Abstract
Dear Editor,
We appreciate the thoughtful comments provided by Dr. Wang regarding our recently published systematic review on oral herbal medicines for radiation-induced esophagitis (RIE). 1 His letter raised important points regarding study quality, heterogeneity, intervention classification, and clinical framing. We welcome this opportunity to clarify our intent, distinguish between the limitations of the evidence base and those of our synthesis, 2 and outline priorities for future research.
Nearly all of the 81 randomized controlled trials (RCTs) included in our review were conducted in China. This geographic concentration limits generalizability, and many trials suffer from poor reporting of randomization, allocation concealment, and blinding. Outcomes were inconsistently defined, and standardized protocols were lacking, all of which undermined confidence in the pooled estimates. These weaknesses reflect the underlying evidence rather than the review itself, and underscore the need for better trial design and reporting.
To address these limitations, we used the GRADE framework. The certainty of evidence for all major outcomes was rated as low or very low, primarily because of risk of bias and inconsistency. The rationale for downgrading was described in detail, and the strength of our conclusions was deliberately tempered to reflect this low certainty. Contrary to concerns that our review maintained strong claims, we deliberately tempered our conclusions to emphasize caution when interpreting pooled effects.
Heterogeneity was substantial, with I² values exceeding 90% for several outcomes. Subgroup analyses, including those reported in the manuscript as well as additional exploratory analyses, did not fully resolve this inconsistency. Moreover, most trials did not report the theoretical rationale or principles of formula construction, which further limited meaningful stratification. As recommended by the CONSORT Extension for Chinese Herbal Medicine Formulas (CONSORT-CHM), 3 details regarding composition, rationale, and standardization were frequently absent, precluding classification by therapeutic approach.
Nevertheless, our review clearly distinguished between preventive and therapeutic applications. For prevention studies, trial characteristics are presented in Supplemental Table 1, herbal compositions in Supplemental Table 4, and outcomes in Figure 3. For treatment studies, trial characteristics are presented in Table 1, herbal compositions in Supplemental Table 3, and outcomes in Figure 4. Within both categories, we further differentiated whether herbal medicine was used as an add-on to conventional radiotherapy or as a sole intervention, thereby clarifying its combined versus independent use. These elements, also described in the main text, ensure that the clinical positioning of herbal interventions is clearly defined and minimize potential confusion regarding their roles in RIE.
We acknowledge the importance of biological plausibility. Our Introduction outlines the pathophysiology of RIE, noting that radiation therapy induces DNA damage, activates stress-related cytokine pathways, and triggers inflammation, edema, and mucosal erosion, leading to esophageal injury. In the Discussion, we linked these mechanisms to preclinical evidence suggesting that certain herbal medicines might mitigate inflammation, oxidative stress, and epithelial injury. While we agree that a stronger integration of mechanistic and clinical research is needed, we believe these aspects were addressed in our review.
Looking forward, we recognize the importance of more rigorous and transparent trials. Future studies should incorporate robust randomization, allocation concealment, blinding (when feasible), and registration. Adherence to CONSORT-CHM 2017 is equally important to ensure that the formula rationale and preparation details are fully documented. Such improvements will enable systematic reviews to conduct more precise subgroup analyses and distinguish between therapeutic strategies. Multicenter and multinational RCTs are needed to enhance generalizability across diverse cultural and regulatory settings.
We thank Dr. Wang for his constructive comments. We acknowledge the limitations of both the primary studies and our synthesis, but maintain that our review contributes by highlighting an underexplored area in supportive oncology and by identifying clear priorities for future research. We hope this exchange will encourage collaborative efforts to generate high-quality evidence on integrative approaches for radiation-induced esophagitis.
Sincerely,
