Semaglutide and Cardiovascular Outcomes in People with Type 2 Diabetes: A SUSTAIN-6 Post Hoc Analysis by Weight Loss Category Authors: Angelo Navas, Joshua Noone, Nathan Laney, Sunny Sheth, Xi Tan, Yuanjie Liang, Paloma Munoz, Caroline Swift Published: 2026 ABSTRACT IntroductionType 2 diabetes (T2D) is a known risk factor for cardiovascular (CV) disease. Glucagon-like peptide-1 receptor agonists (GLP-1RAs), such as semaglutide, improve glycemic control and are associated with weight loss, but the impact of weight loss on major adverse cardiovascular events (MACE) when taking semaglutide is not fully understood. This post hoc analysis of the SUSTAIN-6 trial aimed to examine the correlation between weight loss with semaglutide and the occurrence of MACE (nonfatal myocardial infarction, nonfatal ischemic stroke, and CV death).MethodsThis study is a single-arm, post hoc, observational analysis of the randomized, double-blind, placebo-controlled SUSTAIN-6 cardiovascular outcomes trial (NCT01720446). Participants in the semaglutide group (N = 1479) were stratified into four weight loss categories on the basis of the percentage change in body weight from baseline to 56 weeks: no loss/gain, < 5% loss, 5–10% loss, and > 10% loss. The occurrence of MACE during the follow-up period (weeks 56–104) was compared across weight loss categories using descriptive statistics and logistic regression adjusted for age and sex. Categorical variables were compared across groups using chi-squared tests, and continuous variables were compared using analysis of variance (ANOVA).ResultsPoint estimates of MACE incidence did not differ by weight loss category (P = 0.73). The MACE event rate was comparable across weight loss categories (no loss/gain: 1.05%; < 5% loss: 1.69%; loss 5–10%: 1.06%, > 10% loss: 0.89%). Compared with the no weight loss/gain category, no degree of weight loss was associated with lower adjusted odds (odds ratio) of MACE (loss < 5%: 1.68; P = 0.66; loss 5–10%: 1.13; P = 0.77; loss > 10%: 1.06; P = 0.95).ConclusionsThis post hoc analysis of SUSTAIN-6 did not find an association between the magnitude of weight loss achieved after 56 weeks of semaglutide treatment and subsequent MACE risk in people with T2D, suggesting that weight-independent mechanisms may mediate the cardioprotective effects of semaglutide. Key Summary Points Why carry out this study? People with type 2 diabetes (T2D) carry a high burden of cardiovascular disease; although semaglutide improves glycemic control and is associated with weight loss, it is unclear whether the observed cardiovascular benefit is driven by the magnitude of weight loss.Clarifying this relationship is clinically important for treatment decisions in people with T2D and high cardiovascular risk.This post hoc analysis asked whether the magnitude of predefined weight loss achieved (no loss/gain, < 5%, 5–10%, > 10%) during semaglutide treatment influenced the subsequent risk of major adverse cardiovascular events (MACE) among participants randomized to semaglutide in SUSTAIN-6. What was learned from the study? MACE incidence did not differ across weight loss categories.The observed cardiovascular benefit of semaglutide was independent of the magnitude of weight loss achieved, and broader factors beyond weight loss should be considered in treatment decisions. Why carry out this study? People with type 2 diabetes (T2D) carry a high burden of cardiovascular disease; although semaglutide improves glycemic control and is associated with weight loss, it is unclear whether the observed cardiovascular benefit is driven by the magnitude of weight loss.Clarifying this relationship is clinically important for treatment decisions in people with T2D and high cardiovascular risk.This post hoc analysis asked whether the magnitude of predefined weight loss achieved (no loss/gain, < 5%, 5–10%, > 10%) during semaglutide treatment influenced the subsequent risk of major adverse cardiovascular events (MACE) among participants randomized to semaglutide in SUSTAIN-6. What was learned from the study? MACE incidence did not differ across weight loss categories.The observed cardiovascular benefit of semaglutide was independent of the magnitude of weight loss achieved, and broader factors beyond weight loss should be considered in treatment decisions. Introduction Type 2 diabetes (T2D) substantially increases the risk of cardiovascular (CV) disease, which remains one of the leading causes of morbidity and mortality in this population [1]. Estimates of lifetime risk of CV disease for people with T2D range from 55% to 79% [2, 3], with an even greater risk of disease among people with comorbid obesity [2]. Over the last decade, glucagon-like peptide-1 receptor agonists (GLP-1RAs) have become integral in the management of T2D owing to their capacity to improve glycemic control, promote weight loss, and reduce the incidence of major adverse cardiovascular events (MACE). The