The fundamental biology of ctDNA indicates that tumour burden correlates directly with ctDNA concentrations, as established through mathematical modeling.2 Therefore, clinical scenarios involving low or minimal tumour burden, such as non-muscle invasive bladder cancer (NMIBC), make ctDNA identification more difficult. Wang et al. examined the utility of ctDNA in NMIBC: among 23 patients, 35% were ctDNA-positive, yet 39% underwent cystectomy, resulting in pathological upstaging.3 This implies that while ctDNA exhibits high specificity for aggressive disease biology, the 65% ctDNA-negative rate may reflect assay limitations rather than genuine disease absence. Therefore, under a clinical perspective, a negative ctDNA result in low-burden conditions does not confirm disease absence, highlighting a technical limitation that should warrant clinicians and might worry patients.
Another consideration is when we consider metastatic disease, where tumour-informed assays may provide decreasing value.1 As malignancies evolve and acquire mutations beyond those in the primary specimen, personalised panels become progressively less representative of actual disease burden. The biology of cancer evolution, coupled with the relevant costs of tumour-informed assays, suggests that tumour-agnostic assays may prove to be more practical for advanced disease surveillance, even if prospective studies are necessary to confirm it.
The recently presented “rescue” Imvigor011 trial (ESMO2025) represents the first prospective, randomised confirmation of a ct-DNA directed therapeutic approach in urothelial cancer and the first demonstrating a clear survival advantage with atezolizumab, a result that predecessor trials failed.4 The Imvigor010 trial demonstrated equivalent 18-month disease-free survival (DFS) rates (52%) for adjuvant atezolizumab compared to observation in high-risk patients following radical cystectomy or nephroureterectomy;5 nonetheless, post-hoc ctDNA analysis showed improved OS with atezolizumab in ctDNA-positive patients (HR 0.59), indicating that biomarker-guided selection could induce therapeutic benefit. Similarly, the ALBAN trial did not achieve its primary event-free survival endpoint (p=0.91) when comparing intravenous atezolizumab combined with intravesical BCG against BCG alone in BCG-naïve, high-risk NMIBC, while also showing substantially higher adverse event rates and treatment discontinuation with the combination treatment (28.6% vs 8.8%).6
In IMvigor011, MIBC patients without radiographic disease evidence who tested ctDNA-positive during one-year surveillance were randomised to intravenous atezolizumab or placebo every four weeks up to one year, while persistently ctDNA-negative patients remained untreated. The atezolizumab group achieved the primary endpoint of investigator-assessed DFS (9.9 vs 4.8 months, HR 0.62, p=0.005), median overall-survival (OS) was 32.8 vs 21.1 months, favoring atezolizumab (HR 0.59, p=0.01), representing a 41% OS improvement.4 For patients maintaining ctDNA-negative status, DFS reached 95% at one-year monitoring completion and 88% at two years. This landmark finding demonstrates that molecularly guided adjuvant immunotherapy can substantially enhance survival outcomes.
It is noteworthy that roughly 40% of patients initially ctDNA-negative at baseline converted into positive status during monitoring. This underscores that ctDNA is dynamic rather than binary, and that molecular relapse may emerge beyond the initial testing. Single-time-point assay, therefore, might risk missing late molecular recurrences; therefore, serial ctDNA becomes crucial for precise risk stratification and timely treatment intervention. ctDNA emergence patterns may eventually optimize surveillance intervals and therapeutic decision timeframes, improving patient-tailored medicine, enabling treatment escalation only in those showing molecular evidence of residual or recurrent disease while minimising overtreatment in patients unlikely to benefit. This would shift the treatment paradigm to a biology-driven rather than stage-driven therapy, aligning therapies and their intensity with real-time tumour dynamics.
Despite these advances and a lot of enthusiasm, a critical question remains concerning potential undertreatment. Imvigor011 compared single-agent checkpoint inhibition against placebo in ctDNA-positive patients, whereas contemporary metastatic bladder cancer treatment utilizes enfortumab vedotin plus pembrolizumab with substantially superior outcomes.7 A natural question, therefore, arises: whether ctDNA-positive patients should need a combination therapy instead of monotherapy, paralleling the debate regarding adjuvant pembrolizumab in renal cell carcinoma,8 where optimal treatment intensity for high-risk patients remains undefined. At the same time, we should better define the ideal de-escalation strategy for persistently negative ctDNA individuals. The ongoing four-arm MODERN trial directly investigates these gaps by randomizing ctDNA-positive patients to single-agent nivolumab versus combined nivolumab plus relatlimab therapy, while ctDNA-negative patients are randomised to nivolumab versus placebo.9 This trial may shed some light on both treatment escalation and de-escalation based on ctDNA status, but in the meantime, pending MODERN results, we lack conclusive evidence regarding optimal therapeutic intensity.
ctDNA performance might increase the overall demand for expensive treatments. Colorectal cancer modeling provides some insights about this. Sensitivity analysis of prediction models demonstrated that ctDNA-guided therapy achieved cost-effectiveness only if test costs fell below 1500$, or if ctDNA was predictive of treatment response, or if test performance improved substantially.10 The DYNAMIC study in stage II colon cancer demonstrated that at five years, ctDNA-guided therapies reduced the adjuvant chemotherapy use by approximately 40% at a median follow-up of 59.7 months, without compromising overall survival (93.8% vs 93.3%, HR 1.05, p=0.887) or recurrence-free survival (88% vs 87% for ctDNA-guided and standard treatments, respectively).11 Bladder cancer represents a trickier challenge, considering the greater biological heterogeneity and the evolving therapeutic landscape; additionally, concerns about access to personalised healthcare for low-income countries remain unaddressed.
