Abstract 122: Comparative analysis of small cell lung cancer and other pulmonary neuroendocrine tumors

Abstract

Abstract Small cell lung cancer (SCLC) is accounting for 15% of all lung cancer cases and belongs to the family of pulmonary neuroendocrine tumors. This tumor entity additionally includes large-cell neuroendocrine carcinomas (LCNEC) and pulmonary carcinoids (PCA), which occur in 3% and 2% of all lung cancer patients, respectively. Whereas lung carcinoids are clinically benign, SCLC and LCNEC are characterized as high-grade malignant tumors, which are associated with heavy smoking and a poor 5-year survival rate of less than 5%. We aimed to characterize the genomic alterations in neuroendocrine lung tumors and performed whole genome, whole exome and transcriptome sequencing on up to 110 SCLC cases, over 40 pulmonary carcinoids and over 50 LCNEC tumors. SCLC and LCNEC tumors reveal high mutation rates with an average of 9.5 and 8.6 mutations per megabase, respectively. Our sequencing studies have shown that SCLC tumors are characterized by bi-allelic inactivation of TP53 and RB1 in almost 100% of the cases. Recurrent significant genomic alterations affected among others TP73 (13%) and NOTCH family genes (25%). Further studies in a pre-clinical mouse model confirmed NOTCH as tumor suppressor and as a regulator of neuroendocrine differentiation in SCLC (George et al., 2015). LCNEC tumors showed TP53 and RB1 alterations in up to 20% of the cases. Additionally, LCNECs were found with STK11 and KEAP1 mutations, which occurred mutually exclusive to RB1 alterations (P<0.05). In light of these distinct mutational characteristics, we analyzed the transcriptome sequencing data of neuroendocrine lung tumors in comparison to lung adenocarcinomas (LUAD) and lung squamous cell carcinomas (LUSQ). An unsupervised clustering approach led to the identification of five expression clusters: LUAD, LUSQ and PCA formed distinct transcriptional classes; the majority of SCLC and LCNEC tumors clustered into two subgroups. While a few LCNEC tumors clustered with LUAD, LUSQ and PCA, RB1-mutated LCNEC tumors predominantly clustered with SCLC samples (P <0.05). KEAP1 and STK11 mutated LCNECs did not segregate with a particular histological tumor type. Neuroendocrine lung tumors confirmed high expression levels of neuroendocrine markers. A detailed comparison of LCNEC and SCLC tumors revealed that the majority of the SCLC and a few LCNEC tumors were found in a transcriptional group that showed high expression of the neuroendocrine lineage transcription factor ASCL1, gastrin releasing peptide (GRP) and DLL3, which was less pronounced in transcriptional subsets formed by the majority of LCNEC tumors. Patient tumors that clustered to the transcriptional class marked by higher ASCL1 expression were identified to have a worse overall survival (P<0.05). In summary, these findings point to transcriptional differences in SCLC and LCNEC tumors and emphasize the precise analysis of these histological tumor types with respect to their molecular biology and therapeutic treatment option. Citation Format: Julie George, Lynnette Fernandez-Cuesta, Vonn Walter, Neil Hayes, Roman Thomas. Comparative analysis of small cell lung cancer and other pulmonary neuroendocrine tumors. [abstract]. In: Proceedings of the 107th Annual Meeting of the American Association for Cancer Research; 2016 Apr 16-20; New Orleans, LA. Philadelphia (PA): AACR; Cancer Res 2016;76(14 Suppl):Abstract nr 122.