Abstract
Background and objectives:
Endometrial cancer is a common malignancy and recurrences can be fatal. Although platinum-pretreated endometrial tumors are commonly treated with anthracyclines and taxanes, there is no current standard of care. Both immune checkpoint inhibitors (ICIs) and tyrosine kinase inhibitors (TKIs) have been extensively assessed in this setting, including tumors selected for DNA mismatch repair (MMR)/microsatellite instability (MSI) and programmed death-ligand 1 expression status. This review will provide evidence-based guidance on use of ICIs alone or in combination with TKIs in patients with pretreated advanced, persistent, or recurrent metastatic endometrial cancer.
Data sources and methods:
Randomized phase II–III trials in unselected populations pretreated, recurrent, or metastatic endometrial cancer and phase I–II trials in biomarker selected populations were identified from PubMed as well as conference proceedings using the key search terms ‘immune checkpoint inhibitors’, ‘endometrial cancer’, and ‘advanced’.
Results:
A total of nine eligible studies were identified assessing ICI monotherapy for biomarker-selected or ICI plus TKI combinations and a dual ICI regimen for biomarker-unselected patients with pretreated recurrent or metastatic endometrial cancer. In MMR/MSI-selected tumors, five phase I/II studies evaluated ICI monotherapy indicating benefit in these patients. Only the phase III KEYNOTE-775 trial reported a statistically significant overall survival improvement for the combination of pembrolizumab plus lenvatinib compared with docetaxel or paclitaxel regardless of MMR/MSI status.
Conclusions:
Pembrolizumab plus lenvatinib is indicated for patients with unselected pretreated metastatic endometrial cancer and pembrolizumab monotherapy is a preferred option for patients with MMRd/MSI-H tumors.
Keywords
Introduction
Endometrial cancer is a common malignancy, with nearly 418,000 new cases resulting in approximately 97,000 deaths worldwide in 2020. 1 The global burden of endometrial cancer is increasing, 2 with a 3.7 times higher incidence in developed and developing countries. 1 For example, endometrial cancer is the most common gynecological malignancy in the United States, with an estimated 66,000 new cases resulting in nearly 12,500 expected deaths in 2021 and 2022.3,4 There is a rising incidence of endometrial cancer among young women in developed countries, related to a rise in obesity.5–8 Advanced stage disease is present at diagnosis in approximately 10–15% of patients and is associated with a poor prognosis. 9 Among patients with distant metastases, the 5-year survival rate is only 17–18%.3,4 Furthermore, up to 20% of patients presenting with early-stage disease experience disease recurrence a median of 3 years after diagnosis with the greatest risk in the first 2 years after surgery.10,11
Traditionally, platinum-based chemotherapy has been the standard of care for advanced or recurrent endometrial cancer.12–14 Objective response rates (ORRs) range from 40% to 62% for carboplatin plus paclitaxel, 15 although response duration and long-term survival rates are short. 16 Until recently, there were no standard second line treatments following progression on first-line platinum-based chemotherapy, with low response rates to subsequent cytotoxic regimens. As such ORRs for single agent weekly taxanes or anthracyclines are 27% and 10%, respectively.13,14,17 There has been significant need to identify more effective treatment options for these patients.
In the search for new therapies, a variety of targeted agents including mTOR, PIK3CA, AKT, EGFR, and MEK inhibitors have shown only modest and transient activity in endometrial cancer. 18 However, immune checkpoint inhibitors (ICIs), including programmed death-ligand 1 (PD-L1) and cytotoxic T lymphocyte-associated antigen-4 (CTLA-4) inhibitors, as well as multitargeted antiangiogenic tyrosine kinase inhibitors (TKIs) are of interest based on initial studies. 19
Exploratory analyses to date have been largely unsuccessful in defining subgroups more likely to benefit from particular therapies. This could be due to a variety of factors, including biomarker expression heterogeneity or discordance of biomarker expression from early-stage to recurrent disease. More recently, multi-parametric molecular profiling of endometrial cancer has identified four discrete subgroups of endometrial cancer with distinct molecular and clinicopathologic characteristics and prognosis. The four main molecular groups can be described by the following features: mismatch repair deficient (MMRd)/microsatellite instability-high (MSI-H), DNA polymerase epsilon exonuclease domain mutated, copy number high (with pathognomonic TP53 mutation), and copy number low which now is more commonly referred to as no specific molecular profile.20–23 Microsatellites are composed of short repetitive DNA sequences prone to deletion and expansion mutations due to DNA polymerase pausing and slippage during replication, and mutations in DNA repair systems such as MMR result in tumor cells with high MSI leading to the MMRd/MSI-H subtype. 24 The majority of patients with endometrial cancer have tumors that are MMR proficient (MMRp) and microsatellite stable (MSS), although the less common MMRd/MSI-H subtype (13–33%) is the most thoroughly characterized for targeted therapy approaches.19,25 MMRd/MSI-H endometrial cancer is highly immunogenic with increased tumor-infiltrating lymphocytes and has been associated with enhanced PD-L1 expression.19,25,26 As a result, MMRd/MSI-H along with PD-L1 expression has been used as biomarkers to select patients that might benefit from ICI therapy.
