A Phase 4 interventional study of Metformin and Placebo in Polycystic Ovary Syndrome, Non-Alcoholic Fatty Liver Disease and Metabolic Syndrome, sponsored by Columbia University. Terminated at 1 site in United States. Open to female participants aged 13 Years to 25 Years. Per ClinicalTrials.gov, last updated 2025-07-02.
Sponsored by Columbia University · Phase 4, Interventional, and Treatment
This project, "A double-blind placebo-controlled randomized clinical trial assessing the efficacy of metformin for hepatic fat in adolescents and young adults with polycystic ovary syndrome", proposes exploring the use of novel and noninvasive methodologies in an at-risk adolescent and young adult population with polycystic ovary syndrome (PCOS) who may gain long-term health benefits from early detection and treatment of non-alcoholic fatty liver disease (NAFLD). PCOS is a common condition that frequently presents in adolescence and young adulthood and is defined by elevated androgens (male hormones) in the blood leading to 1. hirsutism and acne and 2. menstrual abnormalities or amenorrhea. Affected individuals are at increased risk of developing insulin resistance (a precursor of diabetes), NAFLD and lipid (cholesterol) abnormalities.These features are all associated with the metabolic syndrome, a rising major public health concern. Recently, an association between PCOS and NAFLD has been noted but has only been superficially studied in the adolescent and young adult population. The susceptibility of certain PCOS patients to developing NAFLD is theorized to be due to having underlying insulin resistance, elevated androgen levels, and a genetic predisposition. Metformin is an insulin sensitizing medication widely used to treat type 2 diabetes mellitus that may have beneficial effects on insulin resistance-related conditions including PCOS and NAFLD. Although widely used in PCOS, its effect on NAFLD in this group has not been previously studied.
The primary aims of this proposal are: 1) To determine whether PCOS with liver fat >/=4.8% treated with metformin for six months will have a decline in percentage liver fat compared to a placebo group. 2) To measure the association of the PNPLA3 I148M allele with NAFLD in PCOS at baseline (n=40). 2b) To measure the association of percentage liver fat with biomarkers of NAFLD, dyslipidemia, insulin resistance and body composition at baseline (n=40) and after a placebo-controlled intervention with metformin in PCOS with liver fat >4.8% (n=20).
The goal of this research proposal is to explore the use of novel and noninvasive technologies in a young and at risk population. Dr. Sopher hopes to use the results of this research to lay the groundwork for the prevention and treatment of NAFLD and other metabolic disorders in adolescents and young adults with PCOS and to prevent lifelong morbidity associated with PCOS.
PCOS subjects will be screened by questionnaire for eligibility with the goal of recruiting 40 subjects who are 14 to 25 years, BMI between the 10th and 95th percentile for age (\<20 years) and 18.5 - 29.9 kg/m2 (20 years or older), and at least two years post-menarche. Oligomenorrheic subjects will complete the protocol during the early follicular phase of the menstrual cycle (days one through seven) and amenorrheic subjects will complete the protocol on any day. The protocol will comprise: a) History and physical exam: Height, weight, body mass index (BMI) (percentile and z-score for \<20 years), blood pressure, physical exam, Ferriman-Gallwey hirsutism scoring, smoking, alcohol and family history; b) Laboratory evaluation will be between 0800 and 0900 after an overnight fast via a 21 gauge intravenous catheter for: 1) General endocrine panel: Thyroid function, prolactin, 17-hydroxyprogesterone; 2) PCOS panel: luteinizing hormone (LH), follicle stimulating hormone (FSH), estradiol, sex hormone binding globulin, testosterone, free testosterone, androstenedione, dehydroepiandrosterone; 3) Liver panel: Liver enzymes (alanine aminotransferase (ALT), aspartate aminotransferase (AST), gamma glutamyl-transpeptidase (GGT)), apoptosis markers (Fas, cytokeratin-18 fragments, M30), pancreatic polypeptide; 4) Cardiovascular risk panel: Cholesterol, triglycerides, HDL, LDL, free fatty acids, c-reactive protein; 5) IR evaluation: hemoglobin A1C, glucose, insulin, c-peptide and two-hour oral glucose tolerance test (OGTT) after a 75 gram glucose challenge with measurement of glucose and insulin at 30, 60, 90 and 120 minutes. IR will be determined by age- and BMI- specific homeostatic measure of insulin resistance (HOMA-IR) cutoffs based on a National Health and Nutrition Examination Survey (NHANES) study of 1,164 nonobese adolescents 12 to 19 years old. Other IR indices that will be evaluated are whole body insulin sensitivity (WBIS) and insulin area under the curve; 6) Genetic evaluation: A blood sample for the PNPLA3 I148M allele (baseline only). c) Magnetic