Surgery alone achieves comparable survival to adjuvant therapy in T4aN0M0 gingival squamous cell carcinoma with clear surgical margins: A registry-based propensity-matched analysis

BACKGROUND: The survival benefit of adjuvant radiotherapy/chemoradiotherapy (RT/CRT) in pT4aN0M0 gingival squamous cell carcinoma (GSCC) with adequate surgical margins remains unclear due to limited subsite-specific evidence. In this retrospective cohort study, we evaluated the associations of adjuvant RT/CRT with survival outcomes and identified prognostic factors in this clinically relevant subgroup. METHODS: We used the Taiwan Cancer Registry (2011-2021) to identify patients with first primary T4aN0M0 GSCC and clear resection margins (≥5 mm). Patients treated with surgery alone were compared with those receiving adjuvant RT/CRT. Cohorts were balanced via 1:1 propensity score matching (PSM) on age, sex, tumor depth, tumor differentiation, and comorbidity burden. Disease-specific survival (DSS) and overall survival (OS) were assessed using Kaplan-Meier estimates and Cox proportional hazards regression models. RESULTS: Among the 530 study patients, 75.3% (n = 399) received adjuvant RT/CRT. In the unmatched cohort, 5-year DSS and OS rates were similar for surgery alone vs. adjuvant RT/CRT (80% vs. 79%, p = 0.27; 67% vs. 72%, p = 0.99, respectively). After PSM (125 pairs), the 5-year DSS and OS rates were comparable, being 81% vs. 83% (p = 0.61) and 68% vs. 78% (p = 0.46), respectively. Multivariable analysis demonstrated that increasing age, moderate/poor differentiation, depth > 14 mm, and Charlson comorbidity index > 1 were independent prognosticators for DSS or OS, whereas adjuvant RT/CRT was not independently associated with better outcomes. CONCLUSION: For patients with T4aN0M0 GSCC and clear surgical margins, adjuvant RT/CRT did not confer a survival advantage over surgery alone, suggesting limited benefit of routine adjuvant therapy in this setting. Prospective studies are needed to validate this finding and to refine evidence-based risk stratification models.