Rana R. McKay, MD, discusses the design and key findings of the OMNIVORE trial, as well as other ongoing adaptive trials in renal cell carcinoma.
Rana R. McKay, MD
Nivolumab (Opdivo) monotherapy followed by the response-based addition of ipilimumab (Yervoy) is not a viable treatment approach for patients with metastatic renal cell carcinoma (mRCC), according to findings from the phase 2 OMNIVORE trial, explained Rana R. McKay, MD.
In the trial, patients with mRCC, irrespective of histology, were treated with nivolumab at a 240-mg every 2 weeks or 480 mg every 4 weeks (n = 83). Patients were stratified into 1 of 2 groups depending on their response to nivolumab induction: confirmed complete response (CR) or partial response (PR) within 6 months (arm A), or confirmed stable disease or progressive disease (arm B). Patients who achieved a CR/PR discontinued nivolumab (n = 12), and patients who experienced stable disease or progressive disease received 2 additional 1 mg/kg doses of ipilimumab on an every-3-week schedule (n = 57).
In arm A, 10 patients achieved a PR within 6 months of nivolumab induction, half of which (n = 5; 42%; 90% CI, 18%-68%) remained off treatment at 1 year following treatment discontinuation. In arm B, 2 patients (4%; 90% CI, 1%-11%) converted to a PR after receiving an additional 2 doses of ipilimumab.
“This really highlights the need for early up-front ipilimumab in the [advanced] RCC,” said McKay. “While a subset of patients treated with nivolumab alone can maintain durable responses off treatment at 1 year, early nivolumab discontinuation in the absence of toxicity can’t be recommended at the present time.”
In an interview with OncLive, McKay, assistant professor of medicine and medical oncologist at the University of California, San Diego, discussed the design and key findings of the OMNIVORE trial, as well as other ongoing adaptive trials in RCC.
OncLive: Could you provide the rationale for the OMNIVORE trial in mRCC?
McKay: The treatment landscape in advanced RCC has rapidly evolved over the past decade. We’ve exited the TKI era and moved into the immune-oncology combination era, which was really launched by the data from the CheckMate-025 trial with nivolumab in previously treated patients with RCC. The results from that trial showed response rates on the order of 23%. In 2018, the data from CheckMate-214 were presented with the combination of nivolumab and ipilimumab in treatment-naive patients with advanced RCC. These results showed an ORR of 39% and a CR rate of 11% in the intent-to-treat population.
However, this increase in efficacy comes at a cost. We also see increased toxicity from CheckMate-025, in which there were grade 3/4 treatment-related adverse effects (TRAEs). TRAEs were observed in 21% of patients. In CheckMate-214, 47% of patients had grade 3/4 [toxicities], and the rate of high-dose steroid usage was upwards of 29%. It’s a balancing act between the benefit and risk of therapy.
We asked whether CTLA-4 blockade necessary for everyone. Can CTLA-4 blockade convert patients who are non-responsive to nivolumab to become responsive? There’s also a big question about treatment discontinuation. For patients who are responding to nivolumab, can they safely discontinue treatment?
We hypothesized that patients who discontinue nivolumab can maintain durable responses at 1 year and that the addition of ipilimumab to nivolumab in non-responders will improve response rates. That was the genesis of our multicenter, phase 2 adaptive trial.
What was the design of the trial?
The trial enrolled patients with mRCC of any histology. Patients could have been treatment-naive or could have received prior therapy. Patients were not allowed to have received a prior checkpoint inhibitor, and they needed to have measurable disease.
All patients underwent treatment with nivolumab monotherapy and imaging assessments at 8, 16, and 20 weeks during the first 6 months of treatment. Based on their response to nivolumab induction, they were allocated to a specific arm. Patients who had a confirmed PR or CR were assigned to arm A and discontinued treatment. Patients who had stable disease or progressive disease were allocated to arm B and had two doses of ipilimumab.
The primary end point for arm A was the proportion of patients with a CR or PR at 1 year posttreatment discontinuation, and the primary end point for arm B was the proportion of patients who converted to a PR or CR. All patients underwent a baseline biopsy and a progression biopsy in order to identify biomarkers of response and resistance.
The trial enrolled 85 patients, 83 of which received at least 1 dose of treatment. Twelve patients were allocated to arm A, and 57 patients were allocated to arm B. The medium follow-up was 19.5 months.
