Utility of quantitative fluorescent in situ hybridization (FISH) to predict non-muscle-invasive bladder cancer (NMIBC) recurrence.

Abstract

286 Background: Urovysion's (Abbot Laboratories, Downers Grove, IL) FISH analysis is used to monitor bladder cancer recurrence in patients with a history of NMIBC and is often reported as a binary variable (normal/abnormal). We investigated whether the percentage of abnormal cells as determined by FISH analysis in patients with a history of NMIBC correlated with risk of recurrence. Methods: At our institution, barbotage FISH analysis is routinely done along with cystoscopy and cytology on both high risk (Ta/T1 high grade or CIS) and low or intermediate risk patients (all others) at every 3-month follow-up for the first year post-resection. We retrospectively reviewed 241 consecutive NMIBC patients and identified 399 FISH analyses for which we had one year follow-up. Normal FISH analyses were defined as 2 or fewer abnormal cells per sample. We calculated the percentage of abnormal cells and correlated that to the number of patients who had a recurrence of NMIBC as defined by positive high grade cytology or tumor on cystoscopy during the first year of follow-up. Results: The sensitivity, specificity, and positive and negative predictive values of FISH analysis if reported as a binary variable was 55, 43, 16 and 89%, respectively. Considering only those patients with abnormal FISH, the average percentage of abnormal cells for patients who were found to have NMIBC recurrence at 1 year was 38% (range 6–100) compared to 21% (range 6–100) for patients who were recurrence-free at 1 year (p<0.0001). High risk patients who recurred within 1 year had a statistically higher percentage of abnormal cells as compared to those who did not recur within 1 year (50% [range 6–100] vs. 25% [range 6– 100], respectively p=0.001). There was no difference in the percentage of abnormal cells for those patients with low or intermediate risk disease based on recurrence within 1 year (22% [range 6–100] vs. 20% [range 6–100], respectively p=0.25). Conclusions: The percentage of abnormal cells in FISH analysis correlates with risk of recurrence for patients with high risk disease and can be used to guide surveillance interval decisions in patients with no other evidence of recurrence. [Table: see text]