Sublobar resection versus lobectomy for early-stage pulmonary carcinoid tumors with three centimeters or less in size: A population-based study.

Abstract

e21062 Background: Pulmonary carcinoids (PCs), comprising of typical carcinoids (TCs) and atypical carcinoids (ACs), are rare low-grade malignant tumors and surgical resection is considered as the standard treatment for early stage PCs. However, it is not clear whether sublobar (Sub-L) resection of PCs would compromise the long-term oncologic outcomes. Here we investigate the equivalency of Sub-L resection versus lobectomy (Lobe) for patients with early stage PCs using the population-based Surveillance, Epidemiology, and End Results (SEER) registry. Methods: The clinical and survival data of early stage PC patients with less or equal 3cm diameter were retrieved. Kaplan-Meier method and log-rank tests were used to assess differences in overall survival (OS). We further performed the subgroup analysis by tumor size (T≤1cm, 1cm < T≤2cm, 2cm < T≤3cm) and patient age ( < 70y and ≥70y). To reduce the bias of inherent retrospective study, we also performed propensity score matching (PSM) analysis by 1:1 fixed ratio nearest neighbor matching using Matchit package in R. Results: In total of 2934 PC patients, including 2741 TCs and 193 ACs, were recruited. 1011(34.46%) and 1923(65.54%) patients received Sub-L resection and lobectomy, respectively. After PSM, there was no significant difference between Sub-L and Lobe group in terms of OS in PC patients (P = 0.33). From the cox regression analysis, it indicated that older age, male gender, larger tumor size, AC histology type and Sub-L surgery were associated with a worse OS. We further performed PSM analysis on TCs and ACs, respectively. For TC patients, it indicated that patients in Lobe group had a much better OS compared to those in Sub-L group (p = 0.0067). In the subgroup analysis, the log-rank tests showed that there was a significant survival advantage for lobectomy in comparison with Sub-L resection for the TC patients with diameter 2cm < T≤3cm (p = 0.0345), but not for those with diameter lower than 2cm. In the subgroup analysis of age, the log-rank test indicated that Lobe group in age < 70 had a better prognosis with statistical difference compared to Sub-L group (p = 0.0032). For ACs, there was no statistic difference between Sub-L and Lobe group after PSM analysis in all the subgroup investigations. Conclusions: Sublobar resection may not significantly compromise the long-term oncologic outcomes in early stage PC patients. However, there is a survival benefit of lobectomy in certain pathological type and population, such as TCs patients with tumor size more than 2cm and age lower than 70-year-old.