BCMA-directed mRNA CAR-T cell therapy for myasthenia gravis: exploratory biomarker analysis of a placebo-controlled phase 2b trial.
Fedak RR, Ruggerie RN, Shan Y, Curvino EJ, de Sousa JF, Daniel S, Ngo-Casi M, Kamboh H, Vu T, Durmuş H, Mozaffar T, Howard JF Jr, English EP, Benson A, Duvernay MT, Singer MS, Kalayoglu MV, Brunn C, Bodansky A, Anderson MS, DeRisi JL, Garcia ST, Yu DJL, Zorn KC, Kurtoglu M, Miljković MD, Stewart CA, Jewell CM, MG-001 Study Team
- DOI
- 10.1038/s41591-025-04170-z
- Record issued
- 2026-08-10
- Engine
- 7.29.0
- 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/deac0a18-ff86-4d05-86a8-a437b68b3c50 is authoritative.
How this rating was calculated
- IntegrityIntegrity concern−0.5★
- ReportingBiological variables not met−0.5★
- ReportingKey resources not met−0.5★
- ReportingStudy design partially met−0.25★
- ReportingEthical approvals partially met−0.25★
- ReportingData & code availability partially met−0.25★
- LinksDead data/code link−0.25★
- References were not verified against Crossref/OpenAlex.
- Statistics were not checked: no recomputable values were found in this text — no test statistic reported with its degrees of freedom, no effect estimate printed with both a 95% CI and a p-value, and no percentage printed with both its count and its denominator.
- 01Biological variables not reported
Biological variables for the human participants and the preclinical mouse model are largely not reported in this manuscript.
“We next assessed functionality in a humanized xenograft model in which immunocompromised NSG mice are engrafted with human malignant PCs.”
ResultsFind in source - 02Key resources not identified
Key biological and chemical resources are incompletely identified: antibodies, cell lines, mycoplasma status, and animal sources are not reported, and the investigational product is named but not fully specified.
“Flow cytometric analysis of absolute change in BCMA expression in plasmablasts at month 1”
Figure 3Find in source - 03Declared data/code link does not resolve
Dead link — nothing to verify.
“https://www.protocols.io/view/derisi-lab-phage-immunoprecipitation-sequencing-ph-czw7x7hn?step=14.1”
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.
The paper presents a well-motivated exploratory biomarker analysis of a phase 2b RCT, with strong scientific premise, transparent reporting of null results, and adequate statistical methods. However, it has significant reporting gaps in biological variables (no demographics), ethics approvals (no IRB/IACUC statements), key resources (missing antibody/cell line details), and data/code availability (no accession numbers or code).
Two independent reviewer runs were synthesized; they agreed on most dimensions but diverged on key resources (fail vs warn), which was resolved to warn based on the checklist percentage. The statistics component checked 0 tests (none recomputable), so statistical correctness is not independently verified. The citation component found no retracted or missing 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
- lowinternal contradictionThe abstract states 'n = 10/15' for the 66.7% responder rate, which is arithmeticially consistent (10/15 = 66.7%); the placebo responder rate of 27.3% is reported without a denominator in this text, so it cannot be verified here but is not contradictory on its face.
In 66.7% of patients ( n = 10/15), transient targeting of BCMA with Descartes-08 ... resulted in durable clinical efficacy.
Abstractreviewer’s wording
Overstated conclusions
1 finding · worst lowConclusions 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.
- Conclusions only partially backed by the presented evidenceAssessed
11 major claims checked against the paper's own evidence: all adequately supported.
- partialReviewer 1Descartes-08 selectively retunes the autoreactive antibody repertoire.The PhIP-Seq data show increased turnover of the autoreactome, but the authors themselves note that the assay does not reveal the specific nature or selectivity of the change, so 'selectively retunes' goes beyond what is directly shown.Evidence: PhIP-Seq autoreactome correlation decreased in DC-08 at month 3 (median r=0.72 vs placebo 0.89, P=0.034) and persisted; no significant change in anti-AChR titer was observed.
“Descartes-08 selectively retunes the autoreactive antibody repertoire”
ResultsFind in source - partialReviewers 1, 2The findings unveil a new type of immune reset and support BCMA-targeted RNA cell therapies for autoimmune disease.The correlative biomarker data support a precision immunomodulatory effect, but 'new type of immune reset' is a mechanistic interpretation not directly tested; the authors acknowledge alternative mechanisms and the need for further study.Evidence: The paper presents on-target pharmacodynamic, cytokine, autoreactome, transcriptomic, and TCR results, and the Discussion explicitly states that potential mechanisms 'warrant future study'.
