Use of Intraoperative Frozen Section to Assess Surgical Margins in HPV-Related Oropharyngeal Carcinoma

Importance: Given the favorable overall prognosis of human papillomavirus (HPV)-related oropharyngeal squamous cell carcinoma (OPSCC) and the morbidity of increased adjuvant therapy associated with positive surgical margins, large-scale studies on the accuracy of frozen sections in predicting final surgical margin status in HPV-related OPSCC are imperative. Final surgical margin status is the definitive assessment of tumor clearance as determined through surgeon-pathologist collaboration based on permanent analysis of frozen section margins, main specimens, and supplemental resections. Objectives: To assess the accuracy and testing properties of intraoperative frozen section histology (IFSH) in assessing final surgical margin status in patients undergoing transoral surgery for HPV-related OPSCC. Design, Setting, and Participants: This retrospective, single-institution cohort study was conducted at the Washington University in Saint Louis School of Medicine and included 299 patients who underwent transoral surgery for HPV-related OPSCC between January 2015 and December 2021 who were identified using an institutional cancer registry. Forty-five patients were excluded due to piecemeal resections, no frozen sections sent during surgery, unknown primaries, or no residual tumor identified during surgery after a diagnostic biopsy. Pathology reports of patients with at least 1 margin assessed by IFSH were reviewed. The data were analyzed between January and February 2024. Main Outcomes and Measures: The accuracy of IFSH for individual margins and overall final surgical margin status was evaluated through calculating sensitivity, specificity, positive predictive values, and negative predictive values from 1482 margins from 254 patients. Results: Of 254 participants, 29 (11.4%) were female, and the mean (SD) age was 60.7 (9.4) years. IFSH demonstrated an accuracy of 97.1%, sensitivity of 72.2%, and specificity of 99.1% for individual margins compared with final pathology results of the same tissue. However, IFSH had a sensitivity of only 21.7% in determining the overall final surgical margin status, with 18 patients (7.1%) having at least 1 positive margin undetected intraoperatively. Positive final surgical margin status was associated with worse disease-specific survival (hazard ratio, 3.26; 95% CI, 1.05-10.13) and higher rates of locoregional recurrence (hazard ratio, 5.02; 95% CI, 1.25-20.19), while no definitive conclusion can be made about the prognostic effect of final tumor specimen histopathology status due to imprecision in the effect size estimates. Conclusion: The study results suggest that despite high accuracy for individual margins, IFSH has limitations in predicting final surgical margin status in HPV-related OPSCC, particularly for base of tongue primaries and deep margins. However, IFSH remains important for guiding intraoperative reresection and is associated with improved oncologic outcomes.