Identification of tumors with NRG1 rearrangement, including a novel putative pathogenic UNC5D-NRG1 gene fusion in prostate cancer by data-drilling a de-identified tumor database

Abstract

The fusion genes containing neuregulin-1 (NRG1) are newly described potentially actionable oncogenic drivers. Initial clinical trials have shown a positive response to targeted treatment in some cases of NRG1 rearranged lung adenocarcinoma, cholangiocarcinoma, and pancreatic carcinoma. The cost-effective large scale identification of NRG1 rearranged tumors is an open question. We have tested a data-drilling approach by performing a retrospective assessment of a de-identified molecular profiling database of 3263 tumors submitted for fusion testing. Gene fusion detection was performed by RNA-based targeted next-generation sequencing using the Archer Fusion Plex kits for Illumina (ArcherDX Inc., Boulder, CO). Novel fusion transcripts were confirmed by a custom-designed RT-PCR. Also, the aberrant expression of CK20 was studied immunohistochemically. The frequency of NRG1 rearranged tumors was 0.2% (7/3263). The most common histologic type was lung adenocarcinoma (n = 5). Also, renal carcinoma (n = 1) and prostatic adenocarcinoma (n = 1) were found. Identified fusion partners were of a wide range (CD74, SDC4, TNC, VAMP2, UNC5D), with CD74, SDC4 being found twice. The UNC5D is a novel fusion partner identified in prostate adenocarcinoma. There was no co-occurrence with the other tested fusions nor KRAS, BRAF, and the other gene mutations specified in the applied gene panels. Immunohistochemically, the focal expression of CK20 was present in 2 lung adenocarcinomas. We believe it should be considered as an incidental finding. In conclusion, the overall frequency of tumors with NRG1 fusion was 0.2%. All tumors were carcinomas. We confirm (invasive mucinous) lung adenocarcinoma as being the most frequent tumor presenting NRG1 fusion. Herein novel putative pathogenic gene fusion UNC5D-NRG1 is described. The potential role of immunohistochemistry in tumor identification should be further addressed.

Keywords: EGF-like domain; ERBB; ERBB3; HER; HER 3; MAPK; NRG1; PIK; carcinoma; data drilling; gene fusion; gene rearrangement; genetics; lung; mRNA sequencing; molecular; neuregulin; next-generation sequencing; receptor heterodimerization; receptor tyrosine kinase.