This review will explore contemporary practice-changing evidence for the use of ICIs either as monotherapy in patients with MMRd/MSI-H or PD-L1-positive endometrial cancer or as combination therapy in unselected patients with recurrent endometrial cancer and disease progression on platinum-based chemotherapy. Randomized trials were prioritized in platinum pretreated advanced endometrial cancer while phase I/II trials were also considered in MMRd/MSI-H or PD-L1-positive populations.
Methods
A search of published and presented literature was conducted to identify potentially practice-changing studies with outcomes assessing ICIs in pretreated metastatic endometrial cancer. Potentially practice-changing investigations included non-randomized studies of monotherapy in populations selected for relevant biomarkers such as MMR, MSI, and PD-L1 expression status and randomized trials of ICI combinations in unselected patients. PubMed (all time to 2, February 2022), the proceedings from the 2020 and 2021 American Society of Clinical Oncology, the European Society for Medical Oncology annual meetings and the 2020, 2021, and 2022 Society of Gynecologic Oncology annual meeting were searched using the key search terms (“immune checkpoint inhibitors”) AND “endometrial cancer” AND “advanced” AND “phase I-III” OR respective aliases. A supplemental bibliographic search of review articles and pooled/meta-analyses was also conducted. In addition, directed searches were performed after the database search cutoff date to ensure that the most up-to-date reports of eligible studies were considered.
English language records were vetted at abstract level and confirmed at full text as needed. Excluded studies included those that were non-original research, preclinical, correlative science, and those not specific to endometrial cancer or with endometrial cancer cohorts with fewer than 15 patients. In addition, retrospective, prospective IIIb, IV or undefined phase, non-randomized phase I–II trials in unselected populations were excluded as were studies without outcomes specific for endometrial cohorts as well as those not addressing ICIs treatments, and duplicate or prior reports.
Findings
The literature search identified a total of 477 records resulting in a total of nine eligible studies. Six investigated ICI monotherapy, including five phase I/II studies in MMRd/MSI-H selected populations3,27–33 and one phase Ib study in patients with PD-L1-positive tumors. 34 Three studies also assessed ICI combinations in unselected patients, including one phase III trial and a randomized phase II study investigating ICI plus TKI combinations35,36 and another assessing a dual ICI combination 37 (Preferred Reporting Items for Systematic Reviews and Meta-Analyses, Figure 1).

PRISMA diagram of eligible studies.
ICI monotherapy in MMRd and/or MSI-H or PD-L1-positive pretreated metastatic endometrial cancer
Four phase I/II studies investigated ICIs as monotherapy for MMRd and/or MSI-H pretreated metastatic, advanced, recurrent, or persistent endometrial cancer.25,27–33,38,39 The phase II KEYNOTE-158 study evaluated pembrolizumab in 79 patients, approximately 60% of which had received more than one line of therapy for recurrent/metastatic disease. 25 With a median follow-up of 42.6 months, pembrolizumab demonstrated an ORR of 48% with median duration of response (DoR) not yet reached (NYR), a median progression-free survival (PFS) of 13.1 months, with overall survival (OS) immature at the time of analysis (Table 1). 32 Approximately half of patients completed the planned 2 years of treatment and 68% of responders estimated to have a DoR ⩾3 years. No new safety signals were seen, no treatment-related deaths were reported (Table 3). A smaller phase II study of pembrolizumab in 24 patients with MMRd/MSI-H recurrent or persistent endometrial cancer (median 1 prior line) reported an ORR of 58% with DoR not reported (NR), a median PFS of 23.5 months and a median OS of 40.0 months. 28 The safety profile was acceptable, and no treatment-related deaths were reported. PD-L1 inhibitors have also been evaluated in two nonrandomized phase II trials in patients with pretreated advanced or recurrent disease.27,29 Durvalumab was assessed in 71 patients with MMRd (n = 36) or MMRp (n = 35) pretreated advanced endometrial cancer, with more patients in the MMRd subgroup receiving durvalumab as first-line therapy (60% versus 8%). 27 With a median follow-up of approximately 20 months, ORRs of 47% versus 3% were seen in MMRd versus MMRp patients, with a median PFS of 8.3 versus 1.8 months and median OS NYR versus 12.1 months, respectively. Treatment was well-tolerated with no treatment-associated deaths reported (Table 3).
