Everolimus/Letrozole with Ribociclib in Recurrent Endometrial Cancer: From All-Comers to CTNNB1-Mutant Subgroups

What earlier evidence established about endocrine therapy and CTNNB1 in recurrent endometrial cancer

Before this trial, endocrine therapy for recurrent endometrial cancer was known to produce modest, transient responses. Everolimus plus letrozole had shown objective response rates of 28–32% and clinical benefit rates of 40–50%, with median PFS of 3–6 months and median OS of 14–31 months [1]. Letrozole monotherapy yields ORRs of only 8–14% and PFS of 3–4 months, and combinations with mTOR or CDK4/6 inhibitors improve these figures to ORRs of 25–32% and PFS up to 9.1 months in ER-positive endometrioid subtypes [7]. The PI3K/AKT/mTOR pathway is activated in up to 90% of endometrioid endometrial cancers, providing a rationale for mTOR inhibition, while CTNNB1 mutations occur in roughly 25% of endometrioid cases and activate WNT/β-catenin signaling with downstream cyclin D1 expression [1].

CTNNB1 mutations are not merely prognostic. In a cohort of 130 patients, CTNNB1 mutations were found in 36.9% of cases and were significantly more common in women aged 40 or younger (68.6% vs 30.5%, p=0.0032), and they were associated with poor response to progestin therapy in fertility-sparing settings [3]. A separate multigene panel study of 38 patients receiving medroxyprogesterone acetate found CTNNB1 mutations in 57.9% of cases, though PTEN and PIK3CA mutations were the alterations significantly associated with treatment outcomes [8]. Preclinical work showed that CD73 sequesters mutant β-catenin at the membrane, and loss of CD73 independently predicts recurrence in CTNNB1-mutant tumors, suggesting that not all CTNNB1 mutations behave identically [5][9]. This heterogeneity matters when interpreting any trial that treats CTNNB1 mutation as a single binary biomarker.

The anchor trial: randomized design, stringent control arm, and a null all-comer result

The anchor phase 2 trial randomized 72 patients with advanced or recurrent endometrioid or mixed endometrioid endometrial cancer to everolimus/letrozole with or without ribociclib after a safety lead-in established the recommended phase 2 dose (everolimus 2.5 mg, letrozole 2.5 mg, ribociclib 200 mg daily) [1]. The control arm received everolimus 10 mg plus letrozole, a deliberately active comparator rather than placebo, which the authors describe as a stringent comparison given everolimus/letrozole's established activity [1]. Prior exposure to endocrine therapy, immunotherapy, targeted therapy, or antibody-drug conjugates was permitted and not counted toward the two-line chemotherapy limit, and ER/PR positivity was not required [1].

In all-comers, there was no significant difference in median PFS (16.7 vs 11.0 months; HR 0.78, 95% CI 0.4–1.4, p=0.38), OS (28.9 vs 31.5 months; HR 0.80, p=0.46), ORR (44.1% vs 25.7%, p=0.11), or CBR (67.7% vs 68.6%, p=0.93) [1]. The authors note that the control arm overperformed relative to historical data, possibly because pure non-endometrioid tumors were excluded—serous histology had shown no responses to everolimus/letrozole in the earlier Slomovitz trial—and this may have limited power to detect a difference [1]. Toxicity was manageable but not identical: neutropenia (any grade 80% vs 25.7%, p<0.001; grade 3 54.3% vs 5.7%, p<0.001) and leukopenia (48.6% vs 14.3%, p=0.002) were significantly higher in the experimental arm, while hypertriglyceridemia and hypercholesterolemia were higher in the control arm [1]. Treatment discontinuation due to toxicity was similar between arms (4 vs 3 patients, p=0.22) [1].

Where the signal emerged: CTNNB1-mutant tumors and prior endocrine therapy

The trial prespecified enrollment of at least 14 patients with CTNNB1-mutant tumors in the experimental arm, and ultimately 32 of 70 patients (46%) had pathogenic exon 3 CTNNB1 mutations [1]. Within the experimental arm, CTNNB1-mutant tumors had significantly longer median PFS than CTNNB1-wild-type tumors (26.7 vs 3.7 months; HR 0.27, 95% CI 0.1–0.7, p=0.006) and significantly longer OS (not reached vs 12.7 months; HR 0.24, 95% CI 0.1–0.7, p=0.004) [1]. The 3-year OS rate was 76.5% for CTNNB1-mutant tumors versus 21.4% for wild-type tumors (p=0.0002) [1]. Clinical benefit rate was also higher in the mutant group (82.4% vs 46.7%, p=0.03) [1]. However, there was no difference in PFS or OS between treatment arms among CTNNB1-mutant tumors, and no PFS or OS difference based on CTNNB1 status within the control arm [1]. This means the CTNNB1 finding is a within-experimental-arm comparison, not a demonstration that ribociclib specifically benefits these patients over everolimus/letrozole alone.