“Our findings unveil a new type of immune reset and support the development of BCMA-targeted RNA cell therapies as a more accessible therapy for autoimmune diseases.”
AbstractFind in source - supportedReviewer 1Descartes-08 resulted in durable clinical efficacy in 66.7% of patients (10/15) after outpatient dosing without lymphodepletion.The paper reports the primary endpoint result and durable responses, referring to the companion trial paper for full clinical details, so the claim is supported as a statement of the trial outcome.Evidence: Main text states the phase 2b RCT met its primary endpoint (66.7% vs 27.3%, P=0.0472) and that responses were durable through 12 months.
“In 66.7% of patients ( n = 10/15), transient targeting of BCMA with Descartes-08 administered in an outpatient setting without lymphodepletion resulted in durable clinical efficacy.”
AbstractFind in source - supportedReviewer 1Descartes-08 preferentially reduces BCMA-high plasma cells/plasmablasts and activated pDCs.Flow cytometry and single-cell data directly show reductions in BCMA expression on plasmablasts and CD86 on pDCs, as well as responder-specific reductions in these populations.Evidence: Fig. 3e-f shows reduced plasmablast BCMA at month 3 (P=0.035) and reduced pDC CD86 at month 1 (P<0.001); Fig. 5 shows on-target transcriptional changes.
“Descartes-08 preferentially reduces BCMA hi PCs and activated pDCs”
ResultsFind in source - supportedReviewer 1Descartes-08 effects occurred without immune suppression, indicated by preserved immunoglobulins and vaccine titers.The presented flow cytometry and titer data show no significant declines in broad immune populations, immunoglobulins, or vaccine titers.Evidence: Fig. 4a-m reports no significant changes in B/T/NK/monocyte frequencies, IgG/IgA/IgM, or vaccine titers at month 3 and month 12.
“These effects occurred without immune suppression, indicated by the lack of decline in vaccine-specific antibodies or hypogammaglobulinemia.”
AbstractFind in source - supportedReviewer 1Descartes-08 remodels the TCR repertoire without dramatic changes to circulating T cell phenotypes.TCR sequencing shows clonal expansion/contraction in the Descartes-08 cohort, while flow cytometry shows only minor T cell subset changes (a decrease in Th2 at month 1), supporting the claim.Evidence: Fig. 6a-d shows expanded/contracted TCR clonotypes in DC-08 vs placebo; Fig. 6e-m shows only a Th2 decrease (P=0.014) among the assessed subsets.
“Descartes-08 remodels the TCR repertoire without dramatic changes to circulating T cell phenotypes”
ResultsFind in source - supportedReviewer 2Transient targeting of BCMA with Descartes-08 resulted in durable clinical efficacy in 66.7% of patients (10/15).The clinical efficacy result is presented as from the phase 2b RCT and is consistent with the companion paper; the biomarker data in this paper do not contradict it.Evidence: The clinical efficacy endpoint (MGC improvement at month 3, 66.7% vs 27.3%, P=0.0472) is reported in the Main text and detailed in the companion clinical paper.
“In 66.7% of patients ( n = 10/15), transient targeting of BCMA with Descartes-08 administered in an outpatient setting without lymphodepletion resulted in durable clinical efficacy.”
AbstractFind in source - supportedReviewer 2Descartes-08 produces a precision retuning of self-reactivity, increasing pro-immune function and decreasing activity of BCMA+ plasma cells and pDCs and reducing disease-associated cytokines such as IL-6.The on-target reductions in BCMA/CD86 and IL-6 are directly measured, and the pro-immune enrichment is shown in bulk/scRNA-seq; the evidence backs the claim.Evidence: Flow cytometry showing reduced BCMA on plasmablasts (P=0.035) and reduced CD86 on pDCs (P<0.001); Olink cytokine panel showing reduced IL-6 (P=0.019); GSEA showing pro-immune enrichment in responders.
