Neoadjuvant pembrolizumab plus chemotherapy followed by adjuvant pembrolizumab compared with neoadjuvant chemotherapy alone in patients with early-stage non-small-cell lung cancer (KEYNOTE-671): a randomised, double-blind, placebo-controlled, phase 3 trial.
Spicer JD, Garassino MC, Wakelee H, Liberman M, Kato T, Tsuboi M, Lee SH, Chen KN, Dooms C, Majem M, Eigendorff E, Martinengo GL, Bylicki O, Rodríguez-Abreu D, Chaft JE, Novello S, Yang J, Arunachalam A, Keller SM, Samkari A, Gao S, KEYNOTE-671 Investigators
- DOI
- 10.1016/S0140-6736(24)01756-2
- 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/65a22d28-9e49-4211-899e-016ecbab2eb6 is authoritative.
How this rating was calculated
- IntegrityIntegrity concern ×2−1★
- The numeric-impossibility checks (GRIM/GRIMMER/DEBIT/SPRITE) did not run: 2 reported means were read, and their group size is not stated where the values are printed. These checks need the count the mean was averaged over, so none was performed.
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 rigorously designed and reported phase 3 randomized controlled trial. The methodology is sound, with clear randomization, blinding, power analysis, and comprehensive reporting of demographics, ethics, resources, statistics, and data availability. Minor reporting gaps include lack of explicit reporting guideline adherence and a few copyedit issues.
Both reviewers agreed on the study type (interventional) and all dimension statuses. The paper is a full-text clinical trial report. The statistics verification component checked only 3 tests (all consistent); other statistics were not machine-verified. The citation check found no retracted or unresolved references.
Numerical inconsistencies
1 finding · worst lowValues 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.
- Internal contradictions in the reported numbersAssessed
Recomputed 3 tests: 3 consistent, 0 inconsistent; 3 via agent-written checks.
- CONSISTENTreported p = .005 · recomputed p = .011Reviewer 1Overall survival hazard ratio p-value from CI
“hazard ratio 0·72; [95% CI 0·56–0·93]; one-sided P=0·0052”
Taken as given: The reported HR is 0.72 with 95% CI 0.56-0.93.; The p-value is one-sided, so the two-tailed p from pCI is halved.; The CI is a 95% confidence interval for the hazard ratio.Method: Compute two-tailed p from HR and CI using pCI, then halve for one-sided.How we recomputed it: pCI(0.72, 0.56, 0.93, 1) - CONSISTENTreported p < .001 · recomputed p = <.001Reviewers 1, 2Event-free survival hazard ratio p-value from CI
“Median event-free survival was 47·2 months (95% CI 32·9 to not reached) in the pembrolizumab group and 18·3 months (95% CI 14·8–22·1) in the placebo group (hazard ratio 0·59 [95% CI 0·48–0·72; ).”
Taken as given: The reported HR is 0.59 with 95% CI 0.48-0.72.; The p-value is one-sided and reported as <0.00001.; The CI is a 95% confidence interval for the hazard ratio.Method: Compute two-tailed p from HR and CI using pCI, then halve for one-sided.How we recomputed it: pCI(0.59, 0.48, 0.72, 1) - CONSISTENTreported p < .010 · recomputed p = .011Reviewer 2Recompute the two-sided p-value for the overall survival hazard ratio from the reported 95% CI.
“hazard ratio 0·72; [95% CI 0·56–0·93]; one-sided P=0·0052”
Taken as given: The reported HR is 0.72 with 95% CI 0.56-0.93.; The CI is a 95% confidence interval for the hazard ratio.; The p-value is two-sided, derived from the CI.Method: Used pCI function to compute two-sided p from HR and 95% CI on log scale.How we recomputed it: pCI(0.72, 0.56, 0.93, 1)
- lowinternal contradictionThe abstract reports 36-month overall survival estimates of 71% and 64%, while the results section reports the same. No contradiction found.
“36-month overall survival estimates were 71% (95% CI 66–76) in the pembrolizumab group and 64% (58–69) in the placebo group”
AbstractFind in source
Overstated conclusions
None foundConclusions 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.
