Alpha¹ by KaimenConfidential

Kaimen Rigor review

Anti-CD38 monoclonal antibody CM313 for primary immune thrombocytopenia: multicentre, randomised, placebo controlled, phase 2 trial.

Chen Y, Xu Y, Dai J, Sun T, Li H, Hua Z, Zhou Z, Zhou H, Yan Z, Zhao X, Xue F, Liu W, Liu X, Fu R, Wang W, Chi Y, Dong H, Ju M, Dai X, Gu W, Pei X, Yang R, Zhang L

DOI: 10.1136/bmj-2025-084314
Record issued: 2026-08-10
Engine: 7.29.0
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/2ee2d639-5c88-4b78-9f54-7f98c02bc24f is authoritative.

Kaimen Rigor review

Kaimen Rigor reviewed

4 / 5

Strong — minor issues to address

Interventional Study

Critical

High-priority

How this rating was calculated

Started at 5★ — deductions

ClaimsEfficacy rests on an unvalidated surrogate endpoint−0.5★
ReportingData & code availability not met−0.5★

Model-reported confidence: highReviewer agreement: moderateEngine: 7.29.0Source: full text

Not verified in this report

References were not verified against Crossref/OpenAlex.
The numeric-impossibility checks (GRIM/GRIMMER/DEBIT/SPRITE) did not run on this paper: the pass that reads its reported means did not complete. No reported mean was checked for arithmetic impossibility.

Biggest issues

01Efficacy rests on an unvalidated surrogate endpoint
highOverreachreviewer-assessedFull detail
The primary efficacy endpoint is the overall response rate, defined by platelet count thresholds (≥30×10^9/L with doubling) and absence of bleeding. Platelet count is a surrogate biomarker for clinical bleeding outcomes. Although the authors report secondary bleeding outcomes and show target engagement (depletion of antibody-secreting cells and immunoglobulin reduction), they do not cite validated evidence linking the platelet-count surrogate to hard clinical outcomes such as major bleeding or mortality. The claim of clinical benefit rests primarily on this surrogate measure.
“The primary outcome was overall response rate (at least two consecutive platelet counts ≥30×10^9 /L, a minimum doubling from baseline, and no bleeding) at week 8.”
Find in source
02Data and code not shared
highDisclosurereviewer-assessedFull detail
No data or code availability statement is provided, and no repository deposit or accession numbers are reported.
“Data availability statement The code used to analyse”
End of manuscriptFind in source

Findings by category

01ArithmeticNone found 202OverreachOverstated conclusions 03AuthenticityNone found 104DisclosureReporting gaps 05AvailabilityNone found 06Copyedit7 minor suggestions

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-designed, well-reported randomised placebo-controlled phase 2 trial with a strong scientific premise, clear ethics approval, sound design, and internally consistent statistics (7/7 recomputed tests consistent). The main weaknesses are an incomplete data/code availability statement and a primary-outcome p-value reported only as a threshold (P<0.001), plus minor reporting gaps (no CONSORT reference, unlabelled ethnicity, missing vendor for the autoantibody kit).

Three independent reviewer runs of the same model were synthesised; reviewers agreed on 6 of 8 statuses and diverged on statistical analysis (pass vs warn) and data code availability (warn vs fail), reconciled here per the scoring rules. The statistics verification recomputed only 7 tests/effect estimates (those with a test statistic + df or effect + CI); threshold-only and exact p-values were not machine-verified. Citation, reproducibility, and claim-audit components found no integrity flags.

Numerical inconsistencies

None found

Values 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 7 tests: 7 consistent, 0 inconsistent; 7 via agent-written checks.

Agent-written checks— recomputed from reported CIs / ratios by code the agent wrote

