Bacillus Calmette-Guerin (BCG) is the gold standard treatment for high-grade nonmuscle invasive (NMI) bladder cancer (BC) and for the in situ carcinoma. Previously, we have demonstrated that BCG induces death of the murine BC cell line MB49, followed by the down-regulation of FGFR3 expression. In vivo, using a murine orthotopic BC model we showed that tumor growth inhibition in response to BCG therapy was associated with FGFR3 down-regulation. The objective was to a) study in vitro the relationship between the expression of FGFR3 and cell viability in response to BCG treatment in human BC cell lines; b) using tumor samples of patients with BC, develop an ex vivo bioassay to analyze FGFR3 modulation in response to BCG and its association to patient response to this immunotherapy. Materials and Methods: a) Human BC cell lines RT4, J82 and T24 treated with +/- BCG (2mg/ml). Cell viability was determined by MTS and FGFR3 expression by quantitative PCR (qPCR). b) Bladder tumors from patients treated in the “Instituto Roffo” (n=33) were enzymatically disaggregated and cells were divided in two cultures: one was treated with medium and was considered as FGFR3 basal expression and the other was treated with BCG (2mg/ml) for 24hs. Then, cultures were collected and FGFR3 expression was quantified by qPCR. FGFR3 modulation by BCG in the bioassay was associated with patient response after BCG therapy. Results: a) BCG increased both, FGFR3 expression and cell viability in RT4 cells. In J82 cell line, FGFR3 was not modified and in T24 it was reduced by BCG treatment, associated with a decrease in cell viability. b) Regarding the ex vivo designed bioassay, we observed a reduction or no variation in FGFR3 expression after BCG
treatment in the 58% of samples (19/33). According to the invasion status, BCG produced diminution or non-variation in 54% of NMI tumors (13/24), been 53% (8/15) low-grade and 60% (5/9) high-grade; and 66.7% (6/9) MI tumors. Five of the nine patients with high-grade NMI tumors were treated with BCG. Two of them
showed a reduction of FGFR3 expression in the bioassay and remained free of disease for at least 2 years. The other three patients presented an increase in FGFR3 and progressed in their disease ending in radical cystectomy. Conclusion: Based on previous and current results, we can suggest that the increase in FGFR3 expression could be related to a greater viability of BC cells and could be used as a predictive marker of response to BCG therapy. Its modulation in response to BCG occurs not only in mice, but also in human BC cell lines and in near 60% of tumor samples from patients analyzed in the bioassay. A large number of high-grade NMI tumor samples and the follow-up of those patients will be necessary to establish the predictive role of FGFR3 and the usefulness of our bioassay as a marker of response to BCG treatment.