Robert S. Alter, MD, provides key takeaways from the CheckMate 9ER study of nivolumab plus cabozantinib as frontline therapy for patients with metastatic renal cell carcinoma.
Martin H. Voss, MD: We’re going to sequence over to the combination of cabozantinib/nivolumab, which was studied in the CheckMate 9ER trial. This phase 3 trial was originally presented at the 2020 ESMO [European Society for Medical Oncology] Virtual Congress. It’s the most recent combination to gain approval in the United States earlier this year. I’ll have Dr. Alter speak to that.
Bob, maybe you can review for us the study design and the high-level overview of the primary efficacy analysis. Also, at this year’s ASCO [American Society of Clinical Oncology] Annual Meeting, we did see some very interesting data in terms of outcomes per clinically relevant subgroups. I’ll let you speak to that.
Robert S. Alter, MD: Absolutely. Thank you very much. Our first exposure to cabozantinib as a first-line agent was from CABOSUN that was presented a multitude of years ago. We have found that cabozantinib has been our first-line regimen as an oral TKI [tyrosine kinase inhibitor] for patients with intermediate- and poor-risk for several years. It’s interesting to now take what we’ve considered to be the maintenance drug of nivolumab and add cabozantinib to it as the first-line agent. This was in the New England Journal of Medicine this past year by Toni Choueiri, MD, et al. It was a phase 3 open-label clinical trial with patients with untreated advanced clear cell renal cell cancer. They were randomized to nivolumab at 240 mg for 2 weeks plus cabozantinib at low dose, 40 mg per day rather than 60 mg as a starting dose. The challenging arm and control arm was Sutent [sunitinib], at the typical 50 mg a day for 4 weeks on, 2 weeks off.
The primary end point was progression-free survival [PFS]. The secondary end points were overall survival [OS], objective response rate, and safety. KPS [Karnofsky Performance Status] was 70 or greater as 1:1 randomization, stratified by the IMDC [International Metastatic RCC Database Consortium] risk categorization, geographical location being United States to Europe or the rest of the world, and the tumor expression of PD-1 [programmed cell death protein 1]. As you said before, in regard to the axitinib/pembrolizumab, this also had a 2-year max of the nivolumab. Afterwards, you were able to continue cabozantinib. Here’s the breakdown on the IMDC categories. There’s 23% of patients with favorable risk and almost 60% of patients with intermediate risk. I believe 25% of the patients were PD-L1 [programmed death-ligand 1]-positive.
The data was presented at an 18.1-month median follow-up for OS. The median progression-free survival in the New England Journal of Medicine article for the combination of nivolumab/cabozantinib was 16.6 months. The Sutent was 8.3 months with a hazard ratio of 0.51. The probability of having progression-free survival at 12 months was around 58% for the nivolumab/cabozantinib arm and around 37% for the Sutent arm, with a 76% to 86% probability of OS at 12 months. That was a hazard ratio of 0.6. The primary end point seemed to be very favorable.
When it comes to the objective response rate, 56% of patients had an objective response rate who received nivolumab/cabozantinib, 27% with Sutent, with an 8% CR [complete response] compared with Sutent, which had a 4.5% CR rate. Clinical benefit rates of nivolumab/cabozantinib were at 88%, compared with Sutent being 69%. The statistic I like most is that progression of disease at first response was almost 5.5% for the combination arm compared to almost 14% in the Sutent arm.
Patients tolerated the therapy well. The duration of response was around 20 months on nivolumab/cabozantinib to around 11 months with Sutent. There were toxicities, so there were dose reductions in both arms: 56% with nivolumab/cabozantinib—primarily with the cabozantinib arm, obviously—and then around 52% in the Sutent arm. The quality-of-life data was significantly better with patients who received the combination of nivolumab/cabozantinib compared with Sutent. PFS, OS, and response rate were consistent across all subgroups, including the IMDC risk categorization, tumor PD-1 expression, as well as the presence or absence of bone metastases. We’ll talk about that shortly. The nivolumab/cabozantinib seemed to be the significant winner of both.
The ASCO meetings did have some updates in regard to patients who had progression-free survival. That number of 16.6 months went up to 17 months for the nivolumab/cabozantinib arm. Sutent remained at 8.3 months. Now they have 18-month data in regard to the probability for overall survival. At 18 months, the numbers were around 79% in nivolumab/cabozantinib, and in Sutent, they remained around 69%, but the hazard ratio remained the same at 0.6. As Tom said before, when more data comes out the numbers of the hazard ratios do change, and the hazard ratio for progression-free survival changed from 0.51 all the way up to 0.52. There seemed to be a little shift, but this was the start of our new approach about how our patients can actually have a little more robust data when it comes to patients with now 3 different risk categorization statuses.
Dr. Apollo presented at the ASCO Annual Meeting a few weeks ago more of a post hoc exploratory analysis looking at PFS, OS, and response rate evaluated across all patient subgroups. When you break down the favorable-risk group, the PFS has a ratio for patients who received nivolumab/cabozantinib of 0.58, which was quite impressive. That’s the first we’ve seen an impressive favorable risk number in regard to a hazard ratio. Progression-free survival was almost double: 25 months compared with 13 months. The response rate was 66% to 44%, and they were both balanced in regard to complete responses. There was no difference between both arms. The overall survival hazard ratio was 0.9, and I think it was too premature to talk.
When it came to the intermediate- and poor-risk, both favored nivolumab/cabozantinib. What really stood out was that the hazard ratio for progression-free survival on nivolumab/cabozantinib was 0.36. Compared with Sutent, the response rates were significantly better for nivolumab/cabozantinib for poor-risk: about 38 months compared with almost 9 to 10 months with Sutent. In the poor-risk group, CRs were more robust in patients who received the nivolumab/cabozantinib compared with Sutent. Therefore, the overall survival hazard ratio for the intermediate- and poor-risk groups were much better. For the intermediate-risk, it was 0.75, and for the poor-risk, 0.45 as favorable for nivolumab/cabozantinib. Across all subgroups, this post hoc analysis was quite impressive. This may give a little more credence to how we can now treat our favorable-risk patients with a little more data as compared with believing this was a population yet to be analyzed.
Martin H. Voss, MD: We will also talk about that later as a group. That’s an excellent point to make.
Transcript Edited for Clarity