Efficacy outcomes of ICI monotherapy trials in biomarker-selected, platinum-pretreated, metastatic endometrial cancer populations.
Efficacy outcomes from single-arm phase I and phase II trials or subgroup analyses of large randomized trials of immune checkpoint and/or TKIs in molecularly selected, advanced endometrial cancer subsets organized by agent class and ordered by date of first report of results.
Efficacy outcomes based on radiographic assessment are per RECIST by IRC or otherwise footnoted.
Type of radiological assessment (investigator or by independent review) not specified.
p Value for the comparison Lynch-like versus MLH1-methylated.
MMRp tumors were defined as those expressing all four MMR proteins (MLH1, MSH2, MSH6, and PMS2) while MMRd tumors as those showing loss of at least one of the four proteins.
By RECIST v.1.1 modified for immune-based therapeutics (iRECIST).
Investigator-assessed.
CI, confidence interval; CT, chemotherapy; EC, endometrial cancer; ICI, immune checkpoint inhibitor; IRC, independent review committee; MMR, mismatch repair; MMRd; mismatch repair deficient; MMRp, mismatch repair proficient; MSI-H, microsatellite instability-high; MSS, microsatellite stable; n, number of patients; NR, not reported; NYR, not yet reached; OS, overall survival; PD, progressive disease; PD-L1, programmed death-ligand 1; PFS, progression-free survival; POLE, polymerase epsilon; QD, daily; qxw, every x weeks.
Avelumab was evaluated in a phase II study involving 31 patients with chemotherapy pretreated MMRd (n = 15) or MMRp (n = 16) recurrent endometrial cancer, 71% of which received ⩾2 prior lines of therapy. With a median follow-up of 18.6 months, ORRs in MMRd versus MMRp subgroups were 26.7% versus 6.3%, with a median PFS of 4.4 versus 1.9 months, respectively. 29 No unexpected toxicities or treatment-associated deaths were reported.
The largest single agent PD-1 inhibitor trial reported to date providing evidence in this setting is the GARNET trial of dostarlimab in patients with advanced solid tumors, which included patients with MMRd/MSI-H (n = 143) and MMRp/MMS (n = 156) advanced or recurrent endometrial cancer having ⩽2 prior lines of treatment. 30 At the third interim analysis with median follow-ups of 27.6 and 33.0 months, ORRs of 45.5% and 15.4% were demonstrated for MMRd/MSI-H and MMRp/MSS subgroups, with median DoRs NYR and 19.4 months, respectively (Table 1). 31 Median PFS was 6.0 and 2.7 months and median OS was NYR and 16.9 months for MMRd/MSI-H and MMRp/MSS subgroups, respectively. Dostarlimab showed a manageable safety profile consistent with previous reports and no treatment-related deaths were reported (Table 3). 30
PD-L1 expression on tumor cells by immunohistochemical assay was employed as a biomarker in the phase Ib KEYNOTE-028 study to select patients that might benefit from pembrolizumab among the cohort of 23 patients with previously treated locally advanced or metastatic PD-L1-positive disease. With a median follow-up of 17.5 months, an ORR of 13.0% was demonstrated with a median PFS of 1.8 months and OS NYR. 34 No new safety signals or treatment-associated deaths were observed.
