Validation of National Comprehensive Cancer Network Guidelines (NCCN) at a single institution.

Abstract

e20556 Background: Invasive mediastinal staging is necessary to identify locally advanced non-small cell lung cancer (NSCLC). We sought to validate NCCN guidelines for invasive mediastinal staging. Methods: We retrospectively reviewed all patients who had curative lung resection for pathologically confirmed NSCLC from October 2018 to December 2019. We excluded patients who had induction therapy without undergoing invasive mediastinal staging first. We evaluated methods of mediastinal staging, staging results, and final pathology. Indications for staging were one or more of the following; mediastinal lymph nodes > 1.0cm in short axis, Standardized Uptake Value of > 3.0 on Positron Emission Tomography, > 50% of tumor on medial side of mid-clavicular line (central vs peripheral), or peripheral lesion > 3.0cm in diameter. Staging methods were mediastinoscopy, endobronchial ultrasound (EBUS), and video-assisted thoracoscopic surgery (VATS) ipsilateral mediastinal staging before resection of main tumor. Results: In total, 457 lung resections were performed. Staging was done in 144/275 indicated cases (52.4%). Mediastinoscopy was completed in 49 patients, with 20.4% (n = 10) N2-positive. The false negative rate of mediastinoscopy was 4.1% (n = 2 at station 7). EBUS was performed in 64 patients and 21.9% (n = 14) were N2 positive. The false negative rate for EBUS was 3.1% (n = 2 at station 7). Two mediastinoscopy and 9 EBUS patients had 0 N2 stations sampled. None of the patients were ultimately N2 positive. Staging of three mediastinal stations (4L, 4R, and 7) was done in 26/49 mediastinoscopies and 15/64 EBUS. Of 20 patients who had VATS ipsilateral mediastinal staging, none were N2 positive and there were 0 false negatives. This left 131 patients who did not received indicated staging. The most common indication in this group was central location, 83.2% (n = 109). Four of these unstaged patients were N2 positive at resection (3.1%). The sole indication for these 4 patients was a centrally located tumor. Clinical tumor sizes were 1.2 cm, 1.3 cm, 1.5 cm, and 1.9 cm. Overall, the N2-positive rate of the staged group was 24/144 (16.7%) vs 4/131 (3.1%) N2 positive in the unstaged group, p < 0.001. Conclusions: The current NCCN staging guidelines accurately reflect the risk of N2 disease for NSCLC. The indication of central location for tumors may benefit from being reevaluated in a larger cohort, particularly given efficacy of adjuvant therapy.