Navigating the RCC Adjuvant Landscape: Examining the Phase 3 RAMPART Trial

The therapeutic paradigm for resected, localized renal cell carcinoma (RCC) continues to evolve with the evaluation of novel adjuvant immunotherapies. In a discussion with CancerNetwork®, Saum Ghodoussipour, MD, provided an analysis of the international phase 3 RAMPART trial (NCT03288532) presented at the 2026 American Society of Clinical Oncology (ASCO) Annual Meeting.¹ Ghodoussipour unpacked the trial’s unique multi-arm design, contrasting the successful disease-free survival (DFS) benefit achieved by combining the PD-L1 inhibitor durvalumab (Imfinzi) with the CTLA-4 inhibitor tremelimumab (Imjudo) against the unexpected failure of durvalumab monotherapy.

While the dual-agent regimen demonstrated clinical efficacy in patients at high risk of recurrence according to their Leibovich scores, it simultaneously introduced substantial toxicities and serious adverse effects (AEs). Ghodoussipour balanced these oncologic outcomes against real-world safety profiles and standard benchmarks, highlighting the critical need for more refined, next-generation circulating biomarkers to help clinicians personalize adjuvant treatment decisions.

Ghodoussipour is the director of the Bladder and Urothelial Cancer Program at Rutgers Cancer Institute and an associate professor of Surgery at Rutgers Robert Wood Johnson Medical School.

CancerNetwork: What was the background behind the study of immune checkpoint inhibition vs active monitoring of RCC in the RAMPART trial?

Ghodoussipour: RAMPART is an international, investigator-led phase 3 clinical trial that had an interesting design––a unique design. It randomly assigned patients with high-risk resected primary renal cell carcinoma, or completely resected M1 oligometastatic disease, to adjuvant immunotherapy with durvalumab plus tremelimumab vs durvalumab alone or active monitoring. They risk-stratified patients according to the Leibovich score, which is a clinical risk stratification model based on clinicopathologic features like T stage, node status, tumor size, grade, and some other histology-specific features. It is a clinically very useful tool.

This was across multiple countries in Europe, so there was a nice representation of patients. As I said, having those 3 arms is nice to have in one trial. They had originally planned to enroll 1750 patients, and they had to cut out 1000 patients because of issues related to enrolling patients during COVID-19, but also because of the pembrolizumab [Keytruda] data that came out after their initial design. However, they were still able to accrue after they adapted the design of the trial, with the primary outcome of disease-free survival as compared with the combination therapy group with durvalumab plus tremelimumab vs active monitoring alone.

What were some of the key findings of the trial?

This year at ASCO, they updated a presentation that was given at the European Society for Medical Oncology (ESMO) Congress 2025.² Last year at ESMO, they focused on the outcomes of patients in the combination group vs active monitoring, and then this year at ASCO, they updated the results to show the outcomes in the durvalumab monotherapy group, putting it all together. [Additionally,] they showed that the combination of durvalumab plus tremelimumab led to a [DFS] benefit compared with active monitoring alone. That [DFS] was 81% vs 73% in those patients, with a hazard ratio of 0.65 that was statistically significant across the 2 groups. However, when they stratified that hazard ratio by intermediate vs high-risk disease according to the Leibovich criteria, all the benefit seemed to come from the patients in the high-risk group. Their hazard ratio was 0.52, and then the intermediate-risk population had a non-significant hazard ratio of 1.19.

This year they presented the durvalumab data. Looking at the entire cohort, it was 790 patients––a bit over 200 in each of the therapy groups, and almost 340 in the active monitoring group. They saw that the 3-year DFS was 80% with the combination, 78% with monotherapy with durvalumab, and 72% with active monitoring. There was not a benefit in terms of [DFS] with durvalumab monotherapy. The hazard ratio there was 0.74, but the confidence interval went from 0.53 and crossed 1, so it was not statistically significant in the group overall, and it was also not significant in the high-risk population.

The overall conclusion is that the combination of durvalumab plus tremelimumab has a [DFS] benefit that seems to be concentrated in those at a high risk of recurrence. However, for the monotherapy, it was surprising to some because of the data we have with pembrolizumab, but there was no benefit in that regard.

Given that non-clear cell RCC often responds differently to immunotherapy, what are your expectations for any upcoming data on non-clear cell subtypes, and would this trial change how you approach adjuvant therapy for these patients?

This trial did report on non-clear cell histology to some degree. This should be contrasted to the data that exists now with the phase 3 KEYNOTE-564 trial (NCT03142334), which was [almost] all clear cell. The problem is, we cannot conclude much here. Eighty-four percent of patients had clear cell, and then there was a very small percentage––it was about 5% or 6% of papillary and chromophobe each––and then there was another bucket of other non-clear cell histologies. I don’t think we can make any meaningful conclusions; the numbers are just not enough for subset analyses.

