Remdesivir Phase III Clinical Trial Fails in China: The Principle of 'Presumption of Guilt' for Drugs Prevails Again

Author: Li Tianquan

According to reports today from the Financial Times and STAT, the results of the Phase III clinical trial of remdesivir conducted in China have been released. The final results will be formally published in a medical paper; although not yet peer-reviewed, they were inadvertently posted online by the World Health Organization (WHO). The published abstract indicates that a total of 237 patients (159 in the treatment group and 78 in the control group) participated in the trial, with 18 patients in the treatment group and 4 in the control group discontinuing therapy due to adverse effects.

Numerically, the medication group showed a slightly higher mortality rate (13.9% vs. 12.8%) and a slightly higher incidence of adverse reactions (65% vs. 64%), but there were no significant differences between the medication group and the placebo control group. In other words, Gilead’s remdesivir failed to reduce viral load in the blood, improve patients’ clinical conditions, or prevent patient deaths. Following the data disclosure, Gilead’s stock price dropped by as much as 8.5%, marking its largest intraday decline since March 16 this year.

This was a Phase III trial conducted in China for patients with severe COVID-19. The study originally planned to enroll 453 participants, but due to difficulties in recruitment, it was announced on April 11 that the trial would be terminated early.

Although Gilead Sciences stated in response to media reports that the study failed to reach its target enrollment and thus lacked sufficient participants to draw meaningful final statistical conclusions, the prospects for success are undoubtedly diminishing.

Remdesivir, also known as “the people’s hope,” has seen that hope shattered today. Although this is disappointing, for a professional, the failure and inefficacy of remdesivir are not unexpected. In science, intuition is often unreliable; drugs that seem quite promising may not withstand rigorous testing. Not only remdesivir, but other “effective drugs” once included in COVID-19 treatment guidelines—such as arbidol and Kaletra (lopinavir/ritonavir)—also failed to demonstrate efficacy in subsequent clinical trials.

Moreover, a recent study in the United States has shown that hydroxychloroquine, once highly regarded, is not effective and may even increase the risk of death. The latest version of the "Guidelines for the Treatment of Coronavirus Disease" by the NIH (National Institutes of Health) does not recommend the use of hydroxychloroquine/chloroquine or remdesivir outside of clinical trials.

As early as February 2, before the clinical trials of this drug had even begun, I authored an article titled “Remdesivir Is Not an Effective Drug for Novel Coronavirus,” which sparked controversy among some readers. I am no oracle capable of predicting clinical trial outcomes; rather, my conclusion was drawn solely from professional common sense—specifically, the principle of “presumption of guilt” in pharmacology.

What is the principle of presumption of guilt for drugs?

We all know that modern legal systems operate on the presumption of innocence. If there is insufficient evidence to prove a suspect’s guilt, they must be acquitted. It is better to err on the side of letting a potentially guilty person go free than to wrongly convict an innocent individual, even if the latter may in fact be guilty.

In contrast, modern drug development adheres to a principle of “presumption of guilt,” whereby a drug is assumed to be harmful and ineffective until rigorously proven otherwise, thereby placing the burden of proof on the drug to demonstrate its safety and efficacy. If efficacy is not conclusively demonstrated, the drug is deemed ineffective; this approach prioritizes avoiding false positives over false negatives, even if the drug might potentially be effective in reality.

Why implement a presumption of guilt, preferring to "wrongly condemn the innocent," while preemptively casting the cure for human diseases as the "villain"?

Let us first examine the history. Throughout thousands of years of global civilization, the principle of “presumption of guilt” for drugs has existed for only a few decades.

Prior to 1962, the United States, like other countries, had very lax drug regulations. Pharmaceutical companies could unilaterally claim efficacy based on cursory research, leading to a proliferation of so-called "miracle drugs" purported to treat various intractable diseases.

What was the FDA doing at that time? Its mandate was then limited to ensuring drug safety. It was precisely due to the steadfastness of Dr. Frances Kelsey at the FDA that thalidomide, a drug with serious safety concerns, was prevented from entering the U.S. market, thereby sparing the United States from experiencing a tragedy akin to Europe’s, where thousands of infants were born with phocomelia.

Due to its exemplary performance in the “thalidomide” incident, the FDA established its authority. In 1962, taking advantage of this momentum, the U.S. Congress passed the Kefauver-Harris Amendments, also known as the “KH Amendments to the Federal Food, Drug, and Cosmetic Act of 1938.” The core provision of the legislation was that all new drug applications must include “substantial evidence” of both safety and efficacy.

So, what constitutes substantial evidence? It is not evidence from “clinical practice,” but rather from “clinical trials.” Typically, this requires randomized, double-blind, controlled, large-sample clinical trials. For further details, please refer to my article, “Who Determines Drug Efficacy? A Superb Popular Science Explanation of the Remdesivir Clinical Trials!”

1962 marked a watershed moment, heralding the emergence of the modern regulatory framework for pharmaceuticals and the widespread adoption of the principle of “presumption of guilt” for drugs. The United States not only imposed this requirement on newly submitted drug applications but also conducted a “drug re-evaluation” for medications marketed prior to 1962. Marketing authorizations were revoked for drugs lacking robust evidence of efficacy. In total, 3,443 drugs were re-evaluated: 2,252 were deemed effective, 1,015 ineffective, and 167 had indeterminate efficacy, resulting in an ineffectiveness rate of 30%. Meanwhile, the indications for certain “panacea”-type drugs were restricted to specific therapeutic uses.

