Estetrol Combined to Progestogen for Menopause or Contraception Indication Is Neutral on Breast Cancer

Abstract

Given the unequivocal benefits of menopause hormone therapies (MHT) and combined oral contraceptives (COC), there is a clinical need for new formulations devoid of any risk of breast cancer promotion. Accumulating data from preclinical and clinical studies support that estetrol (E4) is a promising natural estrogen for MHT and COC. Nevertheless, we report here that E4 remains active on the endometrium, even under a dose that is neutral on breast cancer growth and lung metastasis dissemination. This implies that a progestogen should be combined with E4 to protect the endometrium of non-hysterectomized women from hyperplasia and cancer. Through in vivo observations and transcriptomic analyses, our work provides evidence that combining a progestogen to E4 is neutral on breast cancer growth and dissemination, with very limited transcriptional impact. The assessment of breast cancer risk in patients during the development of new MHT or COC is not possible given the requirement of long-term studies in large populations. This translational preclinical research provides new evidence that a therapeutic dose of E4 for MHT or COC, combined with progesterone or drospirenone, may provide a better benefit/risk profile towards breast cancer risk compared to hormonal treatments currently available for patients.

Keywords: breast cancer; combined oral contraceptive; drospirenone; estetrol; estrogen receptor alpha; menopause hormone therapy; progesterone.

Figure 1

Uterotrophic effect of estetrol (E4), estradiol (E2), progesterone (P4) and drospirenone (DRSP). (A) Treatment protocol schema of the three hormone-dependent breast cancer mouse models: MMTV-PyMT, MCF7 xenograft and patient-derived xenograft (PDX). OVX, ovariectomy; E2/E4/P4/DRSP treatment start pointed by arrows; MCF7, tumor cell injection; PDX, tumor graft; END, mouse sacrifice; W4-W17, 4–17 weeks of age. (B) Representative Ki67 immunostainings on uterus harvested from MMTV-PyMT mice untreated (OVX) or treated with E2, E4 (0.3 or 3 mg/kg/day) combined with or without P4 (1.25 or 4.25 mg/kg/day); scale bar = 500 µm, zoom scale bar = 50 µm. Quantification of (C) uterine wet weight, (D) luminal epithelial height and (E) epithelial cell proliferation (Ki67-positive staining). (F) Representative Ki67 immunostainings on uterus harvested from MMTV-PyMT mice treated with E4 (0.3 mg/kg/day) with or without DRSP (0.06 mg/kg/day); scale bar = 500 µm, zoom scale bar = 50 µm. Quantification of (G) uterine wet weight, (H) luminal epithelial height and (I) epithelial cell proliferation (Ki67-positive staining). Kruskal–Wallis analysis followed by Dunn’s post-tests or Mann–Whitney analysis, n = 6–8 mice/condition. NS: not statistically significant; * or #: p < 0.05; ** or ##: p < 0.01; *** or ###: p < 0.001 and **** or ####: p < 0.0001. * versus OVX, # or NS versus corresponding sham/estrogen-alone treated mice.

Figure 2

Dose-dependent effect of E4 on breast cancer progression. (A) Tumor appearance assessed by the percentage of tumor-free mice in ovariectomized MMTV-PyMT treated with E2 (0.08 mg/kg/day) or E4 (0.3 or 3 mg/kg/day). (B) Tumor growth kinetics (treatment start pointed by the arrow), (C) tumor mass and (D) tumor growth delay. (E) Tumor growth kinetics and (F) tumor mass of MCF7 xenografts from mice untreated (OVX) or treated with E2 (0.08 mg/kg/day) or E4 (0.3 or 3 mg/kg/day). (G) Tumor growth kinetics (treatments started five weeks after engraftment as pointed by the arrow) and (H) tumor mass of PDX from mice untreated (OVX) or treated with E2 (0.08 mg/kg/day) or E4 (0.3 or 3 mg/kg/day) for 5 (W10) or 30 weeks (W35). (I) Hematoxylin/eosin staining of lungs harvested from MMTV-PyMT mice; scale bar = 2.5 mm, zoom scale bar = 250 µm. (J) Percentage of metastasis-positive mice at sacrifice, (K) metastasis number, (L) metastasis size, (M) lung area occupied by metastasis (%). Kruskal–Wallis analysis followed by Dunn’s post-tests, two-way ANOVA analysis followed by Tukey post-tests or Mann–Whitney analysis, n = 5–15 mice/condition. *: p < 0.05; **: p < 0.01; ***: p < 0.001; ****: p < 0.0001. * versus OVX.

