BTA Quantitative Assay and NMP22 Testing Compared with Urine Cytology in the Detection of Transitional Cell Carcinoma of the Bladder

PURPOSE: Both BTA TRAK and NMP22 urine concentrations have shown a sensitivity superior to urine cytology in the detection of bladder cancer. We compared these tumor markers with urine cytology performed on 3 consecutive samples and evaluated by an expert cytopathologist. PATIENTS AND METHODS: The investigations were conducted on 94 patients undergoing a diagnostic cystoscopy for a high suspicion of bladder cancer (group 1) and on 102 patients with previous history of transitional cell carcinoma awaiting a follow-up cystoscopy (group 2). Biopsy specimens were obtained also from tumor negative patients. Immunoassays for BTA TRAK and NMP22 were carried out according to standard methods. The choice of the cut-off was based on the ground of sensitivity and specificity curves intersection. Urine cytology results were expressed as positive, negative and 'dubious'. RESULTS: Overall sensitivity was 56% for NMP22 (cut-off 11 U/ml) and 57% for BTA TRAK (cut-off 60 U/ml). When dubious results were considered as positive cases, urine cytology achieved a sensitivity of 73.3%. Assuming dubious cases as negative results, urine cytology sensitivity resulted 59.3%. When the 2 groups of patients were evaluated separately with different cut-off, there was no significant gain in sensitivity for BTA TRAK and NMP22 over urine cytology. CONCLUSIONS: Urine cytology performed on 3 samples showed the highest sensitivity and specificity. The diagnostic advantage of urine cytology over BTA TRAK and NMP22 was maintained when patients were stratified by tumor grade.