Intraoperative Pathology Consultation in Patients With p16-Positive Unknown Primary Squamous Cell Carcinoma.

Importance: Current guidelines recommend intraoperative frozen section(s) during diagnostic surgery for squamous cell carcinoma for unknown primary tumors (SCCUP).

Objective: To determine the utility of intraoperative pathology consultation during transoral robotic surgery (TORS) in localizing primary tumors and influencing need for adjuvant therapy.

Design, setting, and participants: A retrospective case series including 47 adult patients with human papillomavirus (HPV)-associated SCCUP who underwent TORS/oropharyngectomy between January 2016 and February 2023 was carried out at a single tertiary care hospital. The analysis took place on May 13, 2024.

Exposures: Nodal stage, tonsillectomy history, extranodal extension (ENE).

Main outcomes and measures: Intraoperative pathology consultation and final pathology results were compared with surgical outcomes, including margin revision, need for second procedure and/or radiation with or without chemotherapy.

Results: This study included 47 adult patients. Mean (range) age was 61 (41-79) years; patients were mostly men (37 [79%]). Overall, primary tumors were identified in 37 patients (79%), including all cases with positive nodes involving more than 1 neck level. Patients whose primary tumor was not found tended to have tobacco use history (8/10 vs 13/37 [35%]; difference, 45%; 95% CI, 16%-74%) and absence of ENE (8/10 vs 15/37 [41%]; difference, 39%; 95% CI, 10%-68%). Primary tumor was identified intraoperatively in 18 of 37 patients (49%). SCCs identified intraoperatively were significantly larger than SCCs found on permanent sections only: mean (SE), 1.2 (0.13) cm vs 0.5 (0.1) cm (difference, 0.7 cm; 95% CI, 0.53-1.94). The sensitivity, specificity, positive predictive value, and negative predictive value of intraoperative consultation was 49% (95% CI, 33%-64%), 100% (95% CI, 100%-100%), 100%, and 34% (95% CI, 19%-53%), respectively. Margins were revised in 11 of 18 patients (61%) whose primary tumor was identified intraoperatively (during original procedure) and in 3 of 19 patients (16%) whose primary tumor was identified on permanent pathologic findings only (during a second procedure) (11/18 [61%] vs 3/19 [16%]; difference, 45%; 95% CI, 17%-73%). However, there was no significant difference in the use of adjuvant radiotherapy with or without chemotherapy or need for a second procedure based on intraoperative primary tumor localization.

Conclusion and relevance: In this case series study, the sensitivity and negative predictive value of intraoperative pathology consultation among 47 patients was less than 50%. Given the lack of influence on the need for radiotherapy with or without chemotherapy or second procedure, the practical utility of routine intraoperative frozen section requires further scrutiny.