Cabozantinib regimen ‘beat the doublet in PFS’ in untreated advanced kidney cancer
Click Here to Manage Email Alerts
Cabozantinib plus nivolumab and ipilimumab conferred a significant PFS benefit vs. nivolumab and ipilimumab alone among patients with advanced renal cell carcinoma, according to results of the COSMIC-313 trial presented at ESMO Congress.
Methodology
“COSMIC-313 is the first phase 3 trial using a modern doublet control of nivolumab [Opdivo, Bristol Myers Squibb] and ipilimumab [Yervoy, Bristol Myers Squibb] and asking the question of whether adding cabozantinib [Cabometyx/Cometriq, Exelixis] may lead to improved outcomes and positive results among patients with advanced renal cell carcinoma,” Toni K. Choueiri, MD, director of the Lank Center for Genitourinary Oncology at Dana-Farber Cancer Institute and Jerome and Nancy Kohlberg chair and professor of medicine at Harvard Medical School, told Healio.
The trial included 855 previously untreated patients with advanced renal cell carcinoma at intermediate risk (75%) or poor risk (25%) for survival according to the International Metastatic RCC Database Consortium risk model.
PFS by blinded independent radiology review per RECIST 1.1 among the first 550 randomly assigned patients served as the primary endpoint. OS among all randomly assigned patients served as the secondary endpoint.
Median follow-up was between 17.7 and 20.2 months.
Key findings
Results showed patients who received cabozantinib, a multitargeted tyrosine kinase inhibitor, plus nivolumab and ipilimumab experienced a 27% lower risk for disease progression or death compared with those who received nivolumab and ipilimumab alone (median, not reached vs. 11.3 months; HR = 0.73; 95% CI, 0.57-0.94).
PFS among subgroups appeared mostly consistent with the primary endpoint results, and further analyses suggested those classified as intermediate vs. poor risk experienced greater benefit with the cabozantinib regimen.
Researchers identified no significant survival benefit for the three-drug combination; the OS analysis is ongoing.
“This is the first study in kidney cancer where we see the triplet beat the doublet in PFS,” Choueiri said.
The triplet also conferred a numerically higher objective response rate (43% vs. 36%) and disease control rate (86% vs. 72%) than the doublet, with a generally manageable safety profile consistent with that of each individual treatment. Choueiri noted higher incidence of treatment-related increases in alanine aminotransferase (46% vs. 17%) and aspartate aminotransferase (44% vs. 16%), as well as diarrhea (41% vs. 18%) and skin toxicity, in the cabozantinib vs. control group.
Next steps
The next steps are to see how to integrate this regimen into practice, if approved by regulatory authorities, Choueiri told Healio.
“The treatment landscape in kidney cancer continues to evolve, resulting in more options for advanced disease, but there is still a need for additional, effective first-line treatment options, especially for patients with intermediate- or poor-risk disease. It is essentially finding the right patient population, so we need further research to find the subsets of patients that will benefit,” Choueiri said.
References:
- Choueiri TK, et al. Abstract LBA8. Presented at: European Society for Medical Oncology Congress; Sept. 9-13, 2022; Paris.
- Three-drug combination slows progression of advanced kidney cancer (press release). Available at: www.dana-farber.org/newsroom/news-releases/2022/three-drug-combination-slows-progression-of-advanced-kidney-cancer/. Published Sept. 7, 2022. Accessed Sept. 8, 2022.
Perspective
Back to Top
Pavlos Msaouel, MD, PhD
This year’s ESMO Congress is particularly exciting for renal cell carcinoma, with many interesting trial readouts that will generate a lot of discussion.
The first results of COSMIC-313 are certainly among the top of the list, signaling the advent of triplet therapy combinations for renal cell carcinoma.
At the currently reported follow-up, the study yielded an HR of 0.73 for its primary endpoint of PFS. This is an intriguing but somewhat underwhelming signal of relative treatment effect efficacy. If we account for the exaggeration of relative treatment effects observed in single randomized controlled trials (RCTs), such as COSMIC-313, we can shrink the unbiased reported frequentist estimator using a Bayesian shrinkage prior distribution derived from the primary results of 23,551 RCTs of treatment efficacy in the Cochrane Database of Systematic Reviews. This would yield a posterior mean HR of 0.81, with 95% credible intervals of 0.63 to 1.01 for improved PFS with the triplet combination compared with the control arm, based on the reported data. The posterior probability that the placebo control group had better PFS than the triplet combination would then be 3.1%, a small but not insubstantial number.