resonance spectroscopy (MRS) of the liver for intrahepatic lipid content acquired using standard point resolved spectroscopy sequence (PRESS). Percentage liver fat will be calculated. d) Total body magnetic resonance imaging (MRI) with 10 mm slices and 40 mm interslice gaps for total body adipose tissue and VAT. e) Dual-energy x-ray absorptiometry (DXA): Total body DXA will be performed for total body lean and fat mass and percentage body fat. f) Randomization: Subjects whose liver fat is 4.8% or greater (n≈20) will be randomized to placebo or extended-release metformin 1000 mg (2 tablets) daily. Randomization will be double-blind and will be performed by randomly selected permuted size 2 and size 4 blocks determined by the Columbia University Medical Center (CUMC) Research Pharmacy with the guidance of a statistician. g) Intervention: All subjects will receive identically appearing pill bottles labeled "Study Medication" containing their first three-month supply of either metformin or placebo. Subjects will be instructed to take pills with dinner and to take one pill once a day for one week and subsequently two pills once a day for the remainder of the first three months. Subjects will be asked to call with any side effects that they may experience. They will receive weekly calls from a research coordinator and will be asked to bring their empty pill bottles to their follow-up studies do to ensure compliance. h) Brief interim evaluation: After three months of the intervention phase, subjects will be seen for a general health assessment, which includes weight, blood pressure and a physical examination, and will receive their second three-month supply of their assigned intervention. i) Follow-up exam: At the end of six months, subjects will undergo laboratory and body composition testing identical to those performed in the baseline testing.
Exclusion Criteria:
Adolescents and young adults with PCOS and liver fat greater than or equal to 4.8% will be randomized to metformin or placebo. Metformin ER (extended release) will be administered as two pills of 500 mg each. For the first week subjects will take one pill orally on a daily basis and thereafter will take two pills orally on a daily basis. The intervention will last six months.
Drug: Metformin
Adolescents and young adults with PCOS and liver fat greater than or equal to 4.8% will be randomized to metformin or placebo. Subjects randomized to placebo will be instructed to take one pill daily by mouth for one week and then to take one pill daily by mouth for the remainder of six months.
Drug: Placebo
Metformin ER 500 mg once a day for one week and 1000 mg once a day for the remainder of the six months
Also known as: Fortamet
One placebo pill per day for one week and then two placebo pills per day for the remainder of the six months
Also known as: Placebo pill
Percentage Liver Fat in Participants
To compare percentage liver fat by magnetic resonance spectroscopy in the metformin group and placebo group to baseline and between the groups in order to determine if metformin is efficacious for reducing liver fat compared to placebo in adolescents and young women with Polycystic Ovary Syndrome (PCOS)
Time frame: 6 months
The Association of Percentage Liver Fat by Magnetic Resonance Spectroscopy With Insulin Resistance as Measured by HOMA-IR in Adolescents With PCOS
The association of percent liver fat with insulin resistance as measured by HOMA-IR will be measured by correlation/regression. Change in HOMA-IR with change in percent liver fat following metformin will be assessed using multiple regression analysis.
Time frame: 6 months
The Association of Percentage Liver Fat With M30, a Hepatic Apoptosis Marker
The association of percent liver fat with M30 will be measured by correlation/regression. Change in M30 with change in percent liver fat following metformin will be assessed using multiple regression analysis.
Time frame: 6 months
The Association of Percentage Liver Fat With Pancreatic Polypeptide
The association of percent liver fat with pancreatic polypeptide will be measured by correlation/regression. Change in pancreatic polypeptide with change in percent liver fat following metformin will be assessed using multiple regression analysis.
Time frame: 6 months
The Association of Percentage Liver Fat With Total Body Adipose Tissue
The association of percent liver fat with total body adipose tissue will be measured by correlation/regression. Change in total body adipose tissue with change in percent liver fat following metformin will be assessed using multiple regression analysis.
Time frame: 6 months
The Association of Percentage Liver Fat With Visceral Adipose Tissue
The association of percent liver fat with visceral adipose tissue will be measured by correlation/regression. Change in visceral adipose tissue with change in percent liver fat following metformin will be assessed using multiple regression analysis.
