Regional Nodal Disease (N1) at Time of Segmentectomy Does Not Require Conversion to Lobectomy

R. Leo¹, M. Abdallat¹, M. McAllister¹, M. Silvestri¹, E. Sugarbaker², Y. Xie³, E. Mazzola³, R. Bueno¹, M. Jaklitsch⁴, P. Ugalde Figueroa⁵, D. Wiener⁶, S. Swanson^1
1Brigham and Women's Hospital, Boston, Massachusetts ²Brigham & Women's Hospital, boston, Massachusetts ³Dana-Farber Cancer Institute, Boston, Massachusetts ⁴Brigham and Women's Hospital - Thoracic Surgery, Boston, Massachusetts ⁵Brigham and Women's Hospital, Brookline, Massachusetts ⁶Brigham & Women's Hospital, Div of Thoracic Surgery, Boston, Massachusetts

Purpose: With widespread adoption of segmentectomy for early-stage non-small cell lung cancer (NSCLC)1,2, the need to convert segmentectomy to lobectomy for pathological N1 (pN1) disease comes into question. We set to determine whether lobectomy improves overall survival (OS) and locoregional recurrence-free survival (LRFS) compared to segmentectomy for pN1 disease.

Methods: We conducted a retrospective review of patients who underwent segmentectomy or lobectomy for pN1 NSCLC between 01/2006 and 01/2023 at a single institution. Patients who underwent neoadjuvant treatment, completion pneumonectomy, or additional wedge resections were excluded. Clinicopathologic characteristics were compared with Mann-Whitney, Wilcoxon rank-sum, chi-squared, and Fisher’s exact tests where appropriate. OS and LRFS were evaluated using Kaplan-Meier analysis and Cox proportional-hazards regression. We used a propensity-score weighted analysis to estimate average treatment effect on the treated (ATT), where segmentectomy was the treatment. The weights were based on a propensity score model, calculated using the following variables: age, pathological tumor size (PTS) and percent of predicted FEV1 (FEV1) for OS; PTS, lymphovascular invasion (LVI), and number of lymph nodes sampled during surgery (LN) for LRFS.

Results: 185 patients with pN1 NSCLC met inclusion criteria; 30 (16.2%) underwent segmentectomy and 155 (83.8%) underwent lobectomy. There were 6 lingulectomies, 7 trisegmentectomies, and 17 single-segmentectomies. Median age and FEV1 were similar between groups (p=0.749, p=0.625). Lobectomies had a larger median PTS (cm) than segmentectomies (3.4[IQR:2.30-4.45] vs. 1.95[1.43-2.83], p< 0.005) and higher SUVmax (8.6[5.1-13.05] vs. 5.6[4.4-8.05], p=0.028). Median LN was higher for lobectomies, both in total (9[6-14] vs. segmentectomies: 7[4-9], p< 0.005) and positive count (p=0.011). Median segmentectomy tumor-to-parenchymal margin distance was 2.5cm (IQR:1.2-3.2). Median follow-up was 58.8 months. Before propensity-weighting, there was no difference in OS or LRFS between segmentectomy and lobectomy at 5 years (OS:79.69% vs. 55.60%, log-rank p=0.063; LRFS:78.72% vs. 71.74%, log-rank p=0.47). Segmentectomies trended towards improved OS over lobectomy after propensity-weighting (segment vs. lobe OS:82.28% vs. 66.83%, log-rank p=0.350). In a univariable analysis of OS and LRFS, procedure type was not an independent predictor of outcome. Analysis examining age, number of positive lymph nodes and pulmonary function did not identify subgroups who fared better with lobectomy.

Conclusion: Segmentectomy for pN1 NSCLC confers similar LRFS compared to lobectomy, with a trend towards improved OS. This subgroup of patients will need further study to determine best treatment, but segmentectomy may be reasonable for select patients with small sized tumors.

Identify the source of the funding for this research project: None