Multi-drug resistant tuberculosis (MDR-TB) treatment is evolving. Last month the South Africa National Tuberculosis Program decided to prioritize oral bedaqualine for all rifampin-resistant TB (RR-TB) in place of — the often more toxic — injectable drugs. The question of whether a package of novel oral medications as part of a universal regimen can replace the current approach to TB management divides the TB therapeutic community. In this second article we further explore this divergence of opinion.
Infectious Disease Advisor continues the discussion with 2 authors who advocate opposing views on the issue of a universal TB regimen: Professor Robert Wallis, chief scientific officer of the Aurum Institute in Johannesburg, South Africa, and Dr Jennifer Furin, infectious disease physician and lecturer on global health and social medicine at Harvard Medical School in Boston, Massachusetts.
Infectious Disease Advisor: You mention that it is imperative to use the most effective agents at the start of TB treatment. Would a pan-TB regimen, composed of primarily novel drugs in a fixed combination, be a way of achieving this aim?
Dr Jennifer Furin: Well, that would really depend on the drugs. The best drug we have for TB now is rifampin. And when people are susceptible to rifampin, there is an incredibly effective treatment regimen they can be given. It’s a long regimen, but it is effective and well tolerated. None of the new drugs available have demonstrated efficacy like that seen with rifampin. So with a “universal regimen” you are already talking about denying people access to treatment with rifampin, even individuals who are susceptible to it; you are not using one of the most effective drugs.
In people with resistance to rifampin, there are several drugs that have shown evidence of efficacy in randomized controlled trials. Using these drugs — like linezolid, delemanid, bedaquiline, and clofazimine — up front, and I would also include the quinolones in this list, would be ideal. But this is not the same as a “universal regimen.” So, until we have a drug or drugs that show the same efficacy and safety as rifampin, we cannot even begin to make the claim that we are using the most effective agents.
The need for 3 novel agents with “little or no resistance” also implies that you would have to hold those agents in reserve and not give them to anyone until you have 3 agents that you know would work together, have no negative drug-drug interactions, have limited overlapping toxicity, and would be as effective when used together as the current treatment for DS-TB. Given the crisis of MDR-TB — which will be responsible for a third of deaths from TB due to antimicrobial resistance — withholding effective agents would be ethically questionable, especially since this is an airborne infectious disease. TB drug development is still in a nascent stage, due in part to our past reliance on a universal regimen, and the likelihood that we will develop 3 such agents in the next decade is very small.
Infectious Disease Advisor: New molecular second-line drug susceptibility testing (DST) will hopefully be deployed at scale within a few years. But in the meantime, isn’t a universal regimen, even if it is only a “temporizing measure,” exactly what we need?
Dr Furin: There is not yet a pan-TB regimen, and given what we know about TB clinical trials, it would take at least 10 years for one to be developed. So we are talking about a theoretical temporizing measure, not one that actually exists or is even close to existing. The modelling work done in the Wallis commentary, as well as that done in another recently published paper on this topic by Emily Kendall at the Center for Tuberculosis Research, Johns Hopkins University School of Medicine, Baltimore, Maryland, and colleagues,1 make some huge assumptions about the effectiveness of a pan-TB regimen. The Wallis paper models their arguments based on the NiX trial,2 a single-arm open-label study with a small number of participants who were treated with 3 drugs (bedaquiline, pretomanid, and linezolid). Preliminary results from this study of difficult-to-treat patients look promising, but a significant amount of work — years and years of work — will be needed to show whether this regimen is truly as effective as the study sponsors claim it to be. And I would note here that those data have not yet been published anywhere.
So, there is no universal regimen to implement, and before we have one we will likely have improved diagnostics that we should use to give people drugs they are sensitive to and avoid drugs they are resistant to.
Infectious Disease Advisor: As you mention, the comparative success of pretomanid, linezolid and bedaqualine in extensively drug-resistant patients with TB in the NiX trial is of great interest. What are the drawbacks to using this as a potential universal regimen if efficacy is found to be consistent and replicable?
Dr Furin: The arguments being made by people who are for a pan-TB regimen have not taken the toxicity issue into account. The NiX trial included such high doses of linezolid that almost all the patients had to have a dose reduction or discontinue the medication. That is why the sponsors of NiX are now doing a dose-finding trial of linezolid. We know the doses of linezolid used in NiX are too toxic, but we don’t know whether lower doses of linezolid would be effective.
Toxicity also halted the first faltering efforts of a pan-TB regimen. The highly publicized and much touted STAND Trial, sponsored and developed by the Global Alliance for TB Drug Development, tested a universal regimen for TB that consisted of pretomanid, moxifloxacin, and pyrazinamide (PZA). In individuals with pan-susceptible TB, a number of cases of fulminant hepatitis developed and 3 people died, which put a stop to study enrolment.