Randomized trials of ICI combinations in pretreated metastatic endometrial cancer
Three eligible immunotherapy combination trials enrolled patients pretreated for advanced, metastatic, recurrent, or persistent endometrial cancer without biomarker selection (Table 2). Two trials assessed ICI plus TKI combinations compared with single-agent therapy, either chemotherapy or an ICI, and a third one assessed dual ICIs versus an ICI alone.36,37,40 First, the phase III KEYNOTE-775 trial included 827 patients stratified by MMR status (MMRd tumors, 16% of total population). Patients were randomized to receive pembrolizumab plus lenvatinib (n = 411) or physician choice chemotherapy (doxorubicin or paclitaxel) (n = 416). The primary co-endpoints were PFS and OS. With a median follow-up of 12.2 months for the ICI plus TKI combination and 10.7 months for the chemotherapy arm, the investigational arm significantly outperformed chemotherapy for both median PFS [7.2 versus 3.8 months, hazard ratio (HR): 0.56, 95% confidence interval (CI): 0.47–0.66, p < 0.001) and median OS (18.3 versus 11.4 months, HR: 0.62, 95% CI: 0.51–0.75, p < 0.001). 36 ORRs for the ICI plus TKI combination and chemotherapy were 31.9% and 14.7% (p < 0.0001), and median DoRs were 14.4 and 5.7 months, respectively. Prespecified subgroup analyses indicated OS benefit for the combination over chemotherapy across all subgroups including by MMR status, age, race, ECOG PS status (0 and 1), prior pelvic radiation, endometrioid versus non-endometrioid histology, and number of prior therapies. A pre-planned analysis also assessed this combination in subgroups of patients with either MMRd/MSI-H (n = 130) or MMRp disease (n = 697) disease. 36 Significant improvements were seen for the ICI plus TKI combination versus chemotherapy in both MMRd/MSI-H and MMRp patients, with greater benefits for MMRd/MSI-H patients. Median PFS and OS for the ICI plus TKI combination versus chemotherapy in patients with MMRd/MSI-H disease were 10.7 versus 3.7 months (HR: 0.36, 95% CI: 0.23–0.57, p < 0.001) and NYR versus 8.6 months (HR: 0.37, 95% CI: 0.22–0.62, p < 0.001), respectively.
Overall efficacy outcomes from randomized trials of ICI combinations platinum-pretreated, metastatic endometrial cancer.
Efficacy outcomes of randomized phase II and phase III trials of checkpoint inhibitor combinations with targeted therapy compared to current standards of treatment in advanced endometrial cancer summarized and ordered by date of first report of results.
Efficacy outcomes based on radiographic assessment are per RECIST by IRC or otherwise footnoted.
Median follow-up for the overall trial population.
Type of radiological assessment (investigator or by independent review) not specified.
Significant according to the PFS significance level of 0.10 used for sample size calculation.
Calculated from the reported time in weeks using the conversion factor of 4.35 weeks/month.
CI, confidence interval; CT, chemotherapy; EC, endometrial cancer; HR, hazard ratio; ICI, immune checkpoint inhibitor; IRC, independent review committee; MMRd; mismatch repair deficient; MMRp, mismatch repair proficient; MSI-H, microsatellite instability-high; n, number of patients; NR, not reported; PFS, progression-free survival; QD, daily; qw, weekly; qxw, every x weeks; ST, systemic therapy; TKI, tyrosine kinase inhibitor.
These benefits were more pronounced compared to patients with MMRp tumors (median PFS 6.6 versus 3.8 months, HR: 0.60, 95% CI: 0.50–0.72, p < 0.001 and median OS 17.4 versus 12.0 months, HR: 0.68, 95% CI: 0.56–0.84, p < .001). ORRs were also greater for MMRd/MSI-H subgroup: 40% versus 9.2% for the ICI plus TKI combination versus MMRp patients. The response to chemotherapy was 12% and 5.7% for MMRd/MSI-H and MMRp subgroups, respectively. Median DoRs for the combination versus chemotherapy were NYR versus 4.1 months in those with MMRd/MSI-H tumors and 9.2 versus 5.7 months for MMRp disease. Subgroup analysis by histological subtypes suggested that the small subset of patients with clear cell tumors, which typically are difficult to treat, may benefit more from pembrolizumab plus lenvatinib treatment relative to endometroid or serous subtypes (OS HR: 0.33 versus 0.65 versus 0.68, respectively). 42 In the overall population, no substantial differences in the QLQ-C30 global health status quality-of-life (QoL) scores were observed between arms over time. A post-hoc analysis showed that time to first deterioration was generally similar between arms for most functional scores, with time to definitive deterioration favoring the ICI plus TKI combination for most assessed scales. 43 Grade ⩾3 adverse events (AEs) occurred in 88.9% versus 72.7%, AEs leading to treatment discontinuation occurred in 33.0% versus 8.0%, and deaths due to treatment-related AEs (TRAEs) occurred in 1.5% versus 2.1% of patients, respectively (Table 3). The most common grade ⩾3 TRAEs for the ICI plus TKI combination were hypertension (35.9%), diarrhea (6.2%), weight loss (5.9%), and decreased appetite (5.9%).