Checked — nothing surfaced.
6 major claims checked against the paper's own evidence: all adequately supported.
- supportedReviewers 1, 2Perioperative pembrolizumab significantly improves overall survival compared with neoadjuvant chemotherapy alone.The claim is supported by the primary endpoint result with HR 0.72 and one-sided p=0.0052, which met the pre-specified threshold.Evidence: Overall survival HR 0.72 (95% CI 0.56-0.93), one-sided p=0.0052, threshold p=0.0054.
“The significant overall survival benefit of pembrolizumab plus chemotherapy as neoadjuvant therapy followed by adjuvant pembrolizumab compared with neoadjuvant chemotherapy alone coupled with a manageable safety profile support the use of perioperative pembrolizumab in patients with resectable, early-stage NSCLC.”
AbstractFind in source - supportedReviewers 1, 2Perioperative pembrolizumab improves event-free survival.The claim is supported by the event-free survival result with HR 0.59 and a highly significant p-value.Evidence: Event-free survival HR 0.59 (95% CI 0.48-0.72), one-sided p<0.00001.
“Median event-free survival was 47·2 months (95% CI 32·9 to not reached) in the pembrolizumab group and 18·3 months (95% CI 14·8–22·1) in the placebo group (hazard ratio 0·59 [95% CI 0·48–0·72; ).”
Results ¶4Find in source - supportedReviewers 1, 2Perioperative pembrolizumab does not negatively impact health-related quality of life.The claim is supported by the health-related quality of life analyses showing no significant between-group differences in GHS/QoL and other domains.Evidence: Least squares mean changes in GHS/QoL were not significantly different between groups at week 11 and week 10.
“Least squares mean changes from baseline in the GHS/QoL score were −9·3 points (95% CI −11·7 to −6·9) in the pembrolizumab group and −10·7 points (95% CI −13·1 to −8·4) in the placebo group at week 11 of the neoadjuvant phase (difference 1·4 points [95% CI −1·6 to 4·5]; )”
Results ¶6Find in source - supportedReviewers 1, 2The safety profile of perioperative pembrolizumab is manageable.The claim is supported by the safety data showing a higher rate of grade 3-5 treatment-related adverse events in the pembrolizumab group but no new safety signals.Evidence: Grade 3-5 treatment-related adverse events occurred in 45% vs 38% of participants; treatment-related deaths were 1% in both groups.
“In the as-treated population, grade 3–5 treatment-related adverse events occurred in 179 (45%) of 396 participants in the pembrolizumab group and in 151 (38%) of 399 participants in the placebo group.”
AbstractFind in source - supportedReviewer 1KEYNOTE-671 is the first trial to show a statistically significant overall survival benefit for perioperative immunotherapy in resectable NSCLC.The claim is supported by the trial's positive overall survival result and the context of other trials not yet showing this.Evidence: The trial met its overall survival endpoint; the discussion states it is the first to show this.
“To the best of our knowledge, KEYNOTE-671 is the first of these trials to show a statistically significant overall survival benefit.”
Discussion ¶4Find in source - supportedReviewer 2Perioperative pembrolizumab is a standard-of-care treatment option for resectable NSCLC.The claim is supported by the significant overall survival benefit and manageable safety, though it is a clinical recommendation based on the trial results.Evidence: The trial met its primary endpoint of overall survival with a significant improvement.
“These findings support perioperative pembrolizumab as a standard-of-care treatment option for patients with resectable stage II to IIIB (N2) NSCLC.”
DiscussionFind in source
Efficacy claim is anchored to an adequate endpoint and a meaningful effect.
- ADEQUATESurrogate endpointThe primary efficacy claim is based on overall survival, a hard clinical outcome, with a statistically significant improvement (HR 0.72, p=0.0052). Event-free survival is also a clinical outcome. No surrogate biomarker is used as the primary basis for the efficacy claim.
“The significant overall survival benefit of pembrolizumab plus chemotherapy as neoadjuvant therapy followed by adjuvant pembrolizumab compared with neoadjuvant chemotherapy alone coupled with a manageable safety profile support the use of perioperative pembrolizumab in patients with resectable, early-stage NSCLC.”
