Effect of blue-light cystoscopy on contemporary performance of urine cytology

Objective: To evaluate the performance of urine cytology based on contemporary data, including the effect of enhanced cystoscopic techniques. Materials and Methods: Individual patient data were obtained from three prospective studies: the Photocure (PC) B305 and the PC B308 studies, evaluating the use of blue-light cystoscopy with hexaminolevulinate (BLC-H), and the Cxbladder monitoring study, evaluating the Cxbladder monitor test for the detection of recurrent urothelial carcinoma. The specificity and sensitivity of cytology in each study and for the overall cohort were calculated. Results: A total of 1 487 urine samples from 1 375 patients were included in the analysis; overall 615 tumours were detected correlating to 41% of the cytological specimens. The pooled sensitivity and specificity for cytology were 40.8% and 92.8%, respectively. The pooled sensitivity was 11.4% for low-grade/World Health Organization (WHO) grade 1 disease and 54.3% for high-grade/WHO grade 3 disease. There were no differences in cytology sensitivity based on the type of cystoscopy used, with sensitivity of 41.3% and 40.4% in white-light cystoscopy (WLC) and BLC-H, respectively. Subgroup analysis including carcinoma in situ (CIS) showed a trend towards lower cytology sensitivity in BLC-H (54.5%) vs WLC (69.2%). Conclusions: Based on analysis of contemporary data, the sensitivity of cytology for detecting high-grade tumours and CIS remains low. On a per-patient analysis, cytology sensitivity was not affected by the use of advanced cystoscopic techniques except in patients with CIS. The use of cytology as the main adjunct to cystoscopy in patients at high risk can lead to missed opportunities for early detection of recurrence and for determining which patients are not responding to intravesical therapies such as bacille Calmette-Guérin.