Allogeneic mesenchymal stem cell therapy with laromestrocel in mild Alzheimer's disease: a randomized controlled phase 2a trial.
Rash BG, Ramdas KN, Agafonova N, Naioti E, McClain-Moss L, Zainul Z, Varnado B, Peterson K, Brown M, Leal T, Kopcho S, Carballosa R, Patel P, Brody M, Herskowitz B, Fuquay A, Rodriguez S, Jacobson AF, Leon R, Pfeffer M, Schwartzbard JB, Botbyl J, Oliva AA Jr, Hare JM
- DOI
- 10.1038/s41591-025-03559-0
- Record issued
- 2026-08-15
- Engine
- 7.39.0
- Exported
- 2026-09-20
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/fac83dee-33fa-4bb2-8e17-aec4100e4db3 is authoritative.
How this rating was calculated
- StatisticsImpossible or misreported statistic ×2−2★
- IntegrityIntegrity concern ×3−1.5★
- ClaimsEfficacy rests on an unvalidated surrogate endpoint−0.5★
- ClaimsTreatment effect not shown to be clinically meaningful−0.5★
- ReportingData & code availability partially met−0.25★
A demonstrable critical failure caps the rating at the minimum, regardless of the deductions above.
- The numeric-impossibility checks (GRIM/GRIMMER/DEBIT/SPRITE) did not run: 15 reported means were read, and their group size is not stated where the values are printed (this source has no machine-readable table structure). These checks need the count the mean was averaged over, so none was performed.
- 01Printed percentage does not match its own countdemonstrable
41.7% does not match the reported count 5/13
“5 (41.7)”
Table 1 - 02Printed percentage does not match its own countdemonstrable
8.3% does not match the reported count 1/11
“1 (8.3)”
Table 1 - 03Efficacy rests on an unvalidated surrogate endpoint
The primary efficacy claim is based on a composite score (CADS) that includes left hippocampal volume (a surrogate), and the paper also reports improvements in brain volume and DTI measures (surrogates). The paper does not establish that these surrogates are validated predictors of clinical outcomes, and target engagement at the tested dose is not demonstrated (only a biomarker sTIE2 is mentioned, but not linked to efficacy).
“Laromestrocel improved clinical assessments at 39 weeks compared to placebo, as measured by a composite AD score (secondary end point was met: group 2 versus placebo change: 0.38; 95% CI −0.06–0.82), Montreal cognitive assessment and the Alzheimer’s Disease…”
- 04Treatment effect not shown to be clinically meaningful
The reported effect sizes are small fractions of the normal/reference values (e.g., 48.4% slowing of whole brain atrophy, 61.9% slowing of hippocampal atrophy) and are not anchored to a minimal clinically important difference. The composite score difference of 0.38 is small and the paper itself notes that effect sizes as small as 0.2–0.3 may be considered clinically meaningful, but this is not validated.
“At 39 weeks, Laromestrocel slowed the decline of whole brain volume compared to placebo (n = 10) by 48.4% for all treatment groups combined (groups 2–4: P = 0.005; n = 32) and left hippocampal volume by 61.9% (groups 2–4, P = 0.021; n = 32)”
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 transparently reported phase 2a randomized controlled trial. The main methodological strengths are rigorous randomization/blinding, detailed reporting of demographics and ethics, and proportional conclusions. The primary weakness is the vague data/code availability statement, which limits reproducibility.
Both reviewers independently scored all eight dimensions and agreed on every status; no divergence required reconciliation. The statistics verification component checked only 3 tests (limited coverage) and found no demonstrable errors; the 2 'inconsistent' results were not listed and thus not treated as errors. The citation check found no retracted or unresolved references.
Numerical inconsistencies
2 findings · worst criticalValues 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.
- Summary statistic impossible for the stated N (GRIM/GRIMMER)Recomputed
- Internal contradictions in the reported numbersAssessed
Recomputed 1 test: 1 consistent, 0 inconsistent; 1 via agent-written checks. 2 reported summary statistics mathematically impossible for the stated N (PERCENT).
