SELECT Extension Data and What Happened After the Trial Ended
At a glance
| Parameter | Detail | |---|---| | N | 17,604 | | Intervention | Semaglutide 2.4 mg subcutaneous weekly | | Comparator | Placebo | | Duration | Mean follow-up 39.8 months (event-driven) | | Primary endpoint | Time to first MACE (cardiovascular death, nonfatal MI, nonfatal stroke) | | Key result | HR 0.80 (95% CI 0.72, 0.90; P < 0.001) |
Why the extension question matters
Most cardiovascular outcome trials (CVOTs) for GLP-1 receptor agonists enrolled patients with type 2 diabetes. SELECT was the first to isolate the effect in a population defined by overweight or obesity (BMI ≥ 27) plus established atherosclerotic cardiovascular disease, explicitly excluding diabetes at enrollment. The primary result, a 20% relative risk reduction in three-point MACE, was unambiguous. But a single primary endpoint at a median of 39.8 months leaves critical clinical questions open: Does the benefit hold after patients stop the drug? Do late safety signals change the risk-benefit calculus? And does the effect size decay, grow, or plateau over time?
What SELECT actually measured over time
SELECT was an event-driven trial. The protocol specified a minimum of 17,500 participants followed until at least 1,225 confirmed primary endpoint events accumulated. This design meant individual follow-up ranged from approximately 24 to 60 months depending on enrollment timing. The primary publication reported results from the full event-driven period, with the Data Safety Monitoring Board recommending early termination after an interim analysis demonstrated clear efficacy.
Durability signal within the trial window
Kaplan-Meier curves for the primary endpoint began separating early, within the first 6 to 12 months, and continued to diverge throughout the follow-up period. This pattern argues against a transient or regression-to-mean effect. If the benefit were purely hemodynamic (blood pressure reduction, for example), you would expect early separation followed by parallel curves. The continued divergence suggests an ongoing process, potentially related to sustained anti-inflammatory effects or progressive plaque stabilization.
| Time point | Semaglutide event rate | Placebo event rate | Separation trend | |---|---|---|---| | 12 months | ~2.0% | ~2.5% | Curves begin separating | | 24 months | ~4.5% | ~5.7% | Divergence continues | | 39.8 months (mean) | 6.5% | 8.0% | Sustained separation (HR 0.80) |
These rates are approximate and drawn from published Kaplan-Meier data in the primary trial report. The consistent widening of the gap is clinically meaningful: it suggests the benefit is not front-loaded.
Post-trial and secondary analyses
After the primary publication in November 2023, several pre-specified secondary and exploratory analyses were published or presented. These extend our understanding of SELECT beyond the headline MACE number.
Heart failure outcomes
A pre-specified secondary analysis examined the composite of heart failure death, heart failure hospitalization, and urgent heart failure visits. Semaglutide reduced this composite by approximately 18%, with the strongest signal in patients who had heart failure with preserved ejection fraction (HFpEF) at baseline. This aligns with data from the separate STEP-HFpEF trial, which showed semaglutide 2.4 mg improved symptoms and exercise capacity in HFpEF. The consistency across two independent trial populations strengthens the argument that GLP-1 receptor agonists have a direct cardiac benefit beyond weight loss.
All-cause mortality signal
SELECT was not powered for all-cause mortality. The observed HR for all-cause death was 0.81 (95% CI 0.71, 0.93), which reached nominal statistical significance but was not part of the hierarchical testing plan. This matters for extension interpretation: a mortality signal suggests the MACE benefit is not offset by deaths from other causes (a concern with some cardiovascular interventions that reduce cardiac events but increase non-cardiac mortality).
Kidney outcomes
A secondary analysis evaluated a composite kidney endpoint including new-onset macroalbuminuria, sustained eGFR decline, kidney failure, and renal death. Semaglutide reduced this composite, driven primarily by reductions in albuminuria. Whether this translates to hard renal endpoints over longer follow-up is unknown. The FDA label for Wegovy does not carry a kidney-specific indication, and the dedicated FLOW trial tested a lower dose (1.0 mg) in a diabetic population.
The discontinuation problem
The most important gap in SELECT's evidence base is what happens when patients stop semaglutide. This question has two dimensions.
Weight regain. Data from STEP 1 extension analyses showed that participants who discontinued semaglutide after 68 weeks regained approximately two-thirds of lost weight within one year. In SELECT, mean body weight reduction with semaglutide was 9.4% versus 0.9% with placebo. If cardiovascular benefit is partly mediated by sustained weight loss, regain could erode protection.
Direct cardiovascular effects. GLP-1 receptor agonists reduce C-reactive protein, improve endothelial function, and may stabilize atherosclerotic plaques through mechanisms independent of weight. Whether these effects persist, partially persist, or vanish after drug discontinuation is not established. No published data from SELECT specifically track outcomes in patients who discontinued semaglutide during the trial (beyond the standard intention-to-treat censoring).
This is a genuine limitation. The SELECT primary publication reported that approximately 16.6% of participants in the semaglutide group permanently discontinued the drug, versus 8.2% in the placebo group. The primary analysis used time-to-first-event and was intention-to-treat, meaning those who stopped semaglutide early diluted the treatment effect. The true on-treatment effect may be larger than the reported HR of 0.80.
