The phosphoinositide 3-kinase (PI3K)-AKT signaling pathway is a key regulator of cell growth, survival, and metabolism and is frequently altered in estrogen receptor–positive (ER+) breast cancer. Approximately 30 to 40 percent of advanced ER+ breast cancers have gain of function mutations in PIK3CA, while 5 to 10 percent have loss of function mutations or deletions in PTEN. Genetic activation of AKT-1 is also observed in a subset of these tumors. These alterations promote PI3K-AKT pathway activation and are associated with poor prognosis. The therapeutic relevance of this pathway has been demonstrated by clinical benefit from PI3Kα and AKT inhibitors in combination with endocrine therapy; recent research has shown that capivasertib, an AKT inhibitor, plus fulvestrant increased treatment benefit in PIK3CA-, PTEN-, or AKT-1-altered breast cancer, compared to fulvestrant alone. In this study, Cutano et al1 examined mechanisms of response and resistance to AKT inhibition in ER+ breast cancer models harboring PIK3CA mutations with or without PTEN loss.
Public datasets were analyzed to assess co-occurrence of PIK3CA and PTEN alterations. The frequency of co-occurring PTEN alterations in PIK3CA-altered ER+ breast cancer ranged from 3 to 9 percent. Additionally, hemizygous deletions were found in 18 to 25 percent of ER+, human epidermal growth factor receptor 2–negative (HER2−) breast cancers, including 8 to 16 percent of tumors with PIK3CA mutations.
Responses to AKT and PI3Kα inhibition were evaluated in patient-derived xenografts (PDX) with PIK3CA mutations and varying PTEN status. Two PDX models with PIK3CA mutations and reduced PTEN protein, one with homozygous PTEN deletion and one with hemizygous loss, were less responsive to PI3Kα inhibition with alpelisib, but remained sensitive to AKT inhibition with capivasertib. In these models, the combination of capivasertib and fulvestrant was more effective than either agent alone. In contrast, models with intact PTEN responded well to both inhibitors. These results indicate that reduced PTEN expression in PIK3CA-mutant models impaired response to PI3Kα inhibition but did not diminish sensitivity to AKT inhibition.
To explore the underlying mechanisms, the authors used isogenic ER+ breast cancer cell lines with PIK3CA mutations and PTEN knockout. PTEN protein loss reduced sensitivity to alpelisib but did not impact intrinsic sensitivity to capivasertib. Capivasertib treatment led to suppression of downstream PI3K-AKT signaling markers, such as phosphorylated PRAS40 and S6, in both parental and PTEN-depleted cells. Notably, capivasertib reduced forkhead box protein M1 (FOXM1) expression, a transcription factor associated with cell cycle progression, while alpelisib treatment did not produce the same effect in PTEN-ablated cells.
Further investigation revealed that capivasertib treatment downregulated FOXM1 target genes, particularly those involved in the G2/M checkpoint and E2F signaling. Capivasertib also reduced S-phase deoxyribonucleic acid (DNA) synthesis in PTEN-depleted cells. Additionally, capivasertib-treated cells with PTEN reduction, rather than ablation, showed reduced FOXM1 expression; in patient-derived organoids from tumors with PIK3CA mutations and hemizygous PTEN loss, capivasertib was more effective than alpelisib at reducing cell proliferation and FOXM1 expression. These findings suggest a potential association between FOXM1 suppression and antiproliferative response.
The role of forkhead box O3 (FOXO3), a transcription factor downstream of AKT and negative regulator of FOXM1, was then examined. In PTEN-deficient cells, capivasertib induced nuclear translocation of FOXO3, whereas alpelisib did not. FOXO3 knockout prevented the downregulation of FOXM1 by capivasertib and reduced the drug’s antiproliferative effects. These experiments demonstrated that FOXO3-mediated suppression of FOXM1 is critical for the full response to AKT inhibition. Overexpression of FOXM1 in PTEN-deficient cells reduced sensitivity to capivasertib, further supporting its role in therapeutic resistance.
Chronic exposure to capivasertib led to resistance in ER+ breast cancer cells, with resistant cell lines showing increased FOXM1 expression. In these resistant cells, FOXM1 depletion restored sensitivity to capivasertib, suggesting that sustained FOXM1 expression contributes to long-term resistance. These findings demonstrate that FOXM1 downregulation contributes to sensitivity to capivasertib in resistant models.
The authors conclude that FOXO3 and FOXM1 form a regulatory axis that governs sensitivity to AKT inhibition in ER+ breast cancer models with PIK3CA and PTEN alterations. In models where FOXO3 is activated and translocates to the nucleus upon AKT inhibition, FOXM1 is suppressed, resulting in reduced cell proliferation. However, when FOXO3 expression is lost or its nuclear translocation is impaired, FOXM1 remains active and contributes to drug resistance. These findings support the role of FOXO3 and FOXM1 in regulating cellular response to PI3K-AKT pathway inhibitors and might help explain differences in treatment sensitivity. Their expression and activity might also help guide therapeutic strategies and inform future clinical development of PI3K and AKT inhibitors in ER+ breast cancer.
References
- Cutano V, Chia ML, Wigmore EM, et al. The interplay between FOXO3 and FOXM1 influences sensitivity to AKT inhibition in PIK3CA and PIK3CA/PTEN altered estrogen receptor positive breast cancer. NPJ Breast Cancer. 2025;11(1):36.