Time frame: 6 months
The Association of Percentage Liver Fat With Triglycerides
The association of percent liver fat with triglycerides will be measured by correlation/regression. Change in triglycerides with change in percent liver fat following metformin will be assessed using multiple regression analysis.
Time frame: 6 months
Proportion of PCOS Subjects With the PNPLA3 Allele Comparing Those With Elevated Percentage Liver Fat (>/=4.8%) and Those With Normal Percentage Liver Fat (<4.8%) by Magnetic Resonance Spectroscopy
The proportion of PCOS subjects with the high risk I148M PNPLA3 allele in the PCOS groups with elevated and normal liver fat will be compared using a chi-squared or Fisher's Exact test.
Time frame: 6 months
| Milestone | Metformin | Placebo |
|---|---|---|
| Started | 1 | 2 |
| Completed | 0 | 0 |
| Not completed | 1 | 2 |
To compare percentage liver fat by magnetic resonance spectroscopy in the metformin group and placebo group to baseline and between the groups in order to determine if metformin is efficacious for reducing liver fat compared to placebo in adolescents and young women with Polycystic Ovary Syndrome (PCOS)
| percent | Metformin | Placebo |
|---|---|---|
| Percentage Liver Fat in Participants | 7.4 ± NA | 3.1 ± 2.6 |
The association of percent liver fat with insulin resistance as measured by HOMA-IR will be measured by correlation/regression. Change in HOMA-IR with change in percent liver fat following metformin will be assessed using multiple regression analysis.
No measurements were reported for this outcome.
The association of percent liver fat with M30 will be measured by correlation/regression. Change in M30 with change in percent liver fat following metformin will be assessed using multiple regression analysis.
No measurements were reported for this outcome.
The association of percent liver fat with pancreatic polypeptide will be measured by correlation/regression. Change in pancreatic polypeptide with change in percent liver fat following metformin will be assessed using multiple regression analysis.
No measurements were reported for this outcome.
The association of percent liver fat with total body adipose tissue will be measured by correlation/regression. Change in total body adipose tissue with change in percent liver fat following metformin will be assessed using multiple regression analysis.
No measurements were reported for this outcome.
The association of percent liver fat with visceral adipose tissue will be measured by correlation/regression. Change in visceral adipose tissue with change in percent liver fat following metformin will be assessed using multiple regression analysis.
No measurements were reported for this outcome.
The association of percent liver fat with triglycerides will be measured by correlation/regression. Change in triglycerides with change in percent liver fat following metformin will be assessed using multiple regression analysis.
No measurements were reported for this outcome.
The proportion of PCOS subjects with the high risk I148M PNPLA3 allele in the PCOS groups with elevated and normal liver fat will be compared using a chi-squared or Fisher's Exact test.
No measurements were reported for this outcome.
Collected over At time of randomization, Day 1. Non-serious events are listed at a 0% frequency threshold.
| Group | Deaths | Serious | Other |
|---|---|---|---|
| Metformin | 0/1 (0%) | 0/1 (0%) | 0/1 (0%) |
| Placebo | 0/2 (0%) | 0/2 (0%) | 0/2 (0%) |
| Age, Categorical(Participants) | Metformin | Placebo | Total |
|---|---|---|---|
| <=18 years | 0 | 0 | 0 |
| Between 18 and 65 years | 0 | 2 | 2 |
| >=65 years | 1 | 0 | 1 |
| Sex: Female, Male(Participants) | Metformin | Placebo | Total |
|---|---|---|---|
| Female | 1 | 2 | 3 |
| Male | 0 | 0 | 0 |
| Race and Ethnicity Not Collected(Participants) | Metformin | Placebo | Total |
|---|---|---|---|
| Count of participants | — | — | 0 |
| Region of Enrollment(participants) | Metformin | Placebo | Total |
|---|---|---|---|
| United States | 1 | 2 | 3 |
Documents are hosted by the registry — open the source record to download them.
Plan to share: Undecided — Will consider this option
This study is terminated, as verified in Jun 2025. You cannot join it, but the record below documents what was studied.
Get an email when the registry record changes — status, dates, results — or when someone posts here.
Sign in to followQuestions and observations about this study, from anyone following it. Not medical advice, and not a channel to the study team — their contact details are on the registry record.
Sign in to join the discussion. Reading takes no account; posting does. You choose a display name, and a pseudonym is the default.
Nothing here yet. If you are running this trial, taking part in it, or weighing whether to, this is the place to say so.
Columbia University