The case for conducting first-in-human (phase 0 and phase 1) clinical trials in low and middle income countries

Despite the sustained proliferation of clinical trials in low and middle income countries (LMICs), [1] there has been little examination of whether First in Human (FIH; phase 0 and phase 1) clinical trials should be conducted in LMICs, and if so, under what conditions. Most of the discussion has focused on later phases of clinical trials. The purpose of this paper is to stimulate debate on the merits of FIH trials in LMICs: the default option should be to do the trials where the relevant health conditions present the greatest public health challenge, if, and only if, the necessary capacities exist for participant safety and scientific rigor. We argue, further, that the default presumptions against performing FIH trials in LMICs may be quietly impeding progress in the evolution of the very capacities that these trials require.

Interest in FIH clinical trials has grown recently, in part because of the disastrous experience of the phase I clinical trial of the superagonistic anti-CD28 antibody TGN1412 in the United Kingdom. The drug was developed with the intention to stimulate a specific kind of T-cell while at the same time controlling the production of other T-cells in order to suppress the immune system. In this trial, six healthy volunteers developed a cytokine release syndrome with multi-organ failure and required intensive care [2, 3]. There were mistakes made in that case, and some of the key lessons have been translated into more stringent rules and new guidelines [3‐5].

FIH trials are studies where an investigational medical product (drug, vaccine, medical device, etc), previously developed and assessed through in vitro or animal testing, or through mathematical modeling, is tested on human subjects for the first time [4, 5]. In drug development, such trials involve administering single low, sub-therapeutic doses to a small number of healthy volunteers (10 to 15) to gather preliminary data on the product's pharmacokinetics and pharmacodynamics. These trials help researchers identify the drug candidate with the best pharmacological parameters to take forward for further development and which one to leave out. Traditionally, these early phase trials have been conducted in high income countries (HICs). Until recently they were very rarely, if ever, performed in LMICs even for conditions that are most prevalent in those countries. Why haven't FIH trials been more common in LMICs?

The main historical reasons have been the lack of research and clinical infrastructure, inadequate institutional capacity, and weak regulatory agencies in many LMICs. This has made it difficult for product developers, usually large multinational pharmaceutical companies, and regulatory agencies in developed countries to achieve the necessary standards of safety for trial participants, and clinical and scientific rigor in LMICs [6]. LMICs themselves have often relied on product registration processes by regulatory authorities in developed countries [6]. At the same time, regulatory authorities in developed countries often lack knowledge of the diseases, including many of the so-called Neglected Tropical Diseases (NTDs), and the local conditions most prevalent in LMICs, making it possible that they will make judgments based on inappropriate risk-benefit assessments [7], potentially to the detriment of populations in LMICs. Another reason is that Western companies, which currently develop most new and innovative products, have considered it risky to conduct FIH trials in LMICs for various reasons, including the inability to discriminate adverse events (AEs) caused by the investigational product from the generally much higher frequency of all types of undiagnosed symptoms and untreated morbidities, the cultural obstacles to undertaking autopsies and the fear that they may be perceived to be exploiting vulnerable persons in these countries [8]. The fact that so few companies do FIH trials in low income countries leads to a situation where others are reluctant to start, and this makes more trial sponsors also reluctant to start. The net result is a general presumption against FIH trials in LMICs.

This situation is beginning to change. This shift presents an opportunity to engage in a serious debate about the basic logic of site selection for FIH trials. Here, we are focused on FIH trials for diseases that present important public health challenges for LMICs, including the rapid emergence of a wide range of chronic diseases [9]. In particular, we agree with the ethical argument that trials should be responsive to the health needs of host communities in the sense argued by London and Kimmelman, that is, if they are "part of a program of inquiry that will expand the capacity of health-related social structures in the host community to meet urgent health needs."[10]. For example, it makes little sense to perform FIH trials of malaria vaccines in the United States today, as has happened recently with the live attenuated sporozoite vaccine developed by Sanaria [11], when malaria transmission stopped in the US after World War II [12], but it continues to kill about 1 million children a year in sub-Saharan Africa [13].

There are also strong scientific and pragmatic arguments. The diseases in question are most prevalent in LMICs, where the epidemiology, health services, social determinants of health, compliance patterns, co-morbidities and the genetic make-up of the population also have a bearing on the way in which the health products will ultimately be used and hopefully achieve the desired outcomes [14]. There is evidence that, mutations in genes coding for drug-metabolizing enzymes may result in drugs being metabolized differently in different populations [15], affecting the response and AEs to drugs and vaccines [16]. The gastrointestinal microbiome may also make a difference. Individuals with gastrointestinal infections have been found to respond differently to medical products, for example the live polio vaccine [17]. Thus, testing drugs and vaccines in developed countries, (whether in early or late phase trials), may give misleading results.

An important consideration also relates to the need to conduct both early and late phase clinical trials among subjects who have the target disease. The ethical issues here may need further reflection, but scientifically, there are circumstances and arguments where it would be beneficial to conduct such studies among the patient population. For example, if there's need to evaluate a new product that cannot be assessed in healthy individuals for some reason, including ethical considerations, then it may be necessary to recruit patients with the target diseases. In other cases, it may be difficult to extrapolate the results from healthy individuals to a patient population. In such instances, the results pertaining to a new product maybe either over- or under-estimated [18, 19]. In these examples, it can be extrapolated to product development for conditions that are most prevalent in low income countries, and argued that it would be reasonable for early stage trials to be conducted in these countries where the relevant diseases are most prevalent and it is possible to recruit patients.

The need to reconsider where FIH trials are done first is becoming even more salient as LMICs push to expand their involvement in all phases of drug development, primarily as a source of revenue. More important, though, is the fact that the pipeline of products for conditions that occur predominantly in LMICs is improving significantly [20]. Major philanthropies such as the Bill & Melinda Gates Foundation fund discovery research directly through programs such as the Grand Challenges in Global Health initiative [21], or indirectly through many product-developing Public-Private Partnerships (PPPs) [22], Similarly, large multinational pharmaceutical companies have begun, on their own or through PPPs, to develop more health products for NTDs and for diseases most prevalent in LMICs. Countries such as India [23], China [24, 25], and Brazil [26], now have their own strong and growing pharmaceutical and vaccine manufacturing industries- indeed half of childhood vaccines administered throughout the world by UNICEF are made by one company in India [23].

An important underlying concern in this discussion is, of course, the safety of human research volunteers and the potential for exploitation of vulnerable populations in LMICs (or indeed anywhere else in the world). By definition, the risks in FIH trials, particularly for phase 0, are not known for sure: they could be non-existent, low or, as in the TGN1412 case, high [2, 3, 22, 27, 28]. In addition to posing risks to the participants, FIH trials that do not completely take into account the design of, and data from, preclinical trials may not provide the necessary information to adequately evaluate the results of FIH trials [5]. TGN1412 and other FIH trials in HIC have shown us clearly that risks are inherent in these trials and not unique to LMICs, and so we must be careful not to impose more stringent standards on LMICs (especially those which meet the necessary GCP conditions) than we do on HIC.