Abstract LB-92: Using immunohistochemistry to evaluate FGFR3 gene fusion recurrence and clinical associations in astrocytomas

Abstract

Abstract Aberrant activation of fibroblast growth factor receptor 3 (FGFR3) has been reported to be tumorigenic in various cancer types. When FGFR3 is fused with another gene, most commonly with transforming acidic coiled coil 3 (TACC3), it can escape miRNA-mediated regulation and remain constitutively active most likely due to the dimerization via the domains of TACC3 or another fusion partner. Among brain cancers, gene fusions involving FGFR3 have been detected in glioblastoma and certain low grade gliomas. Furthermore, FGFR3-TACC3 fusion positive glioma cells have responded well to targeted therapies. The endogenous FGFR3 protein levels are extremely low in brain partly due to miR-99a-mediated regulation and gene fusions lead to prominent protein expression. In previous studies, FGFR3 fusion recurrence and fusion partners have been analyzed only in limited patient cohorts and there are no reports about their clinical associations. In this study, we have performed immunohistochemistry for FGFR3 on tissue microarrays with over 750 clinical astrocytoma samples (including grade 1-3 astrocytomas and glioblastomas). FGFR3 positive staining has been detected in subpopulations of all included tumor grades. The observed recurrences (1.8%-6.2%, depending on the grade) are consistent with previous sequencing-based studies. To validate immunohistochemistry-based approach for FGFR3 fusion detection, we are using frozen tumor material to confirm the presence of fusion with Western blotting and with PCR followed by Sanger sequencing. Furthermore, we have associated FGFR3 staining scores with clinical records for all the cases. FGFR3 staining is more common in females than males both in astrocytoma (7.6% vs 2.2%, pearson chi-square p-value 0.004, n=752) and glioblastoma cohorts (8.9% vs 2.1%, pearson chi-square p-value 0.008, n=494). In addition, FGFR3-staining negatively associates with aberrant p53 staining indicating p53 mutation (FGFR3-positive cases: 5.8% vs 1.8% in p53 low vs high groups, respectively, pearson chi-square p-value 0.030, n=399). Altogether, this study evaluates the suitability of immunohistochemical methods for FGFR3 fusion detection as well as investigates their recurrence and clinical significance in astrocytomas. Citation Format: Kirsi J. Granberg, Birgitta Lehtinen, Matti J. Annala, Hannu Haapasalo, Matti Nykter, Wei Zhang. Using immunohistochemistry to evaluate FGFR3 gene fusion recurrence and clinical associations in astrocytomas. [abstract]. In: Proceedings of the 105th Annual Meeting of the American Association for Cancer Research; 2014 Apr 5-9; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2014;74(19 Suppl):Abstract nr LB-92. doi:10.1158/1538-7445.AM2014-LB-92