Safety outcomes from select trials of ICIs regimens in platinum-pretreated, metastatic endometrial cancer.
Safety outcomes from trials of immune checkpoint and/or TKIs in advanced endometrial cancer subsets organized by molecular selection criteria. Selected trials were those randomized that met their primary endpoint or those in MMRd-selected populations with promising clinical activity results (ORR > 20%). Grade ⩾3 AE treatment-related AEs were summarized when available.
All patients treated;
Possibly, probably, or definitely attributable to the study drug.
AE, adverse event; ALT, alanine transaminase; GI, gastrointestinal; iRAE, immune-related AE; MMRd; mismatch repair deficient; MMRp, mismatch repair proficient; MSI-H, microsatellite instability-high; MSS, microsatellite stable; n, number of patients; NR, not reported; ORR, objective response rate; TRAEs, treatment-related AEs.
Second, the NCI-2017-02211 randomized phase II study assessed the combination of nivolumab plus cabozantinib (n = 36) versus nivolumab alone (n = 18) in heavily pretreated patients, with all receiving at least two lines of prior therapy and 54% of patients receiving ⩾3 prior treatments. With the median follow-up not reached at the time of publication, the primary endpoint of PFS was considered statistically significantly longer for the combination (median 5.3 versus 1.9 months, HR: 0.59, 95% CI: 0.35–0.98, p = 0.09; P-level of .10 for sample size calculation), although median OS was not statistically significantly different (13.0 versus 7.9 months, p = 0.62) (Table 2). 40 ORRs were 25.0% versus 11.1% for the combination versus ICI monotherapy, respectively. Notably, a 25% ORR was reported for 20 patients on the combination therapy who had received prior ICI therapy. The ICI plus TKI combination was tolerable, with no grade 4 AEs reported and all grade 3 events associated with the addition of cabozantinib (diarrhea the most common, 11.1%). There was a single treatment-related death in the combination arm (sepsis).
The third study compared dual ICIs targeting both PD-L1 and CTLA-4, durvalumab plus tremelimumab, (n = 27) to durvalumab alone (n = 27) in pretreated recurrent or persistent endometrial cancer. With a median follow-up of 20.0 and 14.6 months, respectively, the primary endpoint of PFS was similar between the two arms (median 1.9 versus 1.7 months; Table 2). 37 ORRs were 11.1% versus 14.8%, with median DoRs of 1.8 and 3.7 months for the combination and monotherapy arms, respectively. No new safety signals were identified, with diarrhea the most common grade 3 TRAE reported in the combination arm and no treatment-associated deaths.
