance for Industry
Additional copies are available from:
Division of Drug Information
Center for Drug Evaluation and Research
Food and Drug Administration
Phone: 855-543-3784 or 301-796-3400
Email: druginfo@fda.hhs.gov
https://www.fda.gov/drugs/guidance-compliance-regulatory-information/guidances-drugs
U.S. Department of Health and Human Services
Food and Drug Administration
Center for Drug Evaluation and Research (CDER)
May 2026
Clinical/Antimicrobial

Contains Nonbinding Recommendations
TABLE OF CONTENTS
I. 	INTRODUCTION............................................................................................................. 1
II. 	BACKGROUND ............................................................................................................... 2
III. 	DEVELOPMENT PROGRAM ....................................................................................... 2
A. 	General Considerations ................................................................................................................. 2
1. Early Phase Clinical Development Considerations......................................................................... 2
a. 	Early bactericidal activity ......................................................................................................... 2
b. 	Phase 2 evaluations................................................................................................................... 3
2. Efficacy Considerations ................................................................................................................... 3
3. Safety Considerations ...................................................................................................................... 3
B. 	Specific Clinical Trial Considerations ......................................................................................... 4
1. Trial Designs.................................................................................................................................... 4
2. Trial Population ............................................................................................................................... 5
3. Inclusion and Exclusion Criteria ..................................................................................................... 6
4. Randomization, Stratification, and Blinding.................................................................................... 7
5. Specific Populations......................................................................................................................... 7
a. 	Pediatric populations................................................................................................................. 7
b. 	Pregnant females ...................................................................................................................... 8
c. 	Other specific populations ........................................................................................................ 9
6. Dose Selection.................................................................................................................................. 9
7. Choice of Comparators .................................................................................................................... 9
8. Efficacy Endpoints ........................................................................................................................... 9
9. Trial Procedures and Timing of Assessments ................................................................................ 11
a. 	Entry visit................................................................................................................................ 11
b. 	Visits during therapy and after therapy completion ............................................................... 12
10. 	Statistical Considerations ......................................................................................................... 12
C. 	Other Considerations................................................................................................................... 13
1. Microbiological Considerations .................................................................................................... 13
a. 	Nonclinical studies.................................................................................................................. 13
b. 	Drug resistance and cross-resistance ...................................................................................... 14
c. 	Types of culture media to identify M. tuberculosis ................................................................ 15
d. 	Differentiate relapse from reinfection or new infection ......................................................... 15
2. Relevant Nonclinical Safety Considerations .................................................................................. 15
3. PK/PD Considerations................................................................................................................... 16
a. 	Phase 1/phase 2 PK trials ........................................................................................................ 16
b. 	Drug interactions .................................................................................................................... 16
c. 	Exposure response .................................................................................................................. 17
d. 	In vitro hollow fiber system models ....................................................................................... 17
4. Foreign Clinical Data .................................................................................................................... 17
5. Data standards for TB ................................................................................................................... 17
REFERENCES............................................................................................................................ 18
APPENDIX .................................................................................................................................. 20

Contains Nonbinding Recommendations
1
Pulmonary Tuberculosis:
Developing Drugs for Treatment
Guidance for Industry1
This guidance represents the current thinking of the Food and Drug Administration (FDA or Agency) on
this topic. It does not establish any rights for any person and is not binding on FDA or the public. You
can use an alternative approach if it satisfies the requirements of the applicable statutes and regulations.
To discuss an alternative approach, contact the FDA staff responsible for this guidance as listed on the
title page.
I. 	INTRODUCTION
The purpose of this guidance is to assist sponsors in the clinical development of investigational
drugs for the treatment of pulmonary tuberculosis (TB) under section 505 of the Federal Food,
Drug, and Cosmetic Act (FD&C Act) (21 U.S.C. 355) and FDA regulations at 21 CFR part 312
and part 601.2 Specifically, this guidance provides the FDA’s current recommendations
regarding the overall development program for a new investigational drug or drugs to be used in
combination with drugs already approved for treatment of pulmonary TB or for other indications
or as a new treatment regimen that includes one or more investigational drugs to support an
indication for the treatment of pulmonary TB. This guidance does not address the development
of drugs for the treatment of latent or extrapulmonary TB infection, nor for prevention of TB.
Sponsors should also refer to the guidance for industry Codevelopment of Two or More New
Investigational Drugs for Use in Combination (June 2013).3 Sponsors are encouraged to discuss
with FDA the programs and pathways facilitating drug development that might be applicable for
their development program.4
This guidance does not contain discussion of the general issues of statistical analysis or clinical
trial design. Those topics are addressed in other guidances such as the International Council for
Harmonisation (ICH) guidances for industry E9 Statistical Principles for Clinical Trials
(September 1998) and E10 Choice of Control Group and Related Issues in Clinical Trials (May
2001) (ICH E10), respectively.
1 This guidance has been prepared by the Division of Anti-Infectives in the Center for Drug Evaluation and
Research at the Food and Drug Administration.
2 For the purposes of this guidance, all references to drugs include both human drugs and therapeutic biological
products unless otherwise specified.
3 We update guidances periodically. For the most recent version of a guidance, check the FDA guidance web page
at https://www.fda.gov/regulatory-information/search-fda-guidance-documents.
4 See the guidance for industry Expedited Programs for Serious Conditions — Drugs and Biologics (May 2014).

Contains Nonbinding Recommendations
2
In general, FDA’s guidance documents do not establish legally enforceable responsibilities.
Instead, guidances describe the Agency’s current thinking on a topic and should be viewed only
as recommendations, unless specific regulatory or statutory requirements are cited. The use of
the word should in Agency guidances means that something is suggested or recommended, but
not required.
II. 	BACKGROUND
Infections caused by Mycobacterium tuberculosis (M. tuberculosis) occur in the United States
and are endemic in many parts of the world. Resistance to multiple drugs and coinfection with
human immunodeficiency virus (HIV) pose challenges in the management of TB. Drugs with
new mechanisms of action, improved safety profiles, and fewer drug-drug interactions, as well as
treatment-shortening combination regimens, are needed to manage TB.
III. 	DEVELOPMENT PROGRAM
A. 	General Considerations
1. 	Early Phase Clinical Development Considerations
Nonclinical evaluations provide valuable information for the development of investigational
drugs (see section III.C.1, Microbiology Considerations, section III.C.2, Relevant Nonclinical
Safety Considerations, and section III.C.3, PK/PD Considerations).
Activity of antimycobacterial drugs can be evaluated in trials of early bactericidal activity (EBA)
and/or in phase 2 trials that evaluate microbiological outcomes at early time points. Treatment of
pulmonary TB requires more than one drug in a treatment regimen, and sponsors may be
developing more than one investigational drug as part of a new combination regimen. For any
combination regimen, the contribution of each drug to the treatment effect must be
demonstrated.5 This can be evaluated in phase 2 clinical development and in nonclinical studies
(see section III.C.1, Microbiology Considerations). Sponsors should consult with the Agency
early in development regarding plans to demonstrate the contribution of the investigational
drug(s) as part of a combination regimen.
a. 	Early bactericidal activity
If applicable to the investigational drug under study, EBA trials evaluating the quantitative
counts of viable M. tuberculosis from daily collections of sputum can provide information on the
bactericidal activity of antimycobacterial drugs. These trials are intended to evaluate
antimycobacterial activity of investigational drugs alone or in combination over a brief time
5 The recommendations in this guidance are relevant to demonstrating the contribution of all individual drugs to the
effect(s) of the combination regimen and are consistent with the requirements of 21 CFR 300.50. Also see the
guidance for industry Codevelopment of Two or More New Investigational Drugs for Use in Combination.

