Making Sense of Medical Science (MSMS)

A couple of weeks ago, headlines screamed that the WHO was stopping a large clinical trial testing the efficacy and safety of treating hospitalized COVID-19 patients with the anti-malaria drug, hydroxychloroquine (HCL). The WHO did this in response to a study published in the May 22 edition of the journal, Lancet, that reported that the drug provided no therapeutic benefit while increasing the risk of heart problems and death in treated patients. The flurry of headlines that resulted was misleading as reported in these pages on May 25. The WHO did not cancel its study; rather they made a mid-study pause so that a team of independent scientists could review the interim data that had been collected to see if it also showed increased risk to study patients as reported in the Lancet study. That interim data analysis is complete and the WHO just announced that the study was re-started. That means that the partial data they have collected did not reveal concerns about study subject safety.

The Lancet study was not a “gold-standard” blinded and controlled clinical trial, but a prospective “chart review” of medical records of 96,000 COVID-19 patients hospitalized in 671 hospitals across six continents. The study claimed to find a 30% increase in mortality in patients treated with the drug and based on this, the World Health Organization paused its global clinical trial of HCL, and many countries banned the drug as a COVID-19 treatment.

Since the Lancet study was published, more than 100 scientists around the world raised serious questions about the reported data and about Surgisphere, the organization that provided the data. They spotted glaring data errors. For example: Obesity and smoking rates in the study patients surprisingly were the same across the six continents. Well established population health data indicates that this is not true. Also, in a letter to Lancet’s editors last week, 120 scientists criticized the study’s sloppiness and aggregation of data from patients who were different in many respects including in the dose of hydroxychloroquine they received, and in the severity of their illness. The complaining scientists also pointed out that in the study Lancet failed to share all patient data. Surgisphere’s CEO, Sapan Desai, one of the study’s authors, claimed his hospital contracts did not allow these data to be shared. This raised more red flags since the reliability and accuracy of the chart data could not be independently confirmed, which is a serious breach of scientific standards.

It now looks like the study may have been based on questionable data from a dubious source. In vitro studies have indicated that hydroxychloroquine can block the virus’s replication, and the drug has been safely used to prevent and treat malaria in millions of people worldwide for 60 years. It is also used to safely treat rheumatoid arthritis and lupus. The Food and Drug Administration has approved it for emergency use for COVID-19 patients. Therefore, the study’s conclusions run counter to decades of experience with the drug.

Wednesday, June 3, Lancet published an “Expression of Concern” about the study and said it would undergo “an independent data audit.” That’s good, but the study's publication may have already done public-health harm. Study doctors are complaining that the negative press has made it difficult to recruit patients in the U.S. for ongoing clinical trials to study the drug’s effectiveness as a prophylactic or for early-stage treatment against Covid-19.

As a result of this brouhaha, the authors of the study recently retracted it. They decided to issue the retraction after Surgisphere Corp. refused to share the full, detailed data set as part of a review after outside researchers raised concerns.  “Based on this development, we can no longer vouch for the veracity of the primary data sources,” said the authors, Mandeep Mehra, Frank Ruschitzka and Amit Patel.

So, the WHO and other studies, including several in the US, such as at New York University, the University of Washington, and elsewhere, continue to study whether hydroxychloroquine can treat people exposed to the virus.

Can HCL treat people with COVID-19? The jury remains out despite this interruption. We will see.

Meanwhile, a different gold-standard, double-blind randomized clinical trial involving 821 subjects was also published last Wednesday in the New England Journal of Medicine and it concluded that the drug is ineffective at preventing infection. Study subjects took the drug or a placebo for 5 days and were followed for 2 more weeks. About 12% of those taking hydroxychloroquine came down with COVID-19 compared to 14% of the placebo group. The difference was not significant. Also, 40% of the subjects taking hydroxychloroquine reported mild side effects while only 17% of the placebo controls did. Nausea, upset stomach and diarrhea were the most common side effects, but none were considered serious.

The bottom line is that the news on HCL and COVID-19 is mixed and confusing. The research continues.

We will see.