Navigating ctDNA-Guided Decision-Making in Muscle-Invasive Bladder Cancer

Molecular residual disease (MRD) assessment via circulating tumor DNA (ctDNA) is reshaping adjuvant decision-making in muscle-invasive bladder cancer (MIBC) following the FDA’s May 2026 approval of atezolizumab (Tecentriq) for patients with ctDNA MRD after cystectomy and the simultaneous approval of the Signatera CDx companion diagnostic.¹

CancerNetwork® spoke with Mamta Parikh, MD, MS, the medical director of Genitourinary Malignancies at the UC Davis Comprehensive Cancer Center and member of the NCCN Guidelines Panel for Bladder Cancer, about how the updated NCCN Clinical Practice Guidelines in Oncology for Bladder Cancer address postoperative MRD testing in an increasingly aggressive perioperative treatment landscape.²

Parikh explained why the guideline’s ctDNA recommendations, anchored to the phase 3 IMvigor011 trial (NCT04660344) population, may not directly translate to patients receiving perioperative immune checkpoint inhibitor regimens such as those investigated in NIAGARA(durvalumab [Imfinzi]; NCT03732677), KEYNOTE-905 (NCT03924895) , and KEYNOTE-B15 (NCT04700124) .^3-6 She discussed how clinicians should approach patients with ctDNA-negative status, the promise and current limits of real-time serial ctDNA and urinary tumor DNA (utDNA) monitoring, practical strategies for operationalizing testing given payer coverage challenges, a possible role for ctDNA in advanced urothelial carcinoma, and how ctDNA dynamics might help refine the definition of a complete responder in bladder-preserving approaches.

CancerNetwork: How do the recent NCCN guideline updates on postoperative MRD assessment fit into the evolving, more aggressive perioperative treatment landscape? Specifically, if a patient receives an intensive perioperative combination therapy, does a positive postoperative ctDNA test carry the same actionable weight for adjuvant escalation as it does for patients who receive standard neoadjuvant chemotherapy or go straight to surgery?

Parikh: This is an excellent question because it is an important opportunity to clarify the current NCCN guidelines regarding MRD monitoring. This is a very specific guideline for patients who did not receive an immune checkpoint inhibitor prior to radical cystectomy, and it is based entirely on the eligibility criteria of patients enrolled on the IMvigor011 study, so these guidelines are narrow in the context of the evolving landscape of [MIBC] management.

We are now taking a much more aggressive perioperative approach that incorporates checkpoint inhibitor therapy both before and after surgery, and in those regimens, by definition, perioperative management implies continuation of therapy postoperatively. In the case of NIAGARA, that involves continuing durvalumab, and in the case of the KEYNOTE-905 and KEYNOTE-B15 studies, that includes continuation of enfortumab vedotin [Padcev] and pembrolizumab [Keytruda]. We do not have enough data about whether ctDNA testing carries the same actionable weight in these settings, but we know these studies were designed as a perioperative approach, and in these trials, regardless of pathologic status at the time of radical cystectomy, patients continued on to receive adjuvant therapy, and all of these studies were positive studies that benefited patients. In the perioperative setting, we need to be mindful that continuation of adjuvant therapy is the current standard, and we do not have sufficient ctDNA data to inform whether we can change that at present.

While the IMvigor011 trial highlighted the survival benefit of treating patients with ctDNA-positive disease, clinical questions remain for those who test negative. How should clinicians handle the ctDNA-negative population in practice?

There are data from both the IMvigor011 study and other studies showing that [patients with] ctDNA-negative [status] tend to do better in general. In IMvigor011, an exploratory analysis of patients who did not enroll in the study but had persistent ctDNA-negative status showed excellent outcomes, with a 1-year disease-free survival of 95% and excellent overall survival. The issue is being able to predict who will have persistently negative ctDNA.

