About this blog

CRIT is no longer adding posts or updating the CRITical Thinking blog.

CRITical Thinking is a blog written by staff, directors, and friends of the Collaboration for Research Integrity and Transparency (CRIT), a joint program of Yale Law School, Yale School of Public Health, and Yale School of Medicine. CRIT's mission is to promote health by improving the integrity and transparency of biomedical and clinical research.

This blog is published by and reflects the personal views of the individual authors, in their individual capacities. It does not purport to represent Yale University's institutional views, if any. No representation is made about the accuracy of the information, which solely constitutes the authors’ personal views on issues discussed. The information contained in this blog is provided only as general information and personal opinions, and blog topics may be updated after being initially posted.


Regulators Must Act In The Public Interest When Considering Real World Evidence: Examples from TB and HIV Treatments

October 16, 2017

In June 2017, Yale CRIT hosted an international conference titled “Ensuring Safety, Efficacy, and Access to Medical Products in the Age of Global Deregulation.” The following blogpost is the ninth installment of a blog series with commentaries from the conference participants. The views and opinions expressed in this blogpost are those of the authors and do not necessarily reflect the position of Yale CRIT. For more blogposts related to this series, see here or click the tag “YaleCRIT17” below.

In general, fears about the potential harms of weakening evidentiary requirements through developments such as the 21st Century Cures Act in the US and so-called adaptive pathways in the EU, are warranted. These legal and regulatory changes increase the likelihood that the public will be exposed to unsafe and/or ineffective medicines. Unfortunately, the deregulation debate is more complicated than this, as shown by three examples relating to tuberculosis (TB) and HIV medicines, which highlight the need for regulatory approaches that allow for consideration of real world evidence when potential benefits outweigh harms.

HIV and TB in South Africa

South Africa is facing dual HIV and TB epidemics. The country has the highest global HIV prevalence, which has fuelled the explosion of TB incidence – now the leading reported cause of death among adults in the country. At the same time, South Africa is grappling with high levels of drug resistant TB. According to the World Health Organization’s Global Tuberculosis Report, 10,000 people were estimated to have rifampicin resistance (a good indicator of drug resistance) in 2015. These patients include patients with multidrug-resistant (MDR) and extremely drug-resistant (XDR) TB, who must undertake lengthy and difficult treatment regimens offering low rates of survival.

In South Africa, the five-year survival rate for people with XDR-TB is less than 30% with current treatments. In such circumstances, the risk-benefit balance is such that many patients would rationally choose to try new experimental drugs. Three experimental drugs have shown particular promise: bedaquiline, delamanid and pretomanid.

Below we highlight the important roles that real world evidence can play in informing regulatory decision and some pitfalls, including reliance on industry goodwill to supply medicines via compassionate access programmes and differing reliability of surrogate markers across disease groups.

Approval of bedaquiline

In 2012, the US FDA gave accelerated approval to the experimental TB drug bedaquiline on the strength of phase II data. The phase II data, however, had some worrying safety signals – in one key trial there were more deaths in the experimental arm than in the control arm. These concerns led the South African Medicines Control Council (MCC) to request more safety data before the drug’s approval in South Africa. This safety data came from an expanded access programme (a form of pre-approval access). Once enough people had been treated in this expanded access programme without further safety concerns, the MCC approved the drug. In other words, non-randomised data, that could be considered a form of real-world data, played a critical role in reassuring the MCC that bedaquiline was safe enough to be approved.

Compassionate access to pretomanid

One ongoing, non-randomised trial called Nix-TB has shown unprecedented cure rates in people with XDR-TB using a regimen containing bedaquiline, linezolid, and pretomanid – so much so, that it would already be rational for patients to choose the Nix-TB regimen over the standard of care – given that the standard of care has extremely poor outcomes.

