How Should Belzutifan Plus Lenvatinib Fit Into Advanced RCC Sequencing?

Belzutifan has moved across the renal cell carcinoma (RCC) treatment continuum in 2026, gaining FDA approval in the adjuvant setting with pembrolizumab and after anti–PD-(L)1 therapy with lenvatinib, while a first-line triplet regimen did not improve outcomes.

Following the 2026 European Society for Medical Oncology (ESMO) Congress, CancerNetwork® spoke with Daniel Y.C. Heng, MD, MPH, FRCPC, head of medical oncology at the Arthur J.E. Child Comprehensive Cancer Centre and clinical professor of medicine at the University of Calgary in Alberta, Canada, about the final analysis of the phase 3 LITESPARK-011 trial (NCT04586231), which he presented at the meeting, and its implications for clinical practice.

Heng began by discussing how subsequent therapy and durable responses may have affected the overall survival findings with belzutifan plus lenvatinib vs cabozantinib, and how a patient's first-line regimen should guide second-line choices. He then outlined his approach to managing anemia, hypoxia, and cardiac adverse events (AEs); explained why toxicity may have undermined belzutifan-based triplet therapy in the phase 3 LITESPARK-012 trial (NCT04736706); and highlighted other ESMO data of interest, including a personalized neoantigen vaccine and kidney injury molecule-1 (KIM-1) as a biomarker. Finally, he discussed how multidisciplinary teams should approach relapse after adjuvant therapy.

CancerNetwork: In LITESPARK-011, more patients in the cabozantinib arm went on to later treatments, and second progression-free survival (PFS2) favored belzutifan plus lenvatinib. How much do you think subsequent treatment narrowed the overall survival difference?

Heng: To set some background, the LITESPARK-011 clinical trial enrolled patients in the second- or third-line setting with metastatic disease or progression after adjuvant immunotherapy, randomly assigning them to lenvatinib plus belzutifan vs cabozantinib. There was a large difference in objective response rate [ORR], 53% vs 40%, and a progression-free survival [PFS] difference of 14.8 months vs 10.8 months.¹ There was an overall survival [OS] difference, but it was not statistically significant. The delta was 5 months, but it is not statistically significant — maybe not yet.

As for why it could be related to subsequent therapy, you are right. There were more progressions in the cabozantinib arm, so more patients in that arm went on to subsequent therapies than in the lenvatinib/belzutifan arm, which can dilute the OS benefit. I also feel that belzutifan has a disease-stabilizing quality, and the duration of response doubled, at 22 months vs [11.9] months. Because of that difference, I wonder whether the OS difference will take a little longer to see. We have to wait for the curves to separate after a year, and then they start separating because that response is quite durable. I look forward to seeing whether we can see that. That is how I explain the data.

The combination's benefit looked largest after first-line ipilimumab plus nivolumab, with a PFS HR of 0.56. How should a patient's first-line regimen shape the choice of second-line treatment?

That is a great question and a great observation. To your point, the PFS and OS curves were further apart in patients treated with first-line ipilimumab/nivolumab than in those treated with an immunotherapy [IO]/VEGF combination. However, the curves were separated in both groups, so there was benefit in patients who had received IO/IO or IO/VEGF therapy in the first-line setting. Because of that, I do not think we can exclude either group. Patients treated with either IO/IO or IO/VEGF can still receive this combination.

[A]lways to use your best treatments first, because you cannot guarantee that a patient will get to second- or third-line therapy. I would choose first-line therapy based on what you need: an adequate response, a strong response, a high complete response rate. Whether it is ipilimumab/nivolumab, lenvatinib/pembrolizumab, cabozantinib/nivolumab, or axitinib/pembrolizumab, you should use what you are used to in the first line. Then, in the second line, if a patient is eligible for lenvatinib/belzutifan, you could use it. If you already used lenvatinib in the first line, you cannot use it in the second line, so I would sequence differently. For example, if you use lenvatinib/pembrolizumab in the first line, I would use cabozantinib in the second line and then single-agent belzutifan in the third line.