Contains Nonbinding Recommendations
3
course (e.g., 7 to 14 days). EBA trials can provide preliminary evidence for the contribution of
each drug to the treatment effect of the combination regimen. Appropriate participants for
enrollment in EBA trials include immunocompetent participants, treatment-naïve adult
participants at low risk of infection with drug-resistant TB, and participants with no evidence of
extrapulmonary disease, who can begin standard-of-care treatment for pulmonary TB at the
completion of the EBA trial.
b. 	Phase 2 evaluations
Sponsors should conduct phase 2 trials to assess the antimycobacterial activity of an
investigational drug regimen. In addition, if feasible, a phase 2 development program should
include a dose-ranging study or studies to assist in determining the most appropriate dose
regimen to be taken into phase 3. Phase 2 exploratory endpoints can include, but are not limited
to, the following: (1) 8-week evaluation for absence of acid-fast bacilli (AFB) in sputum; (2)
time to sputum culture negativity for M. tuberculosis; (3) symptom improvement; and/or (4) a
biomarker intended to predict clinical benefit. The Agency recommends that as part of phase 2
trial designs, sponsors include long-term follow-up with collection of clinical endpoints in
addition to earlier time points.
2. 	Efficacy Considerations
An investigational drug can be evaluated for efficacy when added to combination regimens of
already approved drugs. Additionally, an entirely new combination regimen comprising
investigational drugs can be evaluated for efficacy. A single adequate and well-controlled trial
in participants with pulmonary TB, supported by other confirmatory evidence (e.g., evidence of
antimycobacterial activity from nonclinical data, EBA, and phase 2 trials), may provide evidence
of effectiveness when the single trial demonstrates a clinically meaningful and statistically robust
treatment effect.6 See section III.B, Specific Clinical Trial Considerations, below for further
discussion regarding efficacy considerations.
3. 	Safety Considerations
The evaluation of the safety profile of an investigational drug can be challenging because
patients with pulmonary TB often have comorbid conditions. Sponsors should evaluate potential
drug-drug interactions that may occur during coadministration with other antimycobacterial
drugs or other concomitant medications (e.g., antiretroviral drugs). Hepatotoxicity and QT
interval prolongation are common adverse reactions with antimycobacterial drugs. Sponsors
6 See the draft guidance for industry Demonstrating Substantial Evidence of Effectiveness for Human Drug and
Biological Products (December 2019). When final, this guidance will represent the FDA’s current thinking on this
topic. For the most recent version of a guidance, check the FDA guidance web page at
https://www.fda.gov/regulatory-information/search-fda-guidance-documents.

Contains Nonbinding Recommendations
4
should evaluate investigational drugs for their potential to cause hepatotoxicity, QT
prolongation, and arrhythmias.7,8
Sponsors should discuss the size of the preapproval safety database with the FDA during drug
development. For assessment of risks and benefits of products to meet an unmet medical need
for the treatment of pulmonary TB, a smaller safety database of approximately 300 participants
treated at (or greater/longer than) the proposed intended dose and duration may be sufficient. If
safety signals are identified, a larger safety database may be needed.
B. 	Specific Clinical Trial Considerations
1. 	Trial Designs
Sponsors can use the following trial designs to demonstrate superiority:
• 	A regimen that includes one or more investigational drugs is compared with a standard
regimen, with efficacy demonstrated by showing superiority of the investigational drug
regimen over the standard regimen.9
• 	The investigational drug(s) plus the optimized background regimen (OBR) is compared
with the matching placebo plus the OBR, with efficacy demonstrated by showing
superiority of the investigational drug(s) plus OBR over placebo plus OBR. Optimized
background antimycobacterial treatment should be based on epidemiological information
and in vitro susceptibility testing, when available.
Sponsors can use the following trial designs to demonstrate noninferiority (NI):
• 	An investigational drug regimen is compared with a standard regimen. NI would be
demonstrated by showing that the investigational regimen performs within a prespecified
margin of the performance of the standard regimen.
• 	The investigational drug replaces one of the drugs in a standard combination regimen.
The investigational drug regimen should perform within an acceptable NI margin that is
based on the known quantitative and reliable contribution of the drug that has been
replaced in the standard regimen. This NI trial design determines the efficacy
contribution of the investigational drug to the regimen.
7 See the guidance for industry Drug-Induced Liver Injury: Premarketing Clinical Evaluation (July 2009).
8 See the ICH guidances for industry E14 Clinical Evaluation of QT/QTc Interval Prolongation and Proarrhythmic
Potential for Non-Antiarrhythmic Drugs (October 2005); E14 Clinical Evaluation of QT/QTc Interval Prolongation
and Proarrhythmic Potential for Non-Antiarrhythmic Drugs — Questions and Answers (R3) (June 2017); S7B
Nonclinical Evaluation of the Potential for Delayed Ventricular Repolarization (QT Interval Prolongation) by
Human Pharmaceuticals (October 2005); and M3(R2) Nonclinical Safety Studies for the Conduct of Human Clinical
Trials and Marketing Authorization for Pharmaceuticals (January 2010).
9 See the guidance for industry Codevelopment of Two or More New Investigational Drugs for Use in Combination.

