Urinary cytokines during intravesical Bacillus Calmette-Guerin therapy for superficial bladder cancer: Processing, stability and prognostic value

Purpose: An accurate prognostic indicator to identify nonresponding patients with superficial transitional cell carcinoma of the bladder at an early stage of intravesical bacillus Calmette-Guerin (BCG) therapy is urgently needed. Materials and Methods: The processing conditions and stability of several BCG induced urinary cytokines were analyzed, as was the possible correlation between these cytokines (indicating immune responsiveness to BCG) and bladder tumor recurrence. We studied 23 patients with superficial transitional cell carcinoma of the bladder. Monitoring was performed by serial collection of urine during the first 24 hours after each of the 6 consecutive weekly intravesical BCG instillations. Baseline pre-therapy cytokine levels were 3.9 +/- 4.7 pg./mu mol. creatinine for interleukin-1 beta, 0.0 +/- 0.0 units per mu mol. creatinine for interleukin-2, 8.9 +/- 12.9 pg./mu mol. creatinine for interleukin-6 and 0.1 +/- 0.2 pg./mu mol. creatinine for tumor necrosis factor-ct (all measured by enzyme-linked immunosorbent assay). To investigate the correlation between interleukin-2 and bladder tumor recurrence, patients were stratified into 2 groups based on an early (6 months or less) or late (greater than 6 months) recurrent tumor. For each patient the highest cytokine value measured during the 6-week BCG treatment course was evaluated. Results: The results were positive if the level in urine exceeded 0.34 units interleukin-2 per mu mol. creatinine. A significant correlation between urinary interleukin-2 and tumor recurrence was found (p = 0.003, 23 patients). Of the studied cytokines obtained from BCG treated patients, interleukin-1 beta, 2 and 6 but not tumor necrosis factor-alpha were stable in urine at 4C and 20C. At 37C all cytokines were unstable. Interferon-gamma could only be detected in immediately dialyzed urine and its occurrence correlated most with that of interleukin-2. Processing of urine by centrifugation to remove leukocytes immediately after collection was not required for reliable measurements of interleukins-2 and 6. Based on these results interleukins-2 and 6 were preferred for extensive monitoring of the BCG induced immune reaction. Conclusions: Our study provides significant evidence for a correlation between urinary cytokine induction and clinical response following intravesical BCG therapy. Particularly, monitoring of interleukin-2 may have the potential for prognostic value provided that strict precautions regarding urine collection, such as maximal 2-hour sampling and immediate cooling, are taken.