CONSISTENTreported p < .001 · recomputed p = <.001Reviewers 1, 2, 3
Primary outcome overall response rate at week 8, CM313 vs placebo (Fisher's exact)
“At week 8, the overall response rate was higher in the CM313 group (83%; 25/30) compared with placebo group (20%; 3/15): difference 63.3% (95% confidence interval 33.7% to 81.3%; P<0.001).”
Find in source
Taken as given: The 25 and 3 are the event counts and 30 and 15 the group totals of the CM313 and placebo arms respectively; The comparison is a 2×2 table so df=1; Fisher's exact test is two-sided with no mid-p correction
Method: Two-sided Fisher's exact test on the cell counts (25,5;3,12), p ≈ 0.0001, consistent with P<0.001.
How we recomputed it: pFisher2x2(25,5,3,12,0)
CONSISTENTreported p < .001 · recomputed p = <.001Reviewers 1, 2, 3
Complete response rate within 8 weeks, CM313 vs placebo (Fisher's exact)
“The complete response rate within eight weeks was also higher in the CM313 group (73%; 22/30) compared with placebo group (7%; 1/15), with a difference of 66.7% (95% CI 38.8% to 82.0%; P<0.001).”
Find in source
Taken as given: The 22 and 1 are the event counts and 30 and 15 the group totals of the CM313 and placebo arms respectively; Two-sided Fisher's exact test, no mid-p correction
Method: Two-sided Fisher's exact test on (22,8;1,14), p ≈ 0.00005, consistent with P<0.001.
How we recomputed it: pFisher2x2(22,8,1,14,0)
CONSISTENTreported p < .001 · recomputed p = <.001Reviewer 1
Proportion with ≥2 consecutive PLT counts ≥50×10^9/L within 8 weeks, CM313 vs placebo (Fisher's exact)
“the proportion of patients achieving two consecutive platelet counts ≥50×10 9 /L within eight weeks was higher in the CM313 group (90%; 27/30) compared with placebo group (33%; 5/15), with a difference of 56.7% (95% CI 28.0% to 77.5%; P<0.001).”
Find in source
Taken as given: The 27 and 5 are the event counts and 30 and 15 the group totals of the CM313 and placebo arms respectively; Two-sided Fisher's exact test, no mid-p correction
Method: Two-sided Fisher's exact test on (27,3;5,10), p ≈ 0.00004, consistent with P<0.001.
How we recomputed it: pFisher2x2(27,3,5,10,0)
CONSISTENTreported p = .004 · recomputed p = .004Reviewer 1
Day 5 platelet count ≥50×10^9/L, CM313 vs placebo (Fisher's exact)
“By day 5, 87% (26/30) of patients in the CM313 group achieved platelet counts ≥50×10 9 /L compared with 40% (6/15) in the placebo group (P=0.004).”
Find in source
Taken as given: The 26 and 6 are the event counts and 30 and 15 the group totals of the CM313 and placebo arms respectively; Two-sided Fisher's exact test, no mid-p correction
Method: Two-sided Fisher's exact test on (26,4;6,9), p ≈ 0.004, matching the reported P=0.004.
How we recomputed it: pFisher2x2(26,4,6,9,0)
CONSISTENTreported p < .001 · recomputed p = <.001Reviewer 1
Overall response during treatment period ≤8 weeks, CM313 vs placebo (Fisher's exact)
“During the eight week treatment period, 28 (93%) patients in the CM313 group and five (33%) in the placebo group achieved an overall response rate, with a difference of 60.0% (95% CI 32.1% to 80.0%; P<0.001).”
Find in source
Taken as given: The 28 and 5 are the event counts and 30 and 15 the group totals of the CM313 and placebo arms respectively; Two-sided Fisher's exact test, no mid-p correction
Method: Two-sided Fisher's exact test on (28,2;5,10), p ≈ 0.00003, consistent with P<0.001.
How we recomputed it: pFisher2x2(28,2,5,10,0)
CONSISTENTreported p = .001 · recomputed p = <.001Reviewer 1
Long term sustained PLT count response, CM313 vs placebo (Fisher's exact)
“Long term sustained platelet count response occurred in 18 (60%) patients in the CM313 group and one (7%) patient in the placebo group (P=0.001).”
Find in source
Taken as given: The 18 and 1 are the event counts and 30 and 15 the group totals of the CM313 and placebo arms respectively; Two-sided Fisher's exact test, no mid-p correction
Method: Two-sided Fisher's exact test on (18,12;1,14), p ≈ 0.0009, consistent with reported P=0.001.
How we recomputed it: pFisher2x2(18,12,1,14,0)
CONSISTENTreported p = .001 · recomputed p = <.001Reviewer 2
Two-sided Fisher's exact test for long-term sustained platelet count response
“Long term sustained platelet count response occurred in 18 (60%) patients in the CM313 group and one (7%) patient in the placebo group (P=0.001).”
Find in source
Taken as given: The 18 and 30 are the event count and total for CM313.; The 1 and 15 are the event count and total for placebo.; The test is two-sided Fisher's exact test.
Method: pFisher2x2(18,12,1,14) returns a two-tailed p-value from Fisher's exact test.
How we recomputed it: pFisher2x2(18,12,1,14)

