Continuing Education
for Urology & GU Oncology Clinicians

Washington, DC (  On the final day of the Society of Urologic Oncology Meeting (SUO 2019), the six best abstracts submitted were selected for presentation. One of these highlighted studies from Memorial Sloan Kettering Cancer Center (MSKCC) evaluated the differential response of high grade non-muscle invasive bladder cancer to Bacillus Calmette-Guerin (BCG) on the basis of FGFR3 alterations. Dr. Timothy Clinton, a urologic oncology fellow at MSKCC, presented the group’s findings.

Recently, the FDA approved Erdafitinib, a fibroblast growth factor receptor (FGFR)3 kinase inhibitor, for use in metastatic urothelial carcinoma on the basis of phase II results published this summer in NEJM. In that publication, Loriot et al. demonstrated a 59% response rate in patients with unresectable or metastatic urothelial carcinoma.1 These results have driven interest in expanding the use of FGFR3 kinase inhibitors and looking to FGFR3 alterations for prognostication. Specifically, FGFR3 mutations have been previously associated with lower response rates to immune checkpoint inhibition, are more commonly identified in non-muscle invasive bladder cancer which is specifically less likely to progress. For these reasons, Dr. Clinton postulated that FGFR3 alterations may be able to predict BCG response.

To that end, Dr. Clinton and colleagues identified treatment naïve patients with high-grade nonmuscle invasive bladder cancer (NMIBC) enrolled in a prospective protocol that captured targeted exon sequencing on pretreatment tumor DNA and matched germline DNA. Within the MSKCC cohort there were over 1,400 bladder specimens with a quarter demonstrating genomic alterations in FGFR3. Of these patients, 119 patients with high-grade NMIBC underwent BCG treatment with 51containing a FGFR3 alteration (43%); 39 patients (76%) were cTaHG and 12 (24%) were cT1HG at diagnosis. Interestingly, no patients with CIS had alterations in FGFR3.

Dr. Clinton proceeding by highlight the differential response to BCG between those who were FGFR3 wild-type and those with alterations. Overall, the median follow up was 60 months (IQR 32-75). At that time 39% of the entire cohort was free from high-grade recurrence with a significant difference between the two groups. 65% of wild-type patients were free from high-grade recurrence, were as only 39% of the FGFR3 alteration group was similarly recurrence-free(p<0.05). There was no difference in progression, however, the rate was very low in both groups. Looking more in depth into the FGFR3 altered tumors, the group looked for additional mutations that corresponded to BCG response, identifying significant co-occurrence of FGFR3 mutations with cell cycle genes including CDKN2A (p=0.03) and MDM2 (p=0.03) and no FGFR3 mutated tumor contained an ERBB2 mutation.

Overall, Dr. Clinton concluded that while FGFR3 has been historically associated with favorable prognosis, FGFR3 altered tumors are associated with high recurrence rates following BCG. Given these findings and the recent BCG shortages, patients with these alterations may benefit from alternative therapy with FGFR3 kinase inhibitors. The study represents a major step in predicting BCG response rates and FGFR3 profiling could represent a novel path for personalized medicine in the future. Clinical trials for FGFR3 in NMIBC are currently underway with results pending.


Presented by: Timothy Clinton, MD, Urologic Oncology Fellow at Memorial Sloan Kettering Cancer Center

Written by: Adrien Bernstein, MD, Society of Urologic Oncology Fellow, Fox Chase Cancer Center, Fox Chase Cancer Center, Philadelphia, PA at the 20th Annual Meeting of the Society of Urologic Oncology (SUO), December 4 - 6, 2019, Washington, DC