Our study was conceived within this unresolved space. We did not set out to discover another marker. Instead, we asked a more direct and, in our view, more clinically relevant question: can a well-characterized urinary tumor DNA signal be made more reliable by changing the analytical platform through which it is measured? This distinction matters. Many urine-based biomarker studies are framed around the addition of molecular targets.1,2 Our investigation focused on a different premise: that part of the failure of earlier urinary DNA assays may reflect not the absence of relevant biology, but insufficient analytical fidelity. TERT promoter and FGFR3 hotspot mutations are among the most reproducible molecular events in urothelial carcinoma of the bladder (UCB). The question was whether digital partitioning and absolute molecular quantification could extract more diagnostic information from these established alterations than conventional quantitative PCR (qPCR).
To address this question, we analyzed 239 evaluable urine samples from a prospectively assembled multicenter biobank, including 123 patients with histologically confirmed UCB and 116 patients with benign findings.3 The digital PCR-based Uromonitor assay targeted TERT promoter and FGFR3 hotspot mutations and was compared head to head with the earlier qPCR-based Uromonitor assay and with urine cytology.4 The primary endpoint was sensitivity for histologically confirmed UCB, assessed under blinded conditions and within a standardized diagnostic framework.
The result was unambiguous. Digital droplet PCR (ddPCR) achieved a sensitivity of 75.6%, compared with 51.2% for qPCR and 42.3% for urine cytology, while preserving a specificity of 94.0%. The absolute sensitivity gain over qPCR was 24.4 percentage points, and the gain over cytology was 33.3 percentage points. The positive predictive value was 93.0%, the negative predictive value was 78.4%, and the overall accuracy was 84.5%. The area under the receiver operating characteristic curve was 0.848 for ddPCR, exceeding both qPCR and cytology (vs. 0.726 and 0.660, respectively).
The key message is not simply that ddPCR performed better. The more important point is how it performed better. To our knowledge, this study provides the first clinical evidence in a urine-based DNA test for urothelial carcinoma that ddPCR can be significantly superior to qPCR in sensitivity when the same clinical question is examined within a comparable assay framework. This is not a trivial platform comparison. It is evidence that a molecular signal previously accessible only incompletely by qPCR can become clinically more visible when measured through digital partitioning. The strength of this finding is amplified by the conditions under which it was obtained. A particularly important feature of this comparison is its conservative chronology. The ddPCR assay was performed on archived urine specimens after approximately 2.5 to 4 years of frozen storage, whereas the qPCR data had been generated earlier from the same prospectively collected multicenter sample cohort. This delay is clinically relevant because prolonged storage could plausibly weaken, rather than strengthen, the detectable urinary DNA signal. The superior sensitivity of ddPCR was therefore not achieved under analytically privileged conditions. If anything, the design made the result more stringent. Equally important, the ddPCR assay achieved its superior sensitivity despite a narrower target panel. The earlier qPCR version of Uromonitor assessed TERT, FGFR3, and KRAS alterations, whereas KRAS was deliberately omitted from the ddPCR panel because prior evidence had suggested negligible diagnostic contribution in this setting.4 Thus, the improved performance was not explained by adding molecular targets. It was achieved despite removing one. The implication is clear: the dominant gain came from the measurement architecture itself, not from target expansion.
This observation has implications beyond Uromonitor. A substantial proportion of contemporary urinary DNA and RNA assays in bladder cancer diagnostics relies on qPCR or related amplification-based technologies. Our data should encourage the field to ask a direct translational question: for each of these assays, what diagnostic delta might be obtained if the same molecular targets were transferred to a ddPCR framework? The answer will differ by assay, target biology, specimen quality, and clinical setting. Digital PCR will not rescue every biomarker. But our findings demonstrate that, at least for a mutation-based urinary DNA test in UCB, platform refinement alone can move diagnostic performance from moderate sensitivity toward a range more compatible with risk-adapted clinical use.
The biological rationale is compelling; ddPCR partitions each sample into thousands of individual microreactions, allowing absolute quantification of mutant and wild-type alleles. This is particularly relevant in urine, where tumor-derived DNA is often scarce, fragmented, diluted, or masked by abundant non-neoplastic DNA. Under such conditions, the problem is not necessarily that the tumor signal is absent. The problem may be that conventional qPCR fails to detect it reliably at low allelic fractions. Digital partitioning changes that threshold. The subgroup data support this interpretation; ddPCR showed improved sensitivity across clinically relevant strata, including Ta tumors, low-grade tumors, high-grade tumors, and muscle-invasive disease. In low-grade disease, where cytology is particularly limited, ddPCR achieved a sensitivity of 64.9%, compared with 42.1% for qPCR and 22.8% for cytology. In high-grade tumors, ddPCR reached 84.8%, compared with 59.1% for both comparators. These findings are clinically meaningful because an adjunctive urinary assay has greatest value when it improves detection not only in biologically obvious tumors, but also in lesions that conventional noninvasive assessment may miss. The multivariable analyses further strengthened the clinical interpretation. After adjustment for center, study period, age, sex, smoking status, hematuria, and urinary infection parameters, ddPCR remained strongly and independently associated with histologically confirmed UCB. Its effect size exceeded that of the qPCR assay, whereas cytology did not retain independent significance in the final model. This finding argues against ddPCR positivity as a mere epiphenomenon of urinary inflammation, hematuria, or center-level variation, and instead supports its interpretation as a robust disease-associated molecular signal independently linked to histologically confirmed UCB.
