Long-term weight loss effects of semaglutide in obesity without diabetes in the SELECT trial.
Ryan DH, Lingvay I, Deanfield J, Kahn SE, Barros E, Burguera B, Colhoun HM, Cercato C, Dicker D, Horn DB, Hovingh GK, Jeppesen OK, Kokkinos A, Lincoff AM, Meyhöfer SM, Oral TK, Plutzky J, van Beek AP, Wilding JPH, Kushner RF
- DOI
- 10.1038/s41591-024-02996-7
- Record issued
- 2026-08-16
- Engine
- 7.39.0
- Exported
- 2026-09-21
Prepared by Alpha1. This document is confidential: it is intended for the recipient it was shared with and must not be redistributed. The live record at alpha1science.com/verify/bd220d8e-ba50-4846-8100-65dde67edf52 is authoritative.
How this rating was calculated
- ClaimsEfficacy rests on an unvalidated surrogate endpoint−0.5★
- 01Efficacy rests on an unvalidated surrogate endpoint
The primary efficacy claim is weight loss and anthropometric changes (waist circumference, waist-to-height ratio, BMI category), which are surrogate markers for clinical outcomes. The paper does not demonstrate target engagement at the tested dose (e.g., PK/PD) nor cite validated evidence linking these surrogates to hard clinical outcomes in this context. Although the SELECT trial showed a 20% reduction in MACE, this analysis focuses on weight and anthropometrics, not the hard outcome.
“At 208 weeks, semaglutide was associated with mean reduction in weight (−10.2%), waist circumference (−7.7 cm) and waist-to-height ratio (−6.9%) versus placebo (−1.5%, −1.3 cm and −1.0%, respectively; P < 0.0001 for all comparisons versus placebo).”
This Kaimen Rigor review uses Kaimen Rigor reviewers trained on a curated corpus of high-fidelity and retracted papers, with expert supervision and curation. It can still make mistakes; verify each finding against the source before relying on it.
This is a well-conducted and well-reported prespecified secondary analysis of a large randomized controlled trial. The paper demonstrates strong methodological rigor across all eight dimensions, with only minor reporting gaps (e.g., randomization method not detailed, no explicit reporting guideline statement, assumptions not explicitly verified).
Both reviewers classified the study as interventional, and this was adopted. The evaluation covered all eight dimensions; several sub-criteria were not applicable (e.g., animal-related, cell line, code sharing). The statistics verification checked 4 tests, all consistent; the citation check found no retracted or unresolved references; the reproducibility check confirmed the data link is live.
Numerical inconsistencies
None foundValues that contradict each other or are impossible for the stated sample: recomputed p-values and test statistics, GRIM/GRIMMER checks on summary numbers, percentages against their own counts, totals against their parts, and estimates against their own confidence intervals.
Checked — nothing surfaced.
Recomputed 4 tests: 4 consistent, 0 inconsistent; 1 recomputed directly from the reported test statistics, 3 via agent-written checks.
- CONSISTENTreported p < .001 · recomputed p = <.001Recomputed hazard ratio 0.80 (95% CI 0.72–0.90), reported p<0.001
“hazard ratio 0.80, 95% confidence interval (CI) 0.72 to 0.90; P < 0.001”
Taken as given: 0.72–0.90 is a two-sided 95% confidence interval for the hazard ratio of 0.80, not a range, an IQR, or a different interval level; the hazard ratio is a RATIO measure, so the interval is symmetric on the log scale; p<0.001 is the p for THIS estimate, not for another comparison in the same sentenceMethod: back the two-tailed p out of the log-scale CI width and compare it against the printed pHow we recomputed it: pCI(0.8, 0.72, 0.9, 1) - CONSISTENTreported p < .001 · recomputed p = <.001Reviewers 1, 2Recompute p-value for treatment difference in percentage weight loss at week 208 (in-trial) from reported estimate and 95% CI.
“treatment difference −8.7%; 95% CI −9.42 to −7.88; P < 0.0001”
Taken as given: The treatment difference is -8.7% with 95% CI -9.42 to -7.88.; The CI is two-sided at 95%.; The p-value is two-sided.Method: Used pCI function to derive p-value from estimate and CI assuming normal approximation.How we recomputed it: pCI(-8.7, -9.42, -7.88, 0) - CONSISTENTreported p < .001 · recomputed p = <.001Reviewers 1, 2Recompute p-value for treatment difference in waist circumference at week 208 from reported estimate and 95% CI.
