Progesterone and Breast Cancer

Association of Circulating Progesterone With Breast Cancer Risk Among Postmenopausal Women

IMPORTANCE

The role of endogenous progesterone in the development of breast cancer remains largely unexplored to date, primarily owing to assay sensitivity limitations and low progesterone concentrations in postmenopausal women. Recently identified progesterone metabolites may provide insights as experimental data suggest that 5α-dihydroprogesterone (5αP) concentrations reflect cancer-promoting properties and 3α-dihydroprogesterone (3αHP) concentrations reflect cancer-inhibiting properties.

OBJECTIVE

To evaluate the association between circulating progesterone and progesterone metabolite levels and breast cancer risk.

DESIGN, SETTING, AND PARTICIPANTS

Using a sensitive liquid chromatography-tandem mass spectrometry assay, prediagnostic serum levels of progesterone and progesterone metabolites were quantified in a case-cohort study nested within the Breast and Bone Follow-up to the Fracture Intervention Trial (n = 15 595). Participation was limited to women not receiving exogenous hormone therapy at the time of blood sampling (1992-1993). Incident breast cancer cases (n = 405) were diagnosed during 12 follow-up years and a subcohort of 495 postmenopausal women were randomly selected within 10-year age and clinical center strata. Progesterone assays were completed in July 2017; subsequent data analyses were conducted between July 15, 2017, and December 20, 2018.

EXPOSURES

Circulating concentrations of pregnenolone, progesterone, and their major metabolites.

MAIN OUTCOMES AND MEASURES

Development of breast cancer, with hazard ratios (HRs) and 95% CIs was estimated using Cox proportional hazards regression adjusted for key confounders, including estradiol. Evaluation of hormone ratios and effect modification were planned a priori.

RESULTS

The present study included 405 incident breast cancer cases and a subcohort of 495 postmenopausal women; the mean (SD) age at the time of the blood draw was 67.2 (6.2) years. Progesterone concentrations were a mean (SD) of 4.6 (1.7) ng/dL. Women with higher circulating progesterone levels were at an increased risk for breast cancer per SD increase in progesterone levels (HR, 1.16; 95% CI, 1.00-1.35; P = .048). The association with progesterone was linear in a 5-knot spline and stronger for invasive breast cancers (n = 267) (HR, 1.24; 95% CI, 1.07-1.43; P = .004). Among women in the lowest quintile (Q1) of circulating estradiol (<6.30 pg/mL) elevated progesterone concentrations were associated with reduced breast cancer risk per SD increase in progesterone levels (HR, 0.38; 95% CI, 0.15-0.95; P = .04) and increased risk among women in higher quintiles of estradiol (Q2-Q5; ≥6.30 pg/mL) (HR, 1.18; 95% CI, 1.04-1.35; P = .01; P = .04 for interaction).

CONCLUSIONS AND RELEVANCE

In this case-cohort study of postmenopausal women, elevated circulating progesterone levels were associated with a 16% increase in the risk of breast cancer. Additional research should be undertaken to assess how postmenopausal breast cancer risk is associated with both endogenous progesterone and progesterone metabolites and their interactions with estradiol.

DeepSnap-Deep Learning Approach Predicts Progesterone Receptor Antagonist Activity With High Performance

The progesterone receptor (PR) is important therapeutic target for many malignancies and endocrine disorders due to its role in controlling ovulation and pregnancy via the reproductive cycle. Therefore, the modulation of PR activity using its agonists and antagonists is receiving increasing interest as novel treatment strategy. However, clinical trials using the PR modulators have not yet been found conclusive evidences. Recently, increasing evidence from several fields shows that the classification of chemical compounds, including agonists and antagonists, can be done with recent improvements in deep learning (DL) using deep neural network. Therefore, we recently proposed a novel DL-based quantitative structure-activity relationship (QSAR) strategy using transfer learning to build prediction models for agonists and antagonists. By employing this novel approach, referred as DeepSnap-DL method, which uses images captured from 3-dimension (3D) chemical structure with multiple angles as input data into the DL classification, we constructed prediction models of the PR antagonists in this study. Here, the DeepSnap-DL method showed a high performance prediction of the PR antagonists by optimization of some parameters and image adjustment from 3D-structures. Furthermore, comparison of the prediction models from this approach with conventional machine learnings (MLs) indicated the DeepSnap-DL method outperformed these MLs. Therefore, the models predicted by DeepSnap-DL would be powerful tool for not only QSAR field in predicting physiological and agonist/antagonist activities, toxicity, and molecular bindings; but also for identifying biological or pathological phenomena.