The role of surgery for atypical bronchopulmonary carcinoid tumor: Development and validation of a model based on Surveillance, Epidemiology, and End Results (SEER) database

Abstract

ObjectivesThe rarity of atypical carcinoid (AC) of lung and the lack of prospective clinical trials lead to limited knowledge of its biology, treatment information and prognosis. The current study analyzed AC patients from the Surveillance, Epidemiology, and End Results (SEER) database to better understand the clinical characteristics of this disease and build a prognostic nomogram for clinical practice. Materials and methodsA total of 507 AC patients with pathological confirmation from SEER database were performed with univariate Cox regression analyses for both overall survival (OS) and lung cancer specific survival (LCSS) analyses. Of the 507 observations, 464 were used in the multivariable Cox proportional hazards model as training cohort of new nomogram. A new nomogram was constructed based on the training cohort and validated by an external validation cohort to predict the 3-, 5- and 10-year OS of ACs. The accuracy and clinical practicability were separately tested by Harrell's C-indexes, calibration plots and decision curve analyses (DCA). ResultsLobectomy and segmental resection were found to be protective factors for AC patients. Age, primary tumor size, N stage, M stage, surgery and regional lymph nodes examination were shown as significant prognostic factors in Cox regression analyses and included in the nomogram as predictors. The C-index in the training cohort for 3-, 5-, and 10-year OS were 0.722, 0.737 and 0.712, respectively. The internal and external calibration plots for predictions of the 3-, 5-, and 10-year OS were in excellent agreement. An online webserver was built based on the proposed nomogram for convenient clinical use. ConclusionAC patients with lobectomy or segmental resection tended to have better OS and LCSS. A nomogram was constructed and validated to predict the OS for AC patients and to provide accurate and individualized survival predictions.