cardiovascular outcome trial (CVOT) SUSTAIN-6 demonstrated that once-weekly subcutaneous semaglutide significantly reduced the risk of MACE compared with placebo in adults with T2D and elevated CV risk [4]. Similar findings have been reported with other GLP-1RAs, including liraglutide in LEADER [5] and dulaglutide in REWIND [6]. Weight reduction is recognized as a determinant of improved cardiometabolic health. In the Look AHEAD study, intensive lifestyle interventions leading to weight loss improved glycemic control, blood pressure, and lipid profiles, except for low-density lipoprotein cholesterol levels; however, the long-term impact on CV outcomes was inconclusive [7]. Pharmacologic therapies achieving more robust and sustained weight loss may offer additional benefits. In this context, semaglutide has demonstrated marked effects on body weight, with reductions between 10% and 16% observed in the Semaglutide Treatment Effect in People with Obesity (STEP) program among people with overweight or obesity, with or without diabetes [8–11]. These findings reinforce the role of semaglutide as a weight-lowering agent with consistent efficacy across patient populations. While the evidence base for GLP-1RAs continues to evolve, it remains unclear to what extent weight loss achieved during treatment explains or modifies CV risk reduction. In particular, the association between weight loss during semaglutide treatment and subsequent CV outcomes has not been directly evaluated. This uncertainty is clinically relevant, as understanding whether the CV benefit of semaglutide is independent of the modest weight loss that may occur during treatment could help clarify the extent to which these effects are mediated through mechanisms beyond weight reduction. Potential pathways include improvements in endothelial function, reductions in blood pressure, anti-inflammatory effects, and direct myocardial protection, all of which have been described with GLP-1RA therapy independent of weight change [12–14]. The objective of this post hoc analysis was to examine the incidence of 3-point MACE, including nonfatal myocardial infarction (MI), nonfatal ischemic stroke, and CV death, across different categories of weight loss achieved after 13 months of semaglutide treatment in the SUSTAIN-6 trial, as well as assess the association between weight loss categories and the incidence of MACE outcomes during the subsequent follow-up period (weeks 56–104). Methods Study Design This study is a single-arm, post hoc, observational analysis of the randomized, double-blind, placebo-controlled SUSTAIN-6 CVOT (NCT01720446) (Fig. 1a). Only participants randomized to receive semaglutide at doses of 0.5 mg or 1.0 mg once weekly were included in this analysis. The design and primary outcomes of SUSTAIN-6 have been previously published [4].Fig. 1Study design and participant exclusion flowchart. a Study design schematic detailing the timeline of the study. b Flowchart detailing exclusion criteria and the number of people excluded. *Exclusions denoted with an asterisk (*) were conducted as part of the main SUSTAIN-6 trial, while the remaining exclusions are specific to the present post hoc analysis. ASCVD atherosclerotic cardiovascular disease, CV cardiovascular, CVD cardiovascular disease, HbA1c hemoglobin A1c, MI myocardial infarction, NPH neutral protamine Hagedorn, T2D type 2 diabetes Study design and participant exclusion flowchart. a Study design schematic detailing the timeline of the study. b Flowchart detailing exclusion criteria and the number of people excluded. *Exclusions denoted with an asterisk (*) were conducted as part of the main SUSTAIN-6 trial, while the remaining exclusions are specific to the present post hoc analysis. ASCVD atherosclerotic cardiovascular disease, CV cardiovascular, CVD cardiovascular disease, HbA1c hemoglobin A1c, MI myocardial infarction, NPH neutral protamine Hagedorn, T2D type 2 diabetes Study Population The analysis included all participants with available body weight measurements at baseline and week 56 who were randomized to the semaglutide treatment arm of SUSTAIN-6 (Fig. 1b). Participants who experienced any CV event during the first 13 months of treatment (defined as the weight loss period) were excluded. Participants randomized to placebo and those with missing data for key covariates (sex, age, baseline body weight, weight at week 56, occurrence of CV events during the weight loss period) were also excluded. Eligible participants were adults with T2D aged ≥ 50 years with established CV disease, chronic kidney disease, or New York Heart Association (NYHA) class II–III heart failure, or aged ≥ 60 years with subclinical CV disease. Additional inclusion and exclusion criteria have been described previously. Study Periods The 13-month weight loss period was based on the SUSTAIN-6 visit schedule. Baseline was defined as the time of randomization (week 0). The weight loss period extended