Contains Nonbinding Recommendations
5
Interpretation of the results of an NI trial relies on a justified NI margin. This margin, which is
highly dependent on the specific design of the NI trial, including the control regimen, is based, in
part, on data from previously conducted trials to evaluate for historical evidence of sensitivity to
drug effects (HESDE) and estimate the effect of the active control.10 The Appendix contains an
example of an NI margin justification for a trial of a 4-month regimen for drug-susceptible
pulmonary TB.
The FDA has not estimated an exact numerical treatment effect for the standard regimen of 2
months of treatment with isoniazid, rifampin, pyrazinamide, and ethambutol followed by 4
months of treatment with isoniazid and rifampin (abbreviated terminology: 2HRZE/4HR) for
patients with drug-susceptible pulmonary TB. However, considering the historical data on
management and outcomes of patients with pulmonary TB in the era before antibacterial drug
therapy and the highly successful results following treatment with 2HRZE/4HR, there is support
for the selection of an NI margin based on the large degree of clinical benefit. For example,
given that the success rates of 2HRZE/4HR exceed 90 percent, the numerical treatment effect is
likely to far exceed 10 percent (Nahid et al. 2016). Therefore, based on clinical judgement, a 10
percent NI margin is clinically relevant and has appropriate preservation of the treatment effect
for an NI trial to determine the efficacy of an investigational drug regimen as a whole based on
comparison to the 6-month standard regimen of 2HRZE/4HR.
Depending on the new investigational drug regimen, study design of the NI trial, target
population (e.g., patients with isolates resistant to one or more drugs), potential impact (e.g.,
ability to fulfill an unmet medical need), and safety profile of the regimen, it may be appropriate
to set a wider NI margin and still plan for a trial design that is feasible and provides a reasonable
preapproval safety database. The Agency encourages sponsors to discuss their clinical trial
designs and NI margin justifications with the FDA before study initiation.11
For both superiority trials and NI trials that assess the activity of the investigational drug regimen
as a whole, the sponsor will also need to address the added contribution of the components of the
regimen.12 This may be accomplished through nonclinical studies, EBA studies, phase 2 trials
and/or as part of the pivotal efficacy trials.
2. 	Trial Population
The trial population should include adult participants and if appropriate, pediatric participants
with pulmonary TB. The presence of extrapulmonary disease may require longer durations of
treatment than pulmonary TB and assessment of endpoints that evaluate the extrapulmonary
site(s). Trials can include participants with either drug-susceptible or drug-resistant pulmonary
TB depending on the anticipated effectiveness of the antimycobacterial drugs being evaluated.
10 See ICH E10 for a discussion of HESDE.
11 See also the article Four-Month Moxifloxacin-Based Regimens for Drug-Sensitive Tuberculosis (Gillespie et al.
2014).
12 See the guidance for industry Codevelopment of Two or More New Investigational Drugs for Use in Combination.

Contains Nonbinding Recommendations
6
Sponsors interested in evaluating a pan-TB regimen for both drug-susceptible and drug-resistant
TB in the same trial should discuss the protocol with the Agency before beginning the trial.
Protocols should specify how participants will be handled after in vitro susceptibility results are
available, both in the conduct of the trial and in the analysis of the results.
Enrichment strategies for enrollment for trials for drug-resistant TB can include a focus on close
contacts of participants with drug-resistant TB, participants from areas with a high prevalence of
drug resistance, participants who relapse after previous treatment, and participants with disease
progression on a standard regimen.
3. 	Inclusion and Exclusion Criteria
The FDA recommends the following inclusion criteria for participants with pulmonary TB:
• 	Presence of AFB in a sputum specimen detected by smear microscopy or presence of M.
tuberculosis by another rapid diagnostic test. Microbiological diagnosis of TB should be
confirmed by culture from at least one sputum sample obtained at the time of enrollment.
• 	Chest radiographic findings consistent with active pulmonary TB, for example, cavitary
lesions, apical or other infiltrates, or hilar lymphadenopathy.
• 	A minimum of two of the following signs or symptoms that have been present for at least
2 weeks:
– 	Sputum production
– 	Cough
– 	One or more episodes of hemoptysis
– 	Fever (e.g., oral temperature greater than or equal to 38.0°C on at least two occasions)
– 	Pleuritic chest pain
– 	Weight loss
– 	Night sweats
Use of rapid diagnostic or nonculture tests may help identify a participant for enrollment in a TB
trial. If the tests being used are not FDA cleared, sponsors should provide sufficient information
about the performance characteristics of the tests determined from analytical validation studies.
The FDA recommends the following as exclusion criteria for participants with pulmonary TB:

Contains Nonbinding Recommendations
7
• 	One or more weeks of therapy for the current episode of active TB (unless being enrolled
in a trial targeting drug-resistant TB and there is documented lack of response to therapy
based on clinical and microbiological findings)
• 	Significant concurrent illness other than HIV (e.g., lung cancer) that may affect outcome
assessment
• 	Evidence of extrapulmonary disease
• 	Unwillingness to comply with recommendations from local public health authorities for
the management of patients with pulmonary TB
4. 	Randomization, Stratification, and Blinding
Trials should be randomized and double-blind unless a sponsor can provide a scientifically
adequate explanation why blinding cannot be accomplished. If trials are single blind or open
label, sponsors should discuss potential biases with the FDA and how these biases will be
addressed. If the trial cannot be fully blinded, the sponsor should maintain the maximum
possible level of blinding within the trial with blinded assessors, blinded databases until database
lock, etc.
Sponsors should consider stratification of randomization or adjustment of the analysis based on
certain baseline characteristics that are anticipated to be prognostic for the outcome (e.g., by the
presence or absence of cavitary disease, HIV infection).13
5. 	Specific Populations
a. 	Pediatric populations
The FDA encourages sponsors to begin discussions about their pediatric clinical development
plans as early as is feasible. The additional safeguards of 21 CFR part 50, subpart D, for
enrolling children in clinical investigations, affect the timing and design of trials that support
pediatric drug development. In accordance with these requirements, sponsors can enroll
pediatric participants in trials if sufficient safety, antimycobacterial activity, pharmacokinetic
(PK), and efficacy data in adult participants are available and appropriate dosing regimens for
pediatric participants have been characterized.14 Sponsors can include adolescent participants
with pulmonary TB in phase 3 clinical trials, if appropriate.15
13 See the guidance for industry Adjusting for Covariates in Randomized Clinical Trials for Drugs and Biological
Products (May 2023).
14 For example, see the article Towards Earlier Inclusion of Children in Tuberculosis (TB) Drug Trials: Consensus
Statements from an Expert Panel (Nachman et al. 2015).
15 See the guidance for industry Development of Anti-Infective Drug Products for the Pediatric Population
(December 2021).

Contains Nonbinding Recommendations
8
Sponsors must submit pediatric study plans no later than 60 calendar days after the date of the
end-of-phase 2 meeting or another time as agreed upon by the FDA and the sponsor unless the
investigational drug has been granted an orphan designation.16 Pediatric formulation
development should begin as soon as results from the adult phase 2b trials are known and the
sponsor has determined an appropriate dosing regimen.
Extrapolation of adult efficacy for the treatment of pulmonary TB to pediatric populations is
acceptable for most pediatric populations, with the exception of very young children (i.e., those
younger than 5 years of age), because of differences in clinical manifestations (e.g., increased
likelihood of extrapulmonary disease) and pathophysiologic characteristics. Sponsors should
provide PK and safety information in a sufficient number of pediatric participants to support the
appropriate dose for treatment of children with pulmonary TB. Cohorts for pediatric studies can
be defined based on chronological age or weight-based criteria, particularly for oral drugs.
Studies of drugs across the pediatric spectrum of ages/weights can be conducted in parallel rather
than sequentially unless there are specific safety or PK properties that warrant a different
approach. If existing data have identified potential developmental concerns for target organs ,
additional safety assessment may be appropriate and may include the use of new approach
methodologies. Any proof-of-concept, PK, or safety evaluation should begin with a weight-of-
evidence risk assessment, including the use of adult safety data instead of juvenile animal
toxicity studies to inform pediatric indications. A stepwise approach (a tiered approach
leveraging adult PK and safety data to younger populations in a stepwise fashion) could
eliminate the need for additional nonclinical studies. Sponsors should consult the appropriate
FDA review division regarding the data needed to address specific safety concerns.
Pediatric development plans for new TB investigational drugs could include children living with
HIV provided there are no safety or drug-drug interaction issues that cannot be managed.
b. 	Pregnant females
Sponsors can include pregnant females in clinical trials once all female reproduction toxicity
studies and the standard battery of genotoxicity tests have been conducted.17 Infants born to
female participants in both investigational and control arms should be followed by investigators
for an appropriate prespecified duration.18
16 See section 505B(e)(2)(A) of the FD&C Act (21 U.S.C. 355c(e)(2)(A)). For additional information, see the
guidance for industry Pediatric Study Plans: Content of and Process for Submitting Initial Pediatric Study Plans
and Amended Initial Pediatric Study Plans (July 2020) and the ICH guidance for industry E11 Clinical Investigation
of Medicinal Products in the Pediatric Population (December 2000).
17 See the draft guidance for industry Pregnant Women: Scientific and Ethical Considerations for Inclusion in
Clinical Trials (April 2018). When final, this guidance will represent the FDA’s current thinking on this topic.
18 For recommendations regarding treatment of women during pregnancy or breastfeeding, see the American
Thoracic Society, Centers for Disease Control and Prevention, and Infectious Diseases Society of America
guidelines for treatment of TB (Nahid et al. 2016), available at
https://www.cdc.gov/mmwr/preview/mmwrhtml/rr5211a1.htm.

