Exploring the potential role of ctDNA as a biomarker in kidney cancer

In this interview, Alan Tan, MD, provides an in-depth look into the recent publication, “Longitudinal testing of circulating tumor DNA in patients with metastatic renal cell carcinoma,” for which he served as the senior author.¹ Tan is an associate professor of medicine at the Vanderbilt-Ingram Cancer Center in Nashville, Tennessee.

Alan Tan, MD

This transcript was AI generated and edited by human editors for clarity.

Could you highlight the background/rationale for this study?

In kidney cancer, it's largely believed that circulating tumor DNA (ctDNA) is a low shedder compared to some [other cancers such as] bladder cancer, lung cancer, and melanoma. It was controversial whether ctDNA was going to be a useful tool in [kidney] cancer [for this reason]. ctDNA is a big buzzword these days, and there's more evidence being generated in [different] tumor types. ASCO GI just went by, and [there were] a lot of exciting data on how we're using ctDNA in those tumors. In kidney cancer, we didn't have much data on how we can [use ctDNA] for different purposes, like treatment response monitoring and also in the surveillance setting. Can we survey these patients and look for [minimal residual disease] (MRD) recurrences and predict when these patients might be at highest risk for recurrence? As we all know, in kidney cancer, we have KEYNOTE-564 [NCT03142334], which is an adjuvant study. [It's] the first adjuvant study that showed that immunotherapy, like pembrolizumab [Keytruda], can improve disease-free survival and overall survival.²

[Our study] might have some implications going forward. We need to study this prospectively, because right now, even though these data that we generated are compelling, it's not something that we're endorsing to be used to make decisions across the board. [ctDNA is] commercially available right now, but I think this is a first step in investigating this. Of course, the best way to study this is prospectively and maybe in a randomized, controlled trial.

What were the key findings from this study?

This was a multi-institution study looking at 92 patients in a mixed population. We had a lot of patients who had metastatic RCC in the treatment response monitoring cohort, but we also had patients who were not on therapy at all; they were on surveillance looking to see how their ctDNA results correlated with their clinical outcomes, such as their CT scans, etc. The take-home point from our study of these 92 patients is that in metastatic kidney cancer, there's a high concordance [between] MRD positivity and patients having progressive disease. I would say, if anything, ctDNA is highly prognostic. That's the biggest statement that we can make as far as that.

But also, if you're ctDNA negative, it also concorded with lack of progression and having an indolent outcome. This proves the hypothesis we had that even if we're missing patients who are positive on their scans, if they're ctDNA negative, these patients still have a longer progression-free survival than their counterparts who are at higher risk. We still need to test this prospectively to prove that patients who are ctDNA negative fit into the good risk category that had the longer progression-free survival.

What future work is planned based on this study?

We're studying this prospectively now in a clinical trial (NCT06005818). We've developed a trial that's looking at patients that are receiving IO plus a TKI. As we know, there is a lot of toxicity with continuous TKI therapy. It's my thought that we can de-escalate some of these patients who have good responses. Maybe they start with positive ctDNA in the beginning, they get 6 months of treatment, and then if we clear their ctDNA and their scans look pretty good overall—a complete response or a partial response—maybe these are patients who can take a drug holiday and not be harmed from these treatment breaks. It's already been shown without ctDNA, including the STAR trial, that you could de-escalate patients up to about 3 months, and it doesn't impact their overall survival.³

[However,] patients might have hesitancy to do that without [additional] biomarkers that show us what's happening at a molecular level. Can we detect the DNA of the cancer? If not, maybe there's more equipoise for stopping their TKI treatment but continuing their IO. They can still continue their PD-1 inhibitor for up to 2 years and do an intermittent schedule of the TKI based on clearance or re-emergence of their ctDNA.

Do you foresee any challenges with potentially integrating ctDNA testing into the standard management of kidney cancer?

Yes. We're getting better and better with our ctDNA assays. The ctDNA assay that we used was a tumor-informed 16-plex assay, looking at 16 gene variants and tracking those. As the field is evolving, I think looking at larger panels—50-plus genes or maybe even whole genome sequencing—is going to possibly get us better sensitivity and specificity. I do worry about missing patients that are currently progressing if we don't have a more sensitive assay, especially in the adjuvant setting. [In the] adjuvant setting, we're trying to cure patients. I caution people not to use this to make those type of treatment decisions.

Patients do like having it as an extra eye on their cancer; they love having more information. Some investigators think this is going to cause too much anxiety for patients, but I've been doing this for several years and by and large, most patients like the peace of mind of having this in-between their scans. It may actually relieve some scan-xiety.

Is there anything else that you wanted to add?

It's still up in the air what the right biomarker [is] to follow [in] any cancer. We've talked about KIM-1 as a valuable biomarker [in kidney cancer]. There are some flaws with that assay as well. There may be some issues with the cost of these tests too. These tests, commercially, cost a lot of money. But you can also make the argument that if it tells us valuable information, maybe we're not using these expensive drugs when [they’re] not even working.

There are a lot of debates of how we're going to use this in the future. I think as the years go by and the technology gets better and more prospective studies have results, we're going to figure it out and we will have a biomarker, whether it's ctDNA, tumor informed- or non–tumor-informed methylation, or some other proteomic biomarkers, like KIM-1.

References

1. Basu A, Au C, Kommalapati A, et al. Longitudinal testing of circulating tumor DNA in patients with metastatic renal cell carcinoma. JCO Precis Oncol. 2024:8:e2400667. doi:10.1200/PO-24-00667

2. Choueiri TK, Tomczak P, Park SH, et al. Overall survival with adjuvant pembrolizumab in renal-cell carcinoma. N Engl J Med. 2024;390:1359-1371 doi:10.1056/NEJMoa2312695

3. Brown JE, Royle KL, Gregory W, et al. Temporary treatment cessation versus continuation of first-line tyrosine kinase inhibitor in patients with advanced clear cell renal cell carcinoma (STAR): an open-label, non-inferiority, randomised, controlled, phase 2/3 trial. Lancet Oncol. 2023 Mar;24(3):213-227.doi:10.1016/S1470-2045(22)00793-8