Discussion
Evidence from multiple trials demonstrates the efficacy of ICIs alone or in combination in patients with pretreated advanced, metastatic, recurrent, or persistent endometrial cancer.25,27,28,30,32-34,36,37,40,44,45
Role of ICI monotherapy in MMRd/MSI-H or PD-L1-positive pretreated metastatic endometrial cancer
MMRd/MSI-H disease
The most extensive experience with ICI monotherapy for patients with MMRd/MSI-H pretreated metastatic endometrial cancer is from two phase II studies (KEYNOTE-158 and 1605017712). The larger KEYNOTE-158 study and smaller 1605017712 study showed robust ORRs (48–58%) and prolonged median PFS rates (13.1–23.5 months) for pembrolizumab,27,30 while durvalumab and dostarlimab have shown strong ORRs (47% and 46%, respectively).27,31 Pembrolizumab received Health Canada approval for use as monotherapy in patients with MMRd/MSI-H pretreated metastatic endometrial cancer in March 2020, 46 with the Food and Drug Administration (FDA) following suit for the same indication in March 2022. 47 Dostarlimab received FDA accelerated approval in April 2021 for patients with MMRd pretreated metastatic endometrial cancer, 48 followed by Health Canada approval for MMRd/MSI-H pretreated metastatic disease in December of that year. 49
PD-L1-positive disease
PD-L1 expression has been used to select patients that might benefit from ICIs in pretreated metastatic endometrial cancer. Pembrolizumab, nivolumab, and atezolizumab have been assessed in uterine cancers with high PD-L1 expression,34,50,51 although the largest trial, phase Ib KEYNOTE-028, showed modest activity for pembrolizumab (ORR: 13%). 34 Further research is required before PD-L1 expression should be used to predict benefit from ICIs in patients with endometrial cancer.
Role of immunotherapy combinations in pretreated metastatic endometrial cancer
ICI combinations, including ICI plus TKI combinations36,40 and dual ICI combination, 37 have been assessed in recurrent and metastatic endometrial cancers. Only the completed phase III KEYNOTE-775 trial in biomarker-unselected patients has reported a statistically significant OS improvement for the combination of pembrolizumab plus lenvatinib compared to doxorubicin or paclitaxel, 36 a benefit that was apparent in patients with both MMRp and MMRd patients upon stratified subgroup analysis. This analysis suggested a promising benefit in the MMRd subgroup, demonstrating an ORR of 40%, a median PFS of 10.7 months and median OS NYR in these patients. 36 Although cross-study comparisons must be made with caution due to differences in study populations and study design, outcomes for the pembrolizumab and lenvatinib combination in KEYNOTE-775 were comparable to pembrolizumab alone in KEYNOTE-158 for MMRd/MSI-H disease. However, the combination was associated with increased toxicity (discontinuation due to TRAEs, 26.6% versus 7%) and deaths due to treatment-related (1.5% versus 0) and overall AEs (5.7% versus NR).25,32,36 Available data therefore suggest that ICI monotherapy is appropriate for MMRd/MSI-H pretreated metastatic endometrial cancer.
Despite higher rates of discontinuation of any treatment due to TRAEs reported with the combination of pembrolizumab and lenvatinib compared with chemotherapy (26.6% versus 5.7%) (Table 3), no substantial differences in QoL were measured by the QLQ-C30 global health status scores, 36 and a post-hoc analysis showed that the combination delayed the time to definitive QoL deterioration with benefits observed in most of the scales assessed. 43 However, the QLQ-C30 is designed for heterogeneous cancer populations, thus could lead to measurement non-invariance and biased group comparisons. 52 Finally, the 20 mg starting dose of lenvatinib used in this trial is controversial, a topic which is discussed in greater detail below.
The reported OS improvements support pembrolizumab plus lenvatinib as the new standard in platinum-pretreated recurrent and metastatic endometrial cancer. While the trial included both MMRp/MSS and MMRd/MSI-H patients, this combination was approved for pretreated metastatic MMRp/MSS endometrial cancer by Health Canada in February 2021 and by the United States FDA in July 2021.53,54 The European Medicines Agency approved this combination in November 2021 regardless of MMR/MSI status. 55 While a detailed pharmacoeconomic assessment is outside the scope of this review, it is important to note that combination regimens will likely increase the cost of treatment which may be especially relevant in scenarios where clinical benefits relative to available ICI monotherapy regimens may not be clearly apparent.
What are the safety considerations for ICI monotherapy and ICI plus TKI combinations?