- PERCENT41.7% does not match the reported count 5/13
“5 (41.7)”
Table 1 - PERCENT8.3% does not match the reported count 1/11
“1 (8.3)”
Table 1
- CONSISTENTreported p = .091 · recomputed p = .091Reviewer 2Check p-value for group 2 vs placebo CADS difference using reported CI.
“group 2 ( n = 11) versus placebo ( n = 10) difference 0.38, 95% CI −0.06–0.82; P = 0.091”
Taken as given: The difference is 0.38 with 95% CI -0.06 to 0.82.; The p-value is two-sided from a normal approximation.Method: Recomputed two-sided p from estimate and CI using normal approximation.How we recomputed it: pCI(0.38, -0.06, 0.82, 0)
- lowinternal contradictionThe abstract reports a 48.4% slowing of whole brain volume decline for combined groups 2-4, but the results section reports up to 57% for group 3. This is not contradictory but could be clarified.
“At 39 weeks, Laromestrocel slowed the decline of whole brain volume compared to placebo ( n = 10) by 48.4% for all treatment groups combined (groups 2–4: P = 0.005; n = 32)”
AbstractFind in source - lowinternal contradictionThe abstract states group 2 n=13, but the mITT population table shows group 2 n=12. This is explained by one withdrawal.
group 2 (25 million cells, one infusion followed by three monthly infusions of placebo, n = 13) ... The remaining 48 patients completed treatment with at least one post-baseline efficacy assessment and were analyzed for the secondary and exploratory end points as part of the modified ITT (mITT) population.
Table 1reviewer’s wording - lowinternal contradictionSecondary outcome analysis reports group 2 n=11, while mITT population is n=12 for group 2.
“group 2 ( n = 11) versus placebo ( n = 10) difference 0.38”
ResultsFind in source
Overstated conclusions
3 findings · 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
- Treatment effect not shown to be clinically meaningfulAssessed
- Conclusions only partially backed by the presented evidenceAssessed
7 major claims checked against the paper's own evidence: 3 only partially supported (evidence backs part of the claim; gaps or caveats remain); the rest adequately supported.
- partialReviewer 1Laromestrocel improves clinical assessments at 39 weeks compared to placebo.The secondary endpoint (CADS) showed a positive trend but did not reach the pre-specified significance level of P<0.1 for group 2 (P=0.091). Some exploratory endpoints (MoCA, ADCS-ADL) reached significance.Evidence: CADS difference 0.38 (95% CI -0.06-0.82, P=0.091); MoCA group 2 P=0.009; ADCS-ADL group 4 P=0.04.
“Laromestrocel improved clinical assessments at 39 weeks compared to placebo, as measured by a composite AD score (secondary end point was met: group 2 versus placebo change: 0.38; 95% CI −0.06–0.82)”
AbstractFind in source - partialReviewers 1, 2Laromestrocel may reduce cognitive decline and brain atrophy within 9 months.The evidence shows signals of efficacy but the study was not powered for efficacy and some endpoints were not significant.Evidence: Secondary endpoint CADS did not reach significance; exploratory endpoints showed mixed results.
“Single and multiple intravenous infusions with laromestrocel, an allogeneic cell therapy, is safe in patients with mild AD and may reduce the impact of cognitive decline and brain atrophy within 9 months of treatment.”
AbstractFind in source - partialReviewer 2Laromestrocel improved clinical assessments at 39 weeks compared to placebo.Some secondary endpoints (CADS, MoCA, ADCS-ADL) showed improvement, but not all, and the study was not powered for efficacy.Evidence: Secondary endpoint CADS difference 0.38 (95% CI -0.06-0.82, P=0.091), MoCA P=0.009 for group 2.
“Laromestrocel improved clinical assessments at 39 weeks compared to placebo, as measured by a composite AD score”
AbstractFind in source - supportedReviewers 1, 2Laromestrocel is safe in patients with mild AD.The primary endpoint of safety was met with no significant difference in TE-SAEs between groups.Evidence: Primary endpoint results: TE-SAE rates within 4 weeks were similar across groups, with overlapping CIs.
