Presentation Authors: Bas van Rhijn, Laura Mertens, Amsterdam, Netherlands, Roman Mayr, Regensburg, Germany, Peter Bostrom, Turku, Finland, Mirari Marquez, Madrid, Spain, Ellen Zwarthoff, Joost Boormans, Cheno Abas, Geert van Leenders, Rotterdam, Netherlands, Stefanie Gotz, Regensburg, Germany, Simone Bertz, Erlangen, Germany, Yann Neuzillet, Joyce Sanders, Annegien Broeks, Michiel van der Heijden, Amsterdam, Netherlands, Michael Jewett, Toronto, Canada, Francisco Real, Madrid, Spain, Robert Stohr, Erlangen, Germany, Alexandre Zlotta, Toronto, Canada, Markus Eckstein, Erlangen, Germany, Yanish Scoorojebally, Paris, France, Max Burger, Wolfgang Otto, Regensburg, Germany, Francois Radvanyi, Nanour Sirab, Damien Pouessel, Paris, France, Theo van der Kwast, Toronto, Canada, Nuria Malats, Madrid, Spain, Arndt Hartmann, Erlangen, Germany, Yves Allory, Paris, France, Deric van der Schoot, Breda, Netherlands, Tahlita Zuiverloon*, Rotterdam, Netherlands
Introduction: Fibroblast Growth Factor Receptor 3 (FGFR3) is a potentially actionable target in bladder cancer (BC). FGFR3 mutations are associated with favorable prognosis in non-muscle invasive (NMI) BC and MIBC. Over-expression of FGFR3 was reported in up to 40% of FGFR3 wild-type MIBC. p53 alterations rarely coincide with FGFR3 mutations. We analyzed FGFR3 mutations, protein-expression of FGFR3 and p53 and assessed their prognostic value in a multi-center, multi-laboratory setting.
Methods: We included 1000 cN0M0, chemotherapy-naive patients who underwent radical cystectomy (RC) with pelvic node dissection. Specimens were reviewed by eight uro-pathologists. At seven laboratories, FGFR3 mutation status was examined using PCR-SNaPshot. p53 and FGFR3 expression were determined by immunohistochemistry (IHC). FGFR3 mutation status, p53 and FGFR3 protein expressions were correlated to each-other, clinico-pathological parameters and disease-specific survival (DSS).
Results: FGFR3 mutations were found in 107/1000 RCs (11%), of which 67 were S249C. Over-expression of FGFR3 was found in 279/1000 (28%) of tumors. p53 overexpression (cut-off>10%) was found in 638/926 (69%) of available cases. Among FGFR3 mutant tumors, 73% had FGFR3 over-expression. Among FGFR3 wild-type tumors, 22% had FGFR3 over-expression. FGFR3 mutations were associated with lower pT-stage (P < 0.001), lower grade (G2-WHO 1973) (P < 0.001), absence of CIS (P=0.009), pN0 (P < 0.001), normal p53 (P < 0.001) and prolonged DSS (Plog-rank=0.001). FGFR3 over-expression was associated with lower pT-stage (P < 0.001) and G2 (P < 0.001) but not with absence of CIS (P=0.860), pN0 (P=0.230), normal p53 (P=0.330) nor prolonged DSS (Plog-rank=0.204). We found no significant difference in DSS for patients with FGFR3 mutant tumors comparing normal vs over-expression of FGFR3 (Plog-rank=0.444). Furthermore, we also found no significant difference in DSS for patients with FGFR3 wild-type tumors comparing normal vs over-expression of FGFR3 (Plog-rank=0.754).
Conclusions: FGFR3 mutations identified patients with favorable BC with fewer p53 alterations at RC. FGFR3 over-expression was not associated with DSS in patients with FGFR3 wild-type tumors. Our results suggest that FGFR3 mutations (driver effect) have a distinct functional role than FGFR3 over-expression (passenger effect). Hence, patients with FGFR3 mutations may be more likely to benefit from anti-FGFR3 therapy than patients with only over-expression of FGFR3.