Previous genomic studies have identified two mutually exclusive molecular subtypes of large-cell neuroendocrine carcinoma (LCNEC): the mutated (mostly comutated with ) and the wild-type groups. We assessed whether these subtypes have a predictive value on chemotherapy outcome. Clinical data and tumor specimens were retrospectively obtained from the Netherlands Cancer Registry and Pathology Registry. Panel-consensus pathology revision confirmed the diagnosis of LCNEC in 148 of 232 cases. Next-generation sequencing (NGS) for and genes, as well as IHC for RB1 and P16 was performed on 79 and 109 cases, respectively, and correlated with overall survival (OS) and progression-free survival (PFS), stratifying for non-small cell lung cancer type chemotherapy including platinum + gemcitabine or taxanes (NSCLC-GEM/TAX) and platinum-etoposide (SCLC-PE). mutation and protein loss were detected in 47% ( = 37) and 72% ( = 78) of the cases, respectively. Patients with wild-type LCNEC treated with NSCLC-GEM/TAX had a significantly longer OS [9.6; 95% confidence interval (CI), 7.7-11.6 months] than those treated with SCLC-PE [5.8 (5.5-6.1); = 0.026]. Similar results were obtained for patients expressing RB1 in their tumors ( = 0.001). RB1 staining or P16 loss showed similar results. The same outcome for chemotherapy treatment was observed in LCNEC tumors harboring an mutation or lost RB1 protein. Patients with LCNEC tumors that carry a wild-type gene or express the RB1 protein do better with NSCLC-GEM/TAX treatment than with SCLC-PE chemotherapy. However, no difference was observed for mutated or with lost protein expression. .