Could a New Universal Tuberculosis Regimen Help End the TB Pandemic?

As tuberculosis (TB) is the world’s most lethal infectious disease, there is an urgent need to improve the treatment of this bacterial infection. One approach under consideration is the institution of a universal or “pan-TB” regimen comprising new agents capable of treating multidrug-resistant (MDR) and even extensively drug-resistant (XDR) TB.

The World Health Organization (WHO) target regimen profile for a pan-TB regimen aims to revolutionize the treatment of TB regardless of drug resistance.

The wish list of attributes includes a highly effective, safe, well-tolerated, 3- to 4-drug, once-daily, oral regimen with minimal drug-drug interactions that could be administered to any patient with TB regardless of age, HIV status, pregnancy status, or comorbidity.

Crucially there needs to be no or minimal pre-existing resistance for the regimen, which would allow empirical use without the need for drug susceptibility testing (DST) and therefore eliminating the delay before starting effective treatment.¹

The Lancet Respiratory Medicine in April 2018 staged a debate between proponents and opponents of pan-TB regimens to address the current suboptimal worldwide management of TB resistance.

Proposing this approach, Wallis and colleagues² argue that delayed or missed diagnosis of MDR/XDR TB in many high-burden countries leads frequently to suboptimal or ineffective treatment. Consequently, mortality and onward transmission of resistant isolates are greater than could occur with newer combinations of antimicrobial agents. In support, they cite preliminary data from the Nix-TB trial (ClinicalTrials.gov Identifier: NCT02333799); combining linezolid, pretomanid, and bedaqualine for 6 months demonstrated efficacy in patients with pre-XDR and XDR TB. Moreover, by adapting a TB transmission model² to incorporate the Nix-TB regimen, they anticipate the prevention of 4 million infections and avoiding the loss of half a million lives over the course of a decade in 4 countries with a high burden of TB: South Africa, Mozambique, India, and China (unpublished data).

Opposing this approach, Dheda and colleagues³ raise the concern that such a regimen could rapidly lead to resistance amplification and circulation of strains that are no longer susceptible to new agents in the “universal” regimen. This could lead to greatly diminished options for rescue regimens. As a result, the imagined benefits in terms of lives saved and reduction in transmission would not occur and would instead be overcome by the same barriers faced by current treatment programs (drug costs, import restrictions, inadequate scale-up of field staff, etc). Furthermore, they fear that there would be a market reaction against development of new diagnostics and new drugs, with harmful long-term consequences.

Infectious Disease Advisor discussed the differing points of view with an author of each of the opposing articles: Robert Wallis, MD, Chief Scientific Officer of the Aurum Institute, and Jennifer Furin, MD, PhD, Infectious Disease Physician and Lecturer on Global Health and Social Medicine at Harvard Medical School.

Infectious Disease Advisor: Do you think the roll-out of a pan-TB regimen should be restricted to high-burden countries with limited second-line DST testing or should it be universal to all settings after rifampicin resistance (RIF-R) is confirmed?

Dr Wallis: I think it will depend quite a bit on the specific local situation and the characteristics of the new regimen. South Africa does pretty well at diagnosing RIF-R but not so well in returning results for second-line DST, so that would be a reasonable point at which to intervene. Many high-burden countries have limited capabilities to diagnose RIF-R. In that type of setting, regimen duration, tolerability, and cost might end up being the main factors determining how a pan-TB regimen would be deployed.

Infectious Disease Advisor: Opponents of a universal regimen argue that rapid development of resistance would occur and lead to untreatable strains within a few years. Do you agree that emergence of resistance to newer agents would be more rapid with a pan-TB strategy compared with current approaches?

Prof Robert Wallis: No, I do not. The best way to prevent emergence of resistance to new drugs is to ensure that they are always used in fixed combinations. Thinking about this from a stochastic perspective can be helpful. The frequency of resistant mutants occurring spontaneously in a population with drug-sensitive Mycobacterium tuberculosis  depends on the drug and the drug concentration. For most TB drugs, the frequency of spontaneously resistant mutants ranges from 10^-5 to 10^-10. Some oxazolidinones have particularly low resistance frequencies (10^-10 or better). This is probably why an MDR TB study in Korea,⁴ which was essentially of linezolid monotherapy, had as low a failure rate as it did. In any case, the frequency of 3 independent events all occurring is the product of their individual odds. With 3 drugs, this is likely to be 10^-15 or possibly even 10^-20 (if oxazolidinones are included). Patients with cavitary TB might have 10^-10 or 10^-12 bacilli in total, but not 10^-20. This is an oversimplification, but it illustrates why the risk of resistance is so much greater when only 1 or 2 new drugs are used rather than 3. The present policy of trying to assemble a new regimen for a failing treatment using only 1 or 2 novel drugs (plus several old ones) is much riskier than a regimen comprised solely of novel drugs. 

Infectious Disease Advisor: Should pilot studies be conducted to demonstrate sufficient benefit before a shift in policy is proposed?

