Prospective phase II trial of preresection thoracoscopic mediastinal restaging after neoadjuvant therapy for IIIA (N2) non–small cell lung cancer: Results of CALGB Protocol 39803

ObjectiveAccurate pathologic restaging of N2 stations after neoadjuvant therapy in stage IIIA (N2) non–small cell lung cancer is needed.MethodsA prospective multi-institutional trial was designed to judge the feasibility of videothoracoscopy to restage the ipsilateral nodes in mediastinoscopy-proven stage IIIA (N2) non–small cell lung cancer after 2 cycles of platinum-based chemotherapy and/or 40 Gy or more of radiotherapy. The goals included biopsy of 3 negative N2 node stations or to identify 1 positive N2 node or pleural carcinomatosis.ResultsTen institutions accrued 68 subjects. Of the 68 subjects, 46 (68%) underwent radiotherapy and 66 (97%) underwent chemotherapy. Videothoracoscopy successfully met the prestudy feasibility in 27 patients (40%): 3 negative stations confirmed at thoracotomy in 7, persistent stage N2 disease in 16, and pleural carcinomatosis in 4. In 20 procedures (29%), no N2 disease was found, 3 stations were not biopsied because of unanticipated nodal obliteration. Thus, 47 videothoracoscopy procedures (69%, 95% confidence interval, 57%-80%) restaged the mediastinum. Videothoracoscopy was unsuccessful in 21 patients (31%) because the procedure had to be aborted (n = 11) or because of false-negative stations (n = 10). Of the 21 failures, 15 were right-sided, and 10 had a positive 4R node. The sensitivity of videothoracoscopy was 67% (95% confidence interval, 47%-83%), and the negative predictive value was 73% (95% confidence interval, 56%-86%) if patients with obliterated nodal tissue were included. The sensitivity was 83% (95% confidence interval, 63%-95%) and the negative predictive value was 64% (95% confidence interval, 31%-89%) if those patients were excluded. The specificity was 100%. One death occurred after thoracotomy.ConclusionsVideothoracoscopy restaging was “feasible” in this prospective multi-institutional trial and provided pathologic specimens of the ipsilateral nodes. Videothoracoscopy restaging was limited by radiation and the 4R nodal station.