Accuracy of PET/CT in the diagnosis of pelvic lymph node metastases in melanoma: A subgroup analysis of the evaluation of groin lymphadenectomy extent for metastatic melanoma (EAGLE-FM) trial

Abstract

BACKGROUND: PET/CT is widely used in the staging of patients with melanoma groin lymph node (LN) metastases. This study aims to utilise prospectively collected data to determine the accuracy of this modality in the diagnosis of pelvic metastases in those patients with known inguinal LN involvement. MATERIALS AND METHODS: Data collected as part of the Evaluation of Groin Lymphadenectomy Extent for Metastatic Melanoma (EAGLE-FM) trial was analysed. This was a prospective, randomised, phase III trial. Patients with known inguinal LN metastases were evaluated with 18F-FDG PET/CT. Those without evidence of pelvic LN abnormalities were randomised to inguinal dissection or ilio-inguinal dissection. Those with evidence of abnormality in the pelvic LNs were excluded from the main EAGLE-FM trial but offered ilio-inguinal dissection. Post operative histology results from those patients undergoing ilio-inguinal dissection were analysed to determine the accuracy of PET/CT in the assessment of pelvic LNs RESULTS: There were 101 patients randomised on the EAGLE-FM trial. For this substudy, there were 49 patients who had negative pelvic LNs on PET / CT and 5 patients with positive pelvic nodes on PET / CT included in this analysis. The sensitivity of PET/CT was 25%, specificity 95.6% with positive predictive value (PPV) 50.0% and negative predictive value (NPV) 87.8%. CONCLUSIONS: PET/CT has reasonable NPV but poor PPV. While PET/CT is an important modality in the staging of patients with stage III melanoma, it has limited accuracy in the diagnosis of pelvic LN metastases.