“Comparison of Descartes-08-treated ( n ≤ 19) and placebo ( n ≤ 15) cohorts by flow cytometry, serum profiling, multiplexing cytokine analysis and bulk/single-cell transcriptional analysis reveals a precision retuning of self-reactivity demonstrated by increased pro-immune function, decreased activity of BCMA + plasma cells and plasmacytoid dendritic cells and reductions in disease-associated cytokines, such as IL-6.”
AbstractFind in source - supportedReviewer 2Descartes-08 effects occur without immune suppression, as indicated by lack of decline in vaccine-specific antibodies or hypogammaglobulinemia.The preservation of immunoglobulins and vaccine titers is directly measured and reported as null results, supporting the claim.Evidence: Fig. 4 shows preserved B cells, T cells, NK cells, monocytes, immunoglobulins (IgG/IgA/IgM), and anti-measles/mumps/rubella/diphtheria/tetanus vaccine titers at month 3 and month 12.
“These effects occurred without immune suppression, indicated by the lack of decline in vaccine-specific antibodies or hypogammaglobulinemia.”
AbstractFind in source - supportedReviewer 2Antibody and T cell receptor analysis reveals altered circulating repertoires of self-reactive antibodies and T cell clones among Descartes-08 participants.Both the autoreactome turnover and TCR repertoire remodeling are directly measured with statistical comparisons versus placebo.Evidence: PhIP-Seq autoreactome correlation (r median 0.72 vs 0.89 at month 3, P=0.034) and TCRβ clonal expansion/contraction analysis (Fig. 6).
“Furthermore, antibody and T cell receptor analysis revealed altered circulating repertoires of self-reactive antibodies and T cell clones among Descartes-08 participants.”
AbstractFind in source - supportedReviewer 2Clinical response correlates with enrichment of pro-immune gene pathways important for functional T cell immunity.The GSEA enrichment in responders relative to non-responders is directly reported, supporting the correlation claim.Evidence: Bulk RNA-seq and scRNA-seq GSEA showing enrichment of IFNα, IFNγ, TNF/NF-κB, and IL-6/JAK/STAT3 pathways in responders (n=9 bulk; n=5 scRNA) versus non-responders.
Furthermore, clinical response correlates with enrichment of pro-immune gene pathways that are important in functional T cell immunity.
Discussionreviewer’s wording
Efficacy claim is anchored to an adequate endpoint and a meaningful effect.
- ADEQUATESurrogate endpointThe primary efficacy claim is based on the MG Composite (MGC) score, a validated clinical/functional scale for myasthenia gravis, not a surrogate biomarker. The prespecified threshold of a ≥5-point improvement is a clinically meaningful outcome.
“The trial met the primary endpoint, with 66.7% of participants randomized to Descartes-08 achieving at least a five-point improvement in the MG Composite (MGC) score at month 3 compared to 27.3% of those randomized to placebo (P = 0.0472).”
- ADEQUATEEffect sizeThe effect size is reported as a responder rate of 66.7% for Descartes-08 versus 27.3% for placebo, a 39.4 percentage-point absolute difference, with a statistically significant P value and a prespecified clinically meaningful threshold of ≥5-point MGC improvement.
“66.7% of participants randomized to Descartes-08 achieving at least a five-point improvement in the MG Composite (MGC) score at month 3 compared to 27.3% of those randomized to placebo (P = 0.0472).”
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
5 findings · worst highRequired detail the manuscript never states — study design, biological variables, ethics approval and consent, key resources, statistical reporting, data and code availability, and overall transparency.
- Biological variables not reportedAssessed
- Key resources not identifiedAssessed
- Data/code availability incompleteAssessed
- Ethics/consent reporting incompleteAssessed
- Study-design details incomplete (controls, blinding, power)Assessed
The introduction cites approved CAR-T limitations and prior open-label MG data, and explicitly motivates RNA-based transient CAR-T as addressing those limitations. The rationale links the premise to the study objectives, and limitations of prior approaches are explicitly addressed (inpatient administration, lymphodepletion, toxicity, immunosuppression).
“Engineering of cells using RNA offers distinct advantages toward this goal by exploiting transient CAR expression.”
“This set of features creates a biologic-like approach that involves an initial leukapheresis but is otherwise similar with respect to patient experience.”