- ADEQUATEEffect sizeThe effect size is anchored to a hard clinical outcome: 36-month overall survival improved from 64% to 71% (HR 0.72, 95% CI 0.56–0.93), and median event-free survival improved from 18.3 to 47.2 months. These are clinically meaningful differences in survival outcomes.
“36-month overall survival estimates were 71% (95% CI 66–76) in the pembrolizumab group and 64% (58–69) in the placebo group (hazard ratio 0·72; [95% CI 0·56–0·93]; one-sided P=0·0052)”
Data authenticity concerns
1 finding · worst lowAn 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.
- Other integrity concernAssessed
2 integrity concerns flagged (0 high).
- lowotherThe paper reports that 797 of 1364 screened were randomized, but the trial profile (Figure 1) is not fully described in the text; the numbers appear consistent.
“From April 2018 through December 2021, 797 of 1364 participants screened were randomised”
Results ¶1Find in source
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 historical trials establishing the benefit of neoadjuvant chemotherapy, recent phase 3 trials of PD-1/PD-L1 inhibitors, and identifies the gap that no regimen has shown an overall survival benefit in the intention-to-treat population. The rationale for the perioperative approach is clearly linked to the study objectives. Limitations of prior research are implicitly addressed by the trial design, though not explicitly framed as addressing prior limitations.
“No neoadjuvant or perioperative treatment regimen has shown a significant overall survival benefit in the last 30 years.”
“The ultimate objective for patients with resectable NSCLC is to prolong survival without compromising health-related quality of life.”
“No neoadjuvant or perioperative treatment regimen has shown a significant overall survival benefit in the last 30 years.”
“The KEYNOTE-671 trial is evaluating a perioperative approach of neoadjuvant pembrolizumab plus cisplatin-based chemotherapy followed by surgical resection and adjuvant pembrolizumab for patients with resectable stage II or III NSCLC.”
Randomization was centralized with stratification and block design; blinding was double-blind with masking of participants, investigators, and sponsor personnel. Power analysis was provided with specific assumptions. Inclusion/exclusion criteria were pre-specified and published. The analysis populations (ITT, as-treated, PRO) are clearly defined.
“Randomisation was done centrally using an interactive response technology system and was stratified by disease stage, PD-L1 expression, histology, and geographic region in blocks of four.”
“Participants, investigators, and sponsor personnel were masked to treatment assignments; local pharmacists were unmasked to support treatment preparation.”
“With the occurrence of 386 deaths and four interim analyses and a final analysis, the study has 90% power to detect a difference in overall survival of HR=0·7 at one-sided alpha=0·0148.”
“Randomisation was done using an interactive response system (Almac Clinical Technologies, Souderton, PA, USA) and a participant randomisation list generated by the sponsor and was stratified by disease stage (II vs III), PD-L1 tumour proportion score (TPS; <50% vs ≥50%), tumour histology (squamous vs nonsquamous), and geographic region (east Asia vs other).”
“Participants, investigators, site staff, and sponsor personnel involved in study treatment administration or clinical evaluation were masked to treatment assignments.”
“With the occurrence of 416 event-free survival events and two analyses, the study has 90·1% power to detect a difference in event-free survival of HR=0·7 at one-sided alpha=0·01.”
Sex, age, race, and performance status are reported in Table 1. Both sexes are enrolled, so sex justification is not applicable. Age and health status (ECOG) are reported. Demographics are comprehensive. Species/strain and housing conditions are not applicable for a human trial.
“As previously reported, baseline demographic and disease characteristics were balanced between the two groups ().”
“Female | 118 (30%) | 116 (29%) | | Male | 279 (70%) | 284 (71%)”
“Female | 118 (30%) | 116 (29%) | | Male | 279 (70%) | 284 (71%)”
“Age, years | 63 (58–69) | 64 (58–70)”
“Race | | American Indian or Alaska Native | 1 (<1%) | 0 | | Asian | 124 (31%) | 125 (31%)”
The protocol was approved by the appropriate ethics body for each participating centre, and all participants provided written informed consent. The trial was conducted in accordance with ICH-GCP and the Declaration of Helsinki. Regulatory compliance is stated.
