Lukadi Joseph Lula, Rita Costa, Aimée J P M Franssen, Lin Huang, Emrah Go¨kay O¨zgu¨r, Clara Forcada Barreda, Matic Domjan, Rebecca Weedle, Crt Jasovic, Beatrice Trabalza Marinucci, Mohamed Rebei, Kiarash Ghasemi, Adelino Leite Moreira, Ronan Ryan, G Nural Bekiroglu, V Young, Teresa Gomez, Erik R. de Loos, Gerard J. Fitzmaurice, Kathleen Bennett, Massimiliani Mancini, Gonzálo Varela, Cornel Savu, Alessio Vincenzo Mariolo, Paul Bunn, T STUPNIK, Jozsef Furak, Antonio D’Andrilli, Marcelo Jimenez, Robert Cerfolio, Ciprian Bolca, Alessandro Brunelli, Ramón Rami-Porta, René Horsleben Petersen, Erino Angelo Rendina
OBJECTIVES: To evaluate the impact of common pathologic features, single or in combination, in patients with early-stage lung adenocarcinoma, according to lung resection extent. METHODS: A retrospective, multicentric cohort study including patients with cT1a-bN0M0 lung adenocarcinoma and with at least visceral pleural invasion under the surface (PL1) or up to surface (PL2), lymphovascular invasion, spread through air spaces, necrosis, or neural invasion who underwent lobectomy or segmentectomy with systematic lymph nodal dissection from 2015 to 2021 in 10 European centers. Overall survival (OS), disease-free survival, and lung cancer-specific death between both groups were assessed before and after stabilized inverse probability of treatment-weighting matching. Risk factors for oncologic outcomes were analyzed using parsimonious model Cox proportional hazard regression in both entire and multiple-feature datasets. Kaplan-Meier and cumulative incidence function was used to assess outcome. Log-rank and Gray tests were used to compare the groups. Linearized risk assessed recurrences. RESULTS: Of 1703 patients with cT1a-bN0M0 lung adenocarcinoma, 530 had at least 1 poor pathologic feature and 130 had multiple features. For the 530 patients, 5-year OS: lobectomy was 83.0% and segmentectomy was 89.4%, P = .2; 5-year disease-free survivall obectomy was 78.1% segmentectomy was 83.8%, P = .06; and 5-year lung cancer-specific death lobectomy was 8.9% and segmentectomy was 7.2%, P = .6, which were similar. It was the same in the matched cohort. In multivariable analysis, no poor pathologic feature impacted outcome more than others. Multiple poor features were not associated with any clinical, pathologic trait, but they impacted OS (hazard ratio, 3.24; P = .002). Locoregional recurrence (linearized risk: lobectomy 0.083, segmentectomy 0.086) was similar in the matched entire dataset. CONCLUSIONS: Segmentectomy with systematic lymph nodal dissection can be indicated in patients with stage IA1-2 lung adenocarcinoma suspected to have poor pathologic features. Multiple factors were not predictable but impacted OS.