“The study met its primary end point of safety; the rate of treatment-emergent serious adverse events within 4 weeks of any infusion was similar in all four groups”
AbstractFind in source - supportedReviewer 1Laromestrocel slows brain atrophy and reduces neuroinflammation.Volumetric MRI showed significant slowing of whole brain and hippocampal atrophy in combined treatment groups, and DTI showed reduced neuroinflammation in the cingulate cortex.Evidence: Whole brain volume decline slowed by 48.4% (P=0.005); left hippocampal volume by 61.9% (P=0.021); DTI MD reduced in cingulate cortex (group 2 P=0.048).
“At 39 weeks, Laromestrocel slowed the decline of whole brain volume compared to placebo ( n = 10) by 48.4% for all treatment groups combined (groups 2–4: P = 0.005; n = 32) and left hippocampal volume by 61.9% (groups 2–4, P = 0.021; n = 32), and reduced neuroinflammation as measured by diffusion tensor imaging.”
AbstractFind in source - supportedReviewers 1, 2Reduced brain atrophy correlates with improved cognitive scores.Pearson correlation showed significant correlations between hippocampal volume change and MMSE-2 change.Evidence: R=0.41, P=0.0075 for bilateral hippocampal volume vs MMSE-2.
“The change in bilateral hippocampal volume ( R = 0.41; P = 0.008; n = 42), whole brain volume ( R = 0.35; P = 0.023; n = 42) and bilateral lateral ventricle volume ( R = −0.35; P = 0.0213; n = 42), all correlated significantly with change in MMSE-2 scores”
ResultsFind in source - supportedReviewer 2Laromestrocel slowed brain atrophy and reduced neuroinflammation.Volumetric MRI and DTI showed significant slowing of whole brain and hippocampal atrophy and reduced MD in cingulate cortex.Evidence: Whole brain volume decline slowed by 48.4% (P=0.005), left hippocampal by 61.9% (P=0.021), DTI MD reduced in group 2 (P=0.048).
“Laromestrocel slowed the decline of whole brain volume compared to placebo ( n = 10) by 48.4% for all treatment groups combined (groups 2–4: P = 0.005; n = 32) and left hippocampal volume by 61.9%”
AbstractFind in source
Premise concern: surrogate not validated for clinical benefit; effect size not shown to be clinically meaningful.
- INADEQUATESurrogate endpointThe primary efficacy claim is based on a composite score (CADS) that includes left hippocampal volume (a surrogate), and the paper also reports improvements in brain volume and DTI measures (surrogates). The paper does not establish that these surrogates are validated predictors of clinical outcomes, and target engagement at the tested dose is not demonstrated (only a biomarker sTIE2 is mentioned, but not linked to efficacy).
“Laromestrocel improved clinical assessments at 39 weeks compared to placebo, as measured by a composite AD score (secondary end point was met: group 2 versus placebo change: 0.38; 95% CI −0.06–0.82), Montreal cognitive assessment and the Alzheimer’s Disease Cooperative Study Activities of Daily Living. At 39 weeks, Laromestrocel slowed the decline of whole brain volume compared to placebo (n = 10) by 48.4% for all treatment groups combined (groups 2–4: P = 0.005; n = 32) and left hippocampal volume by 61.9% (groups 2–4, P = 0.021; n = 32), and reduced neuroinflammation as measured by diffusion tensor imaging.”
- INADEQUATEEffect sizeThe reported effect sizes are small fractions of the normal/reference values (e.g., 48.4% slowing of whole brain atrophy, 61.9% slowing of hippocampal atrophy) and are not anchored to a minimal clinically important difference. The composite score difference of 0.38 is small and the paper itself notes that effect sizes as small as 0.2–0.3 may be considered clinically meaningful, but this is not validated.
“At 39 weeks, Laromestrocel slowed the decline of whole brain volume compared to placebo (n = 10) by 48.4% for all treatment groups combined (groups 2–4: P = 0.005; n = 32) and left hippocampal volume by 61.9% (groups 2–4, P = 0.021; n = 32)”
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
1 finding · worst mediumRequired detail the manuscript never states — study design, biological variables, ethics approval and consent, key resources, statistical reporting, data and code availability, and overall transparency.