“The trial protocol and all amendments, which included changes that affected trial design (summarised in the Document History section of the protocol []), were approved by the appropriate ethics body for each participating centre (–).”
“All participants provided written, informed consent.”
“The trial was conducted in accordance with the protocol, the International Council for Harmonisation Good Clinical Practice guidelines, the Declaration of Helsinki ethical principles, and all local regulations.”
“The trial protocol and all amendments, which included changes that affected trial design (summarised in the Document History section of the protocol []), were approved by the appropriate ethics body for each participating centre (–).”
“All participants provided written, informed consent.”
“The trial was conducted in accordance with the protocol, the International Council for Harmonisation Good Clinical Practice guidelines, the Declaration of Helsinki ethical principles, and all local regulations.”
Pembrolizumab is identified as the investigational product with dose and regimen. The PD-L1 IHC assay is identified with vendor and catalog. Chemotherapy agents are named with doses. Software tools (SAS, R) are identified with versions. Antibodies, cell lines, and mycoplasma testing are not applicable as this is a clinical trial without wet-lab assays.
“The neoadjuvant phase comprised four cycles of pembrolizumab 200 mg or placebo (normal saline) given intravenously once every 3 weeks”
“PD-L1 expression in tumour tissue was assessed at a central laboratory during screening (Covance, Indianapolis, IN, USA) using PD-L1 IHC 22C3 pharmDx (Agilent Technologies; Carpinteria, CA, USA).”
“Statistical analyses were done using SAS (version 9.4).”
“The neoadjuvant phase comprised four cycles of pembrolizumab 200 mg or placebo (normal saline) given intravenously once every 3 weeks in combination with cisplatin 75 mg/m 2 given intravenously once every 3 weeks”
“PD-L1 expression in tumour tissue was assessed at a central laboratory during screening (Covance, Indianapolis, IN, USA) using PD-L1 IHC 22C3 pharmDx (Agilent Technologies; Carpinteria, CA, USA).”
“Statistical analyses were done using SAS (version 9.4).”
All statistical tests are named (stratified log-rank, Cox regression, constrained longitudinal data analysis, Miettinen and Nurminen). Assumptions are addressed (proportional hazards sensitivity analysis). Exact p-values are reported (e.g., one-sided P=0.0052). Effect sizes with 95% CIs are reported throughout. Software is identified. Data presentation follows clinical trial conventions (Kaplan-Meier curves, forest plots, CONSORT-like flow). Mathematical plausibility is not applicable for large-N continuous outcomes.
“hazard ratio 0·72; [95% CI 0·56–0·93]; one-sided P=0·0052”
“Between-group differences in overall survival in the overall population were assessed using the stratified log-rank test”
“Median overall survival and the boundaries of the 95% confidence interval were not reached in the pembrolizumab group and were 52·4 months (95% CI 45·7 to not reached) in the placebo group (HR 0·72 [95% CI 0·56–0·93]; one-sided p=0.0052).”
“hazard ratio 0·72; [95% CI 0·56–0·93]; one-sided P=0·0052”
“Median overall survival and the boundaries of the 95% confidence interval were not reached in the pembrolizumab group and were 52·4 months (95% CI 45·7 to not reached) in the placebo group (HR 0·72 [95% CI 0·56–0·93]; one-sided p=0.0052).”
“Between-group differences in overall survival in the overall population were assessed using the stratified log-rank test”
The data sharing statement describes a concrete process for requesting anonymized data through the MSD data sharing website, with conditions and timeframe. Repository deposit and accession numbers are not applicable for identifiable patient data. Code sharing is not applicable as no bespoke code was used.
“The MSD data sharing website (available at: http://engagezone.msd.com/ds_documentation.php ) outlines the process and requirements for submitting a data request.”
“The MSD data sharing website (available at: http://engagezone.msd.com/ds_documentation.php ) outlines the process and requirements for submitting a data request.”
