Comment on “Phase I Randomized,Placebo-Controlled,Cross-Over Dose-Finding Study of Coenzyme Q 10 on Doxorubicin Pharmacokinetics During Breast Cancer Treatment”

Comment on “Phase I Randomized, Placebo-Controlled, Cross-Over Dose-Finding Study of Coenzyme Q₁₀ on Doxorubicin Pharmacokinetics during Breast Cancer Treatment”

We read with interest the article by Greenlee et al ¹ examining coenzyme Q₁₀ (CoQ₁₀) with doxorubicin in breast cancer. While the topic of cardioprotection during anthracycline therapy is important, several methodological concerns temper the authors’ conclusions:

First, the sample size was extremely small (only 6 completed patients, vs a planned 18). The trial was closed early due to poor accrual, resulting in critically low statistical power. Nonetheless, the authors report P values and conclude there were “no clinically significant pharmacokinetic interactions” between CoQ₁₀ and doxorubicin. With such a limited sample, a non-significant finding more likely reflects a type II error (false-negative) than a true absence of effect. Indeed, the authors themselves acknowledge that the lack of observed differences could be due to the small sample size and insufficient power. It is therefore premature to conclude definitively that CoQ₁₀ has no impact on doxorubicin pharmacokinetics.

Second, although described as a “dose-finding Phase I study,” the trial tested essentially 1 dose of CoQ₁₀ (300 mg/day). Only a single patient attempted 600 mg/day before withdrawing, and no patients received the intended 1200 mg dose. There was no systematic dose-escalation or determination of a maximum tolerated dose (MTD). In oncology, Phase I trials are typically designed to escalate through multiple dose levels to establish the MTD or recommended Phase II dose. ² By failing to explore doses beyond 300 mg, this study did not fulfill conventional Phase I objectives. No dose-response data were generated, so the label “dose-finding” is misleading. Essentially, this was a pilot tolerability study of 1 dose level, which limits insight into whether higher CoQ₁₀ doses might have pharmacokinetic or cardioprotective effects.

Third, the cross-over design lacked any washout interval between doxorubicin cycles. Patients crossed over to the opposite arm in the very next chemotherapy cycle, without a washout period. Both doxorubicin and CoQ₁₀ have relatively long elimination half-lives (~20-48 hours for doxorubicin and ~33 hours for CoQ₁₀). With such multi-day persistence of both drugs, carryover effects are a serious concern. In a proper 2-period cross-over, treatments are separated by a washout phase sufficient to eliminate prior drug exposure. ³ Here, any CoQ₁₀ or doxorubicin remaining from Cycle 3 could confound the pharmacokinetics in Cycle 4, especially since doxorubicin’s cumulative effects may last beyond a single cycle. The absence of a washout thus compromises internal validity – one cannot be certain that each patient’s “placebo versus CoQ₁₀” cycles were independent of the preceding cycle’s effects. This design flaw further clouds interpretation of the null interaction finding.

Fourth, the study’s stated aim was partly to explore CoQ₁₀’s potential to mitigate doxorubicin cardiotoxicity, yet no direct cardiac outcomes or biomarkers were assessed. The trial did not evaluate left ventricular ejection fraction (LVEF), cardiac enzymes (eg, troponin or BNP), or other measures of cardiac function/injury. These are standard endpoints for anthracycline cardiotoxicity. ⁴ Instead, the investigators measured plasma total antioxidant capacity (TAC) as a surrogate, reasoning that CoQ₁₀’s antioxidant properties might reduce oxidative stress. TAC is at best an indirect marker and not specific to cardiac myocyte damage. Notably, the authors admit they “lacked data on cardiac biomarkers,” which limited their ability to evaluate CoQ₁₀’s cardioprotective effect. Without echocardiographic or biochemical evidence of cardioprotection, the study cannot substantively address whether CoQ₁₀ prevented heart damage. (Indeed, prior studies of CoQ₁₀ in anthracycline-treated patients have measured changes in LVEF or other cardiac function parameters.) Relying on TAC alone is insufficient to conclude anything about mitigating cardiotoxicity.

Finally, the authors’ broad conclusion that CoQ₁₀ “is safe and does not alter doxorubicin pharmacokinetics” overinterprets very limited data. A sample of 6 patients – with no serious adverse events observed – is reassuring but far too small to definitively prove safety. Rare toxicities or interactions could not be detected in such a pilot cohort. Likewise, the claim of no pharmacokinetic interaction must be viewed in light of the low power and design limitations discussed above. A finding of no significant difference in doxorubicin AUC or C_max in 6 patients does not rule out a modest interaction; it only indicates none was obvious under these conditions. In our view, the conclusions in this report exceed the evidence available. At most, this small study suggests that 300 mg/day of CoQ₁₀ was feasible to give during doxorubicin therapy without immediate apparent harms, and no large pharmacokinetic alterations were observed at that dose. Any firm declaration of safety or lack of interaction, however, is premature. Caution is warranted before generalizing these preliminary results.

In summary, Greenlee et al addressed an important question, but major design and power limitations constrain their conclusions. A larger, adequately powered trial with dose escalation, washout, and cardiac endpoints is needed to determine whether CoQ₁₀ can be safely combined with doxorubicin or offers cardioprotection. Until then, claims of “no significant interaction” should be interpreted cautiously given the risk of false-negative results.

Zhihao Lei, PhD Cornell University, Ithaca, NY, USA