- Data/code availability incompleteAssessed
The introduction cites prior research on AD, anti-amyloid antibody limitations (lecanemab, aducanumab, donanemab), and the rationale for MSC therapy. It acknowledges the limitations of prior work (e.g., ARIAs with anti-amyloid drugs) and builds a logical rationale for laromestrocel. The hypothesis follows from the cited evidence.
Randomization method (block sizes of four, stratified by sex) and unit (patient) are reported. Blinding is described for all personnel. Power analysis is reported but acknowledges the study was not powered for the secondary endpoint. Inclusion/exclusion criteria are detailed. Outlier handling is not explicitly described but the analysis population (ITT/mITT) is defined. Controls (placebo) are appropriate. Independent replication is not applicable for a single trial.
“Randomization was performed electronically using block sizes of four, equally allocated, with each site receiving separate blocks. Randomization was additionally stratified by sex.”
“All study patients, caregivers and all personnel involved in the conduct and interpretation of the study, including investigators, site personnel and sponsor staff, were blinded to treatment group.”
“The study was not powered to detect a statistically significant difference in CADS.”
“Randomization was performed electronically using block sizes of four, equally allocated, with each site receiving separate blocks. Randomization was additionally stratified by sex.”
“All study patients, caregivers and all personnel involved in the conduct and interpretation of the study, including investigators, site personnel and sponsor staff, were blinded to treatment group.”
“The study was not powered to detect a statistically significant difference in CADS.”
Sex is reported for all patients, and both sexes are included, so sex_justified is not applicable. Age, weight, BMI, and health status (AD severity) are reported. Demographics include race, ethnicity, and ApoE status. Species/strain and housing are not applicable for a human trial.
“with a lower proportion of male (44.9%) versus female (55.1%) patients.”
“Age, years, mean (s.d.) | 76.7 (4.40) | 74.9 (7.51) | 70.5 (7.14) | 74.5 (6.07) | 74.1 (6.65)”
“Eligible patients had a clinical diagnosis of mild AD in accordance with the National Institute of Aging and the Alzheimer’s Association criteria at the time of enrollment.”
“Age, years, mean (s.d.) | 76.7 (4.40) | 74.9 (7.51) | 70.5 (7.14) | 74.5 (6.07) | 74.1 (6.65)”
“with a lower proportion of male (44.9%) versus female (55.1%) patients.”
The study reports IRB approval by Western-Copernicus Group IRB with a date. Informed consent is described as written. Regulatory compliance is implied through FDA IND and 21 CFR 1271 for cell products. All applicable criteria are adequate.
“Institutional Review Board (IRB) approval was granted by Western-Copernicus Group Institutional Review Board (WCG IRB) on 24 November 2021.”
“All patients provided written consent on the Western IRB-approved protocol.”
“All bone-marrow collections are regulated under 21 CFR 1271, Human Cells, Tissues, and Cellular and Tissue-Based Products (HCT/Ps).”
“Institutional Review Board (IRB) approval was granted by Western-Copernicus Group Institutional Review Board (WCG IRB) on 24 November 2021.”
“All patients provided written consent on the Western IRB-approved protocol.”
“All bone-marrow collections are regulated under 21 CFR 1271, Human Cells, Tissues, and Cellular and Tissue-Based Products (HCT/Ps).”
Laromestrocel is described with manufacturing details, donor characteristics, and quality control. Antibodies for flow cytometry are identified with catalog numbers and clones. Software (SAS, FlowJo, etc.) is identified. Cell line authentication and mycoplasma testing are not applicable as this is a human trial, but mycoplasma testing is mentioned for the product.
“CD105 PE (cat. no. 560839, clone 266, lots 9155633 and 8137932, 5 μl)”
“Laromestrocel must be negative for Mycoplasma”
“CD105 PE (cat. no. 560839, clone 266, lots 9155633 and 8137932, 5 μl)”
“Laromestrocel must be negative for Mycoplasma”
“All statistical analyses were conducted using SAS software (v.9.4).”