from baseline (week 0) to week 56, during which body weight change was assessed. The follow-up period for CV outcomes extended from week 56 to week 104. Exposure Weight loss was defined as the percentage change in body weight from baseline to week 56. Participants were categorized into one of four mutually exclusive groups on the basis of percentage change in body weight: no weight loss or weight gain, < 5% weight loss, 5–10% weight loss, or > 10% weight loss. Weight loss categories were selected a priori using clinically established thresholds [15, 16]: < 5% (below the threshold for clinically significant weight loss), 5–10% (recommended [17] clinically meaningful weight loss; at threshold for cardiometabolic benefit [16]), and > 10% loss (larger loss; at threshold for CV disease risk reduction [17, 18]). These thresholds aim to aid in clinical interpretation of the present study’s findings and align these analyses with current clinical recommendations of ≥ 5% weight loss [17] and existing clinical data supporting weight loss thresholds for risk reduction of 5–10% for cardiometabolic risk factors [16, 19] and > 10% weight loss for CV disease [17, 18]. Outcomes The primary outcome for this post hoc analysis was the first incidence of 3-point MACE, defined as the composite of CV-related death, nonfatal MI, or nonfatal ischemic stroke. The incidence of 3-point MACE was ascertained from the SUSTAIN-6 prospectively collected, adjudicated event dataset. Investigator reports of MACE events were reviewed and centrally adjudicated by an external, independent, blinded clinical events committee. Only the first adjudicated MACE event per participant that occurred during the follow-up period (weeks 56–104) was included. Statistical Analysis Baseline demographic and clinical characteristics were summarized descriptively overall and according to weight loss category. Categorical variables were compared across groups using chi-squared tests, and continuous variables were compared using analysis of variance (ANOVA). As part of this post hoc analysis, the occurrence of MACE was evaluated across categories of percentage weight change. Event rates were calculated as proportions over the 56-week weight loss period by dividing the number of participants experiencing a MACE event by the total number of participants within each weight loss category. These proportions are presented with corresponding standard deviations (SDs) and 95% confidence intervals (Cls). MACE incidence rates were additionally expressed as events per 1000 person-years of follow-up. Person-years were accrued from randomization until the occurrence of the first MACE event, the end of follow-up, or censoring, whichever occurred first. Incidence rates, SDs, and 95% CIs were calculated and reported per 1000 person-years. Across both analyses, the numbers of participants with and without MACE events were reported for each weight loss category. Logistic regression models, adjusted for age and sex, were used to assess the relationship between weight loss and MACE. Adjusted odds ratios (ORs) were obtained from the output of the logistic regression. Higher odds of developing MACE compared with the reference group (no loss/gain cohort) were defined as an OR > 1. All deidentified data from eligible participants randomized to semaglutide in SUSTAIN-6 were included. Secondary data were accessed through FounData, an internal clinical trial data repository maintained by Novo Nordisk. Analyses were performed using Databricks 14.3 with Python 3.10.12 and statsmodels 0.13.5. Ethics The SUSTAIN-6 trial was conducted in accordance with the principles of the Declaration of Helsinki and Good Clinical Practice guidelines. Approval was obtained from institutional review boards or independent ethics committees at each participating site, and all participants provided written informed consent [4]. This post hoc analysis was determined to be exempt from institutional review board review by WCG IRB under 45 CFR 46.104(d)(4). Results Baseline characteristics for the analysis cohort are presented in Table 1. The cohort comprised 1479 participants, with a mean age of 64.6 years. Overall, 38.4% were female and 61.6% male. The majority of participants were white (84.2%), with smaller proportions of Black (5.9%) and Asian (7.9%) participants. Most of the participants were between the ages of 50 and 79 years, with 51.8% aged 50–64 years and 45.9% aged 65–79 years. The mean duration of T2D was 14.1 years, and baseline hemoglobin A1c (HbA1c) values were distributed as follows: HbA1c < 7%, 3.3%; 7% to < 8%, 33.1%; 8% to < 9%, 31.4%; and ≥ 9%, 32.2%. With regard to baseline body mass index (BMI), 7.2% of included participants were normal or underweight, 29.0% were overweight, 31.1% were in obesity class 1, 19.9% in obesity class 2, and 12.7% in obesity class 3. All participants in the cohort had hypertension (100%) and heart failure (100%), with high rates of antihypertensive (94.7%) and statin (72.9%) use. The proportion of participants varied across weight loss categories by sex (P < 0.001), race (P < 0.05), age (P < 0.05), and HbA1c values (P < 0.001) (Table 1).Table 1Baseline characteristicsVariablesTotal (n = 1479)Loss < 5% (n = 592)Loss 5–10% (n = 378)Loss > 10% (n = 223)No loss/gain (n = 286)P-valueAge, mean (SD), years64.6 (7.1)64.3 (7.0)65.0 (6.7)65.2 (7.1)64.2 (7.5)0.184Sex, n (%)< 0.001*Female568 (38.4)206 (34.8)156 (41.3)116 (52.0)90 (31.5)Male911 (61.6)386 (65.2)222 (58.7)107 (48.0)196 (68.5)Age, n (%)0.047*50–64 years766 (51.8)320 (54.1)192 (50.8)97 (43.5)157 (54.9)65–79 years679 (45.9)257 (43.4)175 (46.3)124 (55.6)123 (43.0)≥ 80 years34 (2.3)15 (2.5)11 (2.9)2 (0.9)6 (2.1)Baseline HbA1c, n (%)< 0.001*< 7%49 (3.3)16 (2.7)11 (2.9)13 (5.8)9 (3.1)7% to < 8%489 (33.1)182 (30.7)147 (38.9)82 (36.8)78 (27.3)8% to < 9%465 (31.4)196 (33.1)112 (29.6)81 (36.3)76 (26.6)≥ 9%476 (32.2)198 (33.4)108 (28.6)47 (21.1)123 (43.0)Baseline BMI, kg/m2, n (%)0.141< 25 normal weight or underweight107 (7.2)40 (6.8)21 (5.6)17 (7.6)29 (10.1)25 to < 30 overweight429 (29.0)165 (27.9)114 (30.2)62 (27.8)88 (30.8)30 to <35 obesity class 1460 (31.1)208 (35.1)110 (29.1)66 (29.6)76 (26.6)35 to <40 obesity class 2294 (19.9)114 (19.3)77 (20.4)43 (19.3)60 (21.0)≥ 40 obesity class 3188 (12.7)65 (11.0)56 (14.8)35 (15.7)32 (11.2)Unknown1 (0.1)0 (0.0)0 (0.0)0 (0.0)1 (0.3)Race, n (%)0.048*White1246 (84.2)515 (87.0)317 (83.9)181 (81.2)233 (81.5)Black88 (5.9)25 (4.2)19 (5.0)22 (9.9)22 (7.7)Asian117 (7.9)46 (7.8)31 (8.2)16 (7.2)24 (8.4)Other28 (1.9)6 (1.0)11 (2.9)4 (1.8)7 (2.4)Ethnicity, n (%)Not Hispanic/Latino1242 (84.0)504 (85.1)307 (81.2)194 (87.0)237 (82.9)0.212Hispanic/Latino237 (16.0)88 (14.9)71 (18.8)29 (13.0)49 (17.1)Duration of T2D, years, mean (SD)14.1 (8.1)13.7 (8.4)14.2 (7.6)15.0 (8.1)14.1 (8.3)0.193History of smoking, n (%)0.924Previous smoker617 (41.7)241 (40.7)160 (42.3)92 (41.3)124 (43.4)Never smoked678 (45.8)272 (45.9)177 (46.8)101 (45.3)128 (44.8)Current smoker184 (12.4)79 (13.3)41 (10.8)30 (13.5)34 (11.9)Insulin and insulin analogues, n (%)0.424Any insulin857 (57.9)352 (59.5)214 (56.6)120 (53.8)171 (59.8)Non-T2D medications, n (%)Antihypertensives1400 (94.7)560 (94.6)362 (95.8)211 (94.6)267 (93.4)0.597Any statins1078 (72.9)424 (71.6)281 (74.3)170 (76.2)203 (71.0)0.445Aspirin939 (63.5)368 (62.2)247 (65.3)143 (64.1)181 (63.3)0.788Thrombin agents4 (0.3)0 (0.0)2 (0.5)1 (0.4)1 (0.3)0.412Non-insulin ADMs, n (%)Biguanides1093 (73.9)454 (76.7)289 (76.5)157 (70.4)193 (67.5)0.011*Sulfonylurea643 (43.5)259 (43.8)172 (45.5)88 (39.5)124 (43.4)0.549Thiazolidinediones31 (2.1)7 (1.2)8 (2.1)8 (3.6)8 (2.8)0.138Alpha glucosidase inhibitors15 (1.0)8 (1.4)2 (0.5)1 (0.4)4 (1.4)0.442Meglitinides41 (2.8)19 (3.2)9 (2.4)6 (2.7)7 (2.4)0.861Dipeptidyl peptidase 4 inhibitors2 (0.1)0 (0.0)1 (0.3)0 (0.0)1 (0.3)0.467Sodium–glucose cotransporter 2 inhibitors (SGLT2i)1 (0.1)0 (0.0)1 (0.3)0 (0.0)0 (0.0)0.405ComorbiditiesAnxiety98 (6.6)41 (6.9)23 (6.1)22 (9.9)12 (4.2)0.079Atrial fibrillation73 (4.9)24 (4.1)23 (6.1)9 (4.0)17 (5.9)0.384Cancer48 (3.2)20 (3.4)11 (2.9)9 (4.0)8 (2.8)0.851Cardiac dysrhythmia2 (0.1)0 (0.0)0 (0.0)2 (0.9)0 (0.0)0.01*Chronic kidney disease95 (6.4)36 (6.1)19 (5.0)19 (8.5)21 (7.3)0.337Chronic obstructive pulmonary disease60 (4.1)27 (4.6)10 (2.6)12 (5.4)11 (3.8)0.340Depression182 (12.3)70 (11.8)47 (12.4)37 (16.6)28 (9.8)0.132Heart failure1479 (100.0)592 (100.0)378 (100.0)223 (100.0)286 (100.0)Hyperlipidemia393 (26.6)155 (26.2)104 (27.5)61 (27.4)73 (25.5)0.930Hypertension1,479 (100.0)592 (100.0)378 (100.0)223 (100.0)286 (100.0)Metabolic dysfunction-associated steatotic liver disease1 (0.1)1 (0.2)0 (0.0)0 (0.0)0 (0.0)0.682Any nonalcoholic steatohepatitis0 (0.0%)0 (0.0%)0 (0.0%)0 (0.0%)Obesity350 (23.7)137 (23.1)88 (23.3)50 (22.4)75 (26.2)0.718Osteoporosis42 (2.8)12 (2.0)16 (4.2)11 (4.9)3 (1.0)0.012*Peripheral vascular disease43 (2.9)17 (2.9)11 (2.9)8 (3.6)7 (2.4)0.9The baseline characteristics of study participants as well as cohorts of people stratified by weight loss categories: loss < 5%, loss 5–10%, loss > 10%, and no loss/gain, based on the percentage change in body weight from baseline to 56 weeks. Comparisons across groups and related P-values were determined using chi-squared tests (categorical values) or ANOVAs (continuous variables). P-values at the following level (P < 0.05) are indicated with an asterisk (*)ADMs antidiabetes medications, ANOVAs analyses of variance, BMI body mass index, HbA1c hemoglobin A1c, SD standard deviation, T2D type 2 diabetes Table 1Baseline characteristicsVariablesTotal (n = 1479)Loss < 5% (n = 592)Loss 5–10% (n = 378)Loss > 10% (n = 223)No loss/gain (n = 286)P-valueAge, mean (SD), years64.6 (7.1)64.3 (7.0)65.0 (6.7)65.2 (7.1)64.2 (7.5)0.184Sex, n (%)< 0.001*Female568 (38.4)206 (34.8)156 (41.3)116 (52.0)90 (31.5)Male911 (61.6)386 (65.2)222 (58.7)107 (48.0)196 (68.5)Age, n (%)0.047*50–64 years766 (51.8)320 (54.1)192 (50.8)97 (43.5)157 (54.9)65–79 years679 (45.9)257 (43.4)175 (46.3)124 (55.6)123 (43.0)≥ 80 years34 (2.3)15 (2.5)11 (2.9)2 (0.9)6 (2.1)Baseline HbA1c, n (%)< 0.001*< 7%49 (3.3)16 (2.7)11 (2.9)13 (5.8)9 (3.1)7% to < 8%489 (33.1)182 (30.7)147 (38.9)82 (36.8)78 (27.3)8% to < 9%465 (31.4)196 (33.1)112 (29.6)81 (36.3)76 (26.6)≥ 9%476 (32.2)198 (33.4)108 (28.6)47 (21.1)123 (43.0)Baseline BMI, kg/m2, n (%)0.141< 25 normal weight or underweight107 (7.2)40 (6.8)21 (5.6)17 (7.6)29 (10.1)25 to < 30 overweight429 (29.0)165 (27.9)114 (30.2)62 (27.8)88 (30.8)30 to <35 obesity class 1460 (31.1)208 (35.1)110 (29.1)66 (29.6)76 (26.6)35 to <40 obesity class 2294 (19.9)114 (19.3)77 (20.4)43 (19.3)60 (21.0)≥ 40 obesity class 3188 (12.7)65 (11.0)56 (14.8)35 (15.7)32 (11.2)Unknown1 (0.1)0 (0.0)0 (0.0)0 (0.0)1 (0.3)Race, n (%)0.048*White1246 (84.2)515 (87.0)317 (83.9)181 (81.2)233 (81.5)Black88 (5.9)25 (4.2)19 (5.0)22 (9.9)22 (7.7)Asian117 (7.9)46 (7.8)31 (8.2)16 (7.2)24 (8.4)Other28 (1.9)6 (1.0)11 (2.9)4 (1.8)7 (2.4)Ethnicity, n (%)Not Hispanic/Latino1242 (84.0)504 (85.1)307 (81.2)194 (87.0)237 (82.9)0.212Hispanic/Latino237 (16.0)88 (14.9)71 (18.8)29 (13.0)49 (17.1)Duration of T2D, years, mean (SD)14.1 (8.1)13.7 (8.4)14.2 (7.6)15.0 (8.1)14.1 (8.3)0.193History of smoking, n (%)0.924Previous smoker617 (41.7)241 (40.7)160 (42.3)92 (41.3)124 (43.4)Never smoked678 (45.8)272 (45.9)177 (46.8)101 (45.3)128 (44.8)Current smoker184 (12.4)79 (13.3)41 (10.8)30 (13.5)34 (11.9)Insulin and insulin analogues, n (%)0.424Any insulin857 (57.9)352 (59.5)214 (56.6)120 (53.8)171 (59.8)Non-T2D medications, n (%)Antihypertensives1400 (94.7)560 (94.6)362 (95.8)211 (94.6)267 (93.4)0.597Any statins1078 (72.9)424 (71.6)281 (74.3)170 (76.2)203 (71.0)0.445Aspirin939 (63.5)368 (62.2)247 (65.3)143 (64.1)181 (63.3)0.788Thrombin agents4 (0.3)0 (0.0)2 (0.5)1 (0.4)1 (0.3)0.412Non-insulin ADMs, n (%)Biguanides1093 (73.9)454 (76.7)289 (76.5)157 (70.4)193 (67.5)0.011*Sulfonylurea643 (43.5)259 (43.8)172 (45.5)88 (39.5)124 (43.4)0.549Thiazolidinediones31 (2.1)7 (1.2)8 (2.1)8 (3.6)8 (2.8)0.138Alpha glucosidase inhibitors15 (1.0)8 (1.4)2 (0.5)1 (0.4)4 (1.4)0.442Meglitinides41 (2.8)19 (3.2)9 (2.4)6 (2.7)7 (2.4)0.861Dipeptidyl peptidase 4 inhibitors2 (0.1)0 (0.0)1 (0.3)0 (0.0)1 (0.3)0.467Sodium–glucose cotransporter 2 inhibitors (SGLT2i)1 (0.1)0 (0.0)1 (0.3)0 (0.0)0 (0.0)0.405ComorbiditiesAnxiety98 (6.6)41 (6.9)23 (6.1)22 (9.9)12 (4.2)0.079Atrial fibrillation73 (4.9)24 (4.1)23 (6.1)9 (4.0)17 (5.9)0.384Cancer48 (3.2)20 (3.4)11 (2.9)9 (4.0)8 (2.8)0.851Cardiac dysrhythmia2 (0.1)0 (0.0)0 (0.0)2 (0.9)0 (0.0)0.01*Chronic kidney disease95 (6.4)36 (6.1)19 (5.0)19 (8.5)21 (7.3)0.337Chronic obstructive pulmonary disease60 (4.1)27 (4.6)10 (2.6)12 (5.4)11 (3.8)0.340Depression182 (12.3)70 (11.8)47 (12.4)37 (16.6)28 (9.8)0.132Heart failure1479 (100.0)592 (100.0)378 (100.0)223 (100.0)286 (100.0)Hyperlipidemia393 (26.6)155 (26.2)104 (27.5)61 (27.4)73 (25.5)0.930Hypertension1,479 (100.0)592 (100.0)378 (100.0)223 (100.0)286 (100.0)Metabolic dysfunction-associated steatotic liver disease1 (0.1)1 (0.2)0 (0.0)0 (0.0)0 (0.0)0.682Any nonalcoholic steatohepatitis0 (0.0%)0 (0.0%)0 (0.0%)0 (0.0%)Obesity350 (23.7)137 (23.1)88 (23.3)50 (22.4)75 (26.2)0.718Osteoporosis42 (2.8)12 (2.0)16 (4.2)11 (4.9)3 (1.0)0.012*Peripheral vascular disease43 (2.9)17 (2.9)11 (2.9)8 (3.6)7 (2.4)0.9The baseline characteristics of study participants as well as cohorts of people stratified by weight loss categories: loss < 5%, loss 5–10%, loss > 10%, and no loss/gain, based on the percentage change in body weight from baseline to 56 weeks. Comparisons across groups and related P-values were determined using chi-squared tests (categorical values) or ANOVAs (continuous variables). P-values at the following level (P < 0.05) are indicated with an asterisk (*)ADMs antidiabetes medications, ANOVAs analyses of variance, BMI body mass index, HbA1c hemoglobin A1c, SD standard deviation, T2D type 2 diabetes Baseline characteristics The baseline characteristics of study participants as well as cohorts of people stratified by weight loss categories: loss < 5%, loss 5–10%, loss > 10%, and no loss/gain, based on the percentage change in body weight from baseline to 56 weeks. Comparisons across groups and related P-values were determined using chi-squared tests (categorical values) or ANOVAs (continuous variables). P-values at the following level (P < 0.05) are indicated with an asterisk (*) ADMs antidiabetes medications, ANOVAs analyses of variance, BMI body mass index, HbA1c hemoglobin A1c, SD standard deviation, T2D type 2 diabetes Primary Analysis of CV Outcomes by Weight Loss Category During the follow-up period (weeks 56–104), 19 participants experienced a first MACE: no weight loss or weight gain (3 of 286), < 5% weight loss (10 of 592), 5–10% weight loss (4 of 378), and > 10% weight loss (2 of 223) (Table 2). The 3-point MACE event rates (including MI, ischemic stroke, and CV-related death) did not differ across weight loss