Contains Nonbinding Recommendations
9
c. 	Other specific populations
Sponsors should include in trials geriatric participants,19 participants with renal insufficiency,
participants with diabetes mellitus, and participants with hepatic impairment, if feasible.
Because of the high incidence of TB in patients coinfected with HIV, participants with HIV
should be included in trials.
6. 	Dose Selection
When selecting a dosing regimen to be evaluated in phase 3 clinical trials, sponsors should
consider target PK/pharmacodynamic (PD) parameters (e.g., area under the curve/minimum
inhibitory concentration (MIC), maximal concentration/MIC, time above the MIC) based on
nonclinical models of TB (see section III.C.1.a), results from early clinical trials (e.g., EBA
and/or trials of AFB clearance from sputum at early time points), and results from exposure-
response evaluations. PK/PD evaluations should include evaluations based on free drug
concentrations.
7. 	Choice of Comparators
The choice of comparator or background regimen depends in part on the participant population
that the sponsor will enroll in the trial (e.g., the likelihood of infection with drug-susceptible or
drug-resistant M. tuberculosis). In general, sponsors should choose comparator regimens that
contain FDA-approved drugs and represent standard of care. Before trial initiation, sponsors
should discuss with the FDA the use of comparator regimens based on local practice outside of
the United States, or the use of drugs that are not FDA-approved.
8. 	Efficacy Endpoints
Sponsors can use the following efficacy endpoints in clinical trials of investigational drugs
intended to treat pulmonary TB:
• 	A primary clinical efficacy endpoint that is composed of survival, evaluation of
M. tuberculosis growth on serial sputum culture examinations at a fixed time point
following randomization for all treatment arms, and an extended disease-free period
of follow-up after completion of the planned treatment period (e.g. 12 to 18 months
after randomization). The FDA defines clinical success and failure as follows:
Clinical success is assigned to participants who are alive, achieved M. tuberculosis
culture negativity on serial sputum examinations, did not experience relapse or recurrence
of pulmonary TB during the posttreatment follow-up period, and otherwise did not meet a
definition of clinical failure. In general, protocol-defined serial sputum examinations
19 See the ICH guidances for industry E7 Studies in Support of Special Populations: Geriatrics (August 1994) and
E7 Studies in Support of Special Populations: Geriatrics: Questions and Answers (February 2012).

Contains Nonbinding Recommendations
10
should occur every 2 to 4 weeks during treatment, and every 3 months following
completion of treatment.20
Clinical failure is defined as having one or more of the following:
– 	Protocol-defined clinical progression of pulmonary disease during treatment
– 	Switch in, or discontinuation of antimycobacterial drugs, unless prespecified in the
protocol, because of tolerability issues or clinical progression of pulmonary TB
– 	Recurrent signs or symptoms of active TB, including radiographic worsening
compared with baseline findings, resulting in reinitiation of antimycobacterial therapy
during posttreatment follow-up21
– 	Death during treatment or follow-up
– 	Growth of M. tuberculosis on sputum culture outlined as follows:
 	Failure to achieve M. tuberculosis culture negativity in serial sputum specimens
during the treatment period
 	Failure to maintain culture negative status after a specific time point defined in
the trial (in general, this is expected to be any time after two consecutive negative
sputum cultures, taken at least 28 days apart) on therapy or in posttreatment
follow-up
– 	Any growth of M. tuberculosis from an extrapulmonary site during the trial
• 	A surrogate endpoint based on results of M. tuberculosis sputum cultures during
treatment. An extended period of follow-up after treatment (e.g., 12 to 18 months post-
randomization) is generally required to meet the clinical efficacy endpoint in TB trials.
To accelerate access to potentially life-saving medications in areas of unmet need (e.g.,
treatment of multidrug-resistant pulmonary TB) demonstration of treatment effect on
sputum culture conversion from positive to negative during treatment could be
considered as a surrogate endpoint reasonably likely to predict clinical benefit under the
accelerated approval statutory and regulatory provisions (Wallis et al. 2010; Wallis et al.
2013; Phillips et al. 2013; Wallis et al. 2015; Wallis and Peppard 2015; Phillips et al.
20 The protocol-defined timing of serial examinations of sputum for culture may differ from clinical practice, which
often depends on local treatment guidelines and respiratory isolation procedures.
21 In some circumstances, antimycobacterial therapy may be restarted though there is diagnostic uncertainty whether
relapse has occurred, but therapy is subsequently stopped when an alternative diagnosis is established. Protocols
should define the duration of retreatment therapy that will be used to define clinical failure to avoid labeling all trial
subjects in this situation as failures.