Figure 3

Dose-dependent effect of E4+progestogen on breast cancer progression. (A) Tumor growth kinetics (treatment start pointed by the arrow) and (B) tumor mass of MMTV-PyMT mice treated with E2 + P4 (4.25 mg/kg/day), E4 (0.3 mg/kg/day) combined with or without P4 (1.25 mg/kg/day) or DRSP (0.06 mg/kg/day). (C) Tumor growth kinetics (treatment start pointed by the arrow) and (D) tumor mass of MMTV-PyMT mice treated with E2 + P4 (4.25 mg/kg/day), E4 (3 mg/kg/day) combined with or without P4 (1.25 or 4.25 mg/kg/day). (E) Tumor growth kinetics and (F) tumor mass of MCF7 xenografts from mice treated with E2 + P4 (4.25 mg/kg/day), E4 (0.3 mg/kg/day) combined with or without P4 (1.25 or 4.25 mg/kg/day) or DRSP (0.06 mg/kg/day). (G) Tumor growth kinetics and (H) tumor mass of MCF7 xenografts from mice treated with E2 + P4 (4.25 mg/kg/day), E4 (3 mg/kg/day) combined with or without P4 (1.25 or 4.25 mg/kg/day). (I) Tumor growth kinetics (treatments started five weeks after engraftment as shown by the arrow) and (J) tumor mass of PDX from mice treated with E2 + P4 (4.25 mg/kg/day), E4 (0.3 mg/kg/day) combined with or without P4 (1.25 or 4.25 mg/kg/day) or DRSP (0.06 mg/kg/day). (K) Tumor growth kinetics (treatment start pointed by the arrow) and (L) tumor mass of PDX from mice treated with E2 + P4 (4.25 mg/kg/day), E4 (3 mg/kg/day) combined with or without P4 (1.25 or 4.25 mg/kg/day). (M) Hematoxylin/eosin coloration of lungs harvested from MMTV-PyMT mice; scale bar = 2.5 mm, zoom scale bar = 250 µm. (N) Percentage of metastasis-positive mice at sacrifice, (O) lung area occupied by metastasis (%), (P) metastasis number, (Q) metastasis size. Kruskal-Wallis analysis followed by Dunn’s post-test, two-way ANOVA analysis followed by Tukey post-tests or Mann Whitney analysis, n = 6–13 mice/condition. *: p < 0.05; **: p < 0.01; ***: p < 0.001; ****: p < 0.0001, * versus E2 + P4.

Figure 4

E4-induced ERα signaling in vitro. PGR mRNA expression normalized to TBP and GAPDH in MCF7 (A) and T47D cells (B) treated with vehicle (EtOH 0.01%), E2 (10⁻¹¹M or 10⁻⁹M) or E4 (ranging from 10⁻¹²M to 10⁻⁵M) for 4 h. n = 3–4 independent experiments. One-way ANOVA analysis followed by Dunnett’s post-test. *: p < 0.05; **: p < 0.01; ***: p < 0.001; ****: p < 0.0001, * versus vehicle. (C) Representative Western Blot of PR (PRA and PRB) and ERα from MCF7 cells treated with vehicle (EtOH 0.01%), E2 (10⁻⁹M) or E4 (10⁻⁷M or 10⁻¹⁰M) for 24h. GAPDH was used as a loading control. (D) Quantification of PR expression normalized to GAPDH level, RI= Relative Intensity, n = 3 independent replicates. (E, F) Representative experiment of cell growth kinetics of MCF7 and T47D cells treated with vehicle (EtOH 0.01%), E2 (10⁻⁹M) or E4 (10⁻¹⁰M, 10⁻⁹M or 10⁻⁷M). (G, H) Proliferation rate after 72 h. Mann–Whitney test. *: p < 0.05; **: p < 0.01; ***: p < 0.001; ****: p < 0.0001, * versus vehicle.