A universal regimen will also leave us with few options in terms of toxicity management. The current World Health Organization (WHO)-recommended shorter regimen for MDR-TB emphasizes a “package regimen” and should not be changed. In one country where I worked, they are keeping patients on the injectable agent even when they have baseline hearing loss or start to lose their hearing because of the “package” nature of the shorter regimen — they feel they can no longer use the shorter regimen if they stop the injectable. What options would there be for toxicity management in a universal regimen? People are overlooking this discussion in their pursuit of a simplistic solution.
Infectious Disease Advisor: Pursuing a universal regimen can be criticised for being under-ambitious and also for not focusing on individualizing care for every patient with TB. Do you regard the universal regimen approach as a long-term alternative to individualised care or a temporary solution to the current state of MDR-TB treatment?
Prof Robert Wallis: I wouldn’t say under-ambitious; I’d say it was exceedingly practical. I would indeed like to see individualized care delivered to all individuals with TB. It’s quite likely, for example, that care can be appropriately lengthened or intensified according to some measures of the extent of disease, whether that is measured by positron emission tomography (PET) scan, TB DNA in sputum, a simple acid-fast bacilli (AFB) smear, or chest X-ray. I also would like to see adjunctive host-directed therapies added in an individualized manner to prevent permanent lung damage and accelerate infection clearance. Knowing the resistance profiles in individual patients is always a good idea. But none of this precludes the standardized use of at least 3 new drugs in fixed combinations. I’m quite convinced this is the best option for patients now and in the future.
Infectious Disease Advisor: Dr Furin suggests that isoniazid, rifampin, pyrazinamide, and ethambutol (HRZE) is a failed universal regimen and that this is a lesson we should learn from and not repeat. Is this your view of the history of TB chemotherapy and should we see this as a cautionary tale?
Prof Wallis: I also heard Dr Castro make that statement [See Could a New Universal Tuberculosis Regimen Help End the TB Pandemic?], but it clearly indicates a lack of understanding of pan-TB or the universal regimen concept. The critical requirement is that the regimen be comprised of new compounds without pre-existing resistance and with a high barrier to new resistance. Isoniazid (the H in HRZE) was originally introduced in the 1950s. It was used together with para-amino salicylic (PAS) and streptomycin, 2 rather marginal drugs in many respects, for more than 2 decades before being incorporated into a standardized regimen. Isoniazid has a low threshold for new resistance; it readily emerges in vitro in a matter of days. HRZE was never a universal regimen; it was only used as one.
Infectious Disease Advisor: Use of a universal regimen in the setting of poor access to DST or in patients treated empirically would deprive susceptible patients of some of the least toxic and efficacious drugs (eg, rifampin). Is this an inevitable consequence?
Prof Wallis: Any decision to substitute a new regimen under those circumstances would certainly be made with the goal of maximizing benefit and minimizing harm. This in turn would strongly depend on the safety and efficacy of the new regimen compared with HRZE. I’d probably only rate HRZE as fair in that regard. Most other modern medicines are held to a much higher standard. We should set our sights on doing much better.
Infectious Disease Advisor: How rapidly do you think a TB regimen could realistically be deployed? Will this be delayed by further trials to find appropriate dosing?
Prof Robert Wallis: The TB drug pipeline is unusually rich at this moment, including some candidates that could quickly move into mid-sized trials that are much needed in the underserved population of patients with pre-extensively drug-resistant (XDR) or extensively drug-resistant TB (XDR-TB). Sutezolid (or possibly other oxazolidinones) is probably the leading candidate in this regard. Pfizer was able to move sutezolid ahead rather quickly once the pharmacokinetics (PK) were known, using a relatively narrow range of doses based on modelling. The relationship between oxazolidinone dose, efficacy, and safety probably can’t be assessed in short-term trials. I think the best approach is to begin studies of new regimens of 2 to 3 months duration. These could be conducted pretty quickly.
Different views prevail regarding how to optimally capitalize on the new medications and technologies in the TB armamentarium. There is nevertheless the prospect of great improvements in outcome with whatever new strategies come to the fore.
- Kendall EA Brigden G, Lienhardt C Dowdy DW. Would pan-tuberculosis treatment regimens be cost-effective? [published online May 30, 2018]. Lancet Respir Med. 2018;6(7):P486-P488.
- Conradie F, Diacon AH, Everitt D, et al. The Nix-TB trial of pretomanid, bedaquiline and linezolid to treat XDR-TB. Conference on Retroviruses and Opportunistic Infections; Boston, Massachusetts, February 13-16, 2017. 80LB.