Testing for urinary hyaluronate improves detection and grading of transitional cell carcinoma

Abstract

Objective The purpose of this study is to establish a method for the diagnosis and grading of transitional cell carcinoma (TCC), which is responsible for 90% of bladder tumors, using a recently developed ultrasensitive assay for the measurement of hyaluronan (HA). Materials and methods Urine samples were collected prior to surgery (cystoscopy, transurethral resection for bladder cancer (TURBT), and cystectomy) in 350 patients. After the procedure, pathologic examination revealed that 160 patients had TCC. HA was measured directly in the urine by a noncompetitive enzyme-linked immunosorbent assay (ELISA)-like fluorometric assay. Using the receiver operator characteristic curve (ROC), t-test, Dunn test, Kruskal-Wallis test, and Mann-Whitney test, we evaluated the differences between groups (those with TCC vs. those without TCC). Results By analyzing the ROC curve, we chose a urinary HA cutoff value of 13.0 μg/l for indicating risk of TCC. Using the value this of 13.0 μg/l, we found that this test had an overall sensitivity of 82.3% and an overall specificity of 81.2%. The positive predictive value of this assay was 78.9%, the negative predictive negative value was 84.2%, and the predictive accuracy was 81.7%. Logistic regression analysis revealed that every 1 μg/l increase in HA increased a patient's likelihood of having TCC by 3.9%. The sensitivity of this test to detect superficial tumors was 76.6%, whereas its sensitivity for detecting invasive tumors was 94.6%. The urinary HA excretion of patients with TCC, classified according to the TNM staging system and the World Health Organization (WHO) grading system, were compared, and a significant difference was observed between the HA levels of patients with superficial tumors compared with invasive tumors ( P = 0.005) as well as between patients with low- vs. high-grade carcinomas ( P < 0.001). Patients with urinary HA levels >35 μg/l had a 4.63 times increased risk of having an aggressive, invasive, high grade tumor ( P = 0.005). Conclusions Our results support the postulate that urinary HA may be used as a tumor marker to aid in the diagnosis and grading of TCC. Additionally, more invasive tumors produce and release more HA in urine than superficial tumors, thus higher HA levels indicate more aggressive disease.