In addition to efficacy, treatment decisions for patients with pretreated metastatic endometrial cancer must be made with careful consideration of the relative safety of available options. It is clear that the ICI plus TKI combination used in KEYNOTE-775 is less tolerable than ICI monotherapies, with nearly all patients in this trial experiencing a TRAE of any grade compared with 68–80% of patients receiving ICI monotherapies across studies (Table 3).27,28,31,32,36,56 AEs occurring with the ICI combination were also more severe, with grade ⩾3 TRAEs (78%) and treatment discontinuation due to TRAEs (27%) considerably more frequent compared with ICI monotherapies (12% to ⩽36% and ⩽4% to 7%, respectively).25,27,29,30,32,36–38,40 A comparison of specific toxicity frequencies for the ICI combination and mean frequencies seen across ICI monotherapy trials showed that the ICI combination was associated with more grade 1/2 and grade ⩾3 TRAEs across studies, almost without exception (Figure 2). The ICI-TKI combination showed substantial rates of any grade hypertension (61.1%), hypothyroidism (54.4%), nausea, diarrhea, and decreased appetite (around 40% each), fatigue (27.8%), proteinuria and vomiting (around 25%) compared with fatigue (approximately 30%), constipation and nausea (approximately 16–17% each) as the most common TRAEs associated with monotherapy based on the average rate across multiple trials. The combination was also associated with grade ⩾3 TRAEs hypertension of 35.9%, 36 underscoring the importance close patient monitoring.

TRAE rates in the phase III trial of pembrolizumab plus lenvatinib and average TRAE rates across multiple single ICI trials. TRAEs included were those reported for the pembrolizumab plus lenvatinib arm of the KEYNOTE-775 trial and/or any TRAE reported in two or more single ICI trials. Any cause AE rate data were used when TRAE rate data were not available (PHAEDRA trial).
The established lenvatinib dose for endometrial cancer settings is 20 mg, although this is associated with the judicious use of dose modifications (interruptions and reductions) as well as concomitant supportive care to manage toxicities. 41 In fact dose modifications (94%) and interruptions (69%) were very common in KEYNOTE-775, confirming that few patients are able to tolerate the full dose regimen (Table 3). 36 Lower doses of lenvatinib have been widely used in other cancer types, for example, renal cell carcinoma 18 mg once daily (OD) in combination with everolimus, 57 hepatocellular carcinoma maximum 12 mg OD as a single agent 58 ; however, a higher dose was selected for use in combinations with ICIs for endometrial cancer. A retrospective study of patients with pretreated metastatic endometrial cancer receiving the ICI combination showed improved safety among 54 patients given a reduced lenvatinib starting dose (83% received 14 mg daily) compared with 16 patients receiving the recommended dosage, without compromising efficacy outcomes. 59 For patients with an EGOG PS 1, the data are insufficient to recommend a lower starting dose although the data are reassuring. For patients with ECOG PS 2 who were not included in KEYNOTE-775, a lower initial dose may be considered, weighed against the unknown impact on outcomes, with the options of titrating to higher doses based on tolerability. Regardless of performance status, physicians and care teams should educate patients on expected side effects and the need to rapidly report issues, unequivocally monitor patients carefully with early and frequent AE assessment, respond quickly to any emerging side effects, and adopt a conservative low threshold for lenvatinib dose reduction.
What is the place in therapy for ICI and/or TKIs in pretreated metastatic endometrial cancer?
Therapy for pretreated metastatic endometrial cancer may involve multiple lines of therapy and should be guided by a multi-disciplinary team of specialists. Results from clinical trials reviewed here present new treatment options for patients for MMRp/MSS and MMRd/MSI-H tumors. For patients with MMRp/MSS disease, the combination of pembrolizumab and lenvatinib is a proven option for patients with ECOG PS of 0 or 1. For patients with MMRd/MSI-H tumors, single-agent pembrolizumab or dostarlimab are appropriate options. Following progression on either combination treatment or monotherapy, additional chemotherapy could be considered, but there are no data to indicate which agents may be most appropriate and generally speaking very modest benefits can be expected given the low activity of second-line chemotherapy previously reported.12,60
What is the ongoing research for ICIs and/or TKIs in endometrial cancer?