In China, a similar “drug re-evaluation” has not yet been conducted. A considerable number of drugs on the market have not undergone rigorous clinical trials, or indeed any clinical trials at all. Currently, the consistency evaluation being carried out in China applies only to chemical generic drugs, leaving a large number of other medications still awaiting re-evaluation. As a licensed pharmacist myself, I often remove medications that I deem ineffective based on my professional judgment from physicians’ prescriptions before purchasing the remaining drugs at pharmacies.

This is not me being overly critical; if you don’t believe me, listen to what the two bureau directors have to say:

In February 2016, at a State Council press conference, Bi Jingquan, then Commissioner of the China Food and Drug Administration, posed a rhetorical question: “We have previously emphasized safety more than efficacy. If a drug is ineffective, does it still need to exist?”

Not only were unsafe and ineffective drugs approved for market launch, but they also brazenly entered the National Reimbursement Drug List (NRDL). In March this year, Hu Jinglin, the current Director of the National Healthcare Security Administration, published an article in Qiushi Journal, pointing out bluntly: “For a long time, ‘miracle drugs’ that are safe yet ineffective have prevailed within the NRDL, with entries but no exits, thereby affecting public access to high-quality pharmaceutical services.”

The formal statements issued by the two bureau directors may not necessarily remove unsafe and ineffective drugs from the market. While there is a consensus that ineffective drugs should be withdrawn, opinions differ on what constitutes an “ineffective drug.” We must transform the professional principle of “presumption of guilt” for drugs into a public consensus.

To reach a consensus, it is essential first to enhance our understanding of why drugs are subject to a "presumption of guilt" and the underlying logic behind this approach. I attempted to review relevant literature but found no existing discussions on this topic. With due humility, I will therefore attempt to systematically outline the rationale for adopting a "presumption of guilt" based on the "four characteristics."

1. From the perspective of efficacy: Prior to validation, the efficacy of a candidate drug is a highly improbable event; therefore, opting for high-probability outcomes is the prudent course of action.

Taking Tu Youyou and her team’s research on antimalarial drugs as an example, they studied more than 2,000 traditional Chinese medicines, identified 640 that showed potential efficacy, but ultimately discovered and validated only one drug—artemisinin.

Even if a drug has undergone systematic preclinical research, including in vitro and animal studies, demonstrating safety and efficacy, it still has only about a 10% chance of passing clinical trials and successfully reaching the market. Given such a low probability of success, the failure of remdesivir was to be expected.

Before a drug’s efficacy has been validated, would you choose to take it based on the 1% small probability of effectiveness, or refuse it given the 99% high probability of ineffectiveness?

Second, from a safety perspective: Without prior verification of medication use, the harms far outweigh the benefits. As the saying goes, “all medicines carry some degree of toxicity.” Typically, drug toxicity accompanies efficacy, whereas absolute safety is often associated with ineffectiveness. The statement “adverse effects are not yet clearly defined” does not mean there are no side effects; nor does “purely natural” equate to non-toxic or harmless (arsenic trioxide, for instance, is also purely natural). Therefore, we should first presume potential harm and require evidence to prove safety. Second, without verified efficacy, we should not expose ourselves to the risk of potential harm; otherwise, we risk suffering losses on both fronts.

Third, from an economic perspective: If the presumption of innocence is adopted and all potentially effective drugs are treated as effective, then as a new drug researcher, why “verify efficacy” when one can simply “claim efficacy”? As a regulator, it is easy to accept “claims of efficacy,” but too difficult to “prove inefficacy”; regulators neither have the capacity nor the responsibility to verify drug efficacy. Meanwhile, for users, drugs with “verified efficacy” are obscured by those merely “claimed to be effective,” leaving them with no real choice. This leads to a significant increase in healthcare costs across society and substantial harm to public health.

4. From the perspective of professionalism: The pharmaceutical and medical field is highly specialized, characterized by significant information asymmetry, making it difficult for the general public to understand and increasing their susceptibility to deception.

In fact, most products involve issues of efficacy, safety, and cost-effectiveness. However, for ordinary consumer goods, consumers themselves can make basic judgments on these three attributes. More importantly, incorrect choices lead to direct adverse consequences. With such feedback and elimination mechanisms in place, even lax regulation poses little problem. It is unnecessary to adopt a “presumption of guilt” principle or judge honorable intentions by petty standards.

However, regarding medications, although users appear to have the right to choose effective drugs, it is difficult for individual doctors and patients to accurately assess drug efficacy due to the complexity of diseases and pharmaceuticals (though they often overestimate their own judgment). Coupled with the divergent interests of physicians and patients, it is not surprising that a safe but ineffective drug can generate annual sales amounting to hundreds of millions of yuan. Over time, this dynamic reverses the principle of “good money driving out bad,” as drugs with “proven efficacy” become overwhelmed by those merely “claimed to be effective,” leading to the rampant proliferation of so-called “miracle cures.”

Therefore, from the perspectives of safety, efficacy, cost-effectiveness, and professionalism, it is more reasonable, safer, and more economical to adopt a "presumption of guilt" rather than a "presumption of innocence" for modern drugs.

Remdesivir is the first anti-SARS-CoV-2 drug worldwide to undergo large-scale, randomized, double-blind, controlled clinical trials. Just as “a true warrior dares to face a bleak life and look squarely at dripping blood,” a genuine medication does not hastily proclaim “Effective! Effective! It is effective!” based on scant research findings; rather, it dares to withstand rigorous trials and confront the possibility of failure.

Remdesivir, although this clinical trial has failed, we still pay tribute to you! Salute to Gilead Sciences, Inc., the researchers, and the patients who participated in the trial!