Sublobar resection for node-negative lung cancer 2-5 cm in size.

OBJECTIVES: Sublobar resection (SLR) is an alternative to lobectomy for non-small-cell lung cancer (NSCLC). Outcomes following SLR for tumours >2 cm are not well described. We sought to determine the utilization of SLR for stage I tumours >2-5 cm in size and to determine predictors of outcome.

METHODS: We utilized the Surveillance, Epidemiology and End Results Program (SEER)-Medicare database to identify NSCLC patients with primary lung cancer ≥66 years old with stage I cancers >2-5 cm in size. We evaluated overall survival and cancer-specific survival among cohorts undergoing lobectomy versus SLR. Propensity score matching was performed. We compared patient characteristics and survival between groups.

RESULTS: For the study time period (2007-2012), among patients with tumours >2 cm and ≤5 cm (n = 4582), 3890 lobectomies (85%) and 692 SLR (15%) were performed. Patients undergoing SLR were older, had smaller tumours and more comorbidities. Patients undergoing lobectomy were much more likely to have any lymph nodes removed (95.6% vs 65.6%, P < 0.001) and to have >10 nodes removed (29.6% vs 7.5%, P < 0.001). All-cause mortality [hazard ratio (HR) 1.65, confidence interval (CI) 1.48-1.85] and cancer-specific (HR 1.63, CI 1.29-2.06) mortality were higher following SLR. At 3 years, overall survival (60.9%, CI 57.0-64.6% vs 54.4%, CI 50.4-58.2%) and cancer-specific survival (87.3%, CI 83.5-90.3% vs 76.5%, CI 71.0-81.1%) favoured lobectomy over SLR. In propensity-matched groups, both all-cause (HR 1.27, CI 1.10-1.47) and cancer-specific (HR 1.54, CI 1.11-2.16) mortality rates were higher with SLR.

CONCLUSIONS: In pathologically staged patients, SLR appears inferior to lobectomy for stage I NSCLC 2-5 cm in size. SLR is associated with less extensive lymphadenectomy and with worse survival than lobectomy in this cohort of patients. However, the 76.5% 3-year cancer-specific survival in patients undergoing SLR may exceed that of other localized treatment options for NSCLC. As such, SLR may be an appropriate option for high-risk patients with carefully staged 2-5 cm N0 tumours.