Comparison of clinical and surgical-pathological staging in IIIA non-small cell lung cancer patients.

Background: The heterogeneous group of IIIA NSCLC patients requires careful preoperative clinical staging as tumor size and lymph node involvement guide treatment. The purpose of our study was to analyze the correctness of clinical staging in IIIA patients.

Methods: Retrospective analysis of all patients resected due to lung cancer that had been staged IIIA either clinically using invasive and noninvasive techniques or surgical-pathologically after surgical resection. Correctness, sensitivity, specificity, and positive and negative predictive values of clinical staging were calculated.

Results: From our tumor database, 49 patients who met the inclusion criteria were identified. The histology of the primary tumor included adenocarcinoma (53%), squamous cell carcinoma (41%), and other (6%). Preoperative clinical staging consisted of computed tomography (CT), integrated positron emission tomography-CT (PET-CT), bronchoscopy, and mediastinoscopy. The predominant surgical procedures performed were lobectomies (57%) and pneumonectomies (29%). Clinical staging for UICC, T and N stage was correct in 36.7, 38.7, and 40.8%, respectively. In terms of T4 stage, sensitivity was 28.5%, specificity was 80.9%, positive predictive value was 20%, and negative predictive value was 87.1%. As for N2 involvement, we found a sensitivity of 66.6% and a specificity of 35.7%. Positive and negative predictive values for N2 involvement were 43.7 and 58.8% in that order.

Conclusions: Despite multimodal preoperative invasive and noninvasive staging techniques, the correctness of clinical staging in IIIA NSCLC patients is low. Hence, in doubt more invasive staging or probatory thoracotomy should be performed not to deny potentially curative surgery in those patients.