Evaluation of a new quantitative point-of-care test platform for urine-based detection of bladder cancer
Ritter R1, Hennenlotter J1, Kühs U1, Hofmann U1, Aufderklamm S1, Blutbacher P1, Deja A1, Hohneder A1, Gerber V1, Gakis G1, Stenzl A1, Schwentner C1, Todenhöfer Urol Oncol. 2013 Dec 11. pii: S1078-1439(13)00404-3. doi: 10.1016/j.urolonc.2013.09.024. [Epub ahead of print]

Author information

1Department of Urology, Eberhard-Karls University of Tuebingen, Tuebingen, Germany. 2Department of Urology, Eberhard-Karls University of Tuebingen, Tuebingen, Germany. Electronic address: tilman.todenhoefer@med.uni-tuebingen.de.

Abstract

OBJECTIVE: Several commercial point-of-care (POC) tests are available for urine-based detection of bladder cancer (BC). However, these tests are restricted to dichotomized results (positive or negative), which limits their diagnostic value. Quantitative protein-based tests offer improved risk stratification but require complex methods restricted to specialized centers. Recently, the first quantitative POC system based on the detection of cytokeratin fragments became available. The aim of the study was to evaluate the diagnostic accuracy of this quantitative POC test.

PATIENTS AND METHODS: A total of 198 patients having symptoms suspicious for BC were included. All patients received urethrocystoscopy and upper-tract imaging. Urine samples were analyzed by the urine BC antigen (UBC) rapid POC system and evaluated both visually and quantitatively using the concile Omega 100 POC reader. For visual evaluation, different thresholds of band intensity for considering a test positive were applied. Moreover, the UBC enzyme-linked immunosorbent assay (ELISA), urine cytology, and the nuclear matrix protein 22 BladderChek were performed. Sensitivities and specifities were calculated by contingency analyses. Optimal cutoffs of quantitative tests were determined by receiver operating characteristic curves.

RESULTS: A total of 61 patients (30.8%) were diagnosed with BC. Visual evaluation of the UBC revealed sensitivities of 38.1% to 71.4% with corresponding specificities of 54.1% to 89.1%, dependent on the threshold of band intensity applied. The quantitative UBC rapid showed a sensitivity of 66.7% and a specificity of 69.1% at optimal cutoff (area under the curve = 0.68). A constant increase of both the probability of BC and high-risk BC with increasing UBC rapid values was observed. UBC concentrations determined by the reader significantly correlated with the UBC ELISA (P<0.001). The UBC ELISA, the nuclear matrix protein22 BladderChek and cytology showed sensitivities of 47.6%, 16.4%, and 51.7% with specificities of 90.9%, 95.3%, and 78.1%, respectively.

CONCLUSION: The UBC rapid in combination with a quantitative POC-reader system for the first time enables quantitative determination of a BC marker under POC conditions. Diagnostic accuracy is at least equivalent to elaborate ELISA-based measurement. The quantitative use of the UBC rapid test facilitates risk prediction compared with conventional nonquantitative dichotomized POC testing.