Researchers at the University of Minnesota Genomics Center set up an automated liquid handler as researchers begin a trial to see whether hydroxychloroquine can prevent or reduce the severity of COVID-19. Photo: Reuters/Craig Lassig.
Note: On May 4, 2020, the Indian Council of Medical Research registered a hydroxychloroquine trial for its use as a prophylactic drug on the Clinical Trial Registry of India. This followed extensive reporting on the issue by Priyanka Pulla for The Wire Science, including the following article.
In a recent video published by ANI, the chief epidemiologist of the Indian Council of Medical Research (ICMR), Raman Gangakhedkar, made a couple of odd statements about hydroxychloroquine. He said that because there was so little evidence of the drug’s efficacy against COVID-19, ICMR was recommending it only as a prophylactic to doctors and contacts of lab-confirmed patients. He also repeatedly referred to a study, as part of which doctors and contacts were taking the drug, but didn’t provide any other details.
These statements are odd because, contrary to Gangakhedkar’s claim, the Government of India has suggested the use of hydroxychloroquine as treatment for COVID-19 patients in intensive care units (ICUs). In its ‘Revised Guidelines on Clinical Management of COVID-19‘, dated March 31, the health ministry says a combination of hydroxychloroquine and azithromycin, an antibiotic, may be administered as “an off-label indication in patients with severe disease and requiring ICU management”.
Second: Although Gangakhedkar’s statement about the lack of data on hydroxychloroquine’s treatment efficacy is correct, there is even less data on the efficacy of hydroxychloroquine as a prophylactic. No completed study in humans thus far has attempted to answer this question, although several are ongoing (see here, here and here). The only human clinical trials whose results are available studied the efficacy of treatment alone, and yielded conflicting results.
Not every drug used for treatment translates to a good prophylactic. Even hydroxychloroquine may impair a healthy person’s ability to fight off COVID-19 by slowing the production of an important immune system molecule called interferon alpha. While this would render the drug a poor prophylactic, it may actually help many sick people who suffer from an over-aggressive immune response (called a cytokine storm) to the virus that triggers COVID-19. The bottom-line: prophylaxis and treatment are not the same thing.
So why would ICMR recommend a drug with unknown efficacy as prophylaxis if the ICMR doesn’t consider it to be good enough for treatment? The rationale is unclear because prophylactics are targeted at healthy people, whose risk of dying from COVID-19 is far lower than that of already sick patients. So if ICMR thinks the drug is not effective enough for sick patients, why would a group of people at lower risk of death from COVID-19 be prioritised over those who are ill?
As I reported earlier, the agency’s guidelines for COVID-19 prophylaxis have led to hundreds of healthy Indians popping the drug. Some states are giving the drugs en masse to doctors as well as nurses and lab technicians. Unlike doctors, these other healthcare workers have no experience prescribing hydroxychloroquine either for malaria, lupus or rheumatoid arthritis. So they may be unaware of the drug’s side-effects, and the risk they personally carry of experiencing these side-effects. Even though ICMR said in its advisory that the drug should be taken by healthcare workers only on the prescription of registered medical practitioners, there is ample evidence that this is not happening in the country.
Study? What study?
There is another peculiar aspect of the ICMR chief’s statements. In the ANI video, Gangakhedkar refers to a “study” to test if hydroxychloroquine works among doctors. He had said something similar at a health ministry press conference on April 1: that once data from a “demonstration study” of the efficacy of hydroxychloroquine in healthcare workers was available, ICMR would decide if the drug could be given as prophylaxis to the general public, instead of being restricted to healthcare workers and contacts of patients.
This statement gives rise to several questions. If ICMR wants to assess the efficacy of hydroxychloroquine as prophylaxis, it would ideally conduct a randomised, controlled and blinded trial. ‘Randomised’ and ‘controlled’ mean that, for the trial, healthcare workers chosen at random would be given the drug, and compared with others (‘controls’), who will receive a placebo. Both arms of this trial will have to practice all currently accepted preventive measures against COVID-19, including the use of personal protective equipment (PPE: masks, gloves, goggles, etc.). ‘Blinded’ means participants of neither arm will be aware of whether they are receiving the drug or a placebo. This ensures that the trial’s results are not sullied by the placebo effect: the biochemical improvement in one’s health that results from the psychological effects of taking a drug.
If such a randomised clinical trial was difficult to do, ICMR could still have taken up the alternative, to determine if the drug was working through an observational study. This would also involve comparing the rates at which healthcare workers on the drug get the disease to those who don’t. However, this study would neither be blinded or randomised, rendering its results less reliable – but still better than nothing.