categories (X2 = 1.31; P = 0.73; data not shown). Among cohorts, the MACE event rate was 1.05% (95% CI 0.22–3.03%) in those with no weight loss or weight gain, 1.69% (95% CI 0.81–3.08%) in those with < 5% weight loss, 1.06% (95% CI 0.29–2.69%) in those with 5–10% weight loss, and 0.89% (95% CI 0.11–3.20%) in those with > 10% weight loss (Fig. 2; Table 2). Expressed per 1000 person-years, event rates were 9.75 (95% CI 2.04–28.15) among individuals treated with semaglutide with no weight loss or weight gain, 15.70 (95% CI 7.52–28.60) among those with < 5% weight loss, 9.84 (95% CI 2.69–24.99) among those with 5–10% weight loss, and 8.27 (95% CI 1.02–29.72) among those with > 10% weight loss (Table 2).Table 2Incidence rates of major adverse cardiovascular events (MACE) by weight loss categoryWeight loss categoryPeople with MACE (n = 19)People without MACE (n = 1460)MACE event rate (% [95% CI])MACE event rate (per 1000 PY)No loss/gainMean baseline weight = 205.2 lb32831.05% (0.22–3.03%)9.75 (2.04–28.15)Loss < 5%Mean baseline weight = 204.8 lb105821.69% (0.81–3.08%)15.70 (7.52–28.60)Loss 5–10%Mean baseline weight = 203.0 lb43741.06% (0.29–2.69%)9.84 (2.69–24.99)Loss > 10%Mean baseline weight = 198.6 lb22210.89% (0.11–3.20%)8.27 (1.02–29.72)The number of people with and without MACE and MACE event rate (% and per 1000 person-years) for semaglutide-treated SUSTAIN-6 participants stratified by weight loss categories: no loss/gain, loss < 5%, loss 5–10%, and loss > 10%PY person yearsFig. 2Forest plot of major adverse cardiovascular events (MACE) rates by weight loss category. The rates of MACE (nonfatal ischemic stroke, nonfatal MI, or cardiovascular death) by weight loss categories (no loss/gain [n = 286], loss < 5% [n = 592], loss 5–10% [n = 378], and loss > 10% [n = 223]) during the weight loss period (week 0–week 56). The black dot indicates the rate of MACE for each group, and the horizontal lines represent the 95% confidence intervals (CIs). MI myocardial infarction Table 2Incidence rates of major adverse cardiovascular events (MACE) by weight loss categoryWeight loss categoryPeople with MACE (n = 19)People without MACE (n = 1460)MACE event rate (% [95% CI])MACE event rate (per 1000 PY)No loss/gainMean baseline weight = 205.2 lb32831.05% (0.22–3.03%)9.75 (2.04–28.15)Loss < 5%Mean baseline weight = 204.8 lb105821.69% (0.81–3.08%)15.70 (7.52–28.60)Loss 5–10%Mean baseline weight = 203.0 lb43741.06% (0.29–2.69%)9.84 (2.69–24.99)Loss > 10%Mean baseline weight = 198.6 lb22210.89% (0.11–3.20%)8.27 (1.02–29.72)The number of people with and without MACE and MACE event rate (% and per 1000 person-years) for semaglutide-treated SUSTAIN-6 participants stratified by weight loss categories: no loss/gain, loss < 5%, loss 5–10%, and loss > 10%PY person years Incidence rates of major adverse cardiovascular events (MACE) by weight loss category No loss/gain Mean baseline weight = 205.2 lb Loss < 5% Mean baseline weight = 204.8 lb Loss 5–10% Mean baseline weight = 203.0 lb Loss > 10% Mean baseline weight = 198.6 lb The number of people with and without MACE and MACE event rate (% and per 1000 person-years) for semaglutide-treated SUSTAIN-6 participants stratified by weight loss categories: no loss/gain, loss < 5%, loss 5–10%, and loss > 10% PY person years Forest plot of major adverse cardiovascular events (MACE) rates by weight loss category. The rates of MACE (nonfatal ischemic stroke, nonfatal MI, or cardiovascular death) by weight loss categories (no loss/gain [n = 286], loss < 5% [n = 592], loss 5–10% [n = 378], and loss > 10% [n = 223]) during the weight loss period (week 0–week 56). The black dot indicates the rate of MACE for each group, and the horizontal lines represent the 95% confidence intervals (CIs). MI myocardial infarction Logistic Regression Analysis Logistic regression adjusted for age and sex did not find an association between the magnitude of weight loss and the odds of 3-point MACE (Table 3). Weight loss of < 5% was not associated with lower odds of 3-point MACE in people who received 13 months of semaglutide treatment (OR, 1.68; 95% CI 0.46–6.17; P = 0.66) compared with those with no weight loss or weight gain. Similarly, the odds of 3‑point MACE did not differ among semaglutide‑treated cohorts with weight loss of 5–10% (OR 1.13; 95% CI 0.25–5.10; P = 0.77) or > 10% (OR 1.06; 95% CI 0.17–6.49; P = 0.95) relative to the no loss/gain cohort.Table 3Odds ratios for weight loss categories and major adverse cardiovascular events (MACE)Weight loss categoryOdds ratio (OR)aConfidence interval (CI)P-value*No loss/gain1.00Ref.Ref.Loss < 5%1.680.46–6.170.66Loss 5–10%1.130.25–5.100.77Loss > 10%1.060.17–6.490.95Odds ratios (ORs) for the association between each weight loss category and major adverse cardiovascular events (MACE) in participants receiving semaglutide stratified by weight loss category: no loss/gain, loss < 5%, loss 5–10%, loss > 10%*P > 0.05 indicates no evidence of higher odds