“We recently reported potent and long-lasting clinical effects in a multicenter, open-label study of Descartes-08 after six once-weekly infusions in generalized myasthenia gravis (MG)”
“The high levels of BCMA expression on these cells and lack of BCMA on other cells, including most B cells, suggest new opportunities to control autoimmunity without the broad immunosuppression associated with targeting ubiquitous B cell antigens”
For a human RCT, the applicable design elements are randomization method, randomization unit, blinding, power analysis, inclusion/exclusion, and outlier/analysis-population handling. Blinding ('double-blind') and the unit (participant) are stated, but no randomization method, a priori power calculation, pre-specified eligibility criteria, or ITT/per-protocol/missing-data approach are reported. This yields roughly one-third of applicable elements adequate, warranting a warning rather than a pass.
“Here and in a companion paper , we report the results of a phase 2b, double-blind, randomized controlled trial (RCT) in patients with MG.”
“Patients underwent a single leukapheresis to collect cells used to prepare Descartes-08”
“we report prespecified exploratory analyses from a successful placebo-controlled, double-blind, randomized phase 2b trial”
For the human trial, sex, age, weight/health status, and demographics are not reported. For the preclinical NSG mouse xenograft component, the strain is named (NSG) but age, sex, weight, and housing conditions are not described, and the source/genetic background is not given. The manuscript therefore fails this dimension.
“We next assessed functionality in a humanized xenograft model in which immunocompromised NSG mice are engrafted with human malignant PCs.”
A ClinicalTrials.gov registration number is given, but this does not substitute for an ethics approval statement. The manuscript does not name an approving IRB/ethics committee, describe informed consent (or a waiver), state IACUC approval for the mouse xenograft work, or cite a regulatory compliance framework. Because applicable ethics elements are absent, the dimension is warned rather than passed.
“ClinicalTrials.gov identifier: NCT04146051 (http://clinicaltrials.gov/study/NCT04146051) .”
Flow cytometry antibodies are used without vendor/catalog/clone information; the MM.1S cell line is named without source or STR authentication; mycoplasma testing is not mentioned; NSG mice are named without vendor/source; and the Descartes-08 product is described functionally but dose/formulation are not given. Software tools are well identified, but that is only one of six applicable criteria.
“Flow cytometric analysis of absolute change in BCMA expression in plasmablasts at month 1”
“In vitro cytotoxicity assay using HD-derived DC-08 or unmodified control CD8 + T cells and MM.1S-GFP PC target cells”
“All analyses were conducted in R (version 4.3+) using Bioconductor (version 3.17+).”
“an autologous, RNA-encoded anti-B cell maturation antigen (BCMA) CAR-T cell therapy”
“All analyses were conducted in R (version 4.3+) using Bioconductor (version 3.17+)”
All main statistical tests are named (t-tests, repeated-measures ANOVA, linear mixed-effects model, DESeq2, GSEA, chi-squared test), software versions are given, and data are shown with individual points and box-whisker plots. Assumptions are only partially addressed (log transformation is stated for non-normal data, but normality testing is not described), and effect sizes with confidence intervals are not consistently reported for the t-test comparisons, though sccomp results are given with 95% credible intervals. No arithmetic impossibilities were detected, so the dimension passes.
“comparisons of placebo to Descartes-08 were performed using a two-sided unpaired t -test, and comparisons within each treatment group were performed using a two-sided paired t -test.”
“Box-and-whisker plots show the median, interquartile range and full range of data, along with individual data points.”
“Unless indicated otherwise, non-normally distributed data were log transformed before performing statistical analyses.”
“comparisons of placebo to Descartes-08 were performed using a two-sided unpaired t -test”
“with 66.7% of participants randomized to Descartes-08 achieving at least a five-point improvement in the MG Composite (MGC) score at month 3 compared to 27.3% of those randomized to placebo ( P = 0.0472)”
The data availability statement is detailed and names a managed-access procedure for patient-level data with a contact and conditions, which is appropriate for clinical trial data. However, the sequencing data (RNA-seq, scRNA-seq, TCR-seq) are not deposited with accession numbers, custom analysis code is not shared, and the source data are described as provided without a persistent repository identifier. Thus only one of four applicable criteria is fully adequate.
“Access to anonymized, individual and trial-level data (analysis datasets) and/or the study protocol will be provided upon request from qualified researchers performing independent, rigorous research, after review and approval of a research proposal and statistical analysis plan and execution of a data sharing agreement.”
“All raw data used to generate the figures are provided as source data, excluding patient-specific data.”
