Opinion|Videos|November 13, 2026

The Biomarker Era: MRD, ctDNA, and Treatment De-Escalation Across Tumor Types

Drs. Kydd and Sun on the state of MRD and ctDNA in GU and head and neck cancer, and where personalization is reshaping neoadjuvant and adjuvant decisions.

Desai:Welcome, everyone, and thank you for joining this OncLive Peer Exchange. I’m Arpita Desai, a genitourinary medical oncologist at UCSF, where I also lead the kidney cancer program. I’m joined today by Dr. Kydd…

Kydd:Hello. I’m Andre Kydd, from Fox Chase Cancer Center.

Desai:…and Dr. Sun.

Sun:I’m Fangdi Sun. I’m a head and neck medical oncologist at Stanford University.

Desai:Today we’re stepping back from any single trial or meeting to look at where oncology is heading — the trends that will shape practice over the next several years, and what excites or concerns us most as clinicians navigating this shift. Let’s start with minimal residual disease, circulating tumor DNA (ctDNA), and dynamic disease monitoring. Dr. Kydd, what’s exciting in the GU space, especially bladder cancer?

Kydd:There is a lot of evolution happening right now around ctDNA. In some ways we’re behind other cancers that have been using MRD and ctDNA, but there has already been a major role for ctDNA in decisions around adjuvant immunotherapy, and more clinicians are now incorporating it into clinical trial design and, ultimately, clinical practice. One area we’ll come back to is the role of personalization in de-escalation strategies in the perioperative setting. With the arrival of perioperative enfortumab vedotin plus pembrolizumab, several clinical trials — including some at Fox Chase — are asking which patients can avoid surgery altogether, and I think ctDNA may play a role there. Urinary tumor DNA is also likely to be incorporated given its potential role in detecting local recurrences.

Kydd:In later disease settings, the role of ctDNA monitoring is very much in flux. How does it compare to conventional imaging when ctDNA is positive but CT or MRI is still negative? Is that an early signal of recurrent disease, and do we act on it, or do we wait for conventional imaging? That is very much in flux in the field, and hopefully clinical trials will catch up and give us evidence-based guidelines for the appropriate use of these technologies.

Desai:Dr. Sun, what are you looking forward to in head and neck cancer?

Sun:It’s actually interesting that Dr. Kydd feels GU may not be as far ahead as other cancers. From where I sit, you are quite far ahead, if anything. In head and neck cancer, the availability of testing is ahead of the prospective data. The area where we most commonly have access to ctDNA testing is in the virus-mediated cancers — HPV-mediated oropharynx cancer and EBV-mediated nasopharynx cancer. Using HPV as an example, there are several commercial tests that give a quantitative result for tumor-associated HPV DNA in the blood, and many clinicians have started to integrate them into disease surveillance.

Sun:The challenge is that what we’ve shown so far is largely that we’re picking up disease about 3 months earlier than we would on imaging or exam — but we have not yet shown that this improves patient outcomes. Patients can be either falsely reassured by a negative result, or anxious with a positive result and no obvious disease. How do you have that conversation? While these tests are readily available, they are not in our guidelines and, in my opinion, not standard of care. They absolutely should be studied, and I think this is an interesting venue going forward — but we need iterations on these tests to make them more sensitive and more specific than what we have today.

Desai:I agree. It’s great to have this modality available as another biomarker for escalation or de-escalation, but it has to be clinically meaningful. The last thing we want is to cause more anxiety. Now let’s turn to treatment de-escalation and personalization — how are you thinking about this space over the next 2 to 3 years?

Kydd:In the perioperative setting in bladder cancer, we have several trials asking which patients, if they have an excellent response — a pathologic complete response, for example — to neoadjuvant enfortumab vedotin can actually avoid surgery. That is just the tip of the iceberg in response-adapted treatment. The VOLGA trial is weighing how much neoadjuvant and how much adjuvant therapy are needed, and what the best selection criteria look like. There are many factors in play, and we need more trial data to resolve them. Hopefully, over the next couple of years, we’ll have more clarity — though we should not assume clarity; sometimes new data raise more questions, and the next 2 to 3 years may ultimately show us what the next generation of trials needs to answer.

Kydd:There’s also the opposite question of who needs intensification, what agents we add, and when to back off — whether we’re guided by ctDNA, imaging response, or something else. The MODERN trial is looking at how we add more immunotherapy, and there are other strategies being tested. So there are many choices in both directions: how to get more therapy to patients who truly need it, and when to back off in patients with excellent responses, or when the trade-off between treatment intensity and quality of life argues for less.

Desai:Dr. Sun, how are you thinking about this in head and neck cancer? Is there a role for de-escalation, and where do you intensify?

Sun:It’s such a good question. The area of head and neck cancer where de-escalation has received the most attention is HPV-associated cancer in the curative setting — and that makes sense, because these are cancers we are highly likely to cure. Part of the reason our treatments get a bad reputation is the long-term toxicity. Patients who receive high-dose curative-intent radiation are often left with neck fibrosis, xerostomia, taste changes, and persistent weight loss. Radiation de-escalation in this radiation-sensitive disease has been a major focus.

Sun:Where it gets tricky is when you look at the actual data. The fraction of patients we can de-escalate in the current state is very small. One cooperative group trial showed that certain intermediate-risk patients who went to upfront surgery could receive 10 Gy less radiation, with comparable long-term outcomes. But those are a tiny fraction of HPV-associated oropharynx cancer patients, and it’s an area of future need — maybe MRD tests, as they become more sensitive, will help us refine it. I want folks to be careful about where de-escalation is implemented in the current state.

Desai:You’ve made the point very well. In a curative setting we want to make sure patients get enough treatment to actually cure the cancer, balancing toxicity as we do.


Related to this article