Prognostic significance of matrix metalloproteinase-1 and tissue inhibitor of metalloproteinase-1 in voided urine samples from patients with transitional cell carcinoma of the bladder.

PURPOSE To study the role of urinary matrix metalloproteinase-1 (MMP-1) and tissue inhibitor of metalloproteinase-1 (TIMP-1) in bladder cancer and their relationship to tumor progression. EXPERIMENTAL DESIGN MMP-1 and TIMP-1 were measured by ELISA in urine samples from 131 patients with bladder tumors (7 cis, 74 Ta, 29 T1, and 21 T2-T4; 46 G(1), 41 G(2), and 37 G(3)), 5 patients with prostate cancer, 33 patients with benign lower urinary tract disorders, and 36 healthy volunteers. Complete clinical data were available for 100 patients with bladder cancer with a median follow-up time of 24 months (range: 4-39 months). RESULTS MMP-1 was detected in urine samples from 21 of 131 (16%) patients with bladder cancer but was undetectable in samples from all other groups (P < 0.0001). Urinary MMP-1 was detected in a higher percentage of patients with T2-T4 tumors and G(3) tumors than patients with cis/Ta/T1 or G(1)-G(2) tumors (P = 0.04 and P = 0.0074, respectively). Patients with detectable concentrations of urinary MMP-1 had higher rates of disease progression (P = 0.04) and death from bladder cancer (P = 0.02) than patients with undetectable urinary MMP-1. All patient groups had higher urinary TIMP-1 concentrations than healthy volunteers (P = 0.02). Patients with muscle-invasive tumors had higher concentrations of urinary TIMP-1 than patients with cis/Ta/T1 tumors (P = 0.037), but there was no association between TIMP-1 and tumor grade. Urinary TIMP-1 levels strongly correlated with tumor size (P = 0.0002). Progression-free survival rates were lower for patients with urinary TIMP-1 concentrations above the median (1.8 ng/ml, P = 0.04), but urinary TIMP-1 levels were not related to disease-specific survival. Patients with T2-T4 tumors and G(3) tumors had significantly lower urinary MMP-1:TIMP-1 ratios than patients with Ta/T1 bladder tumors (P = 0.039) or G(1)-G(2) tumors (P = 0.0415). CONCLUSIONS Where urinary MMP-1 is detectable, the patient is more likely to have a bladder tumor of advanced stage or grade and may be at increased risk of disease progression and death of bladder cancer. The relationship between urinary TIMP-1, muscle-invasion, and disease progression in bladder cancer is at variance with its role as an inhibitor of MMPs and warrants additional evaluation.