Overstated conclusions

2 findings · worst high

Conclusions 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
Conclusions only partially backed by the presented evidenceAssessed

12 major claims checked against the paper's own evidence: all adequately supported.

partialReviewer 1
CM313 reduced bleeding episodes and the requirement for rescue treatment.
Rescue use is formally compared (10% vs 53%, P values reported), but the bleeding reductions are presented descriptively (37% to 10% in CM313 vs 27% to 47% in placebo) without a formal statistical test, so the bleeding component is only partially supported.
Evidence: Results: 'The proportion of patients in the CM313 group with bleeding decreased from 37% (11/30) at baseline to 10% (3/30) at week 8...' and 'Rescue drugs were required in 10% (3/30)... compared with 53% (8/15).'
The proportion of patients in the CM313 group with bleeding decreased from 37% (11/30) at baseline to 10% (3/30) at week 8 and remained at 13% (4/30) at week 24, without requiring rescue treatment.
Resultsreviewer’s wording
partialReviewer 1
CM313 mitigates platelet destruction by targeting plasma cells and regulating antibody-dependent cytotoxic pathways.
The immune data support plasma-cell depletion (reduced antibody-secreting cells and immunoglobulins) and NK-cell reduction, but the macrophage Fcγ-receptor modulation is inferred from correlational observations, so the mechanistic claim reaches somewhat beyond the direct evidence.
Evidence: Results/Changes in immunological indicators: 'antibody secreting cells and serum immunoglobulin levels exhibited a pronounced decline in the CM313 group' and 'the expression of Fcγ receptors (FcγR) I and FcγRIIb on monocytes decreased after administration of CM313.'
“Collectively, these findings highlight a multifaceted mechanism through which CM313 mitigates platelet destruction in patients with immune thrombocytopenia by targeting plasma cells and regulating antibody dependent cytotoxic pathways.”
DiscussionFind in source
partialReviewer 2
CM313 offers a promising alternative with potential for greater effect in immune thrombocytopenia.
The efficacy results are promising, but the small sample size, lack of comparison with other active treatments, and single dosing regimen limit the strength of this claim. The claim is acknowledged as a conclusion.
Evidence: Efficacy results described, but no direct comparison with other treatments.
“Given the challenges associated with achieving long term remission using current treatments for immune thrombocytopenia, CM313 offers a promising alternative with potential for greater effect.”
DiscussionFind in source
partialReviewer 3
CM313 offers a promising alternative with potential for greater effect.
The evidence supports efficacy over placebo, but the claim of 'greater effect' is not directly compared to other treatments and is modestly stated.
Evidence: The paper shows significant improvements over placebo, but does not provide head-to-head comparison with other active treatments.
“CM313 offers a promising alternative with potential for greater effect.”
DiscussionFind in source
supportedReviewers 1, 2
At week 8, the overall response rate was higher in the CM313 group (83%) than the placebo group (20%).
The primary endpoint result is backed by the reported 25/30 vs 3/15, a Fisher exact P<0.001, and a 63.3% (95% CI 33.7-81.3) difference.
Evidence: Abstract and Table 2: 'overall response rate at week 8 | 25 (83) | 3 (20) | 63.3 (33.7 to 81.3) | <0.001'.
“At week 8, the overall response rate was higher in the CM313 group (83%; 25/30) compared with placebo group (20%; 3/15): difference 63.3% (95% confidence interval 33.7% to 81.3%; P<0.001).”
Table 2Find in source
supportedReviewer 1
CM313 rapidly increased platelet counts within three days.
The claim is supported by the reported day-3 mean platelet count being higher in the CM313 group (P=0.01) and the day-5 ≥50×10^9/L response difference.
Evidence: Results: 'As early as day 3, the mean platelet count was higher in the CM313 group than in the placebo group (P=0.01).'
As early as day 3, the mean platelet count was higher in the CM313 group than in the placebo group (P=0.01).
Resultsreviewer’s wording
supportedReviewer 1
CM313 showed a favourable safety profile.
The safety claim is supported by reported AE tables: most TEAEs were grade 1-2, serious AEs were limited (2 patients, 0 drug-related grade ≥3 beyond one cerebral infarction), and no deaths occurred.
Evidence: Table 3 and safety text: 'Any grade ≥3 TEAE | 2 (7) | 0'; 'No treatment emergent adverse events leading to death were reported.'
Serious adverse events were observed in two patients in the CM313 group: one with grade 3 central nervous system inflammation and one with drug related grade 3 cerebral infarction.
Resultsreviewer’s wording
supportedReviewer 2
CM313 induced rapid onset and sustained increases in platelet counts, reduced bleeding episodes, and had a favourable safety profile.
The reported results directly support this claim: 83% overall response at week 8, rapid onset by day 3, sustained response up to week 24, reduced bleeding, and manageable adverse events.
Evidence: Primary and secondary outcomes, safety data.
“CM313 showed a favourable safety profile and encouraging efficacy in adults with persistent or chronic primary immune thrombocytopenia, characterised by rapid increase in platelet count, sustained platelet responses, and manageable adverse events.”
DiscussionFind in source
supportedReviewer 2
Median time to platelet count ≥50×10^9/L was one week in CM313 group vs not reached in placebo (P<0.001).
Data from Table 2 and Kaplan-Meier analysis support this claim.
Evidence: Table 2: median (95% CI) time to first two consecutive PLT counts ≥50×10^9/L: 1.0 (NR) vs NR (5.0 to NR).
The median time to a platelet count of ≥50×10^9/L was one week in the CM313 group but was not reached in the placebo group (P<0.001).
Table 2reviewer’s wording
supportedReviewer 2
CM313 treatment induced a pronounced and sustained reduction in peripheral blood antibody-secreting cells and serum immunoglobulins.
Figure 4 and supplementary figures show these changes compared to placebo.
Evidence: Supplementary figures S4 and S5, Figure 4 showing mean change in total serum IgG.
After treatment, antibody secreting cells and serum immunoglobulin levels exhibited a pronounced decline in the CM313 group compared with placebo group, with levels maintained consistently up to week 24.
Resultsreviewer’s wording
supportedReviewer 2
The efficacy of CM313 was evident as early as day 3.
Figure 2 shows that mean platelet count was higher in CM313 group by day 3 (P=0.01).
Evidence: Figure 2 and text: 'As early as day 3, the mean platelet count was higher in the CM313 group than in the placebo group (P=0.01).'
As early as day 3, the mean platelet count was higher in the CM313 group than in the placebo group (P=0.01).
Results ¶5reviewer’s wording
supportedReviewer 3
CM313 showed a favourable safety profile in adults with persistent or chronic primary immune thrombocytopenia.
Safety data are presented in Tables 3 and 4, showing most adverse events are grade 1-2 and manageable.
Evidence: Tables 3 and 4 detail treatment-emergent adverse events, demonstrating low rates of serious events and manageable infusion-related reactions.
“CM313 showed a favourable safety profile and encouraging efficacy in adults with persistent or chronic primary immune thrombocytopenia, characterised by rapid increase in platelet count, sustained platelet responses, and manageable adverse events.”
AbstractFind in source