“treatment difference of −6.4 cm (95% CI −7.18 to −5.61; P < 0.0001)”
Taken as given: The treatment difference is -6.4 cm with 95% CI -7.18 to -5.61.; The CI is two-sided at 95%.; The p-value is two-sided.Method: Used pCI function to derive p-value from estimate and CI assuming normal approximation.How we recomputed it: pCI(-6.4, -7.18, -5.61, 0) - CONSISTENTreported p < .001 · recomputed p = <.001Reviewer 2Recompute p-value for treatment difference in WHtR at week 208 from reported estimate and 95% CI.
“treatment difference −5.87% points; 95% CI −6.56 to −5.17; P < 0.0001”
Taken as given: The estimate is the treatment difference in percentage points.; The 95% CI is two-sided.; The p-value is for the test that the difference is zero.Method: Two-sided p-value from estimate and 95% CI using normal approximation.How we recomputed it: pCI(-5.87, -6.56, -5.17, 0)
Overstated conclusions
1 finding · worst highConclusions that reach past what the paper's own results support — including a significance claim that no longer holds when the statistic is recomputed, and efficacy resting on an unvalidated surrogate endpoint.
- Efficacy rests on an unvalidated surrogate endpointAssessed
6 major claims checked against the paper's own evidence: all adequately supported.
- supportedReviewers 1, 2Semaglutide produced clinically significant weight loss and improvements in anthropometric measurements versus placebo at 208 weeks.The claim is directly supported by the reported treatment differences and p-values.Evidence: Reported mean weight loss -10.2% vs -1.5% (treatment difference -8.7%, 95% CI -9.42 to -7.88, P<0.0001); WC -7.7 cm vs -1.3 cm (treatment difference -6.4 cm, 95% CI -7.18 to -5.61, P<0.0001); WHtR -6.9% vs -1.0% (treatment difference -5.87%, 95% CI -6.56 to -5.17, P<0.0001).
“At 208 weeks, semaglutide was associated with mean reduction in weight (−10.2%), waist circumference (−7.7 cm) and waist-to-height ratio (−6.9%) versus placebo (−1.5%, −1.3 cm and −1.0%, respectively; P < 0.0001 for all comparisons versus placebo).”
AbstractFind in source - supportedReviewers 1, 2Weight loss was sustained over 4 years.The claim is supported by the trajectory data showing weight loss continued to week 65 and was sustained through week 208.Evidence: Figure 1 shows weight loss trajectory; text states 'the weight-loss trajectory continued to week 65 and then was sustained for the study period through week 208'.
“For those in the semaglutide group, the weight-loss trajectory continued to week 65 and then was sustained for the study period through week 208 (−10.2% for the semaglutide group, −1.5% for the placebo group; treatment difference −8.7%; 95% CI −9.42 to −7.88; P < 0.0001).”
ResultsFind in source - supportedReviewers 1, 2Clinically meaningful weight loss occurred in both sexes and all races, body sizes and regions.The claim is supported by subgroup analyses showing consistent treatment effects across subgroups, though some heterogeneity exists.Evidence: Forest plot (Fig. 5) shows treatment differences across subgroups; text states 'both sexes, all races, all body sizes and those from all geographic regions were able to achieve clinically meaningful weight loss.'
“Furthermore, both sexes, all races, all body sizes and those from all geographic regions were able to achieve clinically meaningful weight loss.”
Discussion ¶1Find in source - supportedReviewers 1, 2Semaglutide was associated with fewer serious adverse events.The claim is supported by reported SAE rates and p-values.Evidence: SAEs reported by 2,941 patients (33.4%) in semaglutide arm vs 3,204 (36.4%) in placebo (P<0.001); SAE rates per 100 years lower in semaglutide across BMI categories.
“We reported in the primary SELECT analysis that serious adverse events (SAEs) were reported by 2,941 patients (33.4%) in the semaglutide arm and by 3,204 patients (36.4%) in the placebo arm ( P < 0.001) .”
