FGFR3 mutations, but not FGFR3 expression and FGFR3 copy-number variations, are associated with favourable non-muscle invasive bladder cancer.

The fibroblast growth factor receptor 3 (FGFR3) is a tyrosine kinase receptor frequently activated by point mutations in bladder cancer (BC). These mutations are associated with genetically stable, Ta and low-grade BC, representing the favourable BC pathway. Conversely, FGFR3 over-expression was recently found in 40 % of muscle invasive BC. We examined FGFR3 mutation status and protein expression in patients originally diagnosed as T1. We also investigated copy-number variations in FGFR3 as a possible alternative mechanism to activate FGFR3. We included 84 patients with T1 BC as their initial diagnosis. A uropathologist reviewed the slides for grade and (sub)stage. The FGFR3 mutation status was examined by PCR-SNaPshot and FGFR3 protein expression by standard immuno-histochemistry (FGFR3-B9). Copy-number status was determined in 69/84 cases with nine probes covering nine exons of the FGFR3 gene (MLPA). Of 27 BC with FGFR3 mutations, 26 (96 %) showed FGFR3 over-expression. Of the 57 wild-type BC, 27 (47 %) BC showed over-expression. Pathological parameters significantly differed (p < 0.01) between mutant and wild-type tumours with the FGFR3 mutation pointing to more favourable BC. However, if the BC exhibited wild-type FGFR3, FGFR3 protein status had no influence on grade and (sub)stage. We found six tumours with more than or equal to three copies of FGFR3. Only 1 of 22 wild-type tumours with over-expression of FGFR3 had more than or equal to three gene copies. In initially diagnosed T1 BC, only the FGFR3 mutation was significantly associated with favourable BC disease characteristics. In addition to almost all FGFR3 mutant BC, 47 % of wild-type BC displayed FGFR3 over-expression, suggesting an alternative mechanism to activate FGFR3. Increased FGFR3 copy number was a rare event and did not account for this mechanism. Nevertheless, FGFR3 wild-type tumours with over-expression of the protein may still represent a subset that might potentially benefit from FGFR3-targeted therapy.