Reply to “Preclinical and Clinical Evidence on Chloroquine or Hydroxychloroquine in COVID‐19”: Achieving Statistical Significance Does Not Mean That Clinical Relevance Can Be Disregarded, and This Must Be Considered When Interpreting Results

Dear Editor,

1.

We thank Dr. Pasquier for his thoughtful comments on our review [1]. We agree that the evaluation of chloroquine (CQ) and hydroxychloroquine (HCQ) in COVID‐19 requires nuance. However, the overall pharmacological and clinical evidence does not support their use as effective antivirals.

First, it is correct that TMPRSS2 is not the sole viral entry pathway for SARS‐CoV‐2, but its predominance in relevant human airway cells remains well established. CQ/HCQ primarily target endosomal acidification, which relative contribution in vivo appears insufficient to translate into clinically meaningful antiviral activity. This is supported by multiple studies showing no efficacy in human airway epithelium and animal models.

Second, the positive findings cited in humanized mice [2], should be interpreted cautiously. Such models do not fully recapitulate human pharmacokinetics or disease complexity, and discordant results across animal systems are common. However, the broader preclinical literature shows a lack of reproducible antiviral efficacy of CQ/HCQ in advanced models.

Third, concerning clinical evidence, we acknowledge that early observational studies and some meta‐analyses yielded heterogeneous results [3]. However, the mortality benefit reported by Di Castelnuovo et al. [3] relied on retrospective cohort studies highly exposed to confounding and survivor bias. The most robust data come from randomized controlled trials (RCTs), in which CQ/HCQ showed no effect on mortality, arguing against any meaningful clinical benefit [3]. Beyond mortality outcomes, multiple studies have reported an increased risk of cardiac arrhythmias with CQ/HCQ use in COVID‐19, mainly in hospitalized patients [4], and this cannot be overlooked.

In the COPCOV prophylaxis trial, no excess cardiac was reported [5]. However, participants were young (median age 29 years) and healthy (< 5% with chronic disease), limiting generalizability to populations at highest risk for severe COVID‐19. While COPCOV suggests a modest signal in prophylaxis (RR ~0.85, borderline significance) [5], this effect is small, non‐clinically relevant (1% vs. 1.2%), and inconsistently observed across trials. Importantly, it did not translate into reductions in severe outcomes, and current evidence‐based guidelines continue to recommend against HCQ for both treatment and prevention. Statistical significance does not imply clinical relevance.

Finally, our central argument is methodological: early drug selection should rely on integrated pharmacokinetic/pharmacodynamic (PK/PD) considerations, validated human‐relevant models, and realistic exposure–response relationships. For CQ/HCQ, the totality of evidence—mechanistic, preclinical, and clinical—remains inconsistent and insufficient to justify their use as antivirals.

We therefore respectfully maintain that CQ/HCQ exemplify the limitations of drug repurposing when pharmacological principles are not rigorously applied from the outset.

Funding

The authors have nothing to report.

Conflicts of Interest

The authors declare no conflicts of interest.