The whole world is thinking about the best sequence, and the honest truth is that I do not know. I do know that not everyone gets to the next line of therapy, so using the best first-line therapies first is important, and I know there are better response rates with combination therapy than without. I look forward to studying this further through the IMDC [International Metastatic RCC Database Consortium] to examine these practice patterns and sequences and see whether one is better than another.

Serious treatment-related AEs were more frequent with the combination, and cardiac concerns led to added cardiac monitoring partway through the trial. How do you manage anemia, hypoxia, and cardiac risk in practice?

With any combination therapy, there are more toxicities. We need to make them tolerable and manageable enough that people can live well on these therapies and maintain a good quality of life. Hypoxia and anemia are on-target AEs of belzutifan. If a patient develops anemia, I think that is a good thing; it means there is enough belzutifan in their body. It is important to monitor it and transfuse as necessary, and you usually start erythropoietin when the hemoglobin reaches 9 g/dL or 90 g/L, depending on the units you use.

For hypoxia, I have my patients buy an oxygen saturation monitor so that they can check their oxygen saturation at home. There was a learning curve for me. At first, I was scared whenever someone was hypoxic; if their oxygen saturation was 88%, I would send them to the emergency department. However, I have learned that belzutifan-induced hypoxia has a few characteristics. First, patients usually do not look dyspneic unless it is late. They do not look short of breath; they look fine. It is just that their oxygen saturation is a little low. Second, if they take deep breaths, the oxygen saturation usually rises back above 90%, whereas with a large pulmonary embolism or pneumonia, it might not.

Obviously, if your patient is very sick, they should go to the emergency department. However, if it is related to belzutifan, you can hold the belzutifan and see whether it improves, which usually happens within 7 days, and then reduce the dose from there. That is how I manage hypoxia. It took me a while to work through that learning curve, but I think it is manageable for most people.

The other interesting aspect of the data was the increase in cardiac toxicity. Partway through the trial, cardiac monitoring was increased, and there was a difference in cardiac toxicity of 15% vs 9% or so [13.0% vs 8.1%]. It is difficult to tease out exactly what the cardiac toxicity was because of the way the electronic case report forms are built in the clinical trial and how coordinators enter the data, so there is only so much we can glean.

However, I think most events are related to asymptomatic declines in ejection fraction. Remember that we used version 5 of the Common Terminology Criteria for Adverse Events for cardiac toxicity. A decrease of 10% is grade 1, and a decrease of 20% is grade 2. It is not hard to qualify for cardiac toxicity based on asymptomatic ejection fraction declines, especially if you are performing multigated acquisition [MUGA] scans often. We want to study that further, but for the most part, I think these are asymptomatic ejection fraction declines.

Belzutifan improved outcomes after surgery in LITESPARK-022 and after immunotherapy in LITESPARK-011, but not when added to first-line pembrolizumab plus lenvatinib in LITESPARK-012. Where does it belong in the treatment sequence?

Belzutifan is helpful in the adjuvant setting when combined with pembrolizumab; there is a [disease-free survival] difference.² It is also helpful when combined with lenvatinib after immunotherapy, and that is now FDA approved. However, it did not help as part of triplet therapy in LITESPARK-012. Belzutifan is very active, but I think the reason it did not work in the first-line trial was toxicity. It turns out that triplet therapy with belzutifan was too difficult and led to many dose reductions. For example, the average number of cycles in the lenvatinib/pembrolizumab control arm was about 16, whereas the average number in patients receiving triplet therapy was about 11.³ That means people had to come off their drugs prematurely because of AEs.

Are we giving too many drugs at once, causing too many AEs, so that we do not give enough of each drug to produce enough benefit? I wonder whether that is why LITESPARK-012 was negative. If we had found a tolerable triplet combination with belzutifan, I think we would have seen a PFS and OS benefit. However, because toxicities were too high with the triplets, I think that is why the trial was negative.

Which other kidney cancer data from ESMO 2026 stood out to you?

The most interesting data for me actually relate to kidney cancer but also do not: the INTerpath-001 trial in melanoma. It has nothing to do with RCC; it evaluated a personalized vaccine created from neoantigens in each patient's melanoma tumor, and there was a substantial [recurrence-free survival] difference.⁴ I think it is exciting because there are INTerpath trials in RCC. Will we do the same thing, and will there be the same benefit in RCC? That remains to be seen. A study published in Nature included [9] patients who received a personalized vaccine and showed good responses in the localized setting.⁵ It will be interesting to see whether we see that in the metastatic setting.