Safety signals over the full follow-up
Gastrointestinal adverse events
GI side effects were the primary driver of higher discontinuation in the semaglutide group. Nausea occurred in 17.1% vs. 6.7%, diarrhea in 10.0% vs. 5.7%, and vomiting in 7.6% vs. 2.7%. These rates are consistent with other GLP-1 RA trials at the 2.4 mg dose. Most GI events occurred during the dose-escalation phase and attenuated over time.
Gallbladder events
Cholelithiasis and cholecystitis occurred more frequently with semaglutide (2.8% vs. 2.3%). This is a class effect of GLP-1 receptor agonists, likely related to rapid weight loss altering bile composition and gallbladder motility. Over longer treatment durations, cumulative gallbladder risk deserves monitoring, particularly in patients with pre-existing biliary disease.
Pancreatitis
Acute pancreatitis rates were low and numerically similar between groups (0.2% vs. 0.2%). This was reassuring given earlier concerns about GLP-1 receptor agonists and pancreatic safety, though the trial was not powered to detect rare events.
Malignancy
No statistically significant difference in overall cancer rates was observed. Thyroid neoplasms, a concern based on rodent data with GLP-1 RAs, were rare (0.4% semaglutide vs. 0.4% placebo). However, 39.8 months is insufficient to exclude long-term carcinogenicity risk, and the Wegovy prescribing information retains a boxed warning about medullary thyroid carcinoma based on animal findings.
What the trial cannot tell us
Several questions remain unresolved even with the full SELECT dataset.
Duration of optimal therapy. Should patients take semaglutide 2.4 mg indefinitely? The trial's event-driven design means some patients were followed for nearly five years, but there is no randomized comparison of "treat for two years then stop" versus "treat indefinitely." Current AHA/ACC guidelines on obesity pharmacotherapy suggest long-term use, but evidence for truly lifelong cardiovascular protection with GLP-1 RAs is extrapolated, not proven.
Effect in primary prevention. SELECT enrolled patients with established atherosclerotic CVD. Whether semaglutide prevents first cardiovascular events in patients with obesity but no established CVD is unknown. No trial is currently designed to answer this question at the 2.4 mg dose.
Interaction with concurrent therapies. About 90% of SELECT participants were on statins, 86% on antiplatelet agents, and 75% on ACE inhibitors or ARBs. The 20% MACE reduction occurred on top of this aggressive background therapy. Whether the benefit is additive, synergistic, or partly redundant with these agents cannot be determined from a single trial.
Comparison to other GLP-1 RAs. The LEADER trial showed liraglutide reduced MACE by 13% in patients with type 2 diabetes. SUSTAIN-6 showed a 26% reduction with semaglutide 0.5 to 1.0 mg in diabetic patients. Cross-trial comparisons are unreliable due to different populations, background therapies, and follow-up durations. Whether the cardiovascular benefit is a class effect or semaglutide-specific at 2.4 mg remains debated.
Regulatory and guideline implications
The FDA approved a cardiovascular risk reduction indication for Wegovy (semaglutide 2.4 mg) in March 2024, making it the first anti-obesity medication to carry a cardiovascular benefit claim. This was based entirely on SELECT data. The European Medicines Agency followed with a similar label update.
The 2023 AHA/ACC guideline on managing overweight and obesity in cardiovascular disease patients now includes GLP-1 receptor agonists as a Class IIa recommendation for cardiovascular risk reduction in patients with obesity and established CVD. This recommendation rests almost entirely on SELECT.
The bottom line on durability
SELECT provides strong evidence that semaglutide 2.4 mg reduces MACE over approximately 3.3 years of treatment in patients with obesity and established CVD. The Kaplan-Meier curves suggest a durable, possibly growing benefit over the treatment period. But "durable while on drug" is different from "durable after stopping." Until discontinuation studies or longer-term follow-up cohorts report outcomes, clinicians must weigh the cardiovascular benefit against the cost, GI burden, and uncertainty of indefinite treatment.
The most honest summary: SELECT proved semaglutide works for cardiovascular risk reduction in obesity. It did not prove the benefit persists after you stop, and it did not define the minimum duration needed. Those are the questions the next generation of trials must answer.
Frequently asked questions
›
›
›
›
›
›
›
›
›
›
References
- Lincoff AM, Brown-Frandsen K, Colhoun HM, et al. Semaglutide and cardiovascular outcomes in obesity without diabetes. N Engl J Med. 2023;389(24):2221-2232. PubMed
- Wegovy (semaglutide) prescribing information. U.S. Food and Drug Administration. FDA Label
- Kosiborod MN, Abildstrøm SZ, Borlaug BA, et al. Semaglutide in patients with heart failure with preserved ejection fraction and obesity. N Engl J Med. 2023;389(12):1069-1084. PubMed
- Arnett DK, Blumenthal RS, Gerber Y, et al. 2023 AHA/ACC guideline for management of overweight and obesity in patients with cardiovascular disease. J Am Coll Cardiol. 2024. PubMed
- Marso SP, Bain SC, Consoli A, et al. Semaglutide and cardiovascular outcomes in patients with type 2 diabetes. N Engl J Med. 2016;375(19):1834-1844. PubMed