P3-211: Number of Surgically Removed Mediastinal Lymph-Nodes (SRMLNs) as prognostic factor for survival in resected early stage non-small-cell lung cancer (NSCLC): a retrospective analysis of a mono-institutional series

The impact on outcome provided by the extent of lymphoadenectomy in patients undergoing lobectomy or pneumonectomy for early stage (I-IIIA) NSCLC is still controversial. Several reports have documented a non significant effect of such procedure on both disease-free and over-all-survival (DFS/OS). Nevertheless, especially in trials of adjuvant chemotherapy, a very wide range of nodes number across all series is re-ported, strongly suggesting a clarification about the eventual supposed prognostic role of this intervention. A retrospective series of resected NSCLC patients was retrieved from the files of the Regina Elena National Cancer Institute. Correlation between known prognostic factors (sex, type of surgery, histology, tumor size, node involvement, grading) plus the number of SRMLNs and clinical outcome (DFS and OS) was then retrospectively explored using both the Cox regression model (considering SRMLNs as quantitative variable; significance cut-off<0.10) as well as classification and regression trees (CART) analysis. Data referring to 122 NSCLC patients who underwent surgery for stage I-IIIA NSCLC were collected. Multivariate analysis identified SRMLNs as significant independent predictors of both DFS and OS, with an hazard ratio of 0.94 (95% CI 0.90, 0.97, p=0.003), and 0.94 (95% CI 0.90, 0.99, p=0.035), respectively, together with nodal involvement, and grading G3. The CART analysis identified 26 removed nodes as the cut-off for better outcome. When considering SRMLNs as categorical variable, patients with more than 26 removed nodes had a significant better DFS and OS, with an hazard ratio of 3.90 (95% CI 1.16, 13.09, p=0.027), and 4.01 (95% CI 0.93, 18.18, p=0.063), respectively, at the multivariate analysis. Data presented herein open the issue that prognosis of early stage of NSCLC can be influenced by the extent of mediastinal lymphoadenec-tomy. Further prospective analyses are needed to confirm this finding.