ResultsFind in source - supportedReviewer 1Semaglutide was associated with increased rates of trial product discontinuation, with discontinuations increasing as BMI class decreased.The claim is supported by the cumulative incidence analysis showing a graded increase in discontinuation for semaglutide, particularly in lower BMI classes.Evidence: Figure 6 shows cumulative incidence of discontinuation by BMI class; text states 'For lower BMI classes, discontinuation rates are higher in the semaglutide group but not the placebo group.'
“Both panels of Fig. depict a graded increase in the proportion discontinuing semaglutide, but not placebo. For lower BMI classes, discontinuation rates are higher in the semaglutide group but not the placebo group.”
ResultsFind in source - supportedReviewer 2Semaglutide was associated with increased rates of trial product discontinuation, and discontinuations increased as BMI class decreased.Figure 6 shows higher discontinuation with semaglutide, especially in lower BMI classes.Evidence: Figure 6 and text describe graded increase in discontinuation with semaglutide, not placebo.
“Semaglutide was associated with increased rates of trial product discontinuation. Discontinuations increased as BMI class decreased.”
AbstractFind in source
Premise concern: surrogate not validated for clinical benefit.
- INADEQUATESurrogate endpointThe primary efficacy claim is weight loss and anthropometric changes (waist circumference, waist-to-height ratio, BMI category), which are surrogate markers for clinical outcomes. The paper does not demonstrate target engagement at the tested dose (e.g., PK/PD) nor cite validated evidence linking these surrogates to hard clinical outcomes in this context. Although the SELECT trial showed a 20% reduction in MACE, this analysis focuses on weight and anthropometrics, not the hard outcome.
“At 208 weeks, semaglutide was associated with mean reduction in weight (−10.2%), waist circumference (−7.7 cm) and waist-to-height ratio (−6.9%) versus placebo (−1.5%, −1.3 cm and −1.0%, respectively; P < 0.0001 for all comparisons versus placebo).”
- ADEQUATEEffect sizeThe effect size is large and clinically meaningful: mean weight loss of 10.2% with semaglutide vs 1.5% with placebo, and 44.2% of semaglutide patients achieved ≥10% weight loss. The paper explicitly states 'clinically meaningful weight loss' and references that ≥5% weight loss is associated with improvements in CV risk factors. The effect is statistically supported and anchored to clinical meaningfulness.
“In SELECT, at 208 weeks, semaglutide produced clinically significant weight loss and improvements in anthropometric measurements versus placebo.”
Data authenticity concerns
None foundAn adversarial read for patterns associated with data that may not be genuine: results that look too clean, implausibly large effects, duplicated data or images, and methods that do not match the results reported.
Checked — nothing surfaced.
Reporting gaps
None foundRequired detail the manuscript never states — study design, biological variables, ethics approval and consent, key resources, statistical reporting, data and code availability, and overall transparency.
Checked — nothing surfaced.
The introduction cites prior work on obesity prevalence, the relationship between body fat and cardiovascular disease, and the development of weight-management medications including semaglutide. It acknowledges the challenge of sustaining weight loss with lifestyle intervention alone and references prior semaglutide trials (e.g., STEP 1). The rationale for this prespecified analysis is clearly stated: to evaluate weight loss and anthropometric outcomes in a diverse population over 208 weeks. Limitations of prior research are implicitly addressed by noting the need for long-term data in diverse populations.
“This prespecified analysis of the SELECT trial investigated weight loss and changes in anthropometric indices in patients with established CVD and overweight or obesity without diabetes, who met inclusion and exclusion criteria, within a range of baseline categories for glycemia, renal function and body anthropometric measures.”
“These data, representing the longest clinical trial of the effects of semaglutide versus placebo on weight, establish the safety and durability of semaglutide effects on weight loss and maintenance in a geographically and racially diverse population of adult men and women with overweight and obesity but not diabetes.”
The trial is described as randomized, double-blind, and placebo-controlled. Randomization method is not detailed in this paper but is referenced to prior publications; however, the design is standard for a large trial. Blinding is stated (double-blind). Power analysis is not reported in this paper, but the trial is a large cardiovascular outcomes trial with a prespecified analysis; the primary trial was adequately powered. Inclusion/exclusion criteria are described (age ≥45, BMI ≥27, established CVD). Outlier handling is not explicitly discussed, but missing data are handled via multiple imputation. Controls are the placebo group. Independent replication is not applicable for a single pivotal trial.