of MACE for this weight loss categoryaOutput of a logistic regression adjusted for sex and age, with people in the no loss/gain category as the reference group. OR > 1 signifies higher odds of developing MACE compared with the reference group Table 3Odds ratios for weight loss categories and major adverse cardiovascular events (MACE)Weight loss categoryOdds ratio (OR)aConfidence interval (CI)P-value*No loss/gain1.00Ref.Ref.Loss < 5%1.680.46–6.170.66Loss 5–10%1.130.25–5.100.77Loss > 10%1.060.17–6.490.95Odds ratios (ORs) for the association between each weight loss category and major adverse cardiovascular events (MACE) in participants receiving semaglutide stratified by weight loss category: no loss/gain, loss < 5%, loss 5–10%, loss > 10%*P > 0.05 indicates no evidence of higher odds of MACE for this weight loss categoryaOutput of a logistic regression adjusted for sex and age, with people in the no loss/gain category as the reference group. OR > 1 signifies higher odds of developing MACE compared with the reference group Odds ratios for weight loss categories and major adverse cardiovascular events (MACE) Odds ratios (ORs) for the association between each weight loss category and major adverse cardiovascular events (MACE) in participants receiving semaglutide stratified by weight loss category: no loss/gain, loss < 5%, loss 5–10%, loss > 10% *P > 0.05 indicates no evidence of higher odds of MACE for this weight loss category aOutput of a logistic regression adjusted for sex and age, with people in the no loss/gain category as the reference group. OR > 1 signifies higher odds of developing MACE compared with the reference group Discussion Previous CVOTs of GLP-1RAs have demonstrated CV benefit that is not fully explained by the magnitude of weight loss, prompting calls to more directly evaluate the relationship between weight reduction and CV outcomes in this therapeutic context [20]. This post hoc analysis used existing clinical trial data to assess whether the extent of weight loss was associated with the incidence of MACE after a prespecified weight loss period in SUSTAIN-6 participants who received semaglutide. In this study, we did not observe a difference in MACE incidence by weight loss category. Although obesity is a well-established risk factor for CV disease, the assumption that greater weight loss uniformly translates into improved CV outcomes may not be accurate in all clinical contexts. The concept of the “obesity paradox” suggests that higher BMI and related anthropometric measures do not consistently predict worse short‑term CV outcomes among individuals with established atherosclerotic CV disease, particularly among those with overweight or class 1 obesity [21]. In studies reporting CV benefit associated with weight loss, effects are often attributed to improvements in metabolic abnormalities common in T2D, such as insulin resistance, hyperglycemia, and dyslipidemia [22], as well as reductions in blood pressure, as observed in the Look AHEAD trial [23]. While similar improvements in cardiometabolic risk factors were observed in the present analysis, these did not translate into differences in MACE incidence across weight loss categories [23]. Semaglutide has been shown to reduce CV risk in people with T2D and established or high risk of CV disease; however, CVOTs such as SUSTAIN-6 were designed to establish CV safety and benefit rather than to evaluate the specific contribution of weight loss to these outcomes [24, 25]. Thus, the results of our study suggest that the CV benefits of semaglutide may be independent of the beneficial effects of weight loss on cardiometabolic factors [26, 27]. Our findings are consistent with emerging evidence suggesting that the CV benefits of GLP-1RAs may occur independently of weight loss and other downstream cardiometabolic changes [20, 28]. In the present analysis, MACE rates were low across all subgroups, and no association was observed between weight loss category and CV outcomes, even among participants achieving > 10% weight loss. Several weight-independent mechanisms may contribute to the cardioprotective effects of GLP-1RAs, including direct vascular and myocardial effects [29], improved plaque stability [29], anti-inflammatory benefits [30], and modulation of prothrombotic [31] and neurohormonal activity [15]. Importantly, the present analysis evaluated the association between the magnitude of weight loss and MACE incidence, rather than whether weight loss itself is the primary driver of cardiometabolic benefit. A prespecified analysis of SELECT, a trial in people with overweight or obesity and CV disease but without diabetes, found that semaglutide reduced MACE across adiposity groups, with reductions in waist circumference only partially mediating the observed benefit [32]. Together, these findings and the present analysis underscore that weight change alone does not fully explain the CV effects of GLP-1RAs and highlight