Premise concern: surrogate not validated for clinical benefit.

INADEQUATESurrogate endpoint
The primary efficacy endpoint is the overall response rate, defined by platelet count thresholds (≥30×10^9/L with doubling) and absence of bleeding. Platelet count is a surrogate biomarker for clinical bleeding outcomes. Although the authors report secondary bleeding outcomes and show target engagement (depletion of antibody-secreting cells and immunoglobulin reduction), they do not cite validated evidence linking the platelet-count surrogate to hard clinical outcomes such as major bleeding or mortality. The claim of clinical benefit rests primarily on this surrogate measure.
“The primary outcome was overall response rate (at least two consecutive platelet counts ≥30×10^9 /L, a minimum doubling from baseline, and no bleeding) at week 8.”
Find in source
ADEQUATEEffect size
The effect size is large and statistically significant: overall response rate 83% (25/30) vs 20% (3/15), difference 63.3% (95% CI 33.7% to 81.3%; P<0.001). The paper also reports reductions in bleeding episodes and rescue drug use, anchoring the effect to a clinically meaningful outcome. The magnitude is comparable to other approved ITP therapies.
“At week 8, the overall response rate was higher in the CM313 group (83%; 25/30) compared with placebo group (20%; 3/15): difference 63.3% (95% confidence interval 33.7% to 81.3%; P<0.001).”
Find in source

Data authenticity concerns

None found

An 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 high

Required 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 and code not sharedAssessed

Eight Dimensions7 pass · 1 fail

Prior work is cited extensively, including the pathophysiology of immune thrombocytopenia, the role of CD38, and previous anti-CD38 antibodies. The rationale for CM313 is clearly linked to the need for better treatments. Limitations of prior treatments are acknowledged, and the current study's design addresses the single-arm nature of the phase 1/2 trial.