But there is no evidence that ICMR is conducting either a randomised trial or an observational study among the many healthcare workers consuming the drug in India. The Clinical Trial Registry of India, on which the drug controller general of India requires all trials to be registered prospectively (i.e. before they’re conducted), shows no Indian studies of hydroxychloroquine as a COVID-19 prophylaxis. So what is the study Gangakhedkar has referred to more than once?
Clinical trials become tough
In fact, ICMR’s advice to healthcare workers to take hydroxychloroquine could have made it harder for other researchers to conduct such studies. For example, scientists from the George Institute for Global Health, India (headquartered in New Delhi) are planning a trial but say they could have a hard time finding enough participants. Vivekanand Jha, executive director of the institute and principal investigator of the study, told The Wire Science that the aim was to evaluate if hydroxychloroquine was more effective than PPE in preventing COVID-19 among healthcare workers. The idea is to recruit between 5,000 and 7,000 people, some of whom will not get the drug.
But Jha said this could prove tough. Once Indian healthcare workers start believing that the drug works, they may refuse to participate in a trial where they do not get it. “For any clinical trial, there should be clinical equipoise – genuine uncertainty in the minds of the community about whether the drug works. But if a weighty body like ICMR – this is not just some random group of people – comes out with a recommendation in favour of hydroxychloroquine, the equipoise may no longer be deemed to exist,” Jha told The Wire Science.
This has already happened in one case. The Aster Malabar Institute of Medical Sciences, Kozhikode, recently launched a randomised clinical trial to determine the efficacy of hydroxychloroquine as prophylaxis. The researchers in charge of it are comparing ICMR’s recommended dose for the drug with an alternative dose of the same drug. However, the trial has no placebo arm.
The reason for this, according to Remesh Bhasi, the trial’s principal investigator and a rheumatologist, is that Aster Malabar’s ethics committee 1All clinical trials conducted in India are legally required to be approved by such ethics committees2 did not allow a placebo arm because ICMR had already recommended hydroxychloroquine to high-risk healthcare workers. Therefore, not giving the drug to a healthcare worker in a clinical trial could be considered unethical because no clinical trial can deny a control arm the best treatment or prophylaxis currently available.
Asked if it was a problem to be forced to consider hydroxychloroquine as the best available prophylaxis today – as a result of ICMR’s recommendation – Bhasi said it was.
Tracking side effects
While it is unclear if and how ICMR is performing the study it claims to, a bigger worry is that no-one seems to be tracking the adverse effects of the doses the ICMR itself has recommended. The prophylactic regimen the agency suggests includes two doses of 400 mg of hydroxychloroquine on the first day – a higher dose than rheumatologists typically prescribe for people with rheumatoid arthritis. Indeed, such a high dose has never been taken by such a large and healthy population, which means unprecedented side-effects could show up.
However, after a doctor recently died in Assam soon after starting on the ICMR-recommended hydroxychloroquine regimen, the hospital where he worked – Guwahati’s Pratiksha Hospital – did not report the death to the Pharmacovigilance Programme of India (PvPI), the Indian agency that tracks drug safety. This, despite the fact that ICMR’s advisory asks people taking the drug to report adverse effects.
The medical superintendent of Pratiksha Hospital told The Wire Science that he wasn’t aware of the need to report adverse effects. When I asked officials at PvPI if they were investigating the cause of the doctor’s death, I didn’t hear back, nor did ICMR respond to questions about whether it was looking into the issue.
Given all this, it’s hard to believe the agency’s claims about the rationale of its hydroxychloroquine policy. The decision to suggest the drug as a preventive hasn’t been without its downsides: already patients of lupus and rheumatoid arthritis, who need to take the drug for years, are unable to find hydroxychloroquine sulphate tablets in pharmacies because demand from an unexpected quarter – healthcare workers taking it as a COVID-19 prophylactic – has soared. Those with plans to perform clinical trials may also experience shortages since they will need large quantities of the drug. The least ICMR can do is explain its unscientific stance. And if it is unable to do so, ICMR should consider reversing it.
Note: 1. A previous version of this article said that according to Gangakhedkar, ICMR had recommended hydroxychloroquine as a prophylactic to doctors and lab-confirmed contacts of patients. The last portion was edited to “contacts of lab-confirmed patients” on April 15, 2020, at 5:15 pm. 2. This article was updated at 12:25 pm on April 18, 2020, to include the Aster Malabar clinical trial.
Priyanka Pulla is a science writer.
The reporting for this story was funded by a public health journalism grant to Priyanka Pulla from The Thakur Family Foundation.