Comment on: ‘alpha‐lipoic acid as a preventive measure in radiation‐induced oral mucositis in head and neck cancer pAtients: a randomized controlled study’—methodological considerations

We read with considerable interest the recent randomized controlled trial by Rizkalla and colleagues, which addresses an important and clinically relevant question regarding the prevention of radiation-induced oral mucositis (RIOM) in patients undergoing radiotherapy for head and neck cancer (HNC).1 RIOM remains one of the most frequent and debilitating acute toxicities of HNC radiotherapy, often resulting in severe pain, impaired oral intake, nutritional compromise and unplanned treatment interruptions that may ultimately influence oncologic outcomes.2-4 In this prospective randomized study, administration of alpha-lipoic acid (ALA) (600 mg twice daily) was associated with a statistically significant reduction in severe mucositis (RTOG Grades 3–4), delayed onset of high-grade mucosal injury and shorter duration of severe mucositis compared with placebo.1 The exploration of inexpensive and widely accessible pharmacologic interventions such as ALA—an agent with documented antioxidant and anti-inflammatory properties5—is therefore commendable. Notwithstanding these encouraging findings, several methodological and interpretative aspects warrant further consideration in interpreting the reported results. First, the single-blind design of the study introduces a potential risk of detection bias in assessing the primary endpoint. Although patients were blinded to treatment allocation, mucositis grading was performed by an oncologist who was aware of the assigned intervention.1 The Radiation Therapy Oncology Group (RTOG) mucositis scale, while widely used in clinical trials, incorporates elements of clinical judgement in distinguishing moderate from severe mucosal injury.2, 6 In supportive care trials in which the primary endpoint relies on clinician-graded toxicity, double-blind designs are generally preferred to minimize the potential influence of observer expectations.7 Even when standardized evaluation protocols are employed, awareness of treatment allocation may inadvertently affect grading thresholds, particularly in relatively small studies. Accordingly, the possibility that detection bias may have contributed, at least in part, to the observed reduction in grade ≥3 mucositis should be considered when interpreting the magnitude of the reported treatment effect. Second, several statistical assumptions underlying the trial design merit consideration when interpreting the magnitude of the reported treatment effect. The assumptions underlying the sample-size calculation appear to rely on a particularly optimistic expected treatment effect. The authors estimated the required sample size based on an anticipated reduction in the incidence of severe mucositis from approximately one third of patients to 5%, extrapolated from data on melatonin-based antioxidant therapy.1 Although modulation of oxidative stress represents a shared biological pathway in RIOM pathogenesis, extrapolating effect sizes across pharmacologically distinct agents may not fully account for differences in pharmacodynamics, bioavailability and clinical efficacy.2, 5 Sample-size calculations based on very large anticipated effects frequently lead to relatively small study populations, increasing the potential risk of the so-called ‘Winner's Curse’, whereby early statistically significant findings from small trials tend to overestimate the true magnitude of treatment benefit.8 Consequently, the observed reduction in severe mucositis—from 41% in the control group to 17% in the ALA group—while statistically significant, should be interpreted cautiously until replicated in larger multicentre studies designed with more conservative assumptions regarding effect magnitude. Third, the clinical implications of the reported reduction in severe mucositis warrant further consideration, given the lack of improvements in patient-centred outcomes. Despite the lower incidence of Grade 3–4 mucositis in the ALA group, the study did not demonstrate significant differences between groups in pain scores, functional oral intake or quality-of-life measures.1 This apparent discordance raises important questions regarding the translation of clinician-graded toxicity reductions into meaningful symptomatic benefit for patients. It is well recognized that treatment-related morbidity during HNC radiotherapy is multifactorial, reflecting not only mucosal ulceration but also xerostomia, swallowing dysfunction, nutritional impairment and psychosocial factors.3, 9 Nevertheless, if severe mucositis were substantially reduced, some degree of improvement in patient-reported outcomes might reasonably be anticipated. The absence of such improvements may therefore underscore the complex determinants of treatment-related symptom burden, but it may also suggest that the intervention's clinical impact is more modest than implied by toxicity grading alone. Taken together, the study by Rizkalla and colleagues provides valuable preliminary evidence supporting the potential role of antioxidant-based interventions in mitigating radiation-induced mucosal injury.1 However, confirmation of these findings will require larger, preferably double-blind, multicentre randomized trials with comprehensive reporting of treatment-related dosimetric parameters and robust patient-reported outcomes.4, 10 Such studies will be essential to determine whether modulation of oxidative stress pathways through agents such as ALA can translate into clinically meaningful improvements in treatment tolerance and quality of life during HNC radiotherapy. We congratulate the authors on their important contribution to supportive care research and hope that future investigations will further clarify the clinical role of antioxidant-based interventions in the prevention of radiation-induced oral mucositis. All authors contributed to the conceptualization of this correspondence, critical analysis of the literature and drafting and revision of the manuscript. All authors approved the final version of the manuscript. The authors have nothing to report. The authors declare no conflicts of interest. No new data were generated or analysed in this study.