Sub-criteria

prior work citedADEQUATE

premise rationaleADEQUATE

limitations addressedADEQUATE

Per reviewer (3) — the votes behind the verdict

PASSReviewer 1· 90% conf

The rationale is well established: ITP pathophysiology, the role of CD38 on antibody-producing cells, prior efficacy of other anti-CD38 antibodies, and the authors' own prior single-arm phase 1/2 trial all motivate this randomised trial.

Evidence

“CD38, a type II glycoprotein abundantly expressed on antibody-producing plasmablasts and plasma cells, represents a promising therapeutic target in autoimmune diseases.”

Introduction ¶1Find in source

“Preliminary results from our previous single arm phase 1/2 trial highlighted the promising efficacy and tolerable safety profile of CM313 in patients with previously treated immune thrombocytopenia.”

Introduction ¶3Find in source

paraphrase[Introduction, paragraph 3]

“Building on these findings, this study presents data from a randomised, placebo controlled, phase 2 trial evaluating the safety and efficacy of CM313.”

PASSReviewer 2· 95% conf

The paper establishes a strong scientific foundation by citing prior work on immune thrombocytopenia and CD38-targeted therapies, providing a logical rationale for the study, and addressing limitations of previous treatments.

Evidence

paraphrase[Introduction, paragraphs 1-3]

“Immune thrombocytopenia is a complex acquired autoimmune disease... CD38, a type II glycoprotein abundantly expressed on antibody-producing plasmablasts and plasma cells, represents a promising therapeutic target in autoimmune diseases.”

“Building on these findings, this study presents data from a randomised, placebo controlled, phase 2 trial evaluating the safety and efficacy of CM313 in patients with immune thrombocytopenia who had received multiple previous treatments.”

IntroductionFind in source

“Relapse is, however, frequent, with many patients progressing to persistent or chronic forms of the disease.”

Introduction ¶2Find in source

PASSReviewer 3· 95% conf

The paper establishes a strong scientific foundation by citing prior work on immune thrombocytopenia, CD38 as a target, and previous phase 1/2 trial results, and provides a logical rationale for the current study.

Evidence

Introduction ¶4Find in source

“CD38, a type II glycoprotein abundantly expressed on antibody-producing plasmablasts and plasma cells, represents a promising therapeutic target in autoimmune diseases.”

Introduction ¶3Find in source

Introduction ¶4Find in source

Randomization used SAS with block size of 3, blinding was maintained with visually indistinguishable treatments and an independent data collection team. A priori power analysis was performed (80% power, alpha=0.05). Inclusion/exclusion criteria are described in supplementary table S1 and in the text. Outlier handling is addressed through the pre-specified analysis population (full analysis set, safety set) and missing data methods (non-response imputation, LOCF).

Sub-criteria

controlsN/A

power analysisADEQUATE

blinding levelsADEQUATE

outlier handlingADEQUATE

randomization unitADEQUATE

inclusion exclusionADEQUATE

randomization methodADEQUATE

replicate distinctionN/A

independent replicationN/A

Per reviewer (3) — the votes behind the verdict

PASSReviewer 1· 85% conf

Key design elements for a randomised phase 2 trial are adequately reported: method and unit of randomisation, double-blind masking, an a priori power calculation, pre-specified inclusion/exclusion criteria, and a defined analysis population with missing-data handling.

Evidence

“Using SAS software, a statistician applied a block size of three to generate the randomisation list for patients.”

MethodsFind in source

“To maintain blinding, both CM313 and placebo were visually indistinguishable.”

MethodsFind in source

“we required a sample size of 39 participants (allocated 2:1) to provide a power of 80% to detect a difference between groups at a two sided α level of 0.05.”

Statistical analysisFind in source

PASSReviewer 2· 95% conf

The study design is rigorous with adequate randomization, blinding, power analysis, inclusion/exclusion criteria, and handling of missing data.

Evidence

“Using SAS software, a statistician applied a block size of three to generate the randomisation list for patients.”

MethodsFind in source

“Based on a previous study, assuming a response rate of 70% for CM313 at week 8 and a spontaneous remission rate of 18% for placebo, we required a sample size of 39 participants (allocated 2:1) to provide a power of 80% to detect a difference between groups at a two sided α level of 0.05.”

Statistical analysisFind in source

“Efficacy outcomes were evaluated based on the full analysis set, which followed the intention-to-treat principle and included all randomised participants who received one dose or more of the study drug and had at least one efficacy assessment.”