the need for further research into mechanisms of GLP-1RA-induced cardioprotection beyond weight loss. The study population consisted predominantly of older adults with T2D and high CV risk, a group in which obesity and age-related changes in body composition, including sarcopenic obesity, are common. In this context, weight loss may not uniformly confer CV benefit and may, in some cases, exacerbate loss of lean mass [33–35]. These considerations are particularly relevant as health systems, payers, and clinicians evaluate access to GLP-1RAs for people with T2D and comorbid obesity. This study highlights that people with T2D who have either established or are at high risk for CV disease might benefit from semaglutide use, regardless of the degree of weight loss they achieve. This could help healthcare professionals avoid placing unnecessary emphasis on weight loss alone and instead shift the focus to other aspects of cardiometabolic health when managing T2D and related CV risk [36]. Clinicians may be able to better stratify their patients’ CV risk and tailor treatment on the basis of potential proven CV benefit of specific GLP-1RAs in the context of this evidence. From a contemporary clinical perspective, these findings also align with the recently proposed cardiovascular–kidney–metabolic (CKM) syndrome framework, which emphasizes the interconnected pathophysiology of metabolic dysfunction, CV disease, and kidney disease across a continuum of risk. The American Heart Association has highlighted the importance of evaluating therapies on the basis of their integrated effects across these domains rather than on isolated outcomes. Within this framework, the absence of an association between weight loss magnitude and CV outcomes in the present analysis supports the concept that semaglutide may confer CV benefit through mechanisms beyond weight reduction alone. It is important to note the limitations of our study when interpreting the results. This was a post hoc analysis; SUSTAIN-6 was not powered to detect differences in MACE incidence between weight loss categories, limiting statistical power, particularly within subgroups. Unmeasured factors associated with a better response to semaglutide treatment (i.e., greater weight loss), such as more favorable social determinants of health, could decrease the likelihood of MACE outcomes. Furthermore, prior research has largely shown that women have a greater therapeutic response to GLP-1RAs in clinical trials [37, 38] and real-world settings [39], whereas data related to the influence of HbA1c values and age on treatment efficacy are less conclusive [37, 38, 40]. While the present study controlled for age and sex, other baseline characteristics, specifically race and HbA1c values, were not randomly distributed across all weight loss categories; therefore, the potential impact of these covariates on CV risk cannot be ruled out. Inclusion of additional covariates in sensitivity analyses produced similar point estimates, but increased confidence intervals. Importantly, multivariate models with many covariates risk overfitting and unreliable parameter estimates, particularly in analyses with a low number of outcome events [41]. Thus, logistic regressions in the present analyses were adjusted only for age and sex. The age- and sex-adjusted results presented herein are intended to be hypothesis-generating estimates, and the possibility of residual confounding cannot be excluded. Additional research should account for and elucidate the dual determinants of weight loss and CV risk reduction in people treated with semaglutide. The present analyses were restricted to participants randomized to the semaglutide arm of SUSTAIN-6 who survived to week 56 and had complete covariate data and excluded participants in the placebo-treated arm. Thus, these associations reflect associations within semaglutide-treated participants only and are not generalizable to between-treatment comparisons or to broader trial populations. Given that people with missing continuous data were excluded from analyses and a complete case approach was used, weighting for selective exclusions or attrition were not conducted in these analyses. However, the present study may be limited by the potential for selection bias given the exclusion of people lost to follow-up. The predominantly older study population may limit generalizability to younger individuals or those with different comorbidity profiles. Finally, the follow-up duration was limited to approximately 2 years, and longer-term studies are needed to further characterize the relationship between weight change and CV outcomes during semaglutide therapy. Conclusions In this post hoc analysis of SUSTAIN-6, weight reduction during the first year of semaglutide treatment was not associated with the incidence of MACE. These findings suggest that the CV benefits of semaglutide may be independent of the degree of weight loss achieved.