2 Clarke Drive
Suite 100
Cranbury, NJ 08512
© 2024 MJH Life Sciences™ and OncLive - Clinical Oncology News, Cancer Expert Insights. All rights reserved.
Guru Sonpavde, MD, discusses challenges with sequencing in advanced renal cell carcinoma and his vision for the field going forward.
Guru Sonpavde, MD
VEGF inhibitors are well-established in the frontline setting for advanced renal cell carcinoma (RCC), according to Guru Sonpavde, MD; however, in the second-line space, Sonpavde says it remains unclear whether continuing with anti-VEGF treatment or switching to PD-1 inhibition is the best option.
Ongoing research is attempting to identify biomarkers and patient characteristics, such as performance status, that can help optimize treatment selection in RCC. Additionally, combination regimens, such as 2 immunotherapy agents or an immunotherapy drug and an anti-VEGF agent, are being explored that could further alter the treatment landscape.
In an interview with OncLive, Sonpavde, associate professor of Medicine and director of Urologic Malignancies at the University of Alabama at Birmingham School of Medicine, discussed challenges with sequencing in RCC and his vision for the field going forward.
OncLive: Where are we currently with sequencing VEGF/PD-1 inhibitors in RCC?
Sonpavde: The major message is that we don’t know the optimal sequencing of agents for RCC. At the moment, VEGF inhibitors are ensconced in first-line regimens, such as sunitinib (Sutent), pazopanib (Votrient), and bevacizumab (Avastin)/interferon. We also have temsirolimus (Torisel) for which we have level 1 evidence for.
Adjuvant use of VEGF inhibitors is where all of the action is; this is where we have all of these new agents approved now. We have nivolumab (Opdivo), cabozantinib (Cabometyx) and lenvatinib (Lenvima) plus everolimus (Afinitor) in the second-line space.
The level one evidence is for nivolumab and cabozantinib, since these were both big phase III trials and they extended both overall survival and progression-free survival (PFS). Lenvatinib/everolimus is a little unique because this combination was approved based on a randomized phase II trial, not a randomized phase III trial. PFS extension was impressive, extending beyond 1 year with the combination compared with everolimus, so it was approved based on the extension of PFS.
Therefore, in the post-VEGF inhibitor space where we have all of these drugs approved, this is where we have to pick from these drugs in the second-line space. We don’t really know the optimal sequencing; there’s not a biomarker. PD-L1 has been looked at to select patients for nivolumab in the phase III trial, but it did not look like it was predictive—it was prognostic. High PD-L1 levels correlated with poor survival in both arms of the study but did not correlate with the improved benefit from nivolumab.
Overall, when you look at the efficacy, toxicity, and the therapeutic index, nivolumab clearly is a more tolerable drug. The toxicity profile is excellent; grade 3/4 toxicities were observed in about 19% with nivolumab. When you look at cabozantinib and lenvatinib/everolimus, their grade 3/4 toxicities occurred in close to 70% of patients. When you’re coming off of first-line sunitinib or pazopanib, most patients have some toxicities from the VEGF inhibitors. Therefore, going to nivolumab—which has a lower toxicity profile and a better therapeutic index—might make sense for most patients.
However, the case for lenvatinib/everolimus or cabozantinib is that these 2 drugs extend PFS in addition to survival. Lenvatinib/everolimus had a better response rate at around 35%, so should we be giving lenvatinib/everolimus or cabozantinib to patients who have more symptomatic, aggressive disease? This is because it seems to have more early benefit with the prolongation of PFS. Overall, when you look at all of these regimens, you want to tailor therapy based on patient preference, toxicity profile, and convenience—nivolumab is intravenous (IV) while the others are oral—so all of these things factor in, as well as physician comfort with the agent.
What do you observe regarding patient preference?
At the moment, we want to go with a VEGF inhibitor in first line—so either sunitinib or pazopanib. There are some patients who are not candidates for VEGF inhibition, such as those with uncontrolled blood pressure and recent cardiovascular comorbidity. In that situation, we might want to try to get nivolumab for such patients as a frontline treatment; everolimus is another option. Some people would say that sorafenib (Nexavar) is a somewhat less toxic VEGF inhibitor and can be good for that setting. In the second-line setting, it boils down to patient and physician preference, the toxicity profile, and convenience of an IV versus an oral drug.
What other steps need to be taken to address the challenge of sequencing?
It’s very challenging at the moment. My strategy going forward is that we should have a two-tiered strategy. We have all of these trials going on that are looking at combinations, such as CTLA-4 plus PD-1 inhibition, compared with sunitinib. We also have VEGF plus PD-1 inhibition as a strategy compared with single-agent, frontline sunitinib.
One of the problems of combination regimens is that they may be less tolerable. Therefore, you might want to pick patients with good performance status—0 to 1—for such combination strategies. With a poor performance status of 2 to 3 comorbidities, you may want to go with sequential single agents. There will probably be a two-tiered strategy going forward—better performance status combination for first-line, combinations for second-line, and poorer performance status going with sequential single agents.
Ideally, you want to have precision medicine playing a big role. We are not there yet, unfortunately, so we still have a ways to go to optimally select patients. It would be nice to incorporate noninvasive biomarker testing, such as circulating tumor cell testing.
What does the future of this field look like?
There are a number of ongoing phase III trials that might change the landscape dramatically over the next 1 to 2 years. We had the nivolumab plus ipilimumab combination strategy that might emerge, which is being compared with sunitinib. Then, we have a number of trials looking at the strategy of VEGF plus PD-1 inhibition; there are about 3 or 4 trials going on in that space. We will see whether the combination immunotherapy strategy or VEGF plus PD-1 will be the big winner. Perhaps they both will win, and then we will have to look at the toxicity profile and see which 1 might confirm a better therapeutic index. There might be a place for both.