Statistical analysisFind in source

PASSReviewer 3· 95% conf

The study design is rigorous with adequate randomization, blinding, power analysis, inclusion/exclusion criteria, and analysis population definitions.

Evidence

“Using SAS software, a statistician applied a block size of three to generate the randomisation list for patients.”

MethodsFind in source

“To maintain blinding, both CM313 and placebo were visually indistinguishable.”

MethodsFind in source

Statistical analysisFind in source

Sex is reported for both groups. Age, BMI, duration of ITP, baseline platelet counts, and comorbidities are reported. The study population is restricted to Chinese patients, which is stated in the limitations.

Sub-criteria

demographicsADEQUATE

sex reportedADEQUATE

sex justifiedN/A

age weight healthADEQUATE

housing conditionsN/A

species strain sourceN/A

Per reviewer (3) — the votes behind the verdict

PASSReviewer 1· 80% conf

Baseline characteristics include sex, age, BMI, disease duration, platelet counts, prior treatments, and bleeding scores; ethnicity is described as Chinese. Mixed-sex study, so a sex justification is not required.

Evidence

“Median age was 38.5 years (interquartile range (IQR) 29.0-52.0) in the CM313 group and 45.0 (25.0-52.0) years in the placebo group, with female patients comprising 18 (60%) and 11 (73%) participants, respectively.”

Table 1Find in source

paraphrase[Results, Patients]

“Female patients comprised 18 (60%) and 11 (73%) of the CM313 and placebo groups respectively, indicating a mixed-sex study.”

PASSReviewer 2· 95% conf

Key biological variables including sex, age, weight, health status, and demographics are adequately reported.

Evidence

paraphrase[Results, Table 1]

“Median (IQR) age (years) | 38.5 (29.0-52.0) | 45.0 (25.0-52.0) | 41.0 (29.0-52.0) | Sex*: Male | 12 (40) | 4 (27) | 16 (36) | Female | 18 (60) | 11 (73) | 29 (64)”

paraphrase[Results, Table 1]

“Baseline platelet count (×10^9/L): <10 | 17 (57) | 4 (27) | 21 (47) | 10 to <30 | 13 (43) | 11 (73) | 24 (53)”

“The study population was restricted to Chinese patients, which may limit the generalisability of the findings to other ethnic groups.”

LimitationsFind in source

PASSReviewer 3· 85% conf

Demographic and clinical characteristics are well-reported, including age, sex, BMI, and disease-specific parameters, though race/ethnicity is not explicitly tabulated.

Evidence

paraphrase[Results, Table 1]

“Male 12 (40), Female 18 (60)”

“Median age was 38.5 years (interquartile range (IQR) 29.0-52.0) in the CM313 group and 45.0 (25.0-52.0) years in the placebo group”

Results ¶1Find in source

“The study population was restricted to Chinese patients, which may limit the generalisability of the findings to other ethnic groups.”

LimitationsFind in source

The trial protocol was approved by the ethics committee of the Institute of Hematology and Blood Diseases Hospital (IIT2023068-EC-1). Written informed consent was obtained from all patients. The study was conducted in accordance with the Declaration of Helsinki.

Sub-criteria

iacuc statementN/A

informed consentADEQUATE

irb ethics statementADEQUATE

regulatory complianceADEQUATE

Per reviewer (3) — the votes behind the verdict

PASSReviewer 1· 90% conf

A named ethics committee approved the protocol with a protocol number, written informed consent was obtained from all patients, and compliance with the Declaration of Helsinki is stated.

Evidence

“The trial protocol received approval from the ethics committees of the Institute of Hematology and Blood Diseases Hospital (IIT2023068-EC-1) and was conducted in strict accordance with the principles of the Declaration of Helsinki.”

EthicsFind in source

“All patients provided written informed consent.”

MethodsFind in source

PASSReviewer 2· 95% conf

Ethical approval, informed consent, and regulatory compliance are adequately reported.

Evidence

EthicsFind in source

“All patients provided written informed consent.”

MethodsFind in source

PASSReviewer 3· 95% conf

The study has explicit ethical approval from a named ethics committee, written informed consent, and adherence to the Declaration of Helsinki.

Evidence

“The trial protocol received approval from the ethics committees of the Institute of Hematology and Blood Diseases Hospital (IIT2023068-EC-1)”

EthicsFind in source

“All patients provided written informed consent.”

MethodsFind in source

“conducted in strict accordance with the principles of the Declaration of Helsinki.”

