Association of Surgical Margin Distance With Survival in Patients With Resected Head and Neck Squamous Cell Carcinoma: A Secondary Analysis of a Randomized Clinical Trial.

Importance: Clear surgical margins reduce the risk of local recurrence, improve survival, and determine decision-making with regard to adjuvant treatment of squamous cell carcinoma of the head and neck (SCCHN). However, the definitions of clear, close, or positive surgical margins vary in both the literature and in practice. Objective: To examine whether the association between surgical margin distance and survival varies by primary tumor site. Design, Setting, and Participants: This was a secondary analysis of a multi-institutional, multinational randomized clinical trial. The trial enrolled patients from January 22, 2007, to March 29, 2013, with stage II to IVA resected SCCHN with extranodal extension (ENE) or positive margins (<5 mm from invasive tumor to the resected margin). The current analysis included those patients with known ENE and margin status and was conducted from April 29, 2022, to December 19, 2022. Interventions: Patients received adjuvant chemoradiotherapy plus either placebo or lapatinib. Main Outcomes and Measures: Overall survival (OS) was calculated to examine association with surgical margin distance, primary site, and survival, with stratification by ENE status. Results: Among 688 patients enrolled in the trial, 630 patients with known ENE and margin status were included. Exact patient ages were not made available; 523 (83%) patients were male, and 415 (66%) patients were White. Patients with 1 high-risk feature (positive margins or ENE) had significantly better OS vs 2 high-risk features (hazard ratio [HR], 0.65; 95% CI, 0.49-0.87), although most other results were not statistically significant. For example, in the cohort with ENE-negative disease, multivariable adjusted analysis showed nonsignificant improvements with shorter surgical margin distance (1- to 5-mm margins), and no association with OS was found in the cohort with ENE-positive status (either >5 mm margins or 1-5 mm margins). The association between survival and margin distance varied based on primary site, human papillomavirus (HPV) status, and ENE status. For example, HPV-positive status was associated with a significant and clinically meaningful increase in survival (adjusted HR, 0.33; 95% CI, 0.11-0.97). The improvement was greatest, although not significantly so, in patients with ENE- and HPV-negative oropharynx (OP), hypopharynx (HP), and larynx cancer (HR, 0.57; 95% CI, 0.30-1.10). No survival benefit was seen in ENE-negative oral cavity cancer (HR, 0.89; 95% CI, 0.45-1.77), nor was an association observed between margins and OS in HPV-positive OP cancer. Conclusions and Relevance: In this secondary analysis of a randomized clinical trial, the presence of high-risk features (extranodal extension, positive margins, or both) was associated with worse survival; longer survival was observed with greater surgical margin distance among patients with oral cavity tumors and human papillomavirus-negative tumors of the OP, larynx, or HP. No other significant differences were found. These findings support variable interpretation of surgical margin distance based on the primary site and HPV status. Trial Registration: ClinicalTrials.gov Identifier: NCT00424255.