Dominique Costagliola is Deputy Director of the Pierre-Louis Institute of Epidemiology and Public Health (Sorbonne University, Inserm). Member of the Academy of Sciences, she participates in the scientific council of the consortium REACTing. She takes stock of the therapies against the Covid-19 still in the race.
The results of studies that have been carried out in Europe and around the world are starting to be published. In the UK, the Recovery trial showed a positive effect of low-dose corticosteroid therapy with dexamethasone, especially in people who were in intensive care. This type of treatment is known to be effective: it was already given to patients who developed acute respiratory distress syndrome (ARDS) in other settings, before infection with the coronavirus SARS-CoV-2. It is not a question of giving it to all patients, but to those who are already in an advanced stage: this has a significant effect, since mortality is reduced after administration.
A few weeks ago, we had talked a lot about Remdesivir, a molecule that has been proven against various viruses (Ebola, Lassa, etc.), because it disrupts the replication of their RNA. What is it?
Remdesivir has been the subject of three randomized studies. A Chinese study was the first published. It aimed to evaluate the use of Remdesivir administered over 10 days. The results were negative: no beneficial effect could be demonstrated. But the inclusions of participants had to be stopped for lack of patients, once the epidemic was under control in China.
Second, the Gilead pharmaceutical company conducted a trial comparing the use of Remdesivir for 5 days with its use for 10 days. This one concluded that there was no significant difference in effect between the two durations of use, without however that one cannot affirm that 5 days is not less than 10 days of treatment.
Finally, a trial conducted by the NIH was stopped early, because the use of Remdesivir made it possible to shorten the length of stay in hospital (by 4 days on the median). The length of hospitalization was reduced from 14 days to 11 days. On the other hand, no effect on mortality could be demonstrated in this study. Not all patients had reached one month of follow-up, which is the length of follow-up used in most inpatient studies. However, this early stopping after a short follow-up for a large proportion of the participants can induce an overestimation of the effect and does not allow to conclude to a benefit on mortality. Despite these limitations, a recommendation for use in hospitalized patients has been issued in the USA and the product has just been granted conditional marketing authorization in Europe, with a request for additional data relating to mortality.
Many clinicians are not convinced by the outcome used, the shortening of the hospital stay. In addition, early stopping did not make it possible to know in which patients this treatment would be most useful. It is all the more annoying that the number of available doses is limited and that the USA have preempted them. How do you know who to treat first?
The evaluation of Remdesivir is continuing to assess its possible impact on mortality in the international trial of the WHO Solidarity and its associated trial Discovery, conducted by Inserm in several European countries (France, Austria, Belgium, Luxembourg for the moment). The question of the interest of Remdesivir remains unanswered, especially since this drug is expensive and it will not be a ” game changer If it does not improve survival.
What are the avenues that have been abandoned by the main ongoing clinical trials?
Of all the molecules that have been tested, hydroxychloroquine clearly does not work for hospitalized people. Lopinavir, an antiretroviral primarily used in the treatment of HIV infection, was also dropped by both the UK Recovery trial and the Solidarity trial. Data collected from Discovery, which is the only trial that tracks safety issues and adverse events in detail, also showed that Lopinavir use was problematic for kidney complications, especially in patients with people with severe forms. This effect was described in the summary of characteristics of Lopinavir, but in HIV infection it is not common.
This point illustrates that the adverse events of a drug can be different depending on the pathology in which it is used, because the pathology itself can have a specific impact on certain organs. However, the SARS-CoV-2 infection is suspected of having consequences on many organs: the lungs of course, but also the heart, the kidney, potentially the nervous system …
And which therapies are still being evaluated?
The Solidarity trial is still testing azithromycin, an antibiotic from the azalide class (macrolide family) alone. This arm has not been stopped, we are still awaiting the results.
Among the interesting leads for which we do not yet have a formal answer is the plasma of convalescent people. Studies of this type of therapy are more complex than when testing a drug whose manufacturing process is well known. This is because plasma is collected from people who have had the disease, but not everyone selects donors in the same way. In France, for example, in the CORIPLASM test, the presence of neutralizing antibodies is checked. This is not necessarily the case in all the studies that are conducted and the timing of administration may also play a role. This should therefore be taken into account when analyzing the results.
So far, some studies have suggested positive results in patients treated with convalescent plasma, but they were not randomized or comparative. Numerous trials around the world are evaluating this solution (including the Recovery trial in the UK). Positive results could pave the way for testing combinations ofmonoclonal antibodies making it possible to envisage a manufacturing process that is simpler and more generalizable than convalescent plasma.
Another avenue explored is to tackle no longer the virus, but the inflammatory storm it triggers: this is the avenue of immunomodulators in general, and anti-interleukins in particular (Editor’s note: interleukins are chemical messengers involved in the immune response and associated inflammation). These medicines include, for example, Tocilizumab, an antibody that blocks the interleukin-6 receptor, used to treat rheumatoid arthritis. An AP-HP team communicated on the subject at the end of April, the results are currently being submitted for publication. Tocilizumab is also being tested in the Recovery trial.
On the basis of recent publications or announcements, injectable or nebulized interferon is also a topical issue (note: interferons are proteins involved in the immune response, particularly antiviral).
Many other hypotheses have also been formulated and are currently being evaluated, without knowing whether the number of participants included in these studies will be sufficient to draw conclusions.
Does this mean that too many clinical trials have been carried out?
The question is complex, and on this point France and the United Kingdom have had two very different approaches. The British have been very drastic. They basically did a really big, simple public trial, Recovery, that they included about 12,000 people. This has already produced results, negative for hydroxychloroquine and Lopinavir, positive for low-dose corticosteroid therapy. More are to come.
In France, on the contrary, there has been an outbreak of studies, without initial coordination. In addition, the Franco-French enthusiasm for hydroxychloroquine has led to a proliferation of trials related to this molecule. This is mainly due to the way the system is organized. The more research university hospitals carry out, the more endowments they receive. However, they are the sponsors of the public trials, precisely. And as long as a team has a sponsor, neither the National Agency for Medicines and Health Products Safety (ANSM) nor the Patient Protection Committee can oppose the start of its trial on the pretext that it There would already be enough (provided that the said test follows the rules relating to the protection of participants, of course).
What is more, the research is organized in such a way as to value the principal investigator of the work. This system encourages plethora and little collaboration, less valued. Over 80 proposals have been submitted to REACTing. The ideal would be to find a happy medium between a single test, as in the United Kingdom, and these 80 proposals… We could have limited ourselves to a few tests in town and in hospitals, for example. But that would have required that people work with each other, and that an organization could be legitimate to prioritize the proposals.
And that would have given rise to another difficulty: when you receive so many proposals, even if some have only a modest rational, how to be sure to give priority to the good ones? It is not an easy exercise in this context.
Finally, we can also note that there has been little cooperation at European level. All of this indicates that lessons will have to be learned from this crisis, both in terms of the organization of clinical trials and national and international cooperation in the context of an emerging infectious disease. And in particular in view of the continuation of the epidemic …
Dominique Costagliola, Epidemiologist and biostatistician, deputy director of the Pierre Louis Institute of Epidemiology and Public Health (Sorbonne University / Inserm), research director, Inserm
This article is republished from The Conversation editorial partner of UP ‘Magazine. Read thearticle original.
–