Although the relative treatment efficacy estimates for ORR are not provided, the reported marginal values of 43% vs. 36% are also underwhelming.
Based on prior experience in oncology RCTs, the underlying mechanism behind the observed relative treatment efficacy in COSMIC-313 is expected to be independence rather than synergy or additivity. It is therefore possible that the lower starting dose of cabozantinib at 40 mg daily may explain, at least in part, the limited efficacy signal. Intermediate endpoints such as PFS derive more benefit from the random treatment assignment in contemporary oncology RCTs, whereas OS is more likely to be biased by mediator-outcome confounding. However, under typical phase 3 RCT designs such as that of COSMIC-313, these challenges in OS estimation are not expected to reverse the conclusions between PFS and OS signals in the contemporary first-line metastatic clear-cell renal cell carcinoma setting. Thus, we should expect OS to either show a positive signal of relative treatment effect efficacy in favor of the triplet combination or at least be inconclusive.
Those who look into imbalances between the treatment groups to explain the observed relative treatment effect estimates will be committing the all-too-common table I fallacy. Similarly, forest plots of subgroup differences in relative treatment effect efficacy on the HR scale are likely to be inconclusive, although we may occasionally be surprised. Prognostic risk biomarkers are typically far more clinically actionable. Accordingly, should this triplet regimen obtain regulatory approval based on the COSMIC-313 results, it will be a reasonable first-line choice in patients with poor-prognosis metastatic clear-cell renal cell carcinoma who are at high risk of being unable to receive a second-line therapy.
An additional key determinant for patient-centered decisions will be the toxicities and impact on patient-reported outcomes from the triplet combination compared with control. The introduction of triplet therapies, and maybe even quadruple therapies in the future, should compel us to put more emphasis on reliable and transportable communication of adverse event estimates in RCTs, eg, by presenting the covariate-specific causal risk differences or by optimizing visual displays. That said, the observed marginal grade 3 or higher treatment-related adverse event (TRAE) frequency of 73% in the triplet arm is consistent with what has previously been observed with combinations of tyrosine kinase inhibitors, such as cabozantinib, with immune checkpoint therapies in metastatic clear-cell renal cell carcinoma. Similarly, the TRAE frequency of 41% in the control arm is consistent with what has been observed with nivolumab plus ipilimumab.
References:
Adashek JJ, et al. Cancer Treat Res Commun. 2020;doi:10.1016/j.ctarc.2020.100166.
Choueiri TK, et al. N Engl J Med. 2021;doi:10.1056/NEJMoa2026982.
Cornelius V, et al. Trials. 2020;doi:10.1186/s13063-020-04903-0.
Cuzick J. JNCI Cancer Spectr. 2018;doi:10.1093/jncics/pky006.
Hahn AW, et al. Ther Adv Med Oncol. 2022;doi:10.1177/17588359221103199.
Kaufman S, et al. J Stat Plan Inference. 2009;doi:10.1016/j.jspi.2009.03.024.
Kotecha RR, et al. J Immunother Cancer. 2021;doi:10.1136/jitc-2021-004059.
Msaouel P. Cancer Invest. 2021;doi:10.1080/07357907.2021.1974030.
Msaouel P, et al. Cancers (Basel). 2021;doi:10.3390/cancers13112741.
Msaouel P, et al. Eur Urol Oncol. 2022;doi:10.1016/j.euo.2021.09.003.
Plana D, et al. Cancer Discov. 2022;doi:10.1158/2159-8290.CD-21-0212.
Rafi Z, et al. BMC Med Res Methodol. 2020;doi:10.1186/s12874-020-01105-9.
Rothman KJ, et al. Modern Epidemiology. 3rd ed.. Wolters Kluwer Health/Lippincott Williams & Wilkins; 2008.
Senn S. Stat Med. 1994;doi:10.1002/sim.4780131703.
van Zwet E, et al. Stat Med. 2021;doi:10.1002/sim.9173.
Collapse
Choueiri TK, et al. Abstract LBA8. Presented at: European Society for Medical Oncology Congress; Sept. 9-13, 2022; Paris.
Click Here to Manage Email Alerts