What have we learned from clinical research against Ebola?

Last October, when I wrote an editorial for Focus Pharmacovigilance (Focus November-December 2014, page 1) about "Ebola exceptions" concerning the rules for assessing the safety and efficacy of drugs, I could never have imagined that within a few days I would find myself personally involved in the matter, and so deeply that it has almost completely absorbed me for the last six months. And I would even less imagined becoming a direct "victim" of the paradox I envisioned in the article: "The testing of potentially promising remedies is only possible in the course of an epidemic, but the difficulties during an Ebola epidemic are such as to render a randomized and controlled trial impossible".
Re-reading those words now, they seem all too prophetic, given the huge confusion that was to be unleashed in the months that followed.
In November, I found myself leaving for Sierra Leone, to help Emergency to design and establish a clinical trial of amiodarone, an anti-arrhythmia medication that had shown a strong activity against Ebola virus in human cells grown in vitro.
Now that the outbreak is abating - and it is now clear that the study cannot be undertaken, because fortunately there are not enough sufficiently ill people anymore - it’s worth taking stock of how things went, and the lessons that can be learned from it, and capitalising on first hand experiences on the ground.
Amiodarone is an old friend: we were using in coronary units in the 1970s, and it continues to be used by millions of people worldwide to this day. Such longevity is rare: after the patent has expired, even the most successful active drugs are quickly replaced by others, which are often very similar, but offer some slight advantage, real or alleged, over the previous one. Like acetylsalicylic acid, metformin and several others, amiodarone is a veteran of the therapeutic armamentarium, whose safety profile is known in detail and is very reassuring, particularly for short-term therapies.
The long time they have been in use also affords us a deeper understanding of their mechanisms of action, and time to think of other potential uses. German and Italian researchers, including the group led by Aldo Baritussio (University of Padova), have exploited the properties of the drug for a long time to block cellular endocytosis, to create a phenotype similar to that of the rare genetic disorder, Niemann-Pick Disease. From there, the idea of testing amiodarone for blocking the Ebola virus which, like other filoviruses, exploits endocytosis like a Trojan horse, for entering cells and replication.
The first in vitro results^1,2 were very encouraging, and, on the basis of WHO recommendations about the ethics of using "unregistered interventions that have shown promise in the laboratory or in animal models, but that have not yet been evaluated for safety and effectiveness", it sounded like testing amiodarone would be a priority too, as when compared to the newer competing drugs - lie the now famous zmapp, a cocktail of humanised monoclonal antibodies - it has the advantage of having been assessed for safety in humans already.
Within a few days, an informal group for the study of amiodarone against Ebola was established. This included Emergency, which runs a treatment centre in Sierra Leone, the Spallanzani Institute for virological expertise, and the IRCCS of Reggio Emilia, for the methodology and data analysis. A formal protocol was written, and subjected to the scrutiny of the ethics committees of two research institutions, which also endorsed the decision to require a randomized control group. It was a difficult decision, because it aggravated the very feasibility of the trial in practice, but in the absence of a proper comparison, the results would be inconclusive.
At this point the trial was registered at the American registry clinicaltrials.gov³ as well as at the Pan-African registry, and approval from the Ethics Committee of Sierra Leone was requested. By mid-November then, when the epidemic was raging with a mortality rate of more than 60%, everything was ready to launch a study that had mortality as the primary endpoint, consisting of the controlled infusion of the drug, the requisite ECG monitoring, laboratory tests and the gathering and transmission of data.
Sadly, things did not go as everyone had hoped.
In its documentation, the WHO Ethics Committee had failed to investigate the benefits of "repositioning" the older drugs in the context of the epidemic.
Perhaps partly as a result of that omission, even the pharmacological working group of the same WHO which had defined the priorities for clinical research against Ebola, spent several months expressing incomprehensible objections about the safety of amiodarone under the conditions of use provided for in the protocol. The same panel submitted erroneous assessments (for example, confusing the dose/response data for amiodarone with that for an antimalarial drug with a similar name: amodiaquine; or disregarding the pharmacokinetic characteristics of a highly lipophilic molecule) to the local Ethics Committee, which therefore continued to postpone study approval until all the misunderstandings and assessment errors had been clarified, at the end of April 2015. By which time, it was too late to launch the study, since in the meantime, fortunately, new cases had become sparse enough to rend it unfeasible.
The clinical trial of amiodarone had been the first to be registered (and perhaps because of this, or because it had no sponsor, it was the subject of such unfounded prejudice?), but it was not the only trial in West Africa that was aborted: of 11 clinical trials registered which should have been assessed as potential treatments for fighting Ebola, few have been launched, and its likely that no-one will be able to produce transferable practical data for possible future outbreaks.⁴
Last autumn, substantial funding was earmarked for research into vaccines and curative interventions against Ebola, thanks mainly to private foundations like the American Bill Gates Foundation and the British Wellcome Trust, but also to public research institutions in various countries: a significant effort which, although late, would have deserved to produce a few concrete results. This failure should be recorded as one of the lessons learned from the epidemic too, along with the others that had been correctly identified and admitted in a courageous, self-critical statement from the WHO.⁵ The main mistake in almost all of the studies initiated was to have abandoned a randomized comparison, by deeming it unethical, when, under conditions of real uncertainty, it was genuinely the exact opposite.
Trials without a comparison generate data from which it is impossible to draw any conclusions, and are therefore a waste of time from any point of view, even from a moral perspective. One of the objections raised against the randomized trials was the impossibility of obtaining truly informed consent. This difficulty is real, even in less dramatic and better-developed settings, but not just for randomized trials.
Any intervention, experimental and, individually or collectively, during a tragic event like the Ebola outbreak, would require an extraordinary amount of informative work and collaborative effort.
Even in these circumstances, discussing, listening, communications, mobilization of people and communities are the key not only to practical success, but to the more general duty to be human as well.