Figure 5

E4 is less potent than E2 to promote ERα signaling in vivo. (A) Representative Western Blot of PRB, pS118-ERα and ERα from MCF7 tumors harvested from mice treated with E2 (0.08 mg/kg/day) or E4 (0.3 or 3 mg/kg/day). GAPDH was used as a loading control. (B) Quantification of PR expression normalized to GAPDH level and (C) quantification of pS118-ERα protein level normalized to ERα. RI= Relative Intensity. (D) Representative immunostainings of ERα, PR, pS118-ERα and Ki67 on MCF7 tumors harvested from mice treated with E2 (0.08 mg/kg/day) or E4 (0.3 or 3 mg/kg/day); scale bar = 100 µm. Quantification of (E) PR, (F) pS118-ERα and (G) Ki67 staining expressed as density by Minimum and Maximum boxes. (H) Western Blot of PR (PRA and PRB), pS118-ERα and ERα from PDX untreated (OVX) or treated with E2 (0.08 mg/kg/day) or E4 (0.3 or 3 mg/kg/day) for 5 (W10) or 30 weeks (W35). GAPDH was used as a loading control. Quantification of (I) PR synthesis normalized to GAPDH level and quantification of (J) pS118-ERα normalized to ERα. (K) Representative immunostainings of ERα, PR, pS118-ERα and Ki67 from PDX untreated (OVX) or treated with E2 (0.08 mg/kg/day) or E4 (0.3 or 3 mg/kg/day) for 5 (W10) or 30 weeks (W35), scale bar = 500 µm. Quantification of (L) PR, (M) pS118-ERα and (N) Ki67 staining expressed as density by Minimum and Maximum boxes. Mann–Whitney tests, n = 8–12 tumors/condition. *: p < 0.05; **: p < 0.01, * versus E2 in MCF7, * versus OVX in PDX.

Figure 6

E2 and E4 transcriptomic profiles. (A) Volcano plot, using a parametric edgeR approach to identify differentially expressed (DE) genes, comparing E2 or E4 (10⁻⁷M) to vehicle from five independent replicates per condition. (B) Venn diagram comparing up- and downregulated genes by E2 or E4 (10⁻⁷M). (C) Correlation between genes regulated by E2 and/or E4 (10⁻⁷M) treatments. (D) Heatmap of gene regulation for each replicate of the different treatments: vehicle, E2 (10⁻⁹M) and E4 (10⁻¹⁰M, 10⁻⁷M). (E) Volcano plot, using a parametric edgeR approach to identify DE genes, comparing E4 (10⁻⁷M) to E2 and E4 (10⁻¹⁰M) to vehicle. (F–L) Venn Diagrams comparing genes regulated by E2 + P4, E2 + R5020, E4 + P4, E4 + R5020, E4* + P4, E4* + R5020 versus vehicle, E4 = 10⁻⁷M E4, E4* = 10⁻¹⁰M E4. (M) Specific genes modulated by E2 + P4, E2 + R5020, E4 + P4, E4 + R5020, E4* + P4, E4* + R5020 in comparison with respective estrogenic treatment alone. The analysis parameters used were: Fc ≥ 2, p-value ≤ 0.01 and power: 97%.

Figure 7

E2- and E4-specific co-regulator binding profiles. (A) Heatmap of interactions between ERα and co-regulators induced by E2 or E4 and represented as the modulation index (MI). MI is expressed as a log of fold-changes relative to the vehicle. (B) Dose-dependent induction by E2 or E4 (10⁻¹²M to 10⁻⁵M) of the ERα interaction with 10 co-regulators. (C) Boxplot comparing the mean of all EC50 (logM) values obtained with E2 or E4.