The worldwide growth in coronavirus infections, followed by the WHO declaration of a global pandemic, led patients and government officials everywhere, concerned about serious and possibly fatal outcomes, to cry out for effective treatments. An assessment of the way in which the research evaluating the roles of chloroquine (CQ) and hydroxychloroquine (HCQ) in treating this disorder evolved might provide us with an opportunity to learn from our mistakes and approach future crises with more scientific vigor.

The Clinical & Scientific Community Response

Early in the pandemic a number of small retrospective or randomized studies were performed and fast-tracked to publication. These showed benefit in patients with varying degrees of disease severity using surrogate endpoints. Over the ensuing months HCQ/CQ prescriptions in the US increased 214% when compared to the same period in 2019. The role of media reports of this ‘science’ and President Trump’s support of HCQ usage in growing this pharmacologic market is uncertain but concerning.

Based upon Newton’s third law of motion (For every action, there is an equal and opposite reaction.), what happened next is not at all surprising with the publication of studies showing no benefit or possible harm. Concerns raised in these studies were amplified by the failure of the original ‘positive’ studies to analyze the potentially serious and life-threatening arrhythmic problems associated with QTc prolongation, which can occur with these drugs especially when combined with other drugs that can also cause QTc prolongation.

However, mathematical models guarantee that the pendulum will continue to swing as demonstrated by the recent retrospective, observational Henry Ford study, showing lower mortality rates among hospitalized COVID-19 patients treated with HCQ or HCQ plus azathioprine when compared to patients not taking either medication.

In an attempt to minimize this confusion, the NIH COVID-19 Treatment Guidelines Panel has recommended against the use of HCQ or CQ in the treatment of COVID-19 except in a clinical trial. Is this too little, too late?

Tough Questions

As of July 04 of this year there are 22,287 COVID publications listed on PubMed and 2447 COVID trials on Emergency supplemental NIH appropriations related to the pandemic have resulted in an additional $3.59 billion appropriation-an 8.6% increase over the approved budget.

While the desire to assist sick or at risk patients is an understandable and noble goal, when clinical decisions are not based upon good scientific data obtained in well-designed prospective, randomized trials, the risk of creating more harm than benefit becomes a real concern. As scientists and clinical leaders we must do better.

Are there really over 20,00 COVID manuscripts worthy of publication?

Have too many authors rushed to publish papers? In the best-case scenario many of the papers seem irrelevant. In the worst-case scenario, might the inaccuracies cause more harm than good?

Have editors, in their wish to help inform clinicians, lowered their standards and caused confusion or harm?

Do we really need 2447 COVID trials or is this an excessive approach that wastes money and may create additional downstream confusion?

Could the extra governmental funding be used in better ways?

Why have so many doctors, health experts, and government officials, as well as journalists and the general public, relied on inaccurate information?

How can we address these questions and establish a way forward in the current crisis and in the next global medical crisis? The first step in creating a better approach requires that we acknowledge that most of the available data is severely flawed, predominantly consisting of small studies limited by poor trial design, selection biases, different patient populations, diverse treatment approaches and an inability to adjust effectively for the multiple confounding variables.

To rectify these problems, I suggest the following, which while focused on the present COVID-19 pandemic situation, could have applicability for considering how we should address future medical crises.

The relevant societies should create position papers outlying the issues to be considered when coordinating prospective trials. Clinical trialists coordinating prospective trials can then consider these issues as they design trials.

Societal registries such as the AHA COVID-19 Registry can help by gathering preliminary data; identifying trends and suggesting research approaches. However this data should NOT be considered as definitive and should not drive policy decisions. Instead, these findings should be viewed as hypothesis generating and inform trialists as they define trial methodology.

Governmental agencies should fund studies; establish programs and generate policies that have the greatest likelihood of providing important, clinically relevant information expeditiously, which will positively impact outcomes.

Responsible physicians and scientists must lead the charge and work together with patients, societies, healthcare providers and payers to generate studies that will truly advance care and improve outcomes.


1 Comment

  1. Did they supplement with zinc? I mean, the chloroquine or the hydroxy may do nothing but allow:

    To see what happens in live humans:

    Lastly, love the study on hospitalized, not that anyone ever considered that even if the chloroquine or hydroxy were alone enough, that maybe it was too late by time of administration. I say that because the paper otherwise shows a rather glaring flaw in the study:

    Hydroxychloroquine-treated patients were more severely ill at baseline than those who did not receive hydroxychloroquine…

    Like saying that antihypertensive drug A was no better than B and never mind that the A group was SP 170 at baseline while the B group was SP 147. And no mention of zinc. Which reminds me of the PROGRESS trial, with its placebo arm, perindopril arm, and perindopril and indapamide arm, with the findings that perindopril alone had near zero effect whereas the combo arm had a statistically significant reduction of recurrent stroke, i.e., is the perindopril useless or does it help big up the indapamide? Here, you need the chloroquine or hydroxy and zinc arm, since without zinc you failed to prove that chloroquine and hydroxy serve the useful purpose of being zinc ionophores and that intracellular zinc can stop viral replication. And see the first paragraph under Rationale here:

    To save you the time, the clinical trials list:

    And see no. 2, since another zinc ionophore, quercitin.

    Now well and truly lastly, this sums the entire affair and puts the piece in the appropriate light:

    Of course if the details of these trials were examined one by one, more using zinc might surface, not listed in interventions nor the title. However it probably gives some sort of reasonable estimate of the ratio of zinc+HCQ trials vs HCQ without zinc, ie about 5%. Since many of these trials are in response to the early anecdotal findings that zinc plus HCQ was a valid treatment, worthy of further testing, it is disappointing that so many of the trials do not include zinc. I am not sure what to make of this exactly but I do share a certain uneasiness about motivations as expressed by comment 6, above.

    Oh, for one more, the Oxford Recovery trial used an initial loading dose of 2400 mg and then 800 mg/day. So thanks for mentioning studies providing a near toxic dose. Strike up one more for Trumpster Derangement Syndrome (see, unease about motivations), since as we all know, this would have played out entirely different were it not for Trump.

Leave a Reply