A critical problem though is that pretomanid is not yet approved in any countries. Even if granted accelerated approval, wider access to pretomanid (and the Nix-TB regimen) is at least a year away. The only realistic ways in which patients can access the drug in the meantime is in clinical trials or through pre-approval access mechanisms such as compassionate use or expanded access. The problem is that pre-approval access mechanisms are dependent on the good will of drug developers and the required investment to make drugs available at no cost. The status quo will thus prevent some patients from accessing a likely life-saving treatment until such time as the drug is registered and marketed. In a situation like this, it is hard to see how strict adherence to the requirement for RCTs prior to allowing wider availability is in the public interest.

Surrogate markers and non-inferiority trials in the approval of HIV drugs

While there are not very good surrogate markers in TB, HIV has good surrogate markers in CD4 count and viral load counts. HIV also already has a series of highly effective drugs. In the case of HIV, it is hard to see how new drugs can be approved on anything but surrogate markers—given the length of time and sizes of studies that will be required to show differences in mortality. Given the large trial sizes that will be required to show superiority over already very good drugs, a good case can also be made for the use of non-inferiority trials. However, what is true for HIV is not true for other diseases, like TB, where surrogate markers are weaker or where the existing standard of care is less effective.


The above examples illustrate that an overly rigid application of traditional regulatory rules may prevent regulators from acting in the public interest. They suggest that in certain circumstances the use of real-world data may be appropriate and that in certain circumstances the use of surrogate markers and non-inferiority trials to approve new medicines may be appropriate. They also highlight a critical flaw in current regulatory systems whereby pre-approval access is dependent on the good will of the drug-maker and where an overly rigid adherence to regulatory rules could cause harm.

Thus, one important challenge is to formulate laws and regulations that, on the one hand, allow regulators the flexibility to respond to situations such as those outlined above, but that on the other hand, prevent the abuse of such flexibilities. The risk is that a law or regulation that would allow for the use of real-world data in the case of a drug like bedaquiline, may in other cases be abused to allow for the use of real-world data for purposes where it is not warranted.

It could, however, be that no set of rules could fully allow for flexibility in warranted cases and fully prohibit such flexibility in non-warranted cases. Either way, existing rules are clearly sub-optimal and the development of better, more nuanced, rules that account for appropriate use of flexibilities must be a priority.

An argument can be made that the fundamental problem with recent regulatory developments, such as 21st Century Cures and adaptive pathways, is not so much the less stringent rules themselves, but the lack of public trust in the FDA and the European Medicines Agency (EMA) to prevent abuse of these less stringent rules. While the new rules, and the context in which they were developed are clearly of concern, it is certainly imaginable that a suitably independent and responsible regulator could use these rules and flexibilities in a responsible way and in line with the public interest—as was the case with the MCC and bedaquiline.

Much of the concern expressed regarding the recent regulatory changes appears to be predicated upon a belief that the FDA and EMA are no longer suitably independent and responsible regulators and that they have become more sensitive to pressures from industry.

It is thus more critical than ever that the independence and transparency of regulatory authorities be protected and that conflicts of interest be strictly managed. This will require both a reaffirmation of the primary mandate of regulators being to protect the public and ensuring that regulators receive enough predictable public funding to ensure organisational stability and independence.

Marcus Low is a former Head of Policy at the Treatment Action Campaign (TAC), a leading South African HIV and TB activist organization that represents users of the public healthcare system in South Africa and litigates critical issues related to the quality of and access to healthcare. He currently edits the South African public health magazine Spotlight. He is a member of the Fix the Patent Laws campaign steering committee, the Life Prize steering committee, and the co-technical lead of the Global Tuberculosis Community Advisory Board. He is a co-founder of the Centre for Health Innovation and Public Interest.

Catherine Tomlinson is an independent advocate for affordable and equitable medicine access. Catherine is based in South Africa where she has undertaken work to improve medicine access for the Treatment Action Campaign, Medicins Sans Frontieres and the Cancer Alliance. She is a co-founder of the Centre for Health Innovation and Public Interest.