Now, we do not know which patients who are ctDNA negative initially will remain persistently ctDNA negative, and that is further complicated by trials like the phase 3 Checkmate 274 study (NCT02632409), a positive study that demonstrated nivolumab (Opdivo) was better than placebo in patients who had ypT2 or higher disease after neoadjuvant chemotherapy, or pT3 or higher disease without neoadjuvant chemotherapy, after radical cystectomy, regardless of ctDNA status.⁷ We still have to offer nivolumab to these patients, and we need more prospective studies looking at further deintensification. These studies are ongoing in trials like [phase 2/3 MODERN trial (NCT05987241)], where [patients with] ctDNA-negative [status] are randomized to receive nivolumab or observation. We will get these results, although they are still a couple of years off.⁸

How close are clinicians to using serial, real-time ctDNA monitoring during neoadjuvant treatment to guide treatment selection or support bladder preservation strategies? How do you anticipate future NCCN guideline updates will address using ctDNA dynamics to scale therapy prior to, or even in place of, radical cystectomy?

I cannot speak to future NCCN guideline updates due to the strict firewalls we need to be mindful of, but personally, real-time serial ctDNA monitoring is an exciting prospect. It needs to be studied more and incorporated into trials moving forward. There are some data from the NIAGARA study, which looked at gemcitabine, cisplatin, and durvalumab vs gemcitabine and cisplatin prior to radical cystectomy. This trial incorporated ctDNA monitoring and released results showing plasma ctDNA clearance was favorably prognostic, though it did not perfectly correlate with a pathologic complete response [pCR]. Investigators also looked at utDNA, which did correlate better with pCR, and retrospective studies have shown that tumor-informed ctDNA may provide more prognostic impact than even pCR. It is exciting, but we are not there yet in terms of standard guidelines we can come up with, given the lack of data we have.

There is a lot of opportunity for bladder-preserving trials to incorporate ctDNA monitoring into their algorithms; the SWOG S2427/BRIGHT trial [NCT07061964] is incorporating ctDNA, and we will get more data in this area.⁹ One thing I need to caution about is that we have a limited set of prospective data in this context, and we must consider the potential anxiety and uncertainty this can cause for patients and providers because we do not have enough data to fully interpret or act upon the results we are receiving. I am hopeful that we will have more prospective data to help guide us.

Do you foresee a role for serial ctDNA monitoring to identify patients with advanced urothelial carcinoma who experience primary progressive disease or rapid failure within 6 months of enfortumab vedotin and pembrolizumab? How might ctDNA kinetics guide clinicians to proactively switch these patients to targeted agents or novel combinations before clinical progression occurs?

I may have a controversial take here: unfortunately, patients who experience primary progressive disease or rapid failure within 6 months of starting enfortumab vedotin and pembrolizumab tend to have very aggressive disease, and they often progress in an overt fashion where serial ctDNA monitoring is not necessarily going to catch the disease progression earlier. Serial ctDNA testing may allow us to further characterize these primary progressive patients so they can be studied as a distinct population because there is a need for novel combinations in particular for these patients. On the other hand, one of the areas in advanced urothelial carcinoma where we do have a lot of data on serial ctDNA monitoring, albeit retrospectively, is identifying patients who are responding favorably to enfortumab vedotin and pembrolizumab.

I am excited about the possibility of trials evaluating either consolidative or deintensification approaches in these patients, giving us a chance to potentially get patients off treatment that is causing toxicity, and possibly give patients a treatment-free interval in the advanced and metastatic stage. That quadrant of patients are the ones who can benefit the most from serial ctDNA monitoring as we get more trials looking at this.

As new bladder preservation trials gain traction, how can clinicians leverage serial ctDNA dynamics to refine the definition of a true complete responder, and how might ctDNA negativity impact how clinical and endoscopic staging is used to select candidates for bladder preservation?

