Brown Examines Lenvatinib/Pembrolizumab Combination in pMMR Endometrial Cancer

Peers & Perspectives in OncologyMay 2024
Volume 2
Issue 7
Pages: 38

During a Case-Based Roundtable® event, Jubilee Brown, MD, discussed the efficacy and safety data from the KEYNOTE-775/Study 309 trial of lenvatinib plus pembrolizumab in patients with mismatch repair proficient advanced endometrial cancer.

Brown, Jubilee

Jubilee Brown, MD

Professor of Gynecologic Oncology

Atrium Health Levine Cancer Institute

Charlotte, NC


  • A postmenopausal woman aged 71 years presented with abnormal uterine bleeding, increasing urinary frequency, and nausea/cramping for approximately 6 months.
  • She has 2 grown children and no known family history of cancer.
  • Her body mass index is 32; she has had type 1 diabetes since childhood, well controlled with medication.
  • Physical examination was notable for enlarged uterus and right lower quadrant abdominal tenderness on palpation.
  • ECOG performance status: 1
  • CT of chest, abdomen, and pelvis showed uterine and bladder masses.
  • Cancer antigen (CA) 125 level: 38.6 U/mL
  • Endometrial biopsy: endometrioid adenocarcinoma, International Federation of Gynecology and Obstetrics stage IVA, grade 3 (poorly differentiated)
  • Notable molecular features: mismatch repair proficient (pMMR), microsatellite stable (MSI-stable), estrogen receptor negative, HER2 negative, NTRK negative
  • Germline testing showed no pathogenic variants.
  • Carboplatin/paclitaxel chemotherapy was initiated; treatment was well tolerated other than grade 1 peripheral neuropathy and vomiting.
    • Four months after initiation, she had no evidence of disease on PET scan, with complete resolution of symptoms.
  • Twelve months post chemotherapy , rising CA 125 level was documented.
  • CT scan of the abdomen and pelvis showed growth of bladder mass and new suspicious peritoneal lymph nodes.
  • PET scan: intense fluorodeoxyglucose F 18–avid lesions in lungs and peritoneum, as well as para-aortic lymph nodes
  • Interventional radiology biopsy revealed recurrent endometrial carcinoma.



  • The patient received lenvatinib (Lenvima) at 20 mg daily plus pembrolizumab (Keytruda) 200 mg every 3 weeks as second-line therapy.

What data support the combination of lenvatinib and pembrolizumab in the pMMR population following initial systemic chemotherapy?

BROWN: If we look at KEYNOTE-775/Study 309 [NCT03517449], this is a phase 3 trial in patients with advanced, metastatic, or recurrent endometrial cancer who had measurable disease with 1 prior platinum-based chemotherapy. This gets back to our National Comprehensive Cancer Network guidelines, where we typically treat with paclitaxel/carboplatin first. Patients had to have had a good performance status and tissue available for MMR testing. They were stratified by their MMR status and by their region, performance status, and history of radiation. They were randomly assigned 1:1 to either lenvatinib and pembrolizumab—remember that the starting dose of lenvatinib in Study 309 was 20 mg/day—vs either doxorubicin or paclitaxel, based on physician’s choice. The primary end point was progression-free survival [PFS] by blinded independent central review and overall survival. There were secondary end points such as overall response rate [ORR], quality of life, pharmacokinetics, and safety, and duration of response [DOR] was a key exploratory end point.

The baseline characteristics were remarkably similar in patients in each arm.1 There weren’t any significant differences between the arms. One thing to call out is the presence of serous and clear cell carcinomas in some of these patients. The majority were endometrioid carcinomas, but there were 20% of patients with serous carcinomas.

With regard to indications and usage, we know that lenvatinib is a kinase inhibitor, indicated for endometrial cancer in combination with pembrolizumab for the treatment of patients with advanced endometrial cancer that is pMMR as determined by an FDA-approved test or not MSI high, so MSI stable. These patients should have disease progression following prior systemic therapy in any setting and are not candidates for curative surgery or radiation.

What were the final efficacy outcomes for this combination?

[Concerning] the final overall survival, if you look just at the pMMR population, the [Kaplan-Meier] curves are quite different.2 These curves are statistically significantly different with 12 months [95% CI, 11.0-14.1] for patients with chemotherapy and 18 months [95% CI, 14.9-20.5] for patients with lenvatinib/pembrolizumab [HR, 0.70; 95% CI, 0.58-0.83]. What I think is interesting in this is that these curves never touch and they never come together over…42 months later.

