Impact of variant microscopic interpretation of the uCyt+ immunocytological urine test for the detection of bladder cancer.

ABSTRACT

BACKGROUND: Urinary marker tests for bladder cancer (BC) detection and surveillance represent a desirable approach to diagnosis and follow-up. The SCIMEDX uCyt+ assay detects antigens expressed by BC cells (mucin glycoprotein and carcinoembryonic antigen) using green and red fluorescence and is interpreted according to specific manufacturer's recommendations. In the present study, we evaluated divergent approaches of numeric and morphological analysis of uCyt+ to generate a rationale for alternative test interpretation strategies. METHODS: A total of 444 patients with hematuria and without history of BC underwent uCyt+ analysis, cystoscopy and histological examination of tissue biopsies. Beside positive cells according to the manufacturer's definition (definitely positive cells, DPC), (i) cells showing borderline character (borderline cells, BLC), and (ii) cells with staining present below defined border (subliminal cells, SLC) were included into the analytical algorithm. Different cut-off levels for cell counts (>0, ≥3, ≥5) were evaluated separately with regard to their diagnostic accuracy. Moreover, the influence of clinical factors on test results were evaluated. RESULTS: Adding BLC at a cut-off of ≥3 cells resulted in Area Under the Curve (AUC) of 0.70 for green and 0.77 for red fluorescence, respectively. Adding SLC led to reduced AUC (0.62 and 0.73, respectively). Male gender was significantly associated with false positive results in the "best AUC" groups (P = .0101). No further correlations to clinical influencing factors were observed. CONCLUSIONS: Adding microscopic BLC as test-positive and adjusting cut-off level for the interpretation of uCyt+ may improve assay performance independent of clinical factors.