“Access to anonymized, individual and trial-level data (analysis datasets) and/or the study protocol will be provided upon request from qualified researchers performing independent, rigorous research, after review and approval of a research proposal and statistical analysis plan and execution of a data sharing agreement.”
The trial is registered with ClinicalTrials.gov (NCT04146051), a Nature reporting summary is referenced, and methods for the omics and immunophenotyping analyses are detailed. Negative/null results are reported (preserved immunoglobulins and vaccine titers, no change in anti-AChR titer, no change in global immune compartment). Limitations are explicitly discussed (e.g., sample size, target cells outside circulation, anti-AChR not a tracking biomarker), conclusions are generally proportional to the data, and funding (NIH grants) plus competing interests are disclosed.
“ClinicalTrials.gov identifier: NCT04146051”
“no significant changes were observed between placebo and Descartes-08 when assessing broad immune populations”
“A caveat to these data is that, although we measured large changes in the autoreactome, the technique does not reveal the specific nature of the change.”
“ClinicalTrials.gov identifier: NCT04146051 (http://clinicaltrials.gov/study/NCT04146051)”
“This work was supported by Cartesian Therapeutics and National Institute of Neurological Disorders and Stroke, National Institutes of Health grants 1R44NS115426 and 1R44NS137943.”
Registered (1 ID: ClinicalTrials.gov). No reporting guideline cited.
Broken references and links
1 finding · worst mediumReferences 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.
- Dead data/code linksRecomputed
3 data/code links checked; 2 live, 1 dead.
- datahttp://clinicaltrials.gov/study/NCT04146051LIVEHTTP 200Resolves, but the content could not be matched to the paper.
- datahttps://www.biorender.com/LIVEHTTP 200Resolves, but the content could not be matched to the paper.
- dataprotocols.ioDEADHTTP 404https://www.protocols.io/view/derisi-lab-phage-immunoprecipitation-sequencing-ph-czw7x7hn?step=14.1Dead link — nothing to verify.
Copyediting
9 minorWording, consistency and formatting errors that need correcting before submission.
No major wording or formatting errors. 9 minor suggestions below.
9 copyedit issues flagged: mostly typo, consistency, clarity.
- MINORtypoExtended Data Fig. 1 title“Quantification of cytokine levels following coincubation coincubation of Descartes-08 and target cells”→ Remove the duplicated word: 'following coincubation of Descartes-08 and target cells'.Duplicated word in figure title.
- MINORconsistencyFig. 5 legend“cDC1, conventional type 1 dendritic cell; cDC2, conventional type 1 dendritic cell”→ Change the second entry to 'cDC2, conventional type 2 dendritic cell'.cDC2 is mislabeled as type 1.
- MINORotherAuthor contributions, Extended data“A full list of members and their affiliations appears in .”→ Insert the reference to the supplementary author list or the relevant table.Incomplete cross-reference.
- MINORgrammarData Availability“Requests can be submitted to trials@cartesiantx.com. are provided with this paper.”→ Remove the orphan clause 'are provided with this paper' or integrate it into the preceding sentence.Sentence fragment at the end of the data availability statement.
- MINORtypoExtended Data Fig. 1 caption“Quantification of cytokine levels following coincubation coincubation of Descartes-08 and target cells.”→ Remove the duplicated word 'coincubation'.Duplicated word in figure caption.
- MINORconsistencyFig. 5 legend“cDC1, conventional type 1 dendritic cell; cDC2, conventional type 1 dendritic cell”→ Change the cDC2 definition to 'conventional type 2 dendritic cell'.cDC1 and cDC2 are both defined as 'type 1'.
- MINORtypoCompeting interests“sponsored by Alexion/AstraZeneca, argenx, Amgen, Cartesians, COUR, Dianthus”→ Change 'Cartesians' to 'Cartesian'.Company name misspelled.
- MINORclarityReferences, ref. 1“N. Engl. J. Med. 379 , 64–73 (2018.”→ Add the missing closing parenthesis: '(2018).'Truncated reference formatting.
- MINORclarityCompeting interests (T.M.)“and the NI.”→ Complete the truncated sentence with the intended funding/body.Sentence ends mid-word.
The published work is moderately robust but has notable reporting gaps that an informed reader should weigh, particularly the absence of participant demographics, ethics approval statements, and key resource identifiers. These gaps do not necessarily invalidate the findings but warrant caution and would benefit from a correction or supplementary documentation.
