Objectives.

To evaluate the diagnostic characteristics of the urinary measurement of cytokeratin tissue polypeptide-specific antigen (TPS) for the detection of bladder cancer.

Methods.

Three hundred thirty-five individuals in five groups were studied: group 1, subjects with microhematuria under suspicion for primary bladder cancer; group 2, patients being followed up with scheduled cystoscopic examinations; group 3, patients in follow-up receiving chemotherapy instillations; group 4, patients with other urologic diseases; and group 5, healthy subjects. Urine samples belonging to subjects from groups 1, 2, and 3 were collected immediately before cystoscopy. Additionally, patients from groups 2 and 3 were monitored with urinary TPS for a minimum period between two cystoscopies. TPS was measured by an enzyme immunosorbent assay.

Results.

Receiver operating characteristic analysis gave a sensitivity of 64% and a specificity of 84% at a threshold value of 279 U/L. The positive and negative predictive value was 66% and 82%, respectively; accuracy was 77%. TPS could discriminate the presence of bladder tumor sooner than the scheduled cystoscopies in 9 of 19 follow-up patients with recurrence. False-positive results during follow-up were found in 112 urine samples, one third of which were associated with urinary tract infections. TPS did not appear to be specific for bladder cancer, with elevated results in 45% of patients from group 4, which might lead to clinical misinterpretation of urinary TPS results.

Conclusions.

Urinary TPS might provide additional information for the detection of bladder cancer as an adjunct to cystoscopy. Considering the false-positive rates, different urologic diseases should be ruled out before making clinical decisions on the basis of elevated urinary TPS results.