This is another area that is currently evolving, and I liken it to the use of AI: ultimately, it is not meant to replace, but to augment. Similarly, ctDNA might augment rather than replace our current staging and selection of patients for bladder preservation. How we define a complete responder is important in this context because if a patient is truly a complete responder with ctDNA negativity, does that mean they should not receive bladder preservation therapy, or does it mean they are the patients most likely to benefit from it? We do not know right now, and the fact that ctDNA monitoring is being incorporated into ongoing bladder preservation trials will answer some questions, though we will likely have even more questions about whether it can be used as a tool to identify patients who need less therapy.

I mentioned utDNA testing before, which might be an exciting prospect that could augment cystoscopy staging and further define what a complete responder is. In the NIAGARA study, utDNA correlated more closely with pCR than plasma ctDNA testing, so it might be that we can augment our definition of a complete responder with some of these tests. It could eventually help us replace endoscopic staging, like cystoscopy, if we can get utDNA testing that is just as accurate. For now, it can be used in an augmented manner to arrive at a tighter definition of ideal candidates for bladder preservation. More trials that incorporate this kind of testing into the patient selection process to better refine exactly how to use it are needed.

Is there anything else you wanted to highlight?

This is a very exciting time for patients with bladder cancer, and we are getting closer to identifying patients who do not need to have their bladder removed, which is exciting in terms of quality of life. We still need more data so we can do this in the safest way possible and make sure we are truly improving outcomes for these patients.

References

1. FDA approves atezolizumab for adjuvant treatment of muscle invasive bladder cancer in patients with molecular residual disease. News release. FDA. May 15, 2026. Accessed July 31, 2026. https://tinyurl.com/msrrrt9u
2. NCCN. Clinical Practice Guidelines in Oncology. Bladder Cancer, Version 2.2026. Accessed July 31, 2026. https://tinyurl.com/3j78syzt
3. Bellmunt J, Gupta S, Durán MÁ, et al. Patient-reported outcomes from IMvigor011: a phase 3 study of circulating tumor (ct)DNA-guided adjuvant atezolizumab vs placebo in muscle-invasive bladder cancer (MIBC). J Clin Oncol. 2026;44(suppl 16):4627. doi:10.1200/JCO.2026.44.16_suppl.4627
4. Van Der Heijden MS, Galsky MD, Joshi R, et al. Urinary tumor DNA (utDNA) and circulating tumor DNA (ctDNA) in patients (pts) with muscle-invasive bladder cancer (MIBC) who received perioperative durvalumab (D) in NIAGARA. J Clin Oncol. 2026;44(suppl 7):636. doi:10.1200/JCO.2026.44.7_suppl.636
5. Vulsteke C, Kaimakliotis HZ, Danchaivijitr P, et al. Perioperative enfortumab vedotin plus pembrolizumab in participants with muscle-invasive bladder cancer who are cisplatin-ineligible: phase 3 KEYNOTE-905 study. Presented at: 2025 ESMO Congress; October 17-21, 2025; Berlin, Germany. Abstract LBA2.
6. PADCEV plus Keytruda significantly improves survival for patients with muscle-invasive bladder cancer regardless of cisplatin eligibility. News release. Astellas Pharma. December 17, 2025. Accessed August 3, 2026. https://tinyurl.com/2u6cfrhs
7. Galsky MD, Gschwend JE, Milowsky MI, et al. Adjuvant nivolumab versus placebo for high-risk muscle-invasive urothelial carcinoma: 5-year efficacy and ctDNA results from CheckMate 274. Ann Oncol. Published online October 17, 2025. doi:10.1016/j.annonc.2025.09.139
8. Testing the role of DNA released from tumor cells into the blood in guiding the use of immunotherapy after surgical removal of the bladder, kidney, ureter, and urethra for urothelial cancer treatment, MODERN study. ClinicalTrials.gov. Updated August 3, 2026. Accessed August 3, 2026. https://tinyurl.com/mr97r4dr
9. Combining immunotherapy and radiation therapy to help patients avoid bladder removal after treatment shrinks muscle invasive bladder cancer, BRIGHT trial. ClinicalTrials.gov. Updated July 30, 2026. Accessed August 3, 2026. https://tinyurl.com/2h839ehb