Contains Nonbinding Recommendations
11
2016; Meyvisch et al. 2018).22 If such an approach is taken, sponsors should obtain serial
cultures at specific time points during treatment (e.g., every 2 to 4 weeks). The time to
sputum culture conversion is the time to the first sterile culture, verified by M.
tuberculosis culture negativity in at least two consecutive sputum specimens taken at
least 28 days apart with subsequent sputum culture negativity while on treatment.
Sponsors considering sputum culture conversion or other surrogate endpoints (e.g., other
biomarkers) that may be reasonably likely to predict clinical benefit should consult with
the Agency as the clinical trial is being planned.
If a drug is considered for approval under the accelerated approval pathway, the confirmatory
trial would consist of the continued evaluation of the same study population in whom the
surrogate endpoint was determined for the duration of follow-up until achievement of the
clinical efficacy endpoint. Sponsors should specify a strategy to account for the statistical
analysis of multiple endpoints, such as hierarchical testing of the surrogate endpoint followed
by testing of the clinical efficacy endpoint.
• 	Secondary and exploratory endpoints. Sponsors should consider the following:
– 	A well-defined and reliable evaluation of symptoms, which can be included in the
clinical trial as a secondary or exploratory endpoint. Of note, symptom evaluation in
certain patient populations may be more difficult to interpret, for example, among
patients coinfected with HIV who may experience immune reconstitution
inflammatory syndrome or non-HIV-infected individuals with paradoxical reactions
(Rangaka et al. 2012).
– 	Molecular evaluations to ascertain whether a positive culture for M. tuberculosis after
drug treatment represents relapse or reinfection (e.g., an exploratory endpoint analysis
that treats relapse of the baseline M. tuberculosis infection as a failure of the original
study treatment and treats reinfection with a new M. tuberculosis isolate as a success
of the original study treatment).
9. 	Trial Procedures and Timing of Assessments
a. 	Entry visit
Sponsors should obtain baseline demographic information, current medications, and complete
physical examinations at the entry visit. In addition, sponsors should obtain the following at
entry:
• 	Clinical signs and symptoms of pulmonary TB (e.g., cough, sputum production, episodes
of hemoptysis, fever, pleuritic chest pain, weight loss, night sweats).
• 	Baseline safety laboratory evaluations.
22 Section 506(c) of the FD&C Act and 21 CFR 314.510 for small molecule drugs and 21 CFR 601.41 for biological
products.

Contains Nonbinding Recommendations
12
• 	HIV serology and, if HIV positive, viral load and CD4 cell count.
• 	Imaging results (standard posterior to anterior view and lateral chest radiographs or
computed tomography scans) describing the extent and severity of pulmonary disease.
• 	Sputum specimens for AFB smears and mycobacterial culture obtained by one of the
following: spontaneous expectoration, induction with hypertonic saline, bronchoscopy, or
gastric lavage (e.g., for pediatric participants). When applicable, baseline specimens for
quantitative cultures should be collected in a standardized manner (e.g., single early
morning induced sputum, pooled 24-hour sputum).
b. 	Visits during therapy and after therapy completion
In general, clinical assessments (including signs and symptoms, adverse events, and laboratory
tests, as appropriate) should occur weekly or biweekly during the first months of therapy,
followed by monthly assessments until therapy completion. Targeted physical examinations
should also be performed. After completion of therapy, assessments should occur approximately
every 3 months until the assessment of the primary efficacy endpoint is complete (e.g., at 12 to
18 months after randomization).
During therapy, sponsors should obtain sputum specimens for AFB smears and culture every 2 to
4 weeks until sputum negativity (two consecutive culture-negative sputum samples taken at least
28 days apart with no subsequent sputum positivity during treatment) has been established.
Depending on the investigational drug regimen and design, a shorter interval between specimen
collections (e.g., 2 weeks) may be appropriate for certain periods of the trial.
If participants are not able to expectorate sputum spontaneously at follow-up visits after therapy
completion, sponsors should consider other methods to obtain sputum (e.g., sputum induction).
10. 	Statistical Considerations
The sponsor should submit a detailed statistical analysis plan stating the trial hypotheses and the
efficacy analysis methods.
Sponsors should consider the following definitions of analysis populations:
• 	Safety population: all participants who received at least one dose of the investigational
drug during the trial.
• 	Intent-to-treat (ITT) population: all randomized participants.
• 	Microbiological ITT (micro-ITT) population: all randomized participants with a
positive culture for M. tuberculosis from a pretreatment prerandomization sample. For
trials intended to focus on participants with drug-resistant TB, sponsors can choose for
the primary analysis a micro-ITT population of all randomized participants with a
positive culture for a drug-resistant isolate of M. tuberculosis in the pretreatment

Contains Nonbinding Recommendations
13
prerandomization sample. For NI trials, the micro-ITT population should only include
participants in whom the baseline pathogen is susceptible to the active control regimen.
In general, the analysis population of greatest interest in the determination of efficacy is the
micro-ITT population. In addition, sponsors should evaluate consistency of results for efficacy
in the ITT population.
All participants should be followed completely for the duration of the trial even if they
discontinue the investigational drug(s). Sponsors should make every effort to minimize the loss
to follow-up throughout the trial. Given that missing data are nonetheless likely to occur, the
protocol should state how missing data will be handled in the primary efficacy analysis.
Additionally, the statistical analysis plan should define additional methods for handling missing
data. The study report should include an assessment of the dependence of the trial results on the
specific method for handling missing data.
To improve the precision of treatment effect estimation and inference, sponsors should consider
stratifying randomization or adjusting for baseline factors that are anticipated to be prognostic of
the outcome.23
C. 	Other Considerations
1. 	Microbiological Considerations
Sponsors of investigational drugs being evaluated for the treatment of TB should have supportive
data from nonclinical microbiological studies. These studies may provide data to inform
selection of the regimen of antimycobacterial drugs to be evaluated in clinical trials and to assess
the contribution of each drug to the investigational drug regimen.
a. 	Nonclinical studies
Nonclinical studies should encompass the following:
• 	Investigations of drug activity (inhibiting growth or killing) against metabolically active,
dormant, and intracellular stages of M. tuberculosis.
• 	Susceptibility testing against metabolically active bacilli from drug-susceptible laboratory
strains, laboratory strains with known patterns of drug resistance, and clinical isolates
representing different geographical regions and M. tuberculosis complex lineages.
23 See the guidance for industry Adjusting for Covariates in Randomized Clinical Trials for Drugs and Biological
Products.

Contains Nonbinding Recommendations
14
• 	Standardized methods for susceptibility testing such as those recommended by the
Clinical Laboratory Standards Institute (CLSI).24
• 	If nonstandard methods are being employed in the trial, prior submission for FDA review
of a complete description of the methods and the performance characteristics of the assay
in the actual laboratory where testing will be done.
• 	Establishment of quality control parameters for susceptibility testing before
determination of in vitro activity.25
If two or more new investigational drugs are under evaluation simultaneously, the sponsor
should conduct factorial design studies evaluating the new investigational drugs and provide the
results to the FDA.9 The FDA encourages testing against multiple strains of M. tuberculosis.
See section III.C.3.d, In vitro hollow fiber system models, for methods of assessment of the
contribution of individual drugs in a combination regimen.
Appropriate models (including NAMs) can serve as an important bridge between the
identification of in vitro antimycobacterial effects of an investigational drug and the initiation of
clinical trials. PK assessments and changes in drug susceptibility in nonclinical studies may
inform clinical trial designs. Sponsors should consider evaluations of the investigational drug,
and/or combinations of investigational drugs, using different models and more than one
strain/isolate of M. tuberculosis to study mycobacterial burden and sterilizing activity. Studies
conducted using a factorial design using clinically relevant exposures can provide information on
the contribution of the individual drugs to the combination regimen.
b. 	Drug resistance and cross-resistance
Sponsors should examine the potential of M. tuberculosis isolates to develop resistance to the
investigational drug in appropriate in nonclinical models and should evaluate the potential for
cross-resistance to drugs in the same class or in other classes used for the treatment of TB. If
resistance is demonstrated, it is important to identify the mechanism(s) of resistance. Sponsors
should attempt to evaluate the clinical significance of any changes in phenotype (e.g., in vitro
susceptibility to the investigational drug) or genotype observed in nonclinical studies by
correlating such changes with efficacy outcomes.
24 For examples, see the guidance for industry and FDA Class II Special Controls Guidance Document:
Antimicrobial Susceptibility Test (AST) Systems (August 2009) and CLSI’s Susceptibility Testing of Mycobacteria,
Nocardiae, and Other Aerobic Actinomycetes; Approved Standard — Third Edition, (available at
https://clsi.org/standards/products/microbiology/documents/m24/). For the most recent version of a class II special
controls guidance document, check the FDA class II special controls guidance document web page at
https://www.fda.gov/medical-devices/guidance-documents-medical-devices-and-radiation-emitting-products/class-
ii-special-controls-documents.
25 For more details, see the guidance for industry Microbiology Data for Systemic Antibacterial Drugs —
Development, Analysis, and Presentation (February 2018).