EthicsFind in source

CM313 is identified with source (Keymed Biosciences) and dose regimen. The autoantibody detection kit is named. Statistical software (SAS 9.4, PASS 2022) is identified. No other key resources are applicable.

Sub-criteria

mycoplasma testingN/A

reagents identifiedADEQUATE

organisms identifiedN/A

antibodies identifiedN/A

cell line authenticationN/A

software tools identifiedADEQUATE

Per reviewer (3) — the votes behind the verdict

PASSReviewer 1· 70% conf

The investigational product (CM313) is named with source and dose regimen, and statistical software is identified with versions; no wet-lab bench reagents needing authentication are used beyond a named detection kit.

Evidence

“CM313 or matching placebo was administered intravenously at a dose of 16 mg/kg weekly for eight weeks.”

AcknowledgementsFind in source

“Statistical analyses were conducted using SAS version 9.4.”

Statistical analysisFind in source

PASSReviewer 2· 90% conf

The investigational drug and software tools are adequately identified.

Evidence

“We thank Keymed Biosciences for providing CM313 and placebo for this study”

AcknowledgementsFind in source

“Statistical analyses were conducted using SAS version 9.4.”

Statistical analysisFind in source

PASSReviewer 3· 85% conf

The investigational product (CM313) is adequately identified with source and dose regimen, and statistical software is identified. The autoantibody detection kit is mentioned but not fully identified.

Evidence

“CM313 or matching placebo was administered intravenously at a dose of 16 mg/kg weekly for eight weeks.”

MethodsFind in source

“Statistical analyses were conducted using SAS version 9.4.”

Statistical analysisFind in source

“Platelet glycoprotein autoantibodies were detected using a commercial kit (Autoantibodies against Platelet-specific Receptors Detecting Kit, Flow cytometry-fluorescein).”

Table 1Find in source

The paper specifies Fisher's exact test, mixed-effects model, Kaplan-Meier method, and Wilcoxon Mann-Whitney test. Effect sizes are reported with 95% CIs. Assumptions are not explicitly tested but standard methods are used. Data presentation includes tables, figures, and Kaplan-Meier curves. Some p-values are reported as <0.001, which is a threshold but acceptable in clinical trial reporting.

Sub-criteria

tests namedADEQUATE

exact p valuesINADEQUATE

effect sizes ciADEQUATE

data presentationADEQUATE

software identifiedADEQUATE

assumptions verifiedADEQUATE

mathematical plausibilityPASS

Per reviewer (3) — the votes behind the verdict

PASSReviewer 1· 80% conf

Statistical tests are named (Fisher's exact, Clopper-Pearson, Miettinen-Nurminen, mixed model for repeated measures, Kaplan-Meier), exact p-values and 95% CIs on differences are reported, software is identified, and per-group n and percentages are internally consistent.

Evidence

“Fisher’s exact test was used for group comparisons and the Miettinen-Nurminen method was used to calculate 95% CIs for differences between groups.”

Statistical analysisFind in source

“difference 63.3% (95% confidence interval 33.7% to 81.3%; P<0.001)”

AbstractFind in source

paraphrase[Results, Efficacy]

“By day 5, 87% (26/30) of patients in the CM313 group achieved platelet counts ≥50×10 9 /L compared with 40% (6/15) in the placebo group (P=0.004).”

WARNReviewer 2· 85% conf

Most statistical aspects are well reported, but the primary outcome p-value is reported imprecisely as P<0.001 instead of an exact value, leading to a warning.

Evidence

“At week 8, the overall response rate was higher in the CM313 group (83%; 25/30) compared with placebo group (20%; 3/15): difference 63.3% (95% confidence interval 33.7% to 81.3%; P<0.001).”

Results ¶1Find in source

paraphrase[Results, Table 2]

“P=0.004, P=0.03, P=0.001, P=0.007, P=0.04, P=0.16”

paraphrase[Results, Table 2]

“Difference, % (95% CI) | 63.3 (33.7 to 81.3) | <0.001”

PASSReviewer 3· 90% conf

Statistical tests are named, effect sizes with confidence intervals are reported, software is identified, and data presentation is appropriate for a clinical trial. Some p-values are reported as thresholds (<0.001) rather than exact values.

Evidence

“Fisher’s exact test was used for group comparisons and the Miettinen-Nurminen method was used to calculate 95% CIs for differences between groups.”

Statistical analysisFind in source

“63.3 (33.7 to 81.3)”

Table 2Find in source

paraphrase[Results, paragraph 1]

“P<0.001”

The paper ends with an incomplete data availability statement ('The code used to analyse') and does not provide any repository link or accession number. No code sharing is mentioned. This is a significant omission for a clinical trial.