“Patients were randomly assigned in a double-blind manner and 1:1 ratio to receive once-weekly subcutaneous semaglutide 2.4 mg or placebo.”
“Eligible patients were aged ≥45 years, with a BMI of ≥27 kg m − 2 and established CVD defined as at least one of the following: prior myocardial infarction, prior ischemic or hemorrhagic stroke, or symptomatic peripheral artery disease.”
“Missing data at the landmark visit, for example, week 104, were imputed using a multiple imputation model and done separately for each treatment arm and included baseline value as a covariate and fit to patients having an observed data point (irrespective of adherence to randomized treatment) at week 104.”
“Patients were randomly assigned in a double-blind manner and 1:1 ratio to receive once-weekly subcutaneous semaglutide 2.4 mg or placebo.”
“Eligible patients were aged ≥45 years, with a BMI of ≥27 kg m − 2 and established CVD defined as at least one of the following: prior myocardial infarction, prior ischemic or hemorrhagic stroke, or symptomatic peripheral artery disease.”
Sex is reported (72.3% male). Age and BMI are reported (mean age 61.6 years, BMI 33.3). Demographics include race and region. Health status is described via inclusion criteria (established CVD, no diabetes). Species/strain and housing conditions are not applicable for a human trial.
“The SELECT study enrolled 17,604 patients (72.3% male) from 41 countries between October 2018 and March 2021, with a mean (s.d.) age of 61.6 (8.9) years and BMI of 33.3 (5.0) kg m − 2”
“Of note, in the lower BMI categories (<30 kg m − 2 (overweight) and 30 to <35 kg m − 2 (class I obesity)), the proportion of Asian individuals was higher (14.5% and 7.4%, respectively) compared with the proportion of Asian individuals in the higher BMI categories (BMI 35 to <40 kg m − 2 (class II obesity; 3.8%) and ≥40 kg m − 2 (class III obesity; 2.2%), respectively).”
The paper states that the protocol was approved by national and institutional regulatory and ethical authorities in each participating country, and all patients provided written informed consent. This satisfies both IRB approval and informed consent requirements. Regulatory compliance is implied by adherence to ethical regulations.
“The protocol for SELECT was approved by national and institutional regulatory and ethical authorities in each participating country.”
“All patients provided written informed consent before beginning any trial-specific activity.”
“The protocol for SELECT was approved by national and institutional regulatory and ethical authorities in each participating country.”
“All patients provided written informed consent before beginning any trial-specific activity.”
Semaglutide is identified as once-weekly subcutaneous semaglutide 2.4 mg, with manufacturer (Novo Nordisk) implied. Dose and regimen are described. Statistical software is identified (SAS version 9.4). Other bench resources are not applicable.
“Patients were randomly assigned in a double-blind manner and 1:1 ratio to receive once-weekly subcutaneous semaglutide 2.4 mg or placebo.”
“All statistical analyses were performed with SAS software, version 9.4 TS1M5 (SAS Institute).”
“All statistical analyses were performed with SAS software, version 9.4 TS1M5 (SAS Institute).”
Statistical tests are named (ANCOVA, logistic regression). Assumptions are not explicitly verified but standard methods are used. Exact p-values are reported (e.g., 1.323762 × 10^-94). Effect sizes with confidence intervals are reported. Software is identified. Data presentation includes figures with error bars and per-group n. Mathematical plausibility checks were not performed due to large N and continuous outcomes.
“Exact P values are 1.323762 × 10 −94 and 9.80035 × 10 −100 for a and b , respectively.”
“treatment difference −8.7%; 95% CI −9.42 to −7.88; P < 0.0001”
“Continuous endpoints were analyzed using an analysis of covariance model with treatment as a fixed factor and baseline value of the endpoint as a covariate.”
“Continuous endpoints were analyzed using an analysis of covariance model with treatment as a fixed factor and baseline value of the endpoint as a covariate.”
“Exact P values are 1.323762 × 10 −94 and 9.80035 × 10 −100 for a and b , respectively.”
“treatment difference −8.7%; 95% CI −9.42 to −7.88; P < 0.0001”
The data availability statement provides a concrete route: data will be shared with bona fide researchers who submit a research proposal approved by an independent review board, with individual patient data shared in deidentified form. This is a managed-access mechanism, which is appropriate for patient-level data. No code is shared, but no bespoke code is mentioned.