Contains Nonbinding Recommendations
15
c. 	Types of culture media to identify M. tuberculosis
Solid media (e.g., Löwenstein-Jensen medium, Middlebrook 7H10 or 7H11 agar media) and
liquid media (e.g., mycobacteria growth indicator tube) are culture assay methods used to
identify and characterize M. tuberculosis. Sponsors can include other newer molecular
methodologies to detect M. tuberculosis and its susceptibility profile in trials for microbiological
evaluations. Sponsors should specify the methods used to culture and identify M. tuberculosis as
well as the in vitro susceptibility testing methods that will be employed in the trial.
For evaluations at baseline as well as during and after treatment, the Agency recommends using
both solid and liquid media. The advantages to this approach are (1) more rapid observation of
mycobacterial growth in liquid media (e.g., less than 2 weeks) and (2) that culture on solid media
is already underway to identify M. tuberculosis and evaluate in vitro susceptibility.
Within a clinical trial, the culture methodologies among trial sites should be consistent to
evaluate all participants in the trial. Other types of culture evaluations can be informative as
secondary or exploratory endpoints (e.g., quantitative culture techniques).
d. 	Differentiate relapse from reinfection or new infection
As a secondary analysis, sponsors should aim to use molecular methods to evaluate whether
clinical failure is caused by relapse of the original infection or by development of a new
infection, especially in participants living in endemic areas. If any of these methods are used in a
clinical trial, the sponsor should include details of the methods used as well as performance
characteristics of all assays in the clinical protocol.
2. 	Relevant Nonclinical Safety Considerations
Combination regimens remain the standard of care for the treatment of TB. Individual drugs
may be developed for treatment of active disease, although they would be used as part of a
combination regimen. Nonclinical studies to characterize the safety profile of individual drugs
or a combination regimen and to support clinical trials and approval of a marketing application
will vary depending on the information available on each drug and the intended patient
population.26 The use of new approach methodologies is strongly encouraged. The Agency
recommends sponsors discuss with the FDA the available toxicology data for each
investigational drug and a proposal for the nonclinical development of the combination regimen.
Alternative, streamlined nonclinical development programs will be considered as part of a
weight of evidence assessment by the FDA when appropriate.
Sponsors may conduct nonclinical toxicology studies of a combination regimen consisting of two
or more investigational (unapproved) drugs before initial administration of that combination
26 For guidance on when to conduct nonclinical combination studies to support clinical trials of combination
regimens, see the following: (1) guidance for industry Nonclinical Safety Evaluation of Drug or Biologic
Combinations (March 2006); (2) ICH M3(R2); (3) ICH guidance for industry S6(R1) Preclinical Safety Evaluation
of Biotechnology-Derived Pharmaceuticals (May 2012); and (4) guidance for industry Codevelopment of Two or
More New Investigational Drugs for Use in Combination.

Contains Nonbinding Recommendations
16
regimen to humans based on the following:
• 	The availability of clinical experience with the individual drugs
• 	The availability of relevant nonclinical toxicology data for each of the individual drugs
for the proposed duration of the combination regimen
• 	The existence of a significant toxicological concern and the safety margin between the no
observed adverse effects level (NOAEL) for each of the individual drugs in the
nonclinical toxicology studies and the proposed human exposure to each of the
investigational drugs in the combination regimen
• 	The potential for drug-drug interactions based on the absorption, distribution,
metabolism, and excretion of each of the drugs
• 	The potential for adverse effects to involve the same organ system (overlapping
toxicities) or synergistic toxicities based on a review of accumulated data from each of
the investigational drugs
3. 	PK/PD Considerations
a. 	Phase 1/phase 2 PK trials
The PK of the investigational drug should be fully characterized in single-dose PK, multiple-
dose PK, and phase 2 PK/PD evaluations. The FDA recommends characterization of PK in
specific populations, including participants who have renal or hepatic impairment, as well as an
evaluation of the drug effect on the QT interval.27
b. 	Drug interactions
Sponsors should conduct in vitro studies to determine the potential of the investigational drug to
act as a substrate, inhibitor, or inducer of major human metabolizing enzymes and relevant
transporters.28 Based on these results, drug interaction evaluations between one or more of the
antimycobacterial drugs used in the planned combination regimen, or with drugs unrelated to the
treatment of TB but likely to be used concomitantly for other indications (e.g., antiretroviral
therapy for treatment of HIV; antiviral therapy for treatment of hepatitis B or C), may be needed
before initiating clinical efficacy trials.29 The Agency strongly recommends that sponsors
consult the FDA during drug development regarding appropriate drug interaction evaluations.
27 See the ICH guidance for industry E14 Clinical Evaluation of QT/QTc Interval Prolongation and Proarrhythmic
Potential for Non-Antiarrhythmic Drugs — Questions and Answers (R2) (June 2017).
28 See the ICH guidance for industry M12 Drug Interaction Studies (August 2024).
29 Ibid.

Contains Nonbinding Recommendations
17
c. 	Exposure response
Sponsors should explore exposure-response relationships during early phases of drug
development to aid in the selection of optimal dosing strategies for evaluation in later trials.30
The FDA encourages sponsors to explore exposure-response relationships for sputum and serum
or plasma drug concentrations and markers of activity (e.g., the time-to-sputum-conversion or
sputum conversion rate at 2 months in participants with pulmonary TB).
d. 	In vitro hollow fiber system models
The results from hollow fiber system models, combined with other sources of nonclinical data,
can help inform the selection of antimycobacterial drug regimens to begin clinical evaluation
(Chilukuri et al. 2015). The hollow fiber system models can be used to simulate PK
characteristics of drugs intended to treat TB and allows for the exploration of concentration-
effect relationships potentially relevant to the treatment of TB in the clinical setting. These
models are expected to provide key information on regimen selection for further evaluation.
These models may also play an important role in evaluating the contribution of each drug (at
clinically relevant exposures) to the treatment effect but are considered supportive and
complementary to data from the relapsing mouse TB model or other validated in vivo systems.
4. 	Foreign Clinical Data31
FDA regulations permit the acceptance of foreign clinical trials in support of a new drug
application or biologics license application approval (21 CFR 312.120). The applicability of
foreign clinical data to the U.S. population must be demonstrated (21 CFR 314.106).
5. 	Data standards for TB
Study data standards describe a standardized way to exchange clinical and nonclinical research
data between computer systems. Data standards have been developed for TB to provide a
consistent general framework for organizing study data, including templates for datasets,
standard names for variables, and standard ways of doing calculations with common variables.32
30 See the guidance for industry Exposure-Response Relationships — Study Design, Data Analysis, and Regulatory
Applications (April 2003).
31 See the guidance for industry and FDA staff FDA Acceptance of Foreign Clinical Studies Not Conducted Under
an IND Frequently Asked Questions (March 2012).
32 See, for example, the TB Therapeutic Area User Guide version 2 available at
https://www.cdisc.org/standards/therapeutic-areas/tuberculosis/tuberculosis-therapeutic-area-user-guide-v2-0 and
FDA’s Study Data Standards Resources web page available at
https://www.fda.gov/ForIndustry/DataStandards/StudyDataStandards/default.htm.