We have a lot more work to be done in the non-clear cell space. At the same time, at the meeting, they showed that pembrolizumab didn’t appear to have benefit in non-clear cell histologies, so there’s a lot more understanding of biology that needs to happen to understand why some of these non-clear cell histologies are so much more immune-cold, and what it is that we have to target in these patients.

Considering the RAMPART results, do you believe that there is a role for durvalumab plus tremelimumab in the RCC treatment landscape?

These data, in light of the existing data and approval of pembrolizumab based on KEYNOTE-564, deserve conversation. Personally, in my practice, it would be hard for me to say that durvalumab plus tremelimumab should replace pembrolizumab. The reason is, while there is benefit that we can see statistically––specifically in that high-risk population––there’s a real toxicity and quality of life impact that patients have to consider. If you’re adding checkpoint inhibitors and CTLA-4-targeted therapy, you’re intensifying that attack of the immune axis.

If you look at the presented data between ESMO and this year at ASCO, you’ll see that only 23% of patients completed their durvalumab as intended, and up to 40% of patients had serious AEs with the combination therapy. These were all immune-related events: colitis, increase in LFTs, and pancreatic enzymes. It’s going to be hard to justify exposing patients to that greater toxicity when we know that single agent pembrolizumab improves DFS, but also OS.

That being said, I like the Leibovich risk criteria. It’s easy for us to calculate in clinic. However, like in all our other disease spaces, these clinicopathologic risk stratification systems are very coarse, and there is room for better biomarkers to guide these therapies. There was a lot of discussion at ASCO about how ctDNA and other circulating biomarkers like KIM-1 can risk-stratify these patients. We’re getting there in bladder cancer…and we’re seeing some utility of these biomarkers, but the actual utility and benefit is not yet quite as clear in kidney cancer.

Were there any unique toxicities with tremelimumab for these patients?

As I mentioned, when you add tremelimumab to durvalumab, you’re intensifying the immune attack, so you’re going to get an intensification of immune-related adverse [effects]. They reported on quality of life, and they said it was no different, but we have to look at how we’re assessing quality of life. Not a lot of the quality-of-life questionnaires that exist are immune-directed in terms of adverse [effects], and sometimes you’re asking these questions after the fact.

The whole thought that we can maybe do these adjuvant therapies better than single agent is why there is reason and rationale to study combination therapies. However, if you're adding immune therapies on top of each other, I think, based on this data, we now see what the effect of that is, and it's why others are looking at immunotherapy plus targeted agents as well. People have looked at belzutifan [Welireg], and there are VEGF inhibitors in an ongoing clinical trial at our center right now as well, in combination with immunotherapy. It might become a very personalized approach in the future of what you're adding to a checkpoint inhibitor, and whether [for] the anticipated toxicity: is the juice going to be worth the squeeze?

What are some potential next steps for research in this RCC population?

The future of medicine, and urological oncology and [genitourinary] oncology for our entire field, is going to be personalized approaches to care. As I said, in bladder cancer, we’re using molecular subtyping and circulating tumor DNA routinely in the clinic, at least in my practice, and we’re using digital pathology with AI-based biomarkers.

In kidney cancer, we know that [RCC] doesn’t shed ctDNA as much as some other cancers, so [clinicians] are developing novel approaches, moving from whole exome to whole genome sequencing and incorporating things like KIM-1, which is a tubular injury-based biomarker. It’s not cancer-specific, but it is circulating much more prevalently than ctDNA. That’s where future investigation will lie, and it’s good that there’s data coming out about combination therapy and single-agent therapy so that in the future we can do the discovery work to identify biomarkers to make personalized medicine a reality in this space.

Is there anything else that you would like to highlight that we might not have discussed?

The results are promising, in that combination therapy can be more effective when you’re looking at oncologic efficacy in certain risk populations, but it’s on us to do a better job of determining who will derive that benefit, and there’s a lot of work going on in that space.

The investigators should also get kudos for looking at quality of life, but [repeatedly], we’re seeing no quality of life-related differences despite differences in toxicity, so maybe we need to look in the mirror and see if we’re assessing quality of life correctly. However, the fact that these questions are being asked is great. It’s awesome, the fact that we had multi-arm trials like this one that are looking at a control group, combination, and monotherapy in one trial is also a great example that I think others can follow, so the future is bright and we're making steps.

References

1. Larkin JM, Powles T, Frangou E, et al. Durvalumab monotherapy versus active monitoring for resected primary renal cell carcinoma in RAMPART: an international, phase 3, randomized controlled trial. J Clin Oncol. 2026;44(suppl 17):LBA4511. doi:10.1200/JCO.2026.44.17_suppl.LBA4511
2. Larkin JM, Powles T, Frangou E, et al. First results from RAMPART: an international phase III randomised-controlled trial of adjuvant durvalumab monotherapy or combined with tremelimumab for resected primary renal cell carcinoma (RCC) led by MRC CTU at UCL. Ann Oncol. 2025;36(suppl 2):S1635. doi:10.1016/j.annonc.2025.09.110