“Data will be made available for request after product approval in the US and EU or after product development is discontinued.”
The trial is registered (NCT03425643). Methods are detailed. Limitations are discussed. Conclusions are proportional to the evidence. Funding and conflicts of interest are disclosed. No reporting guideline checklist is explicitly mentioned, but the paper follows CONSORT-like structure.
“This study is registered at ClinicalTrials.gov (http://ClinicalTrials.gov) , NCT03425643”
“Limitations of the KEYNOTE-671 trial, as well as the other trials of perioperative immune checkpoint inhibition for resectable NSCLC, – include a design that does not permit direct analysis of the relative contributions of the neoadjuvant and adjuvant components of the treatment regimen”
“Funding: Merck Sharp & Dohme LLC, a subsidiary of Merck & Co., Inc., Rahway, NJ, USA.”
“This study is registered at ClinicalTrials.gov (http://ClinicalTrials.gov) , NCT03425643 (https://clinicaltrials.gov/ct2/show/NCT03425643) .”
“Limitations of the KEYNOTE-671 trial, as well as the other trials of perioperative immune checkpoint inhibition for resectable NSCLC, – include a design that does not permit direct analysis of the relative contributions of the neoadjuvant and adjuvant components of the treatment regimen”
“Funding: Merck Sharp & Dohme LLC, a subsidiary of Merck & Co., Inc., Rahway, NJ, USA.”
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 26 references by DOI: 19 verified — 7 no DOI (shown, not verified).
- NO DOITNM Classification of Malignant TumorsNo DOI in the reference — shown for manual review; not independently verifiable (not a fabrication signal).
- NO DOIVP1–2024: RATIONALE-315: Event-free survival (EFS) and overall survival (OS) of neoadjuvant tislelizumab (TIS) plus chemotherapy (CT) with adjuvant TIS in resectable non-small cell lung cancer (NSCLC)No DOI in the reference — shown for manual review; not independently verifiable (not a fabrication signal).
- NO DOINivolumab (NIVO) plus platinum-doublet chemotherapy (chemo) versus chemo as neoadjuvant treatment for resectable non-small cell lung cancer (NSCLC): Health-related quality of life (HRQoL) outcomes from CheckMate 816No DOI in the reference — shown for manual review; not independently verifiable (not a fabrication signal).
- NO DOIHealth-related quality of life (HRQoL) outcomes from CheckMate 77T: neoadjuvant nivolumab (nivo) plus chemotherapy (chemo) followed by adjvuant nivo in resectable NSCLCNo DOI in the reference — shown for manual review; not independently verifiable (not a fabrication signal).
- NO DOIMultiple testing in group sequential trials using graphical approachesNo DOI in the reference — shown for manual review; not independently verifiable (not a fabrication signal).
- NO DOIFDA approves neoadjuvant/adjvuant pembrolizumab for resectable non-small cell lung cancerNo DOI in the reference — shown for manual review; not independently verifiable (not a fabrication signal).
- NO DOIKeytruda - opinion on variation to marketing authorisationNo DOI in the reference — shown for manual review; not independently verifiable (not a fabrication signal).
1 data/code link checked; 1 live.
- datahttp://engagezone.msd.com/ds_documentation.phpLIVEHTTP 200Resolves, but the content could not be matched to the paper.
Copyediting
7 minorWording, consistency and formatting errors that need correcting before submission.
No major wording or formatting errors. 7 minor suggestions below.
7 copyedit issues flagged: mostly consistency, clarity, typo.
- MINORconsistencyAbstract, Methods“Randomisation was done centrally using an interactive response technology system”→ Consider using 'randomization' consistently throughout the manuscript (American spelling) or 'randomisation' (British spelling).The manuscript uses both 'randomisation' and 'randomization' in different sections.
- MINORclarityMethods, Outcomes“The dual primary endpoints were event-free survival, defined as the time from randomisation to the first occurrence of local progression that precluded the planned surgery, unresectable tumour at the time of surgery, progression or recurrence per RECIST version 1.1 by the investigator’s assessment, or death from any cause, and overall survival, defined as the time from randomization to death from any cause.”→ Consider breaking this long sentence into two for clarity.The sentence is long and complex, making it difficult to parse.
