Relationship of ERCC1 genotype variant with mRNA expression and ERCC1 protein levels in advanced urothelial carcinoma (UC).

Abstract

260 Background: DNA repair factors may be predictive for response to chemotherapies that produce DNA damage. While low ERCC1 protein and mRNA levels have been reported as associated with improved outcomes in metastatic UC patients treated with platinum-based chemotherapy, the relationship between genotype, mRNA expression, and protein level is unknown. The ERCC1 germline 19007C>T single-nucleotide polymorphism (SNP) is functionally associated with reduced translation of ERCC1 mRNA. We investigated the relationship between ERCC1 germline SNP, ERCC1 tumor mRNA and protein expression, in a cohort of patients with advanced UC who received first-line, platinum-based chemotherapy. Methods: A cohort of clinically annotated, uniformly-treated advanced UC patients with FFPE primary tumor tissue available was identified through the Hellenic cooperative Oncology Group (HECOG) (N=93). Genomic DNA extraction, nested PCR, and restriction fragment length polymorphism techniques for the 19007C>T SNP were performed to identify C/C, C/T and T/T genotypes. ERCC1 mRNA expression was interrogated using Nanostring nCounter profiling. IHC analysis was performed on tissue arrays using an ERCC1 antibody. Percent of positive nuclear staining was categorized as quartiles using previously identified cut-points. Results: ERCC1 C/T genotype was identified in 30/61 samples (49%) and T/T in 14/61 samples (23%). In 54 patients with both SNP and mRNA data available, T/T genotype was associated with the highest level of mRNA expression, followed by the C/T genotype (p=0.04). Neither ERCC1 genotype (N=44) nor ERCC1 mRNA expression (N=54) was associated with ERCC1 protein expression as measured by IHC (p=0.52 and p=0.13, respectively). Conclusions: ERCC1 19007C>T is associated with increased ERCC1 mRNA expression. However, neither genotype nor mRNA are surrogates for ERCC1 protein detected by IHC in advanced UC tumors. This suggests that while genotype influences mRNA expression of ERCC1, the use of the nucleotide excision repair pathway as a predictive biomarker of platinum-sensitivity may be more complex than previously appreciated and require the integrative use of proteomics, genomics and epigenomics.