Abstract / Summary
Background: Epidermal growth factor receptor (EGFR) inhibitors, including tyrosine kinase inhibitors and anti-EGFR monoclonal antibodies, are widely used targeted agents in oncology, but they can cause acneiform or papulopustular eruptions and other cutaneous adverse events. This review aimed to systematically assess the efficacy and safety of Traditional Chinese Medicine (TCM) for preventing or treating EGFR inhibitor-associated rash. Methods: PubMed, Embase, Cochrane Library, Web of Science, CNKI, and Wanfang were searched from inception through January 20, 2025. Randomized comparative, nonrandomized comparative, and prospective uncontrolled clinical studies were eligible for the descriptive review; comparative meta-analysis required a concurrent control. Dichotomous outcomes were summarized as risk ratios and continuous outcomes as Hedges' g standardized mean differences using REML random-effects models. Results: The search retrieved 190 records, and 27 studies enrolling 2155 participants were included: 21 randomized comparative trials, 4 nonrandomized comparative studies, and 2 uncontrolled single-arm pre/post studies, all conducted in China. Fifteen randomized trials with established rash and compatible favorable-response definitions contributed to the primary meta-analysis (RR = 1.82, 95% CI: 1.45-2.28; I 2 = 82.7%; τ 2 = 0.154; Hartung-Knapp 95% CI: 1.41-2.36). The prediction interval (0.75-4.41) crossed the null, and small-study effects were detected. Three prophylactic randomized trials were summarized separately because their endpoints were not commensurate. Four randomized trials with compatible binary pruritus-response data yielded RR = 3.17 (95% CI: 1.18-8.49; I 2 = 89.7%). TCM was associated with lower total TCM syndrome scores, DLQI scores, IL-1β, and TNF-α in sparse continuous-outcome analyses. Conclusion: TCM may improve established EGFR inhibitor-associated rash on average, but the evidence is highly heterogeneous and predominantly of low or very low certainty. The primary prediction interval included no effect, publication-bias signals were present, and prophylaxis and secondary symptom evidence were sparse. No formulation or combined oral-topical regimen was established as superior.