Tests are named (MMRM, Clopper-Pearson, Pearson correlation). Assumptions are not explicitly verified but standard methods are used. Exact p-values are reported for many outcomes, but some are reported as thresholds (e.g., P < 0.1). Effect sizes with CIs are reported for the primary and secondary endpoints. Software (SAS v.9.4) is identified. Data presentation includes per-group n and error bars. Mathematical plausibility checks were not possible for most statistics due to lack of raw data.
“Change from baseline in CADS was statistically analyzed with an MMRM analysis.”
“group 2 ( n = 11) versus placebo ( n = 10) difference 0.38, 95% CI −0.06–0.82; P = 0.091”
“pre-specified P < 0.1 as a positive outcome due to a small sample size”
“Change from baseline in CADS was statistically analyzed with an MMRM analysis.”
“group 2 ( n = 11) versus placebo ( n = 10) difference 0.38, 95% CI −0.06–0.82; P = 0.091”
“if the two-sided P value for the difference was <0.1, then the difference was considered statistically significant”
The data availability statement says data may be shared under a data-use agreement for IRB-approved research, but does not specify a platform or timeframe. Code availability states SAS code is on file and can be made available under a code use agreement, but no repository is provided. No accession numbers are given.
“All requests for data access should be sent to the corresponding authors, B.G.R. or J.M.H. The minimum dataset, without individual patient data, used for the primary, secondary and exploratory conclusions, may be shared under a data-use agreement for IRB-approved research.”
“SAS code for the minimum dataset is on file and can be made available under a code use agreement for IRB-approved research.”
“All requests for data access should be sent to the corresponding authors, B.G.R. or J.M.H. The minimum dataset, without individual patient data, used for the primary, secondary and exploratory conclusions, may be shared under a data-use agreement for IRB-approved research.”
“SAS code for the minimum dataset is on file and can be made available under a code use agreement for IRB-approved research.”
Methods are detailed enough for replication. Trial registration number is provided. Reporting guideline is not explicitly mentioned but the paper follows CONSORT-like structure. All outcomes are reported, including negative results. Limitations are discussed. Conclusions are proportional. Funding and COI are disclosed.
“ClinicalTrials.gov registration: NCT05233774”
“This study had limitations that warrant mentioning. First, there was a relatively small sample size with a high proportion of patients of Hispanic ethnicity and some variables such as patient education level were not ascertained.”
“This trial was fully funded by Longeveron.”
“ClinicalTrials.gov registration: NCT05233774”
“This study had limitations that warrant mentioning. First, there was a relatively small sample size with a high proportion of patients of Hispanic ethnicity and some variables such as patient education level were not ascertained.”
“This trial was fully funded by Longeveron.”
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 66 references by DOI: 66 verified.
Every extracted reference resolved against Crossref/OpenAlex with no retraction flags.
2 data/code links checked; 2 live.
- datahttps://www.clinicaltrials.gov/study/NCT05233774LIVEHTTP 200Resolves, but the content could not be matched to the paper.
- datahttps://clinicaltrials.gov/ct2/show/NCT05233774LIVEHTTP 200Resolves, but the content could not be matched to the paper.
Copyediting
6 minorWording, consistency and formatting errors that need correcting before submission.
No major wording or formatting errors. 6 minor suggestions below.
6 copyedit issues flagged: mostly consistency, clarity, grammar.
- MINORconsistencyAbstract“group 2 (25 million cells, one infusion followed by three monthly infusions of placebo, n = 13)”→ Ensure consistent use of 'million' vs 'M' notation.Inconsistent use of 'million' and 'M' across the manuscript.
- MINORclarityResults, Secondary outcomes“group 2 ( n = 11) versus placebo ( n = 10) difference 0.38, 95% CI −0.06–0.82; P = 0.091”→ Clarify that the difference is in CADS score change from baseline.The sentence could be clearer about what the difference refers to.