- 1.HIGHethicsAdd a named IRB/ethics committee approval statement with protocol number and a description of informed consent (written or waived) to the Declarations/Compliance section.The paper reports human-subjects research without any ethics approval or consent statement, which is a serious reporting gap that could undermine trust in the study's conduct.
- 2.HIGHethicsAdd an IACUC approval statement and institution for the NSG mouse xenograft studies.The paper includes animal experiments but no animal welfare approval is reported, which is required for publication and reader confidence.
- 3.HIGHreportingReport participant demographics (sex, age, health status, race/ethnicity, comorbidities) in a baseline table, even if deferred to the companion clinical paper.Biological variables are entirely unreported for human participants, which is a major gap for an interventional study and limits assessment of generalizability.
- 4.HIGHreportingState the randomization method (e.g., block/stratified, allocation concealment) and a priori sample-size/power calculation, or explicitly reference the companion paper where these are reported.The trial design is incompletely reported in this paper, and readers need to know how randomization was performed and whether the study was adequately powered.
- 5.HIGHdata codeDeposit sequencing data (bulk RNA-seq, scRNA-seq, TCR-seq) in a public repository with accession numbers (GEO/SRA/ENA) in the Data availability section.The paper generates substantial omics data but provides no accession numbers, preventing independent verification and reuse.
- 6.HIGHdata codeShare the bespoke analysis code (R scripts for DESeq2/fgsea/sccomp) in a version-controlled public repository with a DOI in the Data/code availability section.Custom analysis code is not shared, which limits reproducibility of the reported analyses.
- 7.HIGHrigorIdentify flow-cytometry antibodies by vendor, clone/RRID, and dilution in the Methods; currently only antigen names are given.Antibody identifiers are essential for reproducibility of flow cytometry experiments.
- 8.HIGHrigorAdd mycoplasma testing status and cell-line authentication (STR) for MM.1S, and a source for the NSG mice, in the relevant Methods/preclinical sections.Cell line authentication and mycoplasma testing are standard requirements for preclinical experiments, and their absence raises concerns about experimental validity.
- 9.MEDIUMreportingDescribe the prespecified analysis population (ITT vs per-protocol) and missing-data/imputation approach for the biomarker analyses in the Methods.The paper does not state how missing data were handled or which population was analyzed, which is important for interpreting biomarker results.
- 10.MEDIUMstatisticsAdd a versioned statement of the effect sizes with confidence intervals for the key cytokine/flow comparisons to strengthen reporting beyond p-values.Effect sizes with CIs are only partially reported, and adding them would improve interpretability and meta-analytic use.
- 11.MEDIUMreportingReconcile the number of randomized participants (36) with the analyzed active/placebo cohorts (n≤19 and n≤15) by stating exclusions or missing samples.The discrepancy between randomized and analyzed numbers is not explained, which could confuse readers about the analysis population.
- 12.MEDIUMreportingReport the sex, age, weight, and housing conditions of the NSG mice in the preclinical xenograft methods.Animal characteristics are not reported, which is a gap for the preclinical component.
- 13.LOWcopyeditFix the duplicated word 'coincubation' in Extended Data Fig. 1 title and caption.Typographical error in figure title/caption.
- 14.LOWcopyeditCorrect the cDC2 definition in Fig. 5 legend from 'conventional type 1 dendritic cell' to 'conventional type 2 dendritic cell'.Mislabeling of cDC2 as type 1 is a factual error in the legend.
- 15.LOWcopyeditFix the incomplete cross-reference in Author contributions (Extended data) by inserting the reference to the supplementary author list.Incomplete cross-reference leaves a dangling reference.
- 16.LOWcopyeditRemove the orphan clause 'are provided with this paper' in the Data Availability statement.Sentence fragment at the end of the data availability statement.
- 17.LOWcopyeditCorrect the misspelling 'Cartesians' to 'Cartesian' in the Competing interests section.Company name misspelled.
- 18.LOWcopyeditAdd the missing closing parenthesis in reference 1: '(2018).'Truncated reference formatting.
- 19.LOWcopyeditComplete the truncated sentence 'and the NI.' in the Competing interests (T.M.) section.Sentence ends mid-word, likely a typo or truncation.
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.