When we look at the baseline characteristics of patients who were enrolled on the study, the majority of patients had excellent performance status. Most patients had clear cell histology (96%), 8% had sarcoma differentiation, and 15% had rapid differentiation. The majority of patients had International Metastatic RCC Database Consortium intermediate- or poor-risk disease.
What did the results show?
When we look at the response to nivolumab within the first 6 months of the study period, the ORR was 14%. We didn’t see any CRs. There were 10 PRs and 1 unconfirmed PR, translating to a 12% PR rate. When we look at patients in arm A, 5 patients (42%) remained off of nivolumab at 1-year posttreatment discontinuation. One patient was able to achieve a CR, but that was after 16.9 months of treatment.
When we look at patients in arm B, the rate of conversion to a CR or PR was quite low at 4%. Only 2 patients converted to a PR, and we didn’t see any CRs. Of those 2 patients, both had been previously treated, and both had progressive disease as their best response to nivolumab induction.
At the time of data cutoff, 54 of the 57 patients who were allocated to arm B had discontinued treatment. The primary reason for treatment discontinuation was disease progression.
In terms of overall survival (OS) in the total cohort, the 18-month OS rate was 79%. The median OS has not yet been reached. We broke down TRAEs by the toxicity that was observed when nivolumab was given alone. The rate of grade 3/4 TRAEs was 7%, and 8% of patients discontinued treatment due to AEs. In arm B, the rate of grade 3/4 TRAEs was 25%, and 19% of patients had discontinued treatment for toxicity. The rate of high-dose steroids was 18%.
What other studies have examined similar approaches in RCC?
The TITAN-RCC study was presented at the 2019 ESMO Congress. The investigators took a similar approach of salvage ipilimumab with a nivolumab backbone. The results showed that the rescue rate with ipilimumab was about 10%, so 10% of non-responders were converted to responders with the addition of ipilimumab, when given in 2 sequential boosts of 2 doses at a time. However, there was also a very low CR rate of less than 5%.
My colleague, Michael B. Atkins, MD, presented the results of [HCRN GU16-260] that we were all eagerly awaiting, which is another adaptive-based design to help inform the use of immunotherapy for patients with RCC.
The take-home message from our data is that the use of nivolumab followed by ipilimumab results in a low PR and a low CR conversion rate. We saw a low number CRs with this adaptive approach. It seems that 2 doses of delayed treatment are not sufficient to achieve responses in these patients. We currently cannot recommend a strategy of nivolumab followed by the response-based addition of ipilimumab.
We are eager to explore biomarkers of response and resistance to see how we can tease out who those 42% of patients are who can stay off therapy and do really well. This work was done in collaboration with many investigators and pathologists, and was spearheaded by the team at Dana-Farber Cancer Institute and my mentor, Toni Choueiri, MD.
Is this strategy worth investigating with some of the other immunotherapy and TKI combinations?
I can’t help but highlight the PEDIGREE trial, which is looking at exactly that. That trial is spearheaded by Tian Zhang, MD, and the wonderful ALLIANCE group. In the trial, patients will undergo treatment with nivolumab and ipilimumab. Patients who have a CR will continue on nivolumab maintenance. Patients who have progressive disease will discontinue their immunotherapy and go on to receive cabozantinib (Cabometyx). Everyone who does not have a CR and doesn’t have progressive disease will be randomized to receive nivolumab plus cabozantinib or nivolumab alone.
This trial is going to be critical in trying to answer the question of layering [treatment]. It’s even more important in the context of CheckMate-9ER, [which is a trial looking at] the combination of nivolumab and cabozantinib in the frontline space. The results from CheckMate-9ER are incredibly exciting. They add to the treatment landscape in advanced RCC. It’s yet another combination that is highly effective, demonstrating improvements in OS, which is really dramatic.
We’re seeing that early, up-front combinations are key to enhanced efficacy. While I’m super excited about CheckMate-9ER, I’m even more excited about COSMIC-313, which is evaluating the combination of nivolumab, ipilimumab, and cabozantinib to see whether we can move the needle with regards to long-term durability of response, disease-control, and disease-free intervals.
McKay RR, Xie W, McGregor BA, et al. Optimized management of nivolumab (nivo) and ipilimumab (ipi) in advanced renal cell carcinoma (RCC): a response-based phase II study (OMNIVORE). J Clin Oncol. 2020;38(suppl 15; abstr 5005). doi:10.1200/JCO.2020.38.15_suppl.5005