Metformin to Augment Low Milk Supply in Pre-diabetic Mothers, a Phase I/II Randomized Clinical Trial

Through a progression of research, the PI has developed the central hypothesis that waning insulin secretion in the context of insulin resistance is an important cause of low milk supply. The specific aim of the research described in this protocol is to enact a small-scale randomized placebo-controlled trial (RCT) that will inform a future larger double-masked RCT of adjuvant metformin treatment versus placebo for early postpartum low milk supply in women with evidence of insulin resistance based on the presence of at least one of the following: elevated fasting glucose (FPG, defined as >95 g/dL), history of polycystic ovary syndrome, history of gestational diabetes, or current abdominal obesity. The pilot study is designed to demonstrate feasibility, obtain variance estimates, and test for an trend in the following primary hypothesis: 1) Among eligible women with low milk supply, those randomly assigned to 4 weeks of metformin treatment will experience a greater increase in milk output as compared to the placebo group. The RCT will be preceded by a"process testing phase" in which recruitment and data collection logistics will be confirmed by enacting the study protocol, except without any drug assignment. Upon completion of the process testing phase, the protocol will be amended according to insights gained. Once the revised protocol received IRB approval, the RCT phase will begin. During this phase, mothers meeting Stage 1 eligibility criteria will undergo baseline measurements of cardio-metabolic health and breast milk output. Among mothers meeting stage 2 eligibility criteria, including FPG >95 g/dL, N=30 will be randomly assigned to metformin or placebo using a 2:1 allocation, with replacement of non-completers. All low milk supply participants will receive the standard guidance for increasing milk supply with breast pumping. We will test the following secondary hypotheses: 2) Mammary epithelial cell transcriptomes within the metformin group, but not placebo, will exhibit significantly greater modulation of insulin-stimulated genes between baseline and post treatment. Milk fat globules are a rich source of mammary epithelial cell mRNA. We will isolate milk fat RNA at baseline and post treatment and randomly select a subset for RNA-sequencing. 3) Fasting plasma glucose >95 g/dL will correctly identify low milk supply cases with >75% sensitivity; and <95 g/dL will correctly identify abundant milk supply (comparator group) with >90% specificity. Fasting plasma glucose (FPG) in women with abundant milk supply will be derived from 30 consecutively consenting breastfeeding medicine patients who meet all RCT eligibility criteria except low milk supply (i.e., diagnoses related to infant feeding at the breast such as poor latch, but with abundant milk output). We will combine all available baseline FPG data to determine the sensitivity and specificity of FPG >95 g/dL as biomarker of low milk supply caused by maternal metabolic impairment. 4) Metformin treatment will be safe and adequately tolerated by the lactating mother and her breastfeeding infant.