Detection of telomerase activity in urine as a tool for noninvasive detection of recurrent bladder tumors is poor and cannot be improved by timing of sampling.

To the Editor: Cystoscopy is considered the gold standard as a diagnostic tool for recurrent urothelial cell carcinomas (UCCs). Unfortunately, this procedure is both invasive and costly. As a noninvasive alternative, various biomarkers have been tested in the urine of patients with bladder cancer (1 ). One of the most promising markers is the presence of telomerase activity (1, 2). Following these interesting findings, we analyzed urine samples of 42 patients for the presence of telomerase activity and hTERT mRNA, which codes for the catalytic subunit of telomerase (3 ). All 42 patients were suspected for recurrent UCC by cystoscopy and were identical to the patients presented in the accompanying report (4 ). Second voided morning urines were collected before transurethral resection, processed, and analyzed as described for bladder washing samples (4 ). Because maximum sensitivity was needed, the urothelial cell enrichment procedure with magnetic beads was omitted. After transurethral resection of the suspected tumor, UCC could be confirmed by pathology in 36 patients. Of these patients, 35 and 33 urine samples could be analyzed for telomerase activity and hTERT expression, respectively. Telomerase activity was detected with the TRAPeze method (Intergen), whereas hTERT mRNA was detected with realtime quantitative PCR as described previously (5). The sensitivity in urine was poor for both methods: 10 of 35 (29%) samples were positive for telomerase activity and 8 of 33 (24%) were positive for hTERT mRNA (data not shown). Therefore, in our hands, neither the telomeric repeat amplification protocol (TRAP) nor real-time hTERT PCR were sensitive enough to replace cystoscopy for the detection of recurrent UCC. The low sensitivity we found was similar to results from Dalbagni et al. (6 ), who could detect telomerase activity in only 35% of urine samples from patients with recurrent bladder carcinomas. In the literature, however, sensitivities for the detection of telomerase activity in urine differ tremendously between studies, ranging from 0% to 95% (1, 7–9). These differences may be related in part to patient selection. Alternatively, the (occult) presence of blood in urine samples may have led to false-positive results (10 ), especially because most patients present with hematuria. Interlaboratory differences in sample storage, sample processing, and TRAP performance may also add to inconsistency. An important, often disregarded methodological aspect is the timing of sampling. Kavaler et al. (8 ) suggested that the first voided morning urine may contain more cells than the second voided urine. This may provide higher sensitivity for telomerase activity detection. To test this hypothesis, we collected first and second voided morning urines from 13 patients with confirmed recurrent UCC. Urine samples were immediately put on ice, subsequently processed, and stored within 1 h after collection. No differences in sensitivity for the detection of telomerase activity were observed between first (38%) and second (31%) voided urines (Table 1). Although more cells would be expected in the first morning urine, because of the increased contact time of urine with the bladder wall, degradation of cells, proteins, or RNA may also have occurred over this longer period of time, reducing the effect of the prolonged contact. We conclude that the low sensitivity for the detection of telomerase activity in urine of patients with bladder cancer cannot be increased by the use of the first voided morning urine. Other variables may lead to poor reproducibility between research groups and should be investigated before telomerase activity measurements in urine can be used by the urologist as a noninvasive tool for the (early) detection of recurrent UCC.