Endocrine Abstracts

Pheochromocytomas and paragangliomas (PPGL) are commonly benign catecholamine-producing neuroendocrine tumors (NETs); however, up to 25 % of patients develop distant metastases or aggressive behavior. The current classification of PPGL comprises pseudohypoxia-profile, MAPK-pathway alteration, and Wnt-pathway dysregulation clusters according to their genomic characterization. However, to date, there are no biomarkers to help stratify patients based on their prognosis. Alternative splicing is an emerging cancer feature that has been linked to a more aggressive phenotype in a variety of neoplasms, including NETs. In this context, we have recently discovered alterations in the splicing machinery in other NETs, such as pancreatic and lung NETs. The splicing process has not been studied in detail before in PPGL, but there are reasons to expect that it could be altered. Thus, the aim of this study was to assess the profile of the splicing machinery in PPGL and study its potential relationship with clinical-molecular features. To this end, we studied the expression of 313 splicing-related genes in the data available in the TCGA dataset, which includes 151 patients (29 paragangliomas, PGL, and 122 pheochromocytomas, PCC). Most splicing-related genes were found to be similar in PCC and PGL, but 16 genes, including RBM22, CELF4, and PABPC1, exhibited significant differences. Interestingly, a detailed analysis among the three genomic clusters revealed considerable differences, standing out 143 of 313 splicing-related genes, which were found over- or under-expressed. It is also worth noting that just the expression of CELF4 and API5 was sufficient to clearly distinguish the three clusters: low expression of both genes in pseudohypoxia cluster, high in kinase signaling, and CELF4 high and API5 low expression in Wnt-altered cluster. Furthermore, 27 genes were shown to be associated with aggressive PCC; in particular, categorizing PCC samples based on high or low expression of specific genes, such as LSM4 or SMC1A, allowed us to predict aggressive/metastatic behavior. In addition, aggressive PGLs had differential expression of 25 genes. Altogether, our findings show that the splicing machinery is disrupted in PPGL, which encourage us to explore the splicing process in a larger cohort of PPGL samples and investigate the functional importance of these splicing-related genes in vitro.

Keywords: pheochromocytoma, paraganglioma, metastasis, splicing dysregulation, splicing machinery. This work was supported by MICINN (PID2019-105201RB-I00), Beca GETNE 2019, Fundación Eugenio Rodríguez Pascual, ISCIII (CD19/00255).