Bladder tumor markers for monitoring recurrence and screening comparison of hyaluronic acid–hyaluronidase and BTA‐Stat tests

Abstract

BACKGROUND One of the goals of a noninvasive test for bladder carcinoma screening would be to reduce surveillance cystoscopies among patients with a history of bladder carcinoma. In addition, an accurate bladder carcinoma marker could be used to screen a high‐risk population. The authors examined the efficacy of the hyaluronic acid–hyaluronidase (HA‐HAase) and BTA‐Stat tests to detect and predict bladder carcinoma recurrence and tested their specificity for bladder carcinoma screening. METHODS Over a four year period, the authors prospectively collected 225 urine specimens from 70 bladder carcinoma patients and analyzed them by the HA‐HAase test. Tumors were identified during 178 visits, and in 47 specimens there was no evidence of disease (NED). Twenty six of these 70 patients were randomly selected to have the BTA‐Stat test (111 surveillance visits). In a separate study, 401 former Department of Energy (DOE) workers, who are likely to be at a higher risk for bladder carcinoma, were screened by the HA‐HAase and BTA‐Stat urine tests. RESULTS The HA‐HAase test had an approximately 91.0% sensitivity, 70% specificity, 87% accuracy, 92% positive predictive value (PPV), and 67% negative predictive value (NPV) in the 70 bladder carcinoma patients. There were 14 false‐positives; however, 6 of these had recurred in approximately 5 months. Only 4 out of 33 NED cases recurred in that time period (chi‐square = 5.43; degrees of freedom [DF] = 1; P = 0.0198). Thus, a false‐positive HA‐HAase test carried a significant risk of recurrence within five months (relative risk [RR] = 3.5; odds ratio [OR] = 5.44). In a direct comparison, the HA‐HAase and BTA‐Stat had 94% and 61% sensitivity, 63% and 74% specificity, 87% and 64% accuracy, 89% and 88% PPV, and 77% and 38% NPV, respectively. While 6 of the 10 false‐positive on the HA‐HAase test recurred in 5 months (chi‐square = 9.6; DF = 1; P = 0.004), only 1 of the 7 false‐positives on the BTA‐Stat test recurred in that time period (chi‐square = 0.096; DF = 1; P = 0.756). The RR and OR for the HA‐HAase test were 10.2 and 24, and for the BTA‐Stat, 1.4 and 1.5, respectively. In the DOE worker screening study, the HA‐HAase and BTA‐Stat had 14% (56 out of 401) and 16.7% (67 out of 401) positive rates, respectively. Sixty three percent of the positives on the BTA‐Stat test, but only 25% of the positives on the HA‐HAase test, had benign urologic conditions. None of the biomarker positive cases with clinical follow‐up (n = 29) had evidence of bladder carcinoma. CONCLUSIONS The HA‐HAase test is efficient and superior to the BTA‐Stat for detecting and predicting bladder carcinoma recurrence. Noninvasive tests with low false positive rates could be used for bladder carcinoma screening in high‐risk populations (e.g., those with occupational exposure to carcinogens or smokers). Cancer 2002;95:61–72. © 2002 American Cancer Society. DOI 10.1002/cncr.10652