Other interesting data in RCC were the continued work with KIM-1 levels. KIM-1 circulates when there is a higher tumor burden of kidney cancer. If you use KIM-1 as a biomarker, you could see early whether there is a response or whether there is minimal residual disease. We are working out the cutoffs and learning how to use it. It is not ready for clinical practice yet because we do not know the cutoffs or how to interpret a KIM-1 level, but I think it will be very useful in the future.

When relapse occurs after adjuvant therapy, how should urologists, radiation oncologists, and medical oncologists decide between local treatment, such as metastasis-directed therapy, and starting a combination such as belzutifan plus lenvatinib?

Adjuvant therapy is now widespread for localized, resected, high-risk RCC. I believe that when we use adjuvant therapy, we change the biology of relapse. Instead of relapsing with 20 metastases or something that requires a lot of systemic therapy, we have seen many solitary metastases, or maybe 2 metastases, so patients might relapse more slowly and more gently. We are starting to see data collected on this in small retrospective series, and I look forward to bigger series to see whether that holds.

Often, when people relapse after adjuvant therapy, it is a solitary lesion, and that is when we get our radiation oncologists and surgeons involved. We discuss it at multidisciplinary tumor board: Should we deliver stereotactic body radiation therapy [SBRT] to that lesion, achieve good local control, and then continue surveillance? Usually, that is what we do. In fact, a common criticism of the clinical trials was that not everyone in the control arm received subsequent immunotherapy. I think the reason is that not everyone who relapses after adjuvant pembrolizumab or adjuvant pembrolizumab plus belzutifan needs immunotherapy, or even systemic therapy. If you eliminate the disease with metastasis-directed therapy, you will probably just watch the patient and not give more therapy. Similarly, if a patient needs systemic therapy because they relapsed quickly after adjuvant pembrolizumab, you probably will not give more immunotherapy, because they just relapsed on immunotherapy. That is another reason why some patients might not receive immunotherapy after relapse.

Of course, if it is a large relapse with multiple sites of disease, we use standard-of-care first-line doublets; an IO/VEGF regimen or ipilimumab/nivolumab would be appropriate. It is interesting to document how these relapses happen and what clinicians end up using, whether metastasis-directed therapy or systemic therapy, and we hope to collate much of that to help inform our practice soon.

References

1. Heng DYC, Schmidinger M, McDermott R, et al. Final prespecified analysis of the phase 3 LITESPARK-011 study: belzutifan plus lenvatinib versus cabozantinib for advanced renal cell carcinoma after anti–PD-(L)1 therapy. Ann Oncol. Published online October 25, 2026. doi:10.1016/j.annonc.2026.10.109
2. Choueiri TK, Motzer RJ, Karam JA, et al. Adjuvant pembrolizumab plus belzutifan versus pembrolizumab for clear cell renal cell carcinoma: the randomized phase 3 LITESPARK-022 study. Presented at: 2026 ASCO Genitourinary Cancers Symposium; February 26-28, 2026; San Francisco, CA.
3. Choueiri TK, Powles TB, Voss MH, et al. First-line pembrolizumab plus lenvatinib plus belzutifan or MK-1308A plus lenvatinib in advanced RCC: randomized open-label phase 3 LITESPARK-012 study. Presented at: ESMO Congress 2026; October 23-27, 2026; Madrid, Spain. Abstract 3586O.
4. Long GV, Khattak MA, Gogas H, et al. INTerpath-001: phase 3 study of the mRNA-based individualized neoantigen therapy intismeran autogene plus pembrolizumab vs pembrolizumab alone for adjuvant treatment of resected stage IIB-IV melanoma. Presented at: ESMO Congress 2026; October 23-27, 2026; Madrid, Spain. Abstract LBA1.
5. Braun DA, Moranzoni G, Chea V, et al. A neoantigen vaccine generates antitumour immunity in renal cell carcinoma. Nature. Published online February 5, 2025. doi:10.1038/s41586-024-08507-5