Dr Wallis: I would like to see the first study of a pan-TB regimen start off in patients with high unmet need (XDR, for example), in whom greater risks would be more acceptable. A small study in this population could be highly informative. If things went well, the study could expand progressively to include patients who are not in such dire straits.

Infectious Disease Advisor: Is there concern that drug and precision diagnostics development will be undermined by a universal regimen?

Dr Wallis: Having a pan-TB regimen might actually spur development of new TB diagnostics that do not depend on sputum, such as the highly sensitive test for lipoarabinomannan in urine. The big disadvantage of such tests right now is that they cannot provide any information as to the drug susceptibility of the infection, leaving the patient and the physician in a quandary as to what to do when a test is positive. In any case, the new regimen would not remove the need for TB-control programs to collect isolates, examine drug susceptibility, and perform strain typing to better understand their epidemic. The difference will be that the time sensitivity for getting this information will be removed; patients will not depend on it for their immediate care.

Infectious Disease Advisor: Advocates argue that by using 3 novel drugs in a regimen, the likelihood of resistance developing is greatly reduced. Would this not improve on current approaches, such as during the wait for second-line DST for resistant bacilli that are exposed to only 1 or 2 active agents?

Dr Furin: A universal regimen with 3 novel drugs would be a temporizing measure at best.  Infectious disease practice in general shows that all infectious agents have the potential for resistance to develop.  In fact, it is often naturally occurring mutations — and these exist to every agent we use for TB — that are selected when suboptimal therapy is given.  The problem instead is that people are given weak agents and highly toxic agents that are not effective enough while awaiting DST.

Infectious Disease Advisor: What alternative solutions do you envisage?

Dr Furin: One solution is to improve our diagnostics so that people do not have to wait very long for DST to come back.  And we have many options on the table for doing that, including both the Xpert® expanded platform⁵ as well as rapid and amazing developments in whole genome sequencing that could be field-ready in a matter of a few years.  But when we start talking about taking diagnostics out of the equation, then there is less enthusiasm for people to develop these tools and less action on the part of countries to use and deploy these tools.  This is a major problem.

Second is to offer people the most effective agents at the start of treatment rather than “saving” these for later.  The treatment of DR-TB is based largely on habit, and there is not a single randomized controlled trial (RCT) in people living with DR-TB for any of the “core agents” used in therapy today.  But we do have RCT evidence from people living with DR-TB with 4 drugs: linezolid, bedaquiline, delamanid, and clofazimine (as well as an ongoing trial with levofloxacin).  Why then are we giving people regimens with drugs like ethionamide, kanamycin, and cycloserine?  This is a cost issue due to a perverted notion people have about “protecting drugs” instead of “protecting people.”

Infectious Disease Advisor: Does your criticism of the pan-TB regimen extend to selective application in settings where second-line DST is inadequate?

Dr Furin: The goal where second-line DST is inadequate should be to improve access to second-line DST.  The TB community has been debating for more than 20 years whether or not we need access to DST; we have given up on this, concluding it will be “too complicated.”  Instead we focus on these “one-size-fits-all” regimens that expose people to toxic drugs to which they are resistant and that leaves out effective drugs that could improve their likelihood of relapse-free cure. 

The field of TB has been grossly underambitious and has for decades tried to do the minimum possible and see if we can get away with it.  This approach has clearly failed. Instead, we should make sure DST is available in all settings and that people get the treatment that is best suited for them in terms of drug composition and duration. The prominent TB physician Ken Castro summed it up best for me when he said at a meeting on a universal regimen that “we had a universal regimen — it was called isoniazid (INH), rifampin, pyrazinamide (PZA), and ethambutol.”   The question for me is why would we insist again on trying to use an approach that did not work?  It is short-sighted, violates the human rights of people living with TB, and continues to set a very low bar for drug and diagnostic development.  If a universal regimen is our goal, we are moving backward in time and we will never be able to end TB.

Conclusion

The advent of new drugs and new diagnostics promises an era of shorter courses for MDR-TB treatment. However, the balance between rolling out new effective regimens, tailoring treatment to the individual, and preserving drugs for the future remains challenging. Debate about the optimum approach will no doubt continue.

References

1. Lienhardt C, Nahid P, Rich ML, et al. Target regimen profiles for treatment of tuberculosis: a WHO document. Eur Respir J. 2017;49(1):1602352.
2. Wallis RS, Cohen T, Menzies NA, Churchyard G. Pan-tuberculosis regimens: an argument for. Lancet Resp Med. 2018;6(4):239-240.
3. Dheda K, Gumbo T, Lange C, Horsburgh CR, Furin J. Pan-tuberculosis regimens: an argument against. Lancet Resp Med. 2018;6(4):240-242.
4. Lee M, J Lee J, Carroll M, et al. Linezolid for treatment of chronic extensively drug-resistant tuberculosis. N Engl J Med. 2012;367:1508-1518.
5. Xie YL, Chakravorty S, Armstrong DT, et al. Evaluation of a rapid molecular drug-susceptibility test for tuberculosis. N Engl J Med. 2017;377:1043-1054.