What is also important is that this is true in the all-comer population [HR, 0.65; 95% CI, 0.55-0.77], but it holds true in the pMMR population, who are so notoriously difficult to treat. More new information is that there were similar results observed when patients in the chemotherapy arm who received subsequent lenvatinib or pembrolizumab or patients who received any subsequent checkpoint inhibitor therapy were excluded. This was published later but underscores the lasting improvement of patients who get lenvatinib and pembrolizumab, as in this study. When you say overall survival, you get your wish here.

When we talk about PFS, you can see it’s the same picture. When we look at this, there is a difference of 7.3 months [median PFS favoring the experimental arm in the all-comer population vs a difference of] 3.8 months in the pMMR population, mirroring what we see in the all-comer population.

The ORR was 30% of patients [with lenvatinib/pembro-lizumab] compared with 15% of patients with chemotherapy in the pMMR population.1 That’s [similar to the ORR in] all patients [Table1]. What’s so striking is that not only did we see improvement in the dMMR [mismatch repair deficient] population, but it holds true in the pMMR population too. There was a small percentage of patients who had a complete response. But DOR is important too. We know that in the dMMR population, DOR is not reached with lenvatinib and pembrolizumab, but we see an improvement in the pMMR population, too, a 9-month median DOR [vs 5.7 months with chemotherapy]. Remember that median means that there are patients on either side of that. Either way, it’s better than chemotherapy alone, and they do respond quickly, approximately 2 months until response. Also, importantly in these patients, we look at disease control because a lot of these patients may not be symptomatic when they present, so as long as you can keep them from progressing, that’s a win, and so 71% [with lenvatinib/pembrolizumab] compared with 46% for chemotherapy alone had disease control in this hard-to-treat pMMR population.

Confirmed Tumor Responses in Study 309/KEYNOTE-775 (NCT03517449)


  • At 3-month follow-up, the patient exhibited a good partial response but complained of occasional morning headache. Her vital signs showed hypertension compared with baseline, but electrocardiogram showed no abnormalities. Her abdomen and pelvis were nontender with no evidence of palpable masses. The physician’s impression was grade 2 VEGF receptor inhibitor–associated hypertension.

What adverse events (AEs) were of most concern in this trial?

We know about the AEs of any cause now that we use this so much. Almost everybody, no matter whether they got chemotherapy or lenvatinib/pembrolizumab, had some AE.1 We know that that’s to be expected with any intervention that we give. What stands out is that we do see some hypertension with lenvatinib/pembrolizumab. With the dose that was started, 20 mg/d on this trial, 38% of patients had grade 3 or higher hypertension. We do see some hypothyroidism with the pembrolizumab. We see some diarrhea and some nausea, but these are generally grade 1/2 and easily managed. There is not a lot of grade 3 toxicity, so nothing new or striking there, other than the hypertension that’s a little different from certain other kinds of medications.

A nice graphic that shows adverse responses to lenvatinib and pembrolizumab [was published in The Oncologist], and what’s striking about this is the median time to first onset.3 When do you start looking for [AEs]? Hypertension happens early, approximately 2 weeks after lenvatinib/pembrolizumab is started. Fatigue is the same story; it happens early. Things like stomatitis and nausea tend to happen a month or so later. You’re going to look for the hypothyroidism even up to a couple of months later, so don’t forget about that.


1. Makker V, Colombo N, Casado Herráez A, et al; Study 309–Keynote-775 Investigators. Lenvatinib plus pembrolizumab for advanced endometrial cancer. N Engl J Med. 2022;386(5):437-448. doi:10.1056/NEJMoa2108330

2. Makker V, Colombo N, Casado Herráez A, et al. Lenvatinib plus pembrolizumab in previously treated advanced endometrial cancer: updated efficacy and safety from the randomized phase III Study 309/KEYNOTE-775. J Clin Oncol. 2023;41(16):2904- 2910. doi:10.1200/JCO.22.02152

3. Colombo N, Lorusso D, Monk BJ, et al. Characterization and management of adverse reactions in patients with advanced endometrial cancer receiving lenvatinib plus pembrolizumab. Oncologist. 2024;29(1):25-35. doi:10.1093/oncolo/oyad201

Related Videos
Related Content