The accuracy of intraoperative frozen section examination of sentinel lymph nodes in squamous cell cancer of the vulva

Abstract

<h3>Objectives:</h3> To measure the diagnostic accuracy of intraoperative frozen section examination of sentinel lymph nodes (SLN) in vulvar cancer and to describe associated patient outcomes. <h3>Methods:</h3> This retrospective cohort study included patients with invasive squamous cell carcinoma of the vulva who underwent SLN biopsy with intraoperative frozen section at one cancer centre from Jan 2008 - Feb 2020. Exclusion criteria were tumor size >4cm, multifocal tumor, palpable groin lymph nodes and neoadjuvant therapy. The SLN procedure was performed by injection of the primary vulvar tumor or scar with technetium-99m, ICG and/or blue dye. We compared the intraoperative SLN frozen section diagnosis and the paraffin section report for concordance. Recurrence location was described and recurrence free survival (RFS) was compared using the Kaplan-Meier method. <h3>Results:</h3> A total of 173 patients (258 groins) met inclusion and exclusion criteria. On frozen section, there were 36/258 positive groins on frozen section and 222 negative groins. On final pathology, there were 39/258 positive groin nodes: 30 macro-metastases, 7 micro-metastases, 2 isolated tumor cells (ITCs). There were 219 negative groins. The sensitivity and specificity for frozen section detecting any metastatic disease compared to final pathology, was 89.7% and 99.5%, respectively. The positive-predictive value (PPV) was 97.2% and the negative-predictive-value (NPV) was 98.2%. There were 4 cases of false negative frozen section where final pathology revealed two cases of ITCs, 1 micro-metastasis and 1 macro-metastasis. A total of 30 patients (17.3%) underwent a full inguinal femoral lymphadenectomy during the same operation due to the frozen section results and avoided a second operation. A total of 2 patients (1.2%) required a second operation for lymphadenectomy due to false negative results on frozen section. Median (range) follow up was 38.0 (1-137.8) months. Median RFS was 14.9 (3.6 to 98.2) months. There were 46 recurrences in the study population, of which 29 were confined to the vulva, 9 groin recurrences and 8 distant recurrences. Among patients with a negative SLN biopsy on both frozen section and final pathology, there were 10 (3.9 %) groin and/or distant recurrences. A total of 6 nodal recurrences were ipsilateral, 1 was contralateral, 2 were bilateral and 1 recurrence was distant. After exclusion of patients with local recurrence only, the RFS at 3 years was 91.6% (95% CI 86.2-97.4%) for patients with negative SLN on frozen section and 64.6% (95% CI 46.5-89.7%) for positive SLN on frozen section. On final pathology, patients with negative SLN had a 3-year RFS of 91.7% (95% CI 86.3-97.4%), macro-metastases 58.4% (95% CI 38.5-87.7%), and micro-metastases/ITCs was 100%. <h3>Conclusions:</h3> Intraoperative frozen section of SLNs in vulvar cancer is an accurate intraoperative guide to determine if complete inguinal femoral lymphadenectomy should be performed. SLN biopsy reduced the need for a second surgical procedure and does not appear to compromise patient outcomes.