During a Targeted Oncology™ Case-Based Roundtable™ event, Thomas J. Herzog, MD, discussed the results of the KEYNOTE-775 trial of lenvatinib plus pembrolizumab versus chemotherapy in patients with advanced endometrial cancer.
Targeted OncologyTM: Which trial data supports the use of lenvatinib (Lenvima) and pembrolizumab (Keytruda) in patients with advanced, metastatic, or recurrent endometrial cancer?
The data came out in the New England Journal of Medicine, [and are] very exciting data.1 This is the phase 3 [KEYNOTE-775; NCT03517449] trial with 1:1 randomization, looking at chemotherapy versus lenvatinib/pembrolizumab. Carboplatin/paclitaxel is not the control arm here. Most of these patients have had [prior] carboplatin/paclitaxel.
[The comparator arm] was doxorubicin or paclitaxel, given on a weekly basis for the paclitaxel with 3 weeks on and 1 week off. These were patients with advanced metastatic or recurrent endometrial cancers. The investigators did blinded independent central review [BICR] for assessment of response or progression. They had performance status of 0 or 1 and they all had MMR [mismatch repair] testing. They stratified based on MMR status, region of the world, performance status, and whether they had radiation therapy in their history. The primary end point was progression-free survival [PFS] by BICR with a coprimary end point of overall survival [OS].
Many of the demographic baseline characteristics have prognostic implications for endometrial cancer. Age is about the same [between arms]. Unfortunately, this trial was under-enrolled for Black patients. We’re looking at a target goal of 15% to 20%, and this was very under-enrolled at [4.1% and 3.4% in the experimental and comparator arms, respectively].
This is an area where the FDA has come out and [said] they want diversity plans for all clinical trials moving forward. I’m involved with the GOG Foundation and we take this seriously. We’ve organized a committee around this, with Bhavana Pothuri, MD, from NYU Langone Health who has been named our Diversity Equity Inclusion officer.
We are actively engaged with the FDA as to what success looks like in this area because it varies. In some of these trials, it’s very complicated because some of [the trials] are molecularly based, where patients have to have a certain mutation and there are differences in the rates of mutations in different populations. The bar is always moving, but it’s an area that we need to pay much more attention to, so I’ll put that out there.
Most of the patients were MMR proficient [MMR-P], about 84%, with about 16% being MMR deficient [dMMR]. For performance status, they were all 0 to 1 with most of it being 0. There was a little imbalance [between the arms in terms of] pelvic radiation, but there were 42% [who had received it] in the lenvatinib/pembrolizumab arm and 45% in the chemotherapy arm. The different histologies were pretty well balanced when you look at that as well. Everything overall is pretty well balanced in this trial.
What were the efficacy results of the KEYNOTE-775 trial?
The median PFS in the MMR-P population was 6.6 months [with lenvatinib/pembrolizumab] versus 3.8 months [with chemotherapy], and an HR of 0.6 [favoring lenvatinib/pembrolizumab]. The confidence interval doesn’t come close to crossing 1.00 [95% CI, 0.50-0.72; P < .001], so it was highly statistically significant for the primary end point.
If you look at the overall population, they did slightly better because we’re now including those who are MMR deficient [dMMR] and they tend to respond especially well to the immunotherapy. We have a 7.2 months median PFS for lenvatinib/pembrolizumab versus 3.8 months for chemotherapy, with a very impressive HR of 0.56. All the subgroups benefitted. There appeared to be subgroups that benefitted regardless of age, histologic features, and previously lines of therapy as well. There was always a benefit in all those for the group that received lenvatinib/pembrolizumab over chemotherapy. There are consistent data now.
For OS, the HR is 0.62 and median OS is 18.3 months with lenvatinib/pembrolizumab versus 11.4 months with chemotherapy, so you’re gaining almost 7 months, which is definitely significant. [There was OS] benefit across all the subgroups there.
Having an objective response is important, as a treating physician. A responding patient is one of your shortest visits, and a nonresponding patient is one of your longest visits to your clinic. Having just gotten done with a very long clinic yesterday, I know that to be true. [In the trial], the objective response rate [ORR] in the MMR-P group was 30.3% with lenvatinib/pembrolizumab versus 15.1% with chemotherapy. Lenvatinib/pembrolizumab pretty much doubled [the rate of] complete responses and partial response. For all patients, it was the same thing; ORR [in patients with lenvatinib/pembrolizumab] more than doubled the ORR versus chemotherapy. Then it was impressive in the dMMR group with 40% ORR with lenvatinib/pembrolizumab versus 12% with chemotherapy, with a 14% complete response rate with lenvatinib/pembrolizumab [versus only 3% with chemotherapy] which is very impressive in a trial like this. These are [objective responses] with RECIST criteria, confirmed [as opposed to best response at] just 1 point in time.
The median duration of response [DOR] in [the MMR-P group] was 9.2 months with lenvatinib/pembrolizumab versus 5.7 months with chemotherapy. Then in all patients, 14.4 with lenvatinib/pembrolizumab versus 5.7 months with chemotherapy. [In the dMMR group], the DOR was not reached with lenvatinib/pembrolizumab versus 4.1 months for chemotherapy. These are some pretty big differences.
Time to response was pretty fast overall; lenvatinib/pembrolizumab was a little faster than the chemotherapy. So, if you have someone who is symptomatic, that is something to think about. Disease control [including] stable disease was pretty high, 71.7% with lenvatinib/pembrolizumab versus 46% with chemotherapy [in the MMR-P group], 72% versus 47% with chemotherapy in all patients, and 74% versus 48% [respectively, in the dMMR group]. So it was pretty impressive. I think this was a very important paper.
What was the safety and tolerability observed with this combination regimen?
Looking at the adverse events [AEs] that occurred with this combination…in greater than 25%, there was quite a bit of toxicity with any of these AEs. But investigators saw with grade 3 or higher toxicities that hypertension was much higher in the lenvatinib/pembrolizumab group, which those of us who use it [already] know. Hypothyroidism was more common, most of it being grade 1 or grade 2, but that was extremely common in a lot of our patients. Diarrhea, [which was mostly] grade 1 or grade 2, [occurred] in 54% with lenvatinib/pembrolizumab versus 20% in chemotherapy. For greater than grade 3 diarrhea, it was 7.6% with [lenvatinib/pembrolizumab versus 2.1% with chemotherapy].
Then everything else starts going down fairly significantly after you get past the nausea, decreased appetite, and vomiting. Weight decrease of grade 3 or higher was over 10% with lenvatinib/pembrolizumab, so those are all things that you need to pay attention to in your patient’s nutrition. Anemia was higher in the chemotherapy arm, and neutropenia is quite a bit higher in the chemotherapy arm. Alopecia was also much higher in the chemotherapy arm.
1. Makker V, Colombo N, Casado Herráez A, et al. Lenvatinib plus pembrolizumab for advanced endometrial cancer. N Engl J Med. 2022;386(5):437-448. doi:10.1056/NEJMoa2108330