Abstract
BACKGROUND: Recent preclinical data suggest that neuroendocrine (NE) subtype of small cell lung cancer (SCLC) has strong therapeutic relevance. NE high tumors are associated with immune desert and NE low tumors are considered to have an immune oasis phenotype. Our aim was to investigate the NE phenotypes of surgically resected SCLC tumors according to inter-tumor heterogeneity. METHODS: Expression analysis for 2,560 genes was performed in 32 surgically resected SCLC patients' primary tumors and corresponding lymph node (LN) metastases. To analyze tumor heterogeneity, we examined the differences in the gene expression of primary tumors versus LN metastases. We performed cluster analysis and heat map to divide patients into NE high and low subtypes by using the top NE-associated genes described in preclinical studies. RESULTS: We found 6% (n=154) genes with significant differences and only 13.1% (n=336) of all genes in the panel had a strong correlation between the primary tumor and LN metastases. Cluster analysis clearly distinguished SCLC NE high versus low subtypes both in primary tumor (20 vs. 12, respectively) and LNs (23 vs. 9, respectively). As for inter-tumor heterogeneity, in case of five patients, a change in the NE pattern was observed. Specifically, we found significant downregulation of the NE-associated genes CAV1 (P=0.004), CAV2 (P=0.029) and ANXA3 (P=0.035) in their LN metastases compared to their primary tumor. CONCLUSIONS: Our data confirm the results of preclinical studies and clearly distinguish NE low and high differentiation clusters in SCLC. Moreover, they highlight the gene expression discordance between primary tumors and corresponding LN metastases suggesting that the NE pattern of metastatic LNs might not reflect that of the primary tumor. Altogether, by shedding light on the diversity of SCLC, the current study might help to improve patient selection and treatment in this devastating disease. KEYWORDS: Small cell lung cancer (SCLC); neuroendocrine tumor; lymph node metastasis; tumor heterogeneity; RNA sequencing.