Metformin Non-Responder Profile: Who Doesn't Lose Weight or Control Blood Sugar on Metformin?

Who does not respond to metformin?

Response differs along two separate lines, blood glucose and body weight, and the trials report group-level results rather than a way to sort people into responders and non-responders.

In ADOPT, 4,360 adults with recently diagnosed type 2 diabetes took metformin, glyburide or rosiglitazone for a median of 4.0 years. The cumulative incidence of monotherapy failure at 5 years, defined as a confirmed fasting plasma glucose above 180 mg/dL, was 21% with metformin, 34% with glyburide and 15% with rosiglitazone [2]. About one in five people on metformin met that definition within five years.

In UKPDS 34, 753 overweight adults with newly diagnosed type 2 diabetes were randomized to a conventional diet policy or intensive blood-glucose control with metformin, with a median duration of 10.7 years. Median HbA1c was 7.4% with metformin and 8.0% with conventional treatment [7].

The label frames the drug the same way: an adjunct to diet and exercise for glycemic control in type 2 diabetes, not a weight-loss drug [3].

Missed doses: a measurable factor

Farmer et al. examined UK primary care records from 2004 to 2014 for people newly prescribed metformin, a sulfonylurea, a thiazolidinedione or a DPP-4 inhibitor who continued collecting prescriptions for a year. Nonadherence, defined as a medication possession ratio below 0.8, was 13% in one database and 15% in the other, and the proportion varied by drug from 8.6% with thiazolidinediones to 18.8% with metformin. Nonadherent patients had smaller HbA1c reductions than adherent patients: 0.4 percentage points smaller in CPRD and 0.46 in GoDARTS [4]. This was a retrospective cohort rather than a trial.

For reference, the label says immediate-release tablets start at 500 mg twice daily or 850 mg once daily with meals, increase by 500 mg weekly or 850 mg every 2 weeks, and top out at 2,550 mg per day in divided doses, with doses above 2,000 mg possibly better tolerated three times a day [3]. The extended-release label says to start at 500 mg once daily with the evening meal, go up to a maximum of 2,000 mg once daily, and swallow the tablets whole [8].

GI side effects and stopping

In the immediate-release label trial, 141 patients on metformin up to 2,550 mg/day were compared with 145 on placebo. Diarrhea occurred in 53% versus 12%, nausea or vomiting in 26% versus 8%, flatulence in 12% versus 6%, indigestion in 7% versus 4%, abdominal discomfort in 6% versus 5% and headache in 6% versus 5%. Diarrhea led to discontinuation in 6% [3].

In the extended-release trials, 781 patients on metformin extended-release were compared with 195 on placebo. Diarrhea occurred in 10% versus 3% and nausea or vomiting in 7% versus 2%, with 0.6% stopping because of diarrhea [8].

These numbers come from separate trials with different designs, and the label states that adverse reaction rates from different trials cannot be directly compared [3]. In the Diabetes Prevention Program, GI symptoms were more common with metformin than placebo and declined over time [6].

OCT1 variants and uptake into the liver

Metformin is not metabolized and is excreted unchanged by the kidneys [3]. Getting it into liver cells depends on transporters, and the best studied is organic cation transporter 1 (OCT1).

Shu et al. (J Clin Invest, 2007) found that deleting Oct1 in mouse hepatocytes reduced metformin's effects on AMPK phosphorylation and gluconeogenesis, and that the glucose-lowering effect of metformin was completely abolished in Oct1-deficient mice. They identified seven nonsynonymous OCT1 variants with reduced metformin uptake. The OCT1-420del variant, with an allele frequency of about 20% in white Americans, had reduced activity for metformin. In clinical studies, the effect of metformin during glucose tolerance tests was significantly lower in people carrying reduced-function variants [5].

That is a group-level difference measured in a test setting. It does not predict what will happen to one person, and it does not mean a carrier gets no benefit.

Kidney function that rules the drug out

The label says to check eGFR before starting and at least yearly, not to start metformin when eGFR is 30 to 45 mL/min/1.73 m², to reassess benefit and risk if eGFR later falls below 45, and to stop if it falls below 30 [3]. Metformin is contraindicated below an eGFR of 30 [3].

The label's boxed warning covers lactic acidosis, and names renal impairment first among its risk factors. Others listed are certain drugs, age 65 or older, contrast studies, surgery and other procedures, hypoxic states such as acute congestive heart failure, excessive alcohol intake, hepatic impairment and mitochondrial disease [3].

Other drugs that can lead to loss of glucose control

The label lists medicines that tend to raise blood glucose and may lead to loss of glycemic control: thiazides and other diuretics, corticosteroids, phenothiazines, thyroid products, estrogens, oral contraceptives, phenytoin, nicotinic acid, sympathomimetics, calcium channel blockers and isoniazid. It says to observe closely for loss of glucose control when these are started and for hypoglycemia when they are withdrawn [3].

Who actually loses weight on metformin?

The Diabetes Prevention Program randomized 3,234 adults with elevated fasting and post-load glucose (mean age 51, mean BMI 34.0, 68% women) to placebo, metformin 850 mg twice daily, or an intensive lifestyle program [9]. During the double-blind phase, body weight fell 2.06% with metformin and 0.02% with placebo, and waist circumference fell 2.13 cm versus 0.79 cm [6]. Through 7 to 8 years of open-label follow-up, weight loss remained greater with metformin than placebo (2.0% versus 0.2%) and was related to adherence [6].

A 2019 analysis followed the 1,066 participants who lost at least 5% of body weight in the first year: 289 in the metformin group (28.5%), 640 in the lifestyle group (62.6%) and 137 in the placebo group (13.4%). Among those participants, mean weight loss maintained between years 6 and 15 was 6.2% with metformin, 3.7% with lifestyle and 2.8% with placebo. In the metformin group, predictors of long-term loss were a larger first-year loss, older age and continued metformin use [1]. The analysis looked at non-randomized subsets, so it does not show that metformin produces more weight loss than lifestyle.

For diabetes prevention, metformin cut incidence by 31% versus placebo over an average 2.8 years, compared with 58% for lifestyle [9]. At 15 years, the reductions were 18% for metformin and 27% for lifestyle [10].