Abstract
Dear Editor,
We read with great interest the recently published systematic review entitled “Prevention and treatment of radiation-induced esophagitis with oral herbal medicine: A systematic review of randomized controlled trials” by Ha et al 1 in Integrative Cancer Therapies. The study synthesizes evidence from 81 randomized controlled trials (RCTs), evaluating the efficacy of oral herbal medicine in the prevention and treatment of radiation-induced esophagitis (RIE). While the topic is of considerable clinical relevance and contributes valuable insight into integrative approaches to radiotherapy complications, we would like to offer several critical reflections that extend beyond methodology and also touch upon conceptual framing and clinical interpretation.
First, the risk of bias across included studies is substantial. Only 35.8% of studies reported appropriate random sequence generation, and allocation concealment was described in merely 3 studies. Blinding of participants or assessors was nearly absent. These issues considerably weaken the internal validity of the pooled results. Although a risk-of-bias figure is provided, the implications of these quality limitations are not fully integrated into the authors’ discussion or conclusion. Second, the pooled effects across many outcomes suffer from high heterogeneity. For example, the I² statistic for incidence of RIE was 96%, and for onset time of RIE was 98%. Despite this, the authors maintain strong claims regarding the superiority of herbal interventions without adequate sensitivity or subgroup analysis to explore these sources of heterogeneity. Third, the review merges a broad array of herbal formulations with diverse theoretical underpinnings (eg, heat-clearing, qi-tonifying, yin-nourishing) into a single intervention category. This may obscure important differences between herbal strategies and limits the specificity of clinical recommendations. Fourth, although the GRADE assessment is presented, the rationale for downgrading evidence to “low” or “very low” quality is not adequately explained, particularly in connection with the serious risk of bias and potential publication bias.
Beyond methodology, we wish to raise several conceptual issues. The clinical framing of the review is ambiguous: it is unclear whether the authors position herbal medicine as a supportive adjunct, a primary treatment, or a preventive measure for RIE. Without clear clinical anchoring, the rationale for conducting such a broad and heterogeneous synthesis becomes less compelling. Moreover, the authors do not engage with the biological plausibility of the interventions. Modern pathophysiology of RIE—such as inflammation-mediated mucosal injury, oxidative stress, and epithelial apoptosis—is not discussed, nor are any mechanistic hypotheses offered regarding how herbal medicine might address these processes. Additionally, the review lacks critical reflection on cultural applicability. All 81 included trials are from China, yet the manuscript is positioned within a global oncology journal without acknowledging the cultural, regulatory, or integration challenges of applying Chinese herbal medicine in non-Chinese settings. This limits both the generalizability and the translational value of the findings.
In conclusion, we commend the authors for addressing an important and underexplored topic in supportive cancer care. However, we suggest that future systematic reviews in this field apply stricter inclusion criteria, perform more nuanced subgroup analyses based on herbal classification, clarify the clinical intent of the synthesis, and ensure that both methodological and contextual limitations are fully considered.
Sincerely,