- MINORconsistencyResults, paragraph 2“Among the 396 participants who received ≥1 administration of neoadjuvant pembrolizumab plus chemotherapy, 295 (74%) received all four administrations of neoadjuvant pembrolizumab, 325 (82%) underwent in-study surgery, 290 (73%) received ≥1 administration of adjuvant pembrolizumab, and 191 (48%) completed the regimen”→ Ensure that the percentages are consistent with the counts (e.g., 295/396 = 74.5%, which rounds to 74% or 75%).Minor rounding inconsistencies may exist; verify all percentages.
- MINORconsistencyAbstract, Methods“Randomisation was done centrally using an interactive response technology system”→ Consider using 'interactive response technology system' consistently with the Methods section where it is called 'interactive response system'.Minor terminology inconsistency.
- MINORtypoMethods, Study design and participants“Complete eligibility criteria haven been published”→ Change to 'have been published'.Typographical error.
- MINORconsistencyResults, Health-related quality of life“388 (98%) of 395 participants in the pembrolizumab group and 391 (98%) of 398 participants in the placebo completed the EORTC QLQ-C30 at baseline”→ Add 'group' after 'placebo' for parallel structure.Minor grammatical inconsistency.
- MINORclarityDiscussion, paragraph 5“Notably, use of subsequent therapy, including subsequent PD-1 or PD-L1 inhibitors, was similar With one additional year of follow-up”→ Capitalize 'with' and consider rephrasing for clarity.Capitalization and flow issue.
The published work is robust and well-reported. An informed reader should weigh the minor reporting gaps (no explicit CONSORT statement, minor copyedit issues) and the fact that only a subset of statistical tests were independently verified. No erratum or correction appears warranted based on the available evidence.
- 1.MEDIUMreportingAdd an explicit statement in the Methods or a separate section that the trial was reported in accordance with the CONSORT guideline.Both reviewers noted the absence of an explicit reporting guideline statement, which is a standard expectation for clinical trials.
- 2.MEDIUMreportingIn the Introduction, add a brief paragraph explicitly discussing limitations of prior studies and how the current design addresses them.Reviewer 2 noted that limitations of prior research are not explicitly discussed, which would strengthen the scientific premise.
- 3.MEDIUMdata codeIn the Data Sharing section, clarify the expected timeline for data availability after product approval to meet the 'timeframe' requirement more explicitly.Reviewer 2 suggested clarifying the timeline for data access, which would enhance transparency.
- 4.MEDIUMreportingConsider including a statement about the availability of the statistical analysis plan or protocol in a public repository.Both reviewers suggested this to enhance transparency and reproducibility.
- 5.LOWcopyeditStandardize the spelling of 'randomisation' vs 'randomization' throughout the manuscript.The copyedit pass flagged inconsistent spelling between sections.
- 6.LOWcopyeditBreak the long sentence defining the dual primary endpoints in the Methods, Outcomes section into two sentences for clarity.The copyedit pass noted the sentence is long and complex, making it difficult to parse.
- 7.LOWcopyeditVerify the percentages in the Results, paragraph 2 (e.g., 295/396 = 74.5%) and ensure they are rounded consistently.The copyedit pass flagged potential rounding inconsistencies.
- 8.LOWcopyeditUse consistent terminology for the randomization system: 'interactive response technology system' vs 'interactive response system'.The copyedit pass noted terminology inconsistency between the Abstract and Methods.
- 9.LOWcopyeditFix the typo 'haven been published' to 'have been published' in the Methods, Study design and participants section.The copyedit pass flagged a typographical error.
- 10.LOWcopyeditAdd 'group' after 'placebo' in the Results, Health-related quality of life section for parallel structure.The copyedit pass noted a grammatical inconsistency.
- 11.LOWcopyeditCapitalize 'with' and rephrase the sentence in the Discussion, paragraph 5 for clarity.The copyedit pass flagged a capitalization and flow issue.
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.