Several details deserve particular attention. All patients with combined TERT and FGFR3 positivity by ddPCR had histologically confirmed UCB. The assay preserved high specificity despite substantially higher sensitivity, which is critical in surveillance pathways where false-positive results may trigger additional cystoscopy, imaging, operative procedures, anxiety, and cost. One patient with initially benign histology and isolated TERT promoter positivity subsequently developed a T1 high-grade tumor within three months. This observation cannot establish molecular lead time, but it raises an important hypothesis: in selected patients, urinary tumor DNA may precede visible or histologically confirmed recurrence. At the same time, this study should not be read as a claim that ddPCR can replace cystoscopy. Such an interpretation would be premature and clinically unsound. Digital PCR still missed a subset of tumors, including some high-grade and stage T2 or higher tumors. The proper conclusion is more precise: ddPCR may offer a substantially stronger molecular adjunct to established diagnostic pathways. Its most plausible near-term role lies in risk-adapted surveillance, triage of diagnostically equivocal situations, and refinement of follow-up intensity in patients for whom repeated invasive procedures are burdensome.
The limitations define the next stage of development. The cohort was not a pure surveillance cohort, since most patients with cancer had primary rather than recurrent tumors. The analysis relied on archived samples, and one center was excluded to preserve analytical comparability. Interlaboratory reproducibility, prospective longitudinal performance, cost-effectiveness, scalability, and integration into real-world decision algorithms remain to be established. These are not peripheral issues. They will determine whether ddPCR becomes a robust clinical tool rather than an elegant analytical advance. For that reason, our group is pursuing larger prospective validation in more than 700 patients, with optimized urine processing, improved DNA extraction, and a stronger focus on preanalytical reproducibility. The next generation of studies must go beyond diagnostic accuracy alone. It must define how ddPCR behaves in representative surveillance populations, how it performs over time, how it interacts with cystoscopy and cytology, and whether it can safely inform individualized follow-up strategies. The broader implication of this work is that urinary biomarker development may need to recalibrate its priorities. The field has often assumed that progress requires broader panels, additional markers, or more complex composite algorithms. Our study suggests another path: better measurement of biologically established markers. In urothelial carcinoma, TERT and FGFR3 alterations are not obscure signals. They are recurrent, biologically plausible, and clinically relevant; ddPCR appears to recover more of their diagnostic value.
In conclusion, this multicenter validation study establishes ddPCR-based Uromonitor as a technically mature and clinically promising molecular adjunct for UCB detection.3 Its principal contribution is not the displacement of cystoscopy, but the demonstration that absolute digital quantification can substantially improve the performance of a mutation-based urinary assay under real-world multicenter conditions, even when applied to archived material and a narrower target panel. If confirmed in larger prospective surveillance cohorts, this approach may help move UCB follow-up toward a more individualized, molecularly informed, and patient-centered diagnostic pathway.
Written by: Anton Kravchuk,1 Julio Ruben Rodas Garzaro,1 Stephan Siepmann,1 Christian Gilfrich,1 Ingmar Wolff,2 Sabine Brookman-May,3 Thorsten Ecke,4 and Matthias May1
- Department of Urology, St. Elisabeth Hospital Straubing, Medical Campus Lower Bavaria (MCN), Straubing, Germany
- Department of Urology, University Medicine Greifswald, Greifswald, Germany
- Department of Urology, LMU University Hospital, Ludwig Maximilian University of Munich, Munich, Germany
- Department of Urology, Helios Hospital, Bad Saarow, Germany
- Gontero P, Birtle A, Capoun O, Compérat E, Dominguez-Escrig JL, Liedberg F, Mariappan P, Masson-Lecomte A, Mostafid HA, Pradere B, Rai BP, van Rhijn BWG, Seisen T, Shariat SF, Soria F, Soukup V, Wood R, Xylinas EN. European Association of Urology Guidelines on Non-muscle-invasive Bladder Cancer (TaT1 and Carcinoma In Situ)-A Summary of the 2024 Guidelines Update. Eur Urol. 2024 Dec;86(6):531-549. doi: 10.1016/j.eururo.2024.07.027. Epub 2024 Aug 17. PMID: 39155194.
- Siguencia F, Heard J, Smith SM, Pagano I, Furuya H, Wu X, Ahdoot M, Rosser CJ. Diagnostic Performance of Urinary Biomarker Tests in Detecting Bladder Cancer: A Systematic Review, Meta-analysis, and Network Meta-analysis. Urology. 2026 Apr;210:145-152. doi: 10.1016/j.urology.2026.01.035. Epub 2026 Feb 4. PMID: 41651230.
- Kravchuk A, Garzaro JRR, Wirtz R, Wolff I, Koch S, Schlomm T, Rabien A, Meisl C, Rong D, Brás JP, Vinagre J, Otto T, Barski D, Gössl A, Siepmann S, Gilfrich C, Brookman-May S, Ecke T, May M. Digital PCR-Based Uromonitor for Molecular Detection of Bladder Cancer: A Multicenter Validation Study Comparing Quantitative PCR and Urine Cytology. Clin Genitourin Cancer. 2026 Apr 1:102554. doi: 10.1016/j.clgc.2026.102554. Epub ahead of print. PMID: 42049549.
- Wolff I, Kravchuk AP, Wirtz RM, Schlomm T, Rabien A, Rong D, Hofbauer SL, Labonté FK, Barski D, Otto T, Gössl A, Brookman-May SD, Gilfrich CP, Ecke TH, May M. Real-world performance of Uromonitor® in urothelial bladder cancer detection: a multicentric trial. BJU Int. 2024 Dec;134(6):992-1000. doi: 10.1111/bju.16450. Epub 2024 Jun 24. PMID: 38923777.