Vicky Makker, MD, leads a discussion of efficacy data from the KEYNOTE-775 trial on combination lenvatinib/pembrolizumab in patients with endometrial cancer.
Ramez Eskander, MD: There’s no one better for me to ask than you, of course, being the leader of this clinical trial and transforming the space. But could you take a moment and talk to us a bit about what was the design of KEYNOTE-775 and share with us the data that really informed the approval of this regimen, full approval after initial accelerated approval, helping our patients who are afflicted by recurrent endometrial cancer?
Vicky Makker, MD: Absolutely. So as you know, this was a phase 3 randomized trial of an oral TKI [tyrosine kinase inhibitor] called lenvatinib, which targets VEGF [vascular endothelial growth factor] receptors 1-3, FGFR [fibroblast growth factor receptors] 1-4, PDGFR [platelet derived growth factor receptor] alpha, as well as oncogenes RET and KIT. This agent was combined with a PD-1 [programmed cell death protein 1] inhibitor of pembrolizumab vs a treatment of physician’s choice [TPC] in patients with advanced or recurrent endometrial cancers. With regards to eligibility criteria for this study, the criteria included advanced metastatic or recurrent endometrial cancers. Patients had to have measurable disease by RECIST [response evaluation criteria in solid tumors] criteria and assessments were performed by blinded independent central review on this particular study. Patients had to have 1 but were allowed 2 prior lines of platinum-based chemotherapy as long as 1 line of therapy was administered in the neoadjuvant or adjuvant setting. There was no restriction on prior hormonal exposure for this particular study and patients were stratified by their MMR [mismatch repair] status. And then within the MMR proficient cohort, they were further stratified by ECOG [Eastern Cooperative Oncology Group performance] status, prior history of pelvic radiation therapy, and by region.
Eligible patients were randomized 1:1 to lenvatinib, 20 mg orally, once daily—and that dose stems from a phase 1v2study that established that as the recommended phase 2 dose and the dose that’s worthy of further study—plus pembrolizumab which was flat dosed 200 mg IV [intravenous] every 3 weeks. The question comes up about dose-reducing pembrolizumab and so this is a flat-dose regimen. We don’t dose-reduce it. We discontinue it if there’s an immune-mediated adverse event that requires it. Patients who were on the TPC arm randomized to either doxorubicin 60 mg/m2 IV every 3 weeks, or paclitaxel 80 mg/m2 weekly 3 weeks on-1 week off. And again, these are commonly employed regimens globally speaking and so they were utilized in the TPC arm. So on this study, consistent with the label, pembrolizumab was administered for a maximum of 35 doses, approximately 2 years. Lenvatinib could continue beyond that point if patients were clinically benefiting and the maximum cumulative dose of doxorubicin allowed on trial was 500 mg/m2.
Primary end points were progression-free survival [PFS] by blinded independent central view and overall survival [OS]. Secondary end points included objective response rate and appropriately quality of life analyses. A key exploratory end point was duration of response. In this study, all the major epithelial endometrial cancer histologies were allowed with the exception of carcinosarcomas. Just over 820 patients were randomized between the 2 arms. The arms were very well balanced with regards to age, prior radiation, ECOG status, and importantly race distribution. 84% of patients on both arms were MMR proficient. The full histologic spectrum was balanced between the arms. As expected, endometrial adenocarcinomas were the most common histology; they represented about 60% of the cases in both arms. Importantly, the distribution of low-grade, high-grade endometrioid adenocarcinomas was balanced and comparable between the arms. With regards to the non-endometrioid histology, serous cancers were the most common at just over 25%. Essentially 79% of patients on the lenvatinib/pembrolizumab arm and 76% of patients on the TPC arm had 1 prior platinum-based therapy.
Essentially, the results showed very compellingly both clinically and statistically significant improvements in both progression-free survival and overall survival. The median PFS for the MMR proficient cohort was 6.5 vs 3.8 months for lenvatinib/pembrolizumab vs TPC. In the all-comers, which of course included the dMMR [deficient mismatch repair] patients, the median PFS was 7.2 vs 3.8 months, in favor of lenvatinib/pembrolizumab with a follow-up of 11.4 months overall survival. The PMR [polymyalgia rheumatica] cohort was 17.4 vs 12 months in favor of lenvatinib/pembrolizumab. OS for all comers was 18.3 vs 11.4 months for lenvatinib/pembrolizumab vs TPC. When we did subgroup analyses of treatment effect on PFS and OS for all comers, what we saw was a significant advantage for lenvatinib/pembrolizumab with regards to age, race, region, MMRP [mismatch repair protein] status, prior pelvic radiation, and histology along with prior lines of therapy.
When you look at the dMMR patients, which I know is a very important cohort that we wanted to really drill down and understand more about, what we saw was that the median PFS for this important cohort was 10.7 vs 3.7 months for lenvatinib/pembrolizumab vs TPC. With a median follow-up of 1 year, the OS for the dMMR cohort was not reached vs 8.6 months in favor of lenvatinib/pembrolizumab. The objective response rate in the dMMR cohort was also 40% vs 12%, in favor of lenvatinib/pembrolizumab. Importantly, when quality of life was assessed on this study, no substantial differences in health-related quality of life scores were observed over time between the 2 treatment groups which was really heartening and I think very important to see when one considers the adverse events and toxicities that can come on with this regimen.
Ramez Eskander, MD: Every time I hear you summarize the data, it motivates me. Honestly, I think about a doubling in the median progression-free survival, [and] over a 50% improvement in overall survival in a patient population where not long ago, we were just saying chemotherapy or carboplatin/paclitaxel because you’ve been off of chemotherapy for 8 months and we think that the platinum re-exposure is going to work again which all comes from small retrospective trials which we know are confounded on multiple variables. So I don’t think we can emphasize enough how relevant this is to a cohort of patients that [we] really didn’t have very much to offer.
Transcript edited for clarity.