The role of targeted therapy in endometrial cancer is rapidly evolving with multiple phase III trials underway in all lines of therapy (Table 4). In molecularly unselected patients, first-line trials evaluating ICIs plus TKIs for metastatic disease include the combination of pembrolizumab plus lenvatinib compared to standard platinum–taxane chemotherapy (LEAP-001, NCT03884101 and NCT04865289) and durvalumab with or without olaparib plus platinum–taxane-based chemotherapy followed by maintenance durvalumab with or without olaparib (DUO-E, NCT04269200). ICIs plus chemotherapy are being assessed as first-line, second-line, and early adjuvant therapy. Pembrolizumab plus carboplatin and paclitaxel is being evaluated first line for stage III, stage IVA/B or recurrent endometrial cancer (NCI-2019-02186, NCT03914612) and pembrolizumab plus standard of care chemotherapy is being assessed in the adjuvant setting (KEYNOTE-B21, NCT04634877). Other regimens being assessed first line include atezolizumab (AtTEnd/ENGOT-en7, NCT03603184 and IRFMN-EN-7556, NCT03603184) and dostarlimab (RUBY/ENGOT-en6, NCT03981796), both in combination with carboplatin and paclitaxel. Additional second-line combinations under investigation include the EGFR inhibitor EG-007 plus pembrolizumab and lenvatinib (HPC-EG-007-3.1, NCT05077215) and the poly ADP-ribose polymerase inhibitor niraparib with or without dostarlimab in patients with endometrial carcinosarcoma based on results from the phase II part of the study (ROCSAN/ENGOT-en8, NCT03651206).
Ongoing phase III trials of checkpoint inhibitors in early and advanced endometrial cancer and trials in MMRd/MSI-H-selected endometrial cancer.
Ongoing (trials that are actively recruiting for which efficacy outcomes are not yet available and records have been updated in the last 2 years) phase III trials of checkpoint and/or TKIs for early or advanced endometrial cancer and trials in MMRd/MSI-H-selected endometrial cancer listed at CT.gov on 25, February 2022 ordered by treatment setting and estimated primary completion date.
CT, chemotherapy; DFS, disease-free survival; DLT, dose-limiting toxicity; EBRT, external beam radiotherapy; EC, endometrial cancer; ICI, immune checkpoint inhibitor; IMRT, intensity modulated radiotherapy; MMRd; mismatch repair deficient; MSI-H, microsatellite instability-high; ORR, objective response rate; OS, overall survival; PCD, primary completion date; PFS, progression-free survival; RR, response rate; RT, radiotherapy; SR, surgical resection; TCR, T-cell receptor; TEAE, treatment-emergent adverse event; TT, targeted therapy.
In patients selected for MMRd/MSI-H status, second-line regimens being assessed include nivolumab plus ipilimumab (NCI-2021-11881, NCT05112601). First-line trials are assessing single-agent pembrolizumab (ENGOT-en15, NCT05173987) and dostarlimab (DOMENICA, NCT05201547), and pembrolizumab (NCI-2019-08602, NCT04214067); durvalumab (RAINBO, NCT05255653) and dostarlimab (20-491, NCT04774419) are being evaluated in the adjuvant setting for high-risk disease following resection.
Summary
Therapy for pretreated metastatic endometrial cancer may involve multiple lines of therapy and should be guided by a multi-disciplinary team of specialists. Recent trials in pretreated metastatic endometrial cancer have reported benefit for ICIs alone and in combination with TKIs. Combination therapy with pembrolizumab plus lenvatinib has shown efficacy in this population regardless the molecular status and is presently indicated for MMRp/MSS tumors. For patients with MMRd/MSI-H-selected disease, ICI monotherapy is highly active and well tolerated. Research to find additional ICI combinations as well as to overcome ICI resistance is ongoing.
Supplemental Material
sj-docx-1-tam-10.1177_17588359231157633 – Supplemental material for A rapidly evolving landscape: immune checkpoint inhibitors in pretreated metastatic endometrial cancer
Supplemental material, sj-docx-1-tam-10.1177_17588359231157633 for A rapidly evolving landscape: immune checkpoint inhibitors in pretreated metastatic endometrial cancer by Anna V. Tinker, Neesha C. Dhani, Prafull Ghatage, Deanna McLeod, Vanessa Samouëlian, Stephen A. Welch and Alon D. Altman in Therapeutic Advances in Medical Oncology
Footnotes
Acknowledgements
We would like to thank Ilidio Martins and Paul Card from Kaleidoscope Strategic Inc. for their research and editorial support as well as Eisai Limited (CA), GlaxoSmithKline (GSK), and Merck for funding this review.
Authors’ note
No discussion or viewing of review content was permitted with sponsors prior to publication acceptance.
Disclaimer
This review was prepared according to ICMJE standards with editorial assistance from Kaleidoscope Strategic Inc.
Declarations
Supplemental material
Supplemental material for this article is available online.
References
Supplementary Material
Please find the following supplemental material available below.
For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.
For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.