On top of biological and cost-related considerations, we must consider patient experience: does earlier molecular detection of disease recurrence improve both survival and quality of life? Having an earlier diagnosis of molecular residual disease might translate into earlier adjuvant therapies, meaning a higher chance of experiencing treatment toxicities (neurologic, dermatologic, gastrointestinal, cardiovascular, etc), while a negative ctDNA might increase patients’ anxiety and fear of cancer recurrence, reflecting a poor quality of life. The Imvigor011 and the MODERN quality-of-life endpoints will be crucial to determine whether earlier molecular intervention translates into overall well-being.
Finally, among patients who remained ctDNA-negative, only 11% had disease recurrence or death by two years of follow-up. A negative predictive value of 88% is remarkably high compared with traditional pathological or imaging-based risk stratification, although it remains uncertain whether this NPV is adequate to meet all patient care requirements. Additionally, nowadays we lack standardization across platforms, and we should integrate the use of ctDNA with complementary biomarkers such as radiomic or transcriptomic immune signatures to enhance sensitivity.
To conclude, ctDNA is a powerful and promising tool that will most probably reshape MIBC management, but it is not ready for widespread routine adoption outside of clinical trials. Careful counseling and patient education on the uncertainty of molecular surveillance is mandatory, and further improvement of sensitivity, timing, and integration with other biomarkers is needed. As personalised precision medicine advances, the goal should not be only survival, but smarter, less toxic, and more equitable healthcare, ensuring treatment of the right patient, at the right time, and with the right intensity.
Written by: Clara Cerrato,1 and Maria Carmen Mir2
- Department of Urology, Nottingham University Hospital NHS Trust, Nottingham, UK
- Department of Urology, Mount Sinai Icahn School of Medicine, Elmhurst/Queens Hospitals, New York, USA
- Dong Q, Chen C, Hu Y, Zhang W, Yang X, Qi Y, et al. Clinical application of molecular residual disease detection by circulating tumor DNA in solid cancers and a comparison of technologies: review article. Cancer Biol Ther 2023;24.
- Avanzini S, Kurtz DM, Chabon JJ, Moding EJ, Hori SS, Gambhir SS, et al. A mathematical model of ctDNA shedding predicts tumor detection size. Sci Adv 2020;6.
- Wang B, Davis LE, Weight CJ, Abouassaly R, Bukavina L. Real-World Experience with a Commercial Circulating Tumor DNA Assay in Non-muscle-invasive Bladder Cancer. Eur Urol Oncol 2025;8:883–7.
- Powles T, Kann AG, Castellano D. ctDNA-Guided Adjuvant Atezolizumab in Muscle-Invasive Bladder Cancer. N Engl J Med 2025.
- Powles T, Assaf ZJ, Degaonkar V, Grivas P, Hussain M, Oudard S, et al. Updated Overall Survival by Circulating Tumor DNA Status from the Phase 3 IMvigor010 Trial: Adjuvant Atezolizumab Versus Observation in Muscle-invasive Urothelial Carcinoma. Eur Urol 2024;85:114–22.
- Roupret M, Bertaut A, Pignot G, Neuzillet Y, Houede N, Mathieu R, et al. ALBAN (GETUG-AFU 37): A phase 3, randomized, open-label, international trial of intravenous atezolizumab and intravesical Bacillus Calmette-Guérin (BCG) versus BCG alone in BCG-naive high-risk, non-muscle invasive bladder cancer (NMIBC). Annals of Oncology 2025.
- Powles T, Valderrama BP, Gupta S, Bedke J, Kikuchi E, Hoffman-Censits J, et al. Enfortumab Vedotin and Pembrolizumab in Untreated Advanced Urothelial Cancer. New England Journal of Medicine 2024;390:875–88.
- Cerrato C, Pradere B, Mir MC. Re: Toni K. Choueiri, Thomas Powles, Laurence Albiges, et al. Cabozantinib plus Nivolumab and Ipilimumab in Renal-Cell Carcinoma. N Engl J Med 2023;388:1767–78. Eur Urol 2023;84:e93.
- Lindskrog SV, Dyrskjøt L. Towards circulating tumor DNA-guided treatment of muscle-invasive bladder cancer. Transl Androl Urol 2024;13:1056–60.
- Kramer A, Greuter MJ, Schraa S, Vink GR, Phallen J, Velculescu V, et al. 164P Early evaluation of effectiveness and cost-effectiveness of ctDNA-guided selection for adjuvant chemotherapy in stage II colon cancer. Annals of Oncology 2023;34.
- Tie J, Wang Y, Lo SN. Circulating tumor DNA analysis guiding adjuvant therapy in stage II colon cancer: 5-year outcomes of the randomized DYNAMIC trial. Nat Med 2025;31:1509–28.