Sub-criteria

code sharingNOT REPORTED

accession numbersN/A

repository depositNOT REPORTED

data availability statementNOT REPORTED

Per reviewer (3) — the votes behind the verdict

WARNReviewer 1· 60% conf

A data availability statement exists but is truncated/vague in the provided text ('The code used to analyse'), with no concrete repository, accession, or managed-access route stated; no code repository with a persistent identifier is given.

Evidence

“Data availability statement The code used to analyse”

Data availabilityFind in source

absence[Data availability statement]

No code repository with a persistent identifier is reported.

FAILReviewer 2· 90% conf

No data or code availability statement is provided, and no repository deposit or accession numbers are reported.

Evidence

“Data availability statement The code used to analyse”

End of manuscriptFind in source

absence[Entire manuscript]

No code repository or sharing mentioned.

FAILReviewer 3· 90% conf

The data availability statement is incomplete and does not provide a concrete access route. Code sharing is not reported. For a clinical trial, a managed access plan should be described.

Evidence

“Data availability statement The code used to analyse”

Data availabilityFind in source

absence[Data availability statement]

No code repository shared

Methods are complete enough for replication. Trial registration is provided (NCT06199089). All outcomes are reported, including negative results. Limitations are discussed. Conclusions are proportional to the evidence. Funding and conflicts of interest are stated. However, no reporting guideline (e.g., CONSORT) is referenced.

Sub-criteria

funding coiADEQUATE

trial registrationADEQUATE

reporting guidelineNOT REPORTED

methods completenessADEQUATE

all outcomes reportedADEQUATE

limitations discussedADEQUATE

conclusions proportionalADEQUATE

Per reviewer (3) — the votes behind the verdict

PASSReviewer 1· 80% conf

Methods are detailed, the trial is registered (NCT06199089), limitations are explicitly discussed, conclusions are proportional, and funding/COI are disclosed. The only gap is that no reporting guideline (e.g. CONSORT) is explicitly referenced.

Evidence

“Trial registration ClinicalTrials.gov NCT06199089 .”

AbstractFind in source

“This study has several limitations, including a small sample size, a limited follow-up period, and the lack of patient health related quality of life measurements.”

LimitationsFind in source

paraphrase[Funding / Competing interests]

“The funders of the study had no role in study design, data collection, data analysis, data interpretation, or writing of the report.”

PASSReviewer 2· 90% conf

The manuscript is generally transparent, with detailed methods, trial registration, funding/COI, limitations, and proportional conclusions, but lacks a reporting guideline checklist.

Evidence

paraphrase[Abstract and Methods]

“Trial registration ClinicalTrials.gov NCT06199089.”

“This study has several limitations, including a small sample size, a limited follow-up period, and the lack of patient health related quality of life measurements.”

LimitationsFind in source

paraphrase[Funding and Competing interests]

“This study received support from the Chinese Academy of Medical Sciences innovation fund for medical sciences... Competing interests: All authors have completed the ICMJE uniform disclosure form...”

PASSReviewer 3· 90% conf

The paper is well-reported with trial registration, methods detailed, limitations discussed, and proportional conclusions. However, no explicit reference to a reporting guideline (e.g., CONSORT) is made.

Evidence

“ClinicalTrials.gov NCT06199089”

AbstractFind in source

“This study has several limitations, including a small sample size, a limited follow-up period, and the lack of patient health related quality of life measurements.”

LimitationsFind in source

paraphrase[Funding and Competing interests]

“This study received support from the Chinese Academy of Medical Sciences innovation fund for medical sciences... All authors have completed the ICMJE uniform disclosure form...”

Registered (2 IDs: ClinicalTrials.gov). No reporting guideline cited.

ClinicalTrials.gov

NCT06199089

ClinicalTrials.gov

NCT06594146

Broken references and links

None found

References 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.

3 data/code links checked; 3 live.

datahttps://public.flourish.studio/visualisation/25195771/LIVEHTTP 200
Resolves, but the content could not be matched to the paper.
datahttps://public.flourish.studio/visualisation/25276109/LIVEHTTP 200
Resolves, but the content could not be matched to the paper.
datahttps://public.flourish.studio/visualisation/25299115/LIVEHTTP 200
Resolves, but the content could not be matched to the paper.

Copyediting

7 minor

Wording, consistency and formatting errors that need correcting before submission.