Accuracy of frozen section for HPV-Associated squamous cell carcinoma of unknown primary

INTRODUCTION: Current guidelines for the management of metastatic squamous cell carcinoma of unknown primary (SCCUP) recommend submission of suspicious primary sites for frozen section analysis (FSA). This study aims to investigate the diagnostic accuracy of FSA for identification of HPV-associated SCCUP. METHODS: A retrospective cohort study of patients with biopsy-proven p16-positive SCCUP who underwent diagnostic operation at two tertiary care institutions was performed. Sensitivity, specificity, PPV, and NPV of diagnostic FSA were assessed. RESULTS: 77 patients were included in analysis. 66 patients underwent definitive TORS (diagnostic TORS operation with subsequent neck dissection after identification of the occult primary tumor), 7 patients underwent diagnostic TORS (TORS to identify occult primary tumor, no neck dissection), and 4 patients underwent direct laryngoscopy and biopsy only. Primary tumors were identified in 63 patients (82%) with a mean tumor size of 1.1 cm. There was no significant difference in size between patients whose tumor was identified on FSA (mean 1.1 cm) and on permanent only (mean 0.9 cm) (p = 0.26). The sensitivity, specificity, PPV, and NPV of FSA for SCCUP was 86%, 100%, 100%, and 86%, respectively. Diagnostic frozen specimens included 52 direct laryngoscopy biopsies and 69 TORS excisions. In the biopsies, sensitivity was 100% and NPV was 100%, whereas in the TORS-excised specimens, sensitivity was 77% and NPV was 77%. CONCLUSIONS: In this case series of 77 patients with SCCUP, the sensitivity and NPV of FSA for identification of the primary tumor was over 85%. FSA is valuable during diagnostic operation for SCCUP.