“Data will be shared with bona fide researchers who submit a research proposal approved by the independent review board. Individual patient data will be shared in data sets in a deidentified and anonymized format. Information about data access request proposals can be found at https://www.novonordisk-trials.com/ .”
“Data will be shared with bona fide researchers who submit a research proposal approved by the independent review board. Individual patient data will be shared in data sets in a deidentified and anonymized format. Information about data access request proposals can be found at https://www.novonordisk-trials.com/ .”
The trial is registered (NCT03574597). Limitations are discussed in the Discussion. Conclusions are proportional to the evidence. Funding and competing interests are disclosed. Reporting guideline is not explicitly mentioned, but the paper follows CONSORT-like reporting for a trial.
“ClinicalTrials.gov identifier: NCT03574597 (https://clinicaltrials.gov/ct2/show/NCT03574597) .”
“The SELECT study has some limitations. First, SELECT was not a primary prevention trial, and the data should not be extrapolated to all individuals with overweight and obesity to prevent major adverse CV events.”
“D.H.R. declares having received consulting honoraria from Altimmune, Amgen, Biohaven, Boehringer Ingelheim, Calibrate, Carmot Therapeutics, CinRx, Eli Lilly, Epitomee, Gila Therapeutics, IFA Celtics, Novo Nordisk, Pfizer, Rhythm, Scientific Intake, Wondr Health and Zealand Pharma;”
“ClinicalTrials.gov identifier: NCT03574597 (https://clinicaltrials.gov/ct2/show/NCT03574597) .”
“The SELECT study has some limitations. First, SELECT was not a primary prevention trial, and the data should not be extrapolated to all individuals with overweight and obesity to prevent major adverse CV events.”
Registered (1 ID: ClinicalTrials.gov). No reporting guideline cited.
Broken references and links
None foundReferences checked against Crossref, OpenAlex and Retraction Watch for retractions and resolvability, plus declared data and code links probed for whether they resolve to content matching the paper.
Checked — nothing surfaced.
Checked 32 references by DOI: 23 verified — 9 no DOI (shown, not verified).
- NO DOIObesity and overweightNo DOI in the reference — shown for manual review; not independently verifiable (not a fabrication signal).
- NO DOIPharmacologic treatment of overweight and obesity in adultsNo DOI in the reference — shown for manual review; not independently verifiable (not a fabrication signal).
- NO DOIWegovy (semaglutide) summary of product characteristicsNo DOI in the reference — shown for manual review; not independently verifiable (not a fabrication signal).
- NO DOIWEGOVY (semaglutide) prescribing informationNo DOI in the reference — shown for manual review; not independently verifiable (not a fabrication signal).
- NO DOIOzempic (semaglutide) summary of product characteristicsNo DOI in the reference — shown for manual review; not independently verifiable (not a fabrication signal).
- NO DOIOZEMPIC (semaglutide) prescribing informationNo DOI in the reference — shown for manual review; not independently verifiable (not a fabrication signal).
- NO DOIThe Asia-Pacific perspective: redefining obesity and its treatmentNo DOI in the reference — shown for manual review; not independently verifiable (not a fabrication signal).
- NO DOIObesity EpidemiologyNo DOI in the reference — shown for manual review; not independently verifiable (not a fabrication signal).
- NO DOIObesity classificationNo DOI in the reference — shown for manual review; not independently verifiable (not a fabrication signal).
1 data/code link checked; 1 live.
- datahttps://www.novonordisk-trials.com/LIVEHTTP 200Resolves, but the content could not be matched to the paper.
Copyediting
3 minorWording, consistency and formatting errors that need correcting before submission.
No major wording or formatting errors. 3 minor suggestions below.
3 copyedit issues flagged: mostly consistency, typo.
- MINORtypoAbstract“waist-to-height ratio (−6.9%) versus placebo (−1.5%, −1.3 cm and −1.0%, respectively; P < 0.0001 for all comparisons versus placebo).”→ Ensure the list of placebo values aligns with the semaglutide values (weight, WC, WHtR).The sentence lists three semaglutide values but only two placebo values before the parenthetical; likely a formatting issue.