- MINORgrammarDiscussion“Laromestrocel is not expected to directly target β-amyloid or p-Tau, but instead is aimed at the neurovascular and inflammatory components of AD and related disorders.”→ Consider rephrasing for clarity.Minor grammatical issue.
- MINORconsistencyAbstract“group 2 (25 million cells, one infusion followed by three monthly infusions of placebo, n = 13)”→ Ensure group 2 n is consistent with mITT population (n=12) elsewhere.Abstract reports n=13 for group 2, but Results state mITT n=12 for group 2.
- MINORtypoResults, Secondary outcomes“group 2 ( n = 11) versus placebo ( n = 10) difference 0.38, 95% CI −0.06–0.82; P = 0.091”→ Check if n=11 is correct for group 2 in mITT.Potential inconsistency in group 2 sample size across sections.
- MINORclarityMethods, Statistical analysis“assuming a two-sided α of 0.1, the power to detect a 50% slowing of disease progression based on CADS comparing three active arms combined ( n = 36) versus placebo ( n = 12) was 36% at 6 months.”→ Clarify the power calculation basis and whether it was post-hoc.Power calculation is described but may be confusing.
The published work is robust and generally trustworthy, but an informed reader should weigh the limited statistical verification coverage and the vague data/code availability. No erratum is warranted based on the evidence; however, the authors should consider clarifying the data-sharing mechanism and reporting exact p-values to improve transparency.
- 1.CRITICALstatisticsCorrect or explain the statistically impossible value: PERCENT: 41.7% does not match the reported count 5/13Demonstrable critical failure — blocks the verdict from passing.
- 2.CRITICALstatisticsCorrect or explain the statistically impossible value: PERCENT: 8.3% does not match the reported count 1/11Demonstrable critical failure — blocks the verdict from passing.
- 3.HIGHdata codeIn the Data availability section, specify a concrete data-sharing mechanism (e.g., a named repository like Vivli or YODA, or a data-access committee) with conditions and a timeline for access.The current 'available on request' statement is vague and does not meet modern reproducibility expectations.
- 4.HIGHdata codeIn the Code availability section, deposit the SAS analysis code in a public repository (e.g., GitHub, Zenodo) with a DOI, or provide a detailed code description.Sharing code is essential for reproducibility and is currently only available on request.
- 5.MEDIUMreportingIn the Methods or a reporting summary, explicitly state adherence to a reporting guideline such as CONSORT.Explicitly naming the reporting guideline improves transparency and reader confidence.
- 6.MEDIUMstatisticsIn the Statistical analysis section, report exact p-values for all outcomes instead of thresholds like 'P < 0.1' where possible.Exact p-values allow readers to assess evidence strength more precisely.
- 7.MEDIUMstatisticsIn the Statistical analysis section, describe how outliers were handled and clarify missing data imputation methods.Outlier handling and missing data methods are currently not described, which is a reporting gap.
- 8.MEDIUMstatisticsIn the Statistical analysis section, add a statement on assumptions verification (e.g., normality, homogeneity of variance) or justify why they are not needed.Assumption verification is not explicitly reported, which is a common reviewer concern.
- 9.MEDIUMreportingIn the Abstract and Results, reconcile the group 2 sample size (n=13 in abstract vs n=12 in mITT table vs n=11 in secondary analysis) and clarify the reasons for each discrepancy.Inconsistent sample sizes across sections could confuse readers and undermine trust.
- 10.MEDIUMreportingIn the Results, clarify that the reported difference (e.g., 0.38) refers to the change from baseline in CADS score.The current phrasing is ambiguous about what the difference represents.
- 11.LOWcopyeditIn the Abstract, ensure consistent use of 'million' vs 'M' notation for cell doses.Inconsistent notation is a minor copyedit issue that could be confusing.
- 12.LOWcopyeditIn the Discussion, rephrase the sentence about laromestrocel's mechanism for clarity.Minor grammatical issue that affects readability.
- 13.LOWreportingIn the Methods, clarify whether the power calculation was post-hoc and describe the basis for the 50% slowing assumption.The power calculation description is confusing and could be misinterpreted.
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.