Contains Nonbinding Recommendations
18
REFERENCES	1
2
Chilukuri, D, O McMaster, K Bergman, et al., 2015, The Hollow Fiber System Model in the	3
Nonclinical Evaluation of Antituberculosis Drug Regimens, Clin Infect Dis, 61 (Suppl	4
1):S32-S33.	5
6
DerSimonian, R and N Laird, 1986, Meta-Analysis in Clinical Trials, Controlled Clin Trials,	7
7(3):177–188.	8
9
East African/British Medical Research Council, 1981, Controlled Clinical Trial of Five Short-	10
Course (4-Month) Chemotherapy Regimens in Pulmonary Tuberculosis, Am Rev Respir Dis,	11
123(2):165–170.	12
13
East and Central Africa/British Medical Research Council Fifth Collaborative Study, 1983,	14
Controlled Clinical Trial of 4 Short-Course Regimens of Chemotherapy (Three 6-Month and	15
One 8-Month) for Pulmonary Tuberculosis, Tubercle, 64(3):153 – 166.	16
17
Farrington, CP and G Manning, 1990, Test Statistics and Sample Size Formulae for Comparative	18
Binomial Trials With Null Hypothesis of Non-Zero Risk Difference or Non-Unity Relative Risk,	19
Stat Med, 9(12):1447–1454.	20
21
Gillespie, SH, AM Crook, TD McHugh, et al., 2014, Four-Month Moxifloxacin-Based Regimens	22
for Drug-Sensitive Tuberculosis, New Engl J Med, 371(17): 1577– 1587.	23
24
Johnson, JL, DJ Hadad, R Dietze, et al., 2009, Shortening Treatment in Adults With Noncavitary	25
Tuberculosis and 2-Month Culture Conversion, Am J Respir Crit Care Med, 180(6):558–563.	26
27
Meyvisch, P, C Kambili, K Andries, et al., 2018, Evaluation of Six Months Sputum Culture	28
Conversion as a Surrogate Endpoint in a Multidrug Resistant-Tuberculosis Trial, PLoS One,	29
13(7):e0200539.	30
31
Nachman, S, A Ahmed, F Amanullah, et al., 2015, Towards Earlier Inclusion of Children in	32
Tuberculosis (TB) Drug Trials: Consensus Statements from an Expert Panel, Lancet Infect Dis,	33
15(6):711–720.	34
35
Nahid, P, SE Dorman, N Alipanah, et al., 2016, Official American Thoracic Society/Centers for	36
Disease Control and Prevention/Infectious Diseases Society of America Clinical Practice	37
Guidelines: Treatment of Drug-Susceptible Tuberculosis, Clin Infect Dis, 63(7):853–867.	38
39
Nunn, AJ, PPJ Phillips, and SH Gillespie, 2008, Design Issues in Pivotal Drug Trials for Drug	40
Sensitive Tuberculosis (TB), Tuberculosis; 88(Suppl 1):S85–S92.	41
42
Phillips, PPJ, CM Mendel, DA Burger, AM Crook, AJ Nunn, R Dawson, et al., 2016, Limited	43
Role of Culture Conversion for Decision-making in Individual Patient Care and for Advancing	44
Novel Regimens to Confirmatory Clinical Trials, BMC Med, 14:19.	45
46

Contains Nonbinding Recommendations
19
Phillips, PPJ, K Fielding, and AJ Nunn, 2013, An Evaluation of Culture Results During	47
Treatment for Tuberculosis as Surrogate Endpoints for Treatment Failure and Relapse, PLoS	48
ONE, 8(5):e63840.	49
50
Rangaka, MX, RJ Wilkinson, JR Glynn, et al., 2012, Effect of Antiretroviral Therapy on the	51
Diagnostic Accuracy of Symptom Screening for Intensified Tuberculosis Case Finding in a	52
South African HIV Clinic, Clin Infect Dis, 55(12):1698–1706.	53
54
Singapore Tuberculosis Service/British Medical Research Council, 1986, Long-Term Follow-up	55
of a Clinical Trial of Six-Month and Four-Month Regimens of Chemotherapy in the Treatment	56
of Pulmonary Tuberculosis, Am Rev Respir Dis, 133(5):779–783.	57
58
Wallis, RS, and T Peppard, 2015, Early Biomarkers and Regulatory Innovation in Multidrug-	59
Resistant Tuberculosis, Clin Infect Dis, 61 (Suppl 3):S160–S163.	60
61
Wallis, RS, C Wang, TM Doherty, P Onyebujoh, M Vahedi, H Laang, et al., 2010, Biomarkers	62
for Tuberculosis Disease Activity, Cure, and Relapse, Lancet Infect Dis,10(2): 68–69.	63
64
Wallis, RS, C Wang, D Meyer, and N Thomas, 2013, Month 2 Culture Status and Treatment	65
Duration as Predictors of Tuberculosis Relapse Risk in a Meta-Regression Model PLoS ONE,	66
8(8): e71116.	67
68
Wallis, RS, T Peppard, and D Hermann, 2015, Month 2 Culture Status and Treatment Duration	69
as Predictors of Recurrence in Pulmonary Tuberculosis: Model Validation and Update, PLoS	70
ONE, 10(4):e0125403.	71
72

