Alterations in DNA damage response (DDR) genes, excluding ATM, correlate with improved outcomes in relapsed/advanced urothelial cancer (UC).
Across three independent cohorts totaling 301 patients, overall survival (OS) was longer in patients with DDR mutations than in those without DDR mutations, and a trend was observed for longer OS with an increasing number of DDR mutations, reported Monika Joshi, MD, at the Genitourinary Cancers Symposium in San Francisco.
DDR gene defects play an important role in UC tumorigenesis. A plethora of DDR mutations exist, each of which has potentially distinct implications on the functional impact of DDR proteins, she said. However, the prognostic and predictive roles of DDR gene alterations in patients with advanced UC remain unclear.
“Previously, we pooled a dataset between three institutions and looked at survival, and found that mutations in ATM and Rb1 were negative prognostic indicators,” said Dr. Joshi, associate professor of medicine and genitourinary oncologist at Penn State Cancer Institute, Hershey. “We performed this study because we wanted to expand our knowledge and further validate outcomes in patients with relapsed or advanced UC who have DNA damage repair gene defects excluding ATM abnormality.”
The investigators used 81 patients from the City of Hope, Cleveland Clinic, and Penn State who had FoundationOne tumor tissue genomic sequencing as a discovery cohort. Findings were validated in two separate cohorts, one consisting of 91 patients from The Ohio State University with FoundationOne testing and another cohort of 129 patients with relapsed/refractory UC from The Cancer Genomic Atlas. OS was measured from time of initial UC diagnosis to death or last follow-up.
Also see: Pyruvate kinase appears to be bladder cancer marker
A panel of 32 DDR genes, excluding ATM, was used for analyses. Most of the patients in all three cohorts had relapsed UC.
DDR mutations were present in 76.5% (62/81), 40.7% (37/91), and 51.2% (66/129) of patients in the three datasets. DDR mutations were associated with longer OS in all three datasets.
In the discovery cohort, median OS was 50.8 months for those with DDR mutations and 22.7 months in those without DDR mutations, corresponding to an adjusted hazard ratio (HR) for death of 0.39 (95% CI: 0.21–0.73, p=.01).
In validation cohort 1, median OS was 60.7 months versus 30.7 months in patients with and without DDR mutations, respectively, translating to an adjusted HR for death of 0.51 (95% CI: 0.26–1.03, p=.06) in those with DDR mutations.
In validation cohort 2, median OS was 21.8 months versus 19.4 months in patients with and without DDR mutations, respectively, corresponding to an adjusted HR for death of 0.62 (95% CI 0.39–0.97, p=.08).
Continue to the next page for more.Across the three cohorts, the OR for death was 0.61 (95% CI: 0.38-0.98) for any DDR mutation compared to no DDR mutation.
Patients with DDR mutations were more likely to have an objective response to platinum-based treatment than those without DDR mutation, and the odds increased with more DDR mutations. Among 144 patients pooled from the three cohorts who received any platinum-based treatment, the objective response rates were 29% in those without a DDR mutation, 39% in those with one or two DDR mutations, and 60% in those with three or more DDR mutations. For any DDR mutation, the odds ratio for objective response was 1.81 (95% CI: 0.85-3.92) compared with no DDR mutation, and for those with three or more DDR mutations, the OR for an objective response was 3.65 (95% CI: 0.91-14.7) compared with no DDR mutations.
Similarly, the risk of death was lowest among patients with three or more DDR mutations (OR: 0.49; 95% CI: 0.19-1.27) compared with no DDR mutations.
The findings suggest that presence of DDR mutations is correlated with improved outcome in relapsed/advanced bladder cancer.
“Our results from a separate retrospective study... suggest that patients with DDR mutations have better outcomes with immunotherapy. However, the presence of ATM alteration was again suggestive of poor prognosis in that study.
Read: Deep-learning algorithm developed for bladder Ca detection
“Hence, one way to possibly treat this particular patient population would be to think about targeting the ATM pathway with an inhibitor,” Dr. Joshi said.
Dr. Joshi’s institution has received funding from AstraZeneca, and several of her co-authors have disclosures with Foundation Medicine and one or more pharmaceutical companies.