A separate signal appeared among patients with prior endocrine therapy: ORR was 61.5% in the experimental arm versus 0.0% in the control arm (p<0.001) [1]. This subgroup finding is hypothesis-generating and based on a small number of patients, but it aligns with the biological rationale that CDK4/6 inhibition may overcome endocrine resistance. The trial also found that grade 3 tumors had shorter PFS and OS within both arms, and among grade 1/2 tumors in the experimental arm, CTNNB1-mutant tumors had significantly longer PFS (HR 0.15, p=0.002) and OS (HR 0.16, p=0.01) than wild-type tumors, a benefit not observed in the control arm [1]. These grade-stratified results suggest the CTNNB1 effect may be concentrated in low-grade disease, consistent with the known association of CTNNB1 mutations with low-grade endometrioid histology [3].

How this trial compares with prior CDK4/6 and mTOR combination evidence

Most prior trials of CDK4/6 inhibitors plus endocrine therapy in endometrial cancer were single-arm studies with fewer than 10 patients with CTNNB1-mutant tumors, limiting mechanistic insight [1]. This trial's randomized design and prespecified biomarker enrichment distinguish it from that earlier frontier. Preclinical work had shown that dual PI3K/AKT/mTOR and CDK4/6 inhibition suppresses survivin in uterine dedifferentiated endometrial carcinoma, providing a mechanistic rationale for the triplet [2]. The anchor trial is the first clinical test of combining a CDK4/6 inhibitor, an mTOR inhibitor, and an aromatase inhibitor in endometrial cancer [1].

The broader molecular classification literature supports the idea that NSMP tumors—which include most CTNNB1-mutant cases—are the most favorable subgroup for hormonal therapy, with the highest complete response rates and lowest recurrence rates in fertility-sparing settings, likely due to preserved estrogen and progesterone receptor signaling [6]. CTNNB1 mutations occur in 70.8% of NSMP tumors in one cohort [3]. However, the fertility-sparing evidence also shows that CTNNB1 mutations are associated with poor progestin response and longer time to complete response, which appears to conflict with the anchor trial's finding of benefit from the ribociclib-containing regimen [3]. The difference may lie in the therapeutic context: progestin monotherapy in early-stage disease versus combined CDK4/6 and mTOR inhibition in recurrent disease, where the triplet may target cyclin D1-driven proliferation downstream of β-catenin [1]. This is an interpretation, not a proven mechanism.

Limitations, open questions, and what would be needed to change practice

The anchor trial is a phase 2 study with 35 evaluable patients per arm, and the all-comer result was null [1]. The CTNNB1 and prior endocrine therapy findings come from subgroup analyses without adjustment for multiple comparisons, and the confidence intervals are wide [1]. The authors explicitly state that the experimental regimen in the CTNNB1-mutant population should be investigated in larger confirmatory studies [1]. The trial was open-label, and investigators were not blinded to allocation during treatment and outcome assessment [1]. The control arm's overperformance relative to historical data may have reduced statistical power, and the sample size was calculated based on older PFS data from a trial that included serous histology [1].

The MITO END-3 trial experience illustrates the risk of overinterpreting subgroup findings in endometrial cancer: a preplanned analysis based on MMR status showed a signal, but the authors cautioned about interpretation of subgroup analyses in a phase 2 randomized setting [4]. The same caution applies here. The CTNNB1 biomarker itself is heterogeneous—CD73 expression modifies the oncogenic activity of mutant β-catenin, and not all CTNNB1-mutant tumors recur or respond identically [5][9]. The fertility-sparing literature shows that PTEN and PIK3CA mutations, not CTNNB1, were significantly associated with treatment outcomes in one multigene panel study, suggesting that other pathway alterations may confound the CTNNB1 signal [8]. A phase 3 trial with prespecified CTNNB1 stratification, centralized molecular testing, and adequate power to detect interaction effects would be required before adopting CTNNB1 as a predictive biomarker for this regimen [1].

About These Sources

This research page is built on 9 peer-reviewed studies — published from 2021 to 2026, 7 from 2024 or later, collectively cited 63 times — selected as the most relevant from 10 studies that passed quality screening, drawn from 59 papers retrieved from a database of over 500 million.

Sources used in this answer