A Treatment-Line Framework for Preventing Preterm Delivery in PCOS Pregnancies: Metformin and Vaginal Progesterone as Complementary, Not Competing, Interventions

A prior hypothesis paper proposed that preterm delivery in PCOS pregnancies arises from placental progesterone insufficiency relative to fetal and uterine demand, with cervical shortening as a downstream marker rather than a cause (Sounderrajan, 2026; Zenodo DOI: 10.5281/zenodo.22162331). This paper extends that hypothesis by proposing a specific, three-pathway mechanistic model for how PCOS lowers placental progesterone capacity — placental vascular compromise, a pathological steroidogenesis shunt diverting progesterone toward androgen synthesis, and functional progesterone-receptor resistance driven by hyperinsulinemia and inflammation — and uses this model to resolve an apparent conflict in current clinical guidance between metformin and progesterone as competing PCOS-pregnancy interventions. We argue that metformin and vaginal progesterone act at different levels of the same causal chain: metformin upstream, by reducing the androgen shunt at the level of steroidogenic enzyme activity, and progesterone downstream, by directly replacing the resulting deficit regardless of its cause. This reframing predicts that combined therapy should outperform either intervention alone, and that single-agent trials of either drug should show only the modest, inconsistent benefit each has, in fact, shown in the existing literature. We state this as a seventh falsifiable claim (H7), propose a three-arm comparative trial design to test it, and propose a mechanism-informed treatment-line framework for clinical use pending that evidence. This is a hypothesis-generating paper; no original laboratory or clinical data are presented.