Eli Lilly and Company, along with partner institutions in the US and United Kingdom, describe how short-term pharmacologic intervention does not appear to have lasting effects for most tirzepatide patients. People with obesity who lost weight on tirzepatide often regained at least 25% of that loss within a year of stopping treatment, along with reversals in improvements in waist circumference, blood pressure, lipids, and glycemic measures.
Chronic obesity is a disease associated with an elevated risk for cardiometabolic complications. Treatment goals extend beyond weight reduction toward lowering that risk across blood pressure, lipid levels, glucose regulation, and insulin resistance.
Tirzepatide is a dual-action synthetic hormone that regulates blood sugar, appetite, and digestion. Drugs with tirzepatide are currently the top selling drugs in the world (Eli Lilly brands Mounjaro and Zepbound).
Current guidelines recommend obesity management medications in combination with a healthy diet and exercise to achieve and maintain weight reduction and to improve cardiometabolic parameters. Calibrating how healthy a diet is, and how effective exercise routines are, can be difficult under pharmacologically assisted weight loss.
Results can be positive without a change or even a less healthy lifestyle. Consequently, those that see weight loss while on weight loss medications frequently regain the weight when they stop using.
In the study, “Cardiometabolic Parameter Change by Weight Regain on Tirzepatide Withdrawal in Adults With Obesity,” published in JAMA Internal Medicine, researchers performed a post hoc analysis on a select group of patients within a larger trial to assess changes in cardiometabolic parameters by degree of weight regain after withdrawal of tirzepatide.
SURMOUNT-4 enrolled 783 participants, with 670 who completed the initial 36-week open-label period being randomized to either continue tirzepatide (n=335) or switch to a placebo (n=335) for the 52-week double-blind phase. The current study followed 308 of the “switch to a placebo” group, eliminating those that did not lose at least 10% of their weight while on treatment.
During tirzepatide treatment
Weight loss during the 36 week tirzepatide exposure was large across later weight regain groups. Mean percentage change in weight from week 0 to week 36 was −21.9%.
Body mass index and waist circumference followed similar trajectories. Mean body mass index at week 36 had a mean reduction of 8.3 units and a mean waist circumference reduction of 18.3 cm
Cardiometabolic parameters improved for all categories during the lead-in. Systolic and diastolic blood pressure, triglycerides, non–high density lipoprotein cholesterol, HDL cholesterol, hemoglobin A1c, fasting serum glucose, fasting insulin, HOMA2 insulin resistance, and HOMA2 β cell function all decreased between week 0 and week 36.
After stopping
Weight trajectories diverged once tirzepatide stopped at week 36 and placebo replaced it while lifestyle measures continued. Weight regain between week 36 and week 88 was defined as a percentage of the earlier weight reduction achieved from week 0 to week 36. That percentage was determined by four groups at week 88.
One group of 54 participants showed less than 25% weight regain relative to the initial loss. A second group of 77 participants showed a 25% to less than 50% regain. A third group of 103 participants showed a 50% to less than 75% regain. A fourth group of 74 participants showed 75% or more regain.
Around half of participants regained 50% or more of the earlier weight reduction and approximately one in four regained at least 75% of it. Almost 9% surpassed their starting weight with more than 100% weight regain.
Approximately 4% continued to lose weight under lifestyle intervention alone after tirzepatide withdrawal. Greater weight regain at week 88 was associated with smaller initial reductions in weight, body mass index, and waist circumference during the tirzepatide phase.
Cardiometabolic parameters tracked closely with these weight patterns during withdrawal. At week 36, systolic blood pressure averaged 114.3 mm Hg with a standard deviation of 12.0 mm Hg and diastolic blood pressure averaged 76.2 mm Hg with a standard deviation of 8.6 mm Hg.
From week 36 to week 88, systolic blood pressure increased in all weight regain groups. Least squares mean increases reached 6.8 mm Hg in the less than 25% group, 7.3 mm Hg in the 25% to less than 50% group, 9.6 mm Hg in the 50% to less than 75% group, and 10.4 mm Hg in the 75% or more group. Diastolic blood pressure increased by 2.8 mm Hg, 1.6 mm Hg, 3.9 mm Hg, and 4.3 mm Hg respectively.
Waist circumference rose in step with weight regain. From week 36 to week 88, mean change was 0.8 cm in the less than 25% group, 5.4 cm in the 25% to less than 50% group, 10.1 cm in the 50% to less than 75% group, and 14.7 cm in the 75% or more group.
Triglycerides and non–high density lipoprotein cholesterol also moved upward after withdrawal. From week 36 to week 88, estimated percentage changes in triglycerides were 5.5%, 5.3%, 29.6%, and 18.9% across categories from lowest to highest regain. Estimated percentage changes in non–high density lipoprotein cholesterol were −0.4%, 1.6%, 8.4%, and 10.8% across the same four groups.
Glycemic measures followed a similar pattern. Hemoglobin A1c increased from week 36 to week 88 by least squares mean values of 0.14%, 0.15%, 0.27%, and 0.35% in the ascending weight regain categories. Fasting serum glucose increased by 3.8 mg/dL, 6.9 mg/dL, 9.2 mg/dL, and 9.0 mg/dL across categories.
Markers of insulin resistance changed along with glycemia and lipids. Fasting insulin decreased by 4.0% in the less than 25% group and 15.4%, 46.2%, and 26.3% in the 25% to less than 50%, 50% to less than 75%, and 75% or more groups, respectively. HOMA2 insulin resistance increased by 0.6%, 12.6%, 39.5%, and 27.0%.
Participants who limited weight regain to less than 25% showed no statistically significant changes between week 36 and week 88 in waist circumference, triglycerides, non–high density lipoprotein cholesterol, fasting insulin, and HOMA2 insulin resistance. Several lipid and insulin markers remained similar to values at the end of tirzepatide treatment and continued to differ from baseline.
HDL cholesterol followed a distinct sequence. Levels decreased during tirzepatide treatment, then rose in all weight regain categories after withdrawal. From week 36 to week 88, estimated percentage increases were 19.7%, 16.7%, 13.6%, and 13.4% across the four categories.
HOMA2 β cell function decreased during tirzepatide treatment and decreased further after withdrawal in some categories. That pattern differs from earlier post hoc analyses in adults with type 2 diabetes and obesity that reported increases in HOMA2 β cell function during tirzepatide therapy, and the current work calls for further investigation of insulin resistance and β cell function in adults with obesity under tirzepatide treatment and after discontinuation.
Across the full week 0 to week 88, participants with 75% or more weight regain showed cardiometabolic parameters at week 88 similar to baseline values.
Short-term pharmacologic intervention did not appear to have long-term benefits for most tirzepatide patients who ceased taking the treatment. Continued treatment for those living with chronic obesity may be needed to sustain weight reduction and cardiometabolic benefits, according to the study authors.
Limitations include the post hoc selection structure, absence of body composition and objective diet and physical activity measurements, and potential for conflict of interest and conclusion biasing.