- MINORconsistencyResults, Safety and tolerability“Rates (events per 100 years of observation) of SAEs were 43.23, 43.54, 51.07 and 47.06 for semaglutide and 50.48, 49.66, 52.73 and 60.85 for placebo”→ Verify that the order of BMI categories matches the order in the text (e.g., <30, 30-<35, 35-<40, ≥40).The order is consistent with the text, but a reader might appreciate explicit labeling.
- MINORconsistencyResults, Safety and tolerability“Rates (events per 100 years of observation) of SAEs were 43.23, 43.54, 51.07 and 47.06 for semaglutide and 50.48, 49.66, 52.73 and 60.85 for placebo”→ Verify that the order of BMI categories matches the order in the text.The order of BMI categories is not explicitly stated in this sentence; ensure consistency.
The published work is robust and well-reported. An informed reader should weigh the minor reporting gaps (randomization method not detailed, no explicit reporting guideline statement, assumptions not explicitly verified) as minor limitations, but they do not undermine the validity of the findings. No erratum or correction appears warranted based on the checks performed.
- 1.MEDIUMreportingAdd an explicit statement in the Methods or Reporting Summary that the study adheres to the CONSORT reporting guideline.Both reviewers noted the absence of an explicit reporting guideline statement, which is a standard expectation for trial reports.
- 2.MEDIUMreportingIn the Methods, describe the randomization method (e.g., central randomization with a web-based system) or cite the primary publication where it is detailed.The randomization method is not detailed in this paper, and both reviewers flagged this as a minor reporting gap.
- 3.MEDIUMstatisticsAdd a brief note in the Statistical analysis section on verification of statistical assumptions (e.g., normality) or state that robust methods were used.Reviewer 1 flagged assumptions_verified as 'reported_but_inadequate'; explicitly addressing assumptions would strengthen the statistical reporting.
- 4.LOWreportingClarify in the Methods or Discussion that the trial was powered for the primary cardiovascular outcome, and that this secondary analysis is exploratory.Power analysis for secondary endpoints is not reported; clarifying the primary powering would preempt reader questions.
- 5.LOWdata codeConsider sharing the statistical analysis code in a public repository (e.g., GitHub) to enhance reproducibility, even though data are controlled access.Both reviewers suggested code sharing as an optional improvement; it would increase transparency.
- 6.LOWcopyeditIn the Abstract, ensure the list of placebo values aligns with the semaglutide values (weight, WC, WHtR) — the sentence lists three semaglutide values but only two placebo values.The copyedit pass flagged a potential formatting issue that could confuse readers.
- 7.LOWcopyeditIn the Results, Safety and tolerability section, explicitly label the order of BMI categories for the SAE rates to ensure consistency with the text.The copyedit pass noted that the order of BMI categories is not explicitly stated, which could lead to misinterpretation.
The star rating is the report’s one-glance summary. Every paper starts at 5★ and loses stars for the concrete problems the review finds — so a rating is never a vague average, it’s a running total you can read line by line under “How this rating was calculated.”
- Reporting — 8 dimensionseach dimension that fully fails−½★
- each dimension partially met−¼★
- Statistics · Integrity · Claimseach serious problem−1★
- each medium problem−½★
- Citationseach retracted or unverifiable reference−¼★
- Copyeditonly when the manuscript needs a full edit−½★
The rating never drops below 1★, and a demonstrable critical failure (an impossible statistic, a proven ethics violation) caps it at 1★ on its own — so the stars can never look healthy when the verdict is CRITICAL.
The rating draws on a panel of agents. Three independent Kaimen Rigor reviewers grade the eight dimensions below across several independent passes (the shown verdict is their majority vote — steadier than any single run), isolate the paper’s major claims and check its own evidence backs them, and flag integrity concerns. Alongside them, a citation agent resolves every reference against Crossref, OpenAlex, and Retraction Watch; a statistics agent recomputes reported tests; and rule-based checks verify that declared data/code links actually resolve. Full text is required — an abstract-only submission is not analyzed.
Graded against NIH, MDAR, ARRIVE 2.0, CONSORT, EQUATOR, and RRID guidelines. A dimension that doesn’t apply to the study type is skipped, never penalized.
This Kaimen Rigor review is model-assisted and is not a substitute for formal expert review. It complements human evaluation by surfacing potential methodological concerns — verify each finding against the source.