Contains Nonbinding Recommendations
20
APPENDIX	73
Example of a Justification for a Noninferiority Margin	74
in a Treatment-Shortening Clinical Trial of Pulmonary Tuberculosis	75
76
This appendix provides an example of a noninferiority (NI) margin justification. As stated in	77
this guidance, NI margin justifications are dependent on the specific design of the NI trial. This	78
justification is for a specific NI trial that would compare an investigational drug regimen	79
consisting of a new investigational drug plus the first 4 months of the standard regimen to the	80
standard 6-month regimen in participants with drug-susceptible tuberculosis (TB). The effect of	81
the investigational drug essentially replaces the effect of Months 5 and 6 of the standard regimen.	82
Using historical data, this justification determines the effect of these 2 months of therapy	83
(historical evidence of sensitivity to drug effects (HESDE)) to determine if the new	84
investigational drug is effective based on the results of the NI trial. Additional information is	85
available regarding a complete discussion of NI trials and justifications of margins.1	
86
87
We identified two trials that allowed for an estimate of the effect of Months 5 and 6 in the	88
standard regimen for drug-susceptible TB, based on a comparison of the standard-of-care	89
regimen (2 months of treatment with ethambutol (or streptomycin), isoniazid, rifampin, and	90
pyrazinamide followed by 4 months of treatment with isoniazid and rifampin, which is often	91
described in abbreviated terminology as 2EHRZ/4HR or 2SHRZ/4HR) to a 4-month regimen of	92
2EHRZ/2HR or 2SHRZ/2HR.2 The endpoint of unfavorable outcome was defined as one of the	93
following: (1) participants who never become sputum culture negative for Mycobacterium	94
tuberculosis while on therapy; (2) participants who had microbiological confirmation of relapse	95
of pulmonary TB within a 12-month period of observation following therapy completion; or (3)	96
participants who died at any time within the clinical trial drug administration period and 12-	97
month period of observation following therapy completion.	98
99
Table A, below, contains the results from the two trials among participants randomized to	100
receive the 6-month regimen or the 4-month regimen. A comparison of the two regimens gives	101
an estimate of the effect of the final 2 months of the 6-month regimen of 8.4 percent with a lower	102
bound of the 95 percent confidence interval of 4.8 percent; 4.8 percent can be used as a	103
conservative estimate of the treatment effect of Months 5 and 6 of treatment.	104
105
1 See the guidance for industry Non-Inferiority Clinical Trials to Establish Effectiveness (November 2016). We
update guidances periodically. For the most recent version of a guidance, check the FDA guidance web page at
https://www.fda.gov/regulatory-information/search-fda-guidance-documents.
2 See Singapore Tuberculosis Service/British Medical Research Council 1986; East and Central Africa/British
Medical Research Council Fifth Collaborative Study 1983; and East African/British Medical Research Council
1981.

Contains Nonbinding Recommendations
21
Table A: The Results of Two Treatment-Shortening Studies *	
106
107
Study** 	6-Month
Regimen
Unfavorable
Outcome
4-Month
Regimen
Unfavorable
Outcome
Treatment Effect (4-
Month Regimen
Minus 6-Month
Regimen) and 95% CI
1 	2SHRZ/4HR(Z) 	1.2% (2/158) 	2SHRZ/2HR(Z) 	9.6% (15/156) 	8.4% (3.8%, 14.2%)
2 	2SHRZ/4HR 	4.7% (8/172) 	2SHRZ/2HR(Z) 	13.2% (28/212) 	8.6% (2.4%, 14.6%)
Summary Estimate and 95% CI*** 	8.4% (4.8%, 12.1%)
* CI = confidence interval; 2SHRZ/4HR(Z) = 2 months of treatment with streptomycin, isoniazid, rifampin, and	108 pyrazinamide followed by 4 months of treatment with isoniazid and rifampin (and pyrazinamide); 2SHRZ/2HR(Z) =	109 2 months of treatment with streptomycin, isoniazid, rifampin, and pyrazinamide followed by 2 months of treatment	110 with isoniazid and rifampin (and pyrazinamide).	111 ** The number of deaths is unknown for Study 1 and therefore is not included in the outcome. The 6-month and 4-	112 month regimens in Study 2 are from separate trials; however, they were similarly designed and conducted and	113 occurred close in time. Study 1: Singapore Tuberculosis Service/British Medical Research Council, 1986, Long-	114 Term Follow-up of a Clinical Trial of Six-Month and Four-Month Regimens of Chemotherapy in the Treatment of	115 Pulmonary Tuberculosis, Am Rev Respir Dis, 133(5):779–783. Study 2: East and Central Africa/British Medical	116 Research Council Fifth Collaborative Study, 1983, Controlled Clinical Trial of 4 Short-Course Regimens of	117 Chemotherapy (Three 6-Month and One 8-Month) for Pulmonary Tuberculosis, Tubercle, 64(3):153–166; and East	118 African/British Medical Research Council, 1981, Controlled Clinical Trial of Five Short-Course (4-Month)	119 Chemotherapy Regimens in Pulmonary Tuberculosis, Am Rev Respir Dis, 123(2):165–170.	120 *** Random effect model per DerSimonian, R and N Laird, 1986, Meta-Analysis in Clinical Trials, Controlled Clin	121 Trials, 7(3):177–188.	122
123
In an NI trial in participants with drug-susceptible pulmonary TB where a treatment-shortening	124
regimen is compared with a standard 6-month regimen, the selection of an NI margin of 4.8	125
percent can be supported by the historical data. The NI margin justification presented here is a	126
modification of the justification presented in Nunn et al. 2008.3	
127
128
Although an NI margin of 4.8 percent may seem overly conservative, the fact that a very high	129
proportion of participants achieve a successful primary efficacy outcome with standard of care	130
provides for a reasonable estimate of the sample size for an NI trial. Additionally, given the high	131
proportion of participants achieving a successful outcome, there is interest in maintaining this	132
high proportion in new investigational drug regimens. For example, we identified a trial	133
(Johnson et al. 2009)4 that described halting of the trial by a data monitoring committee based on	134
an approximately 5 percentage point estimate difference between the standard regimen and a	135
treatment-shortening regimen, indicating that there is a clinical expectation that there should be a	136
high proportion of participants achieving successful outcomes in both treatment groups, making	137
the selection of an NI margin of 4.8 percent a feasible consideration.	138
139
The following example provides a framework for discussion with the FDA about sample size	140
estimation for an NI trial evaluating a treatment-shortening regimen (Farrington and Manning	141
3 Nunn, AJ, PPJ Phillips, and SH Gillespie, 2008, Design Issues in Pivotal Drug Trials for Drug Sensitive
Tuberculosis (TB), Tuberculosis; 88(Suppl 1):S85–S92.
4 Johnson, JL, DJ Hadad, R Dietze, et al., 2009, Shortening Treatment in Adults With Noncavitary Tuberculosis and
2-Month Culture Conversion, Am J Respir Crit Care Med, 180(6):558–563.

Contains Nonbinding Recommendations
22
1990).5 The total sample size of enrolled participants is approximately 528 participants per arm	142
based on the following assumptions: (1) the identification of Mycobacterium tuberculosis in 90	143
percent of enrolled participants (primary analysis population is approximately 475 participants	144
per arm); (2) a two-sided type I error of 0.05 and power of 90 percent; (3) for both arms, a rate of	145
5 percent of participants who have the endpoint of failure to convert to negative sputum cultures,	146
or who experience relapse of TB, or death at a 12-month period of observation; and (4) an NI	147
margin of 4.8 percent.	148
149
Sponsors should discuss with the FDA appropriate NI margins for specific NI trials being	150
proposed.	151
5 Farrington, CP and G Manning, 1990, Test Statistics and Sample Size Formulae for Comparative Binomial Trials
With Null Hypothesis of Non-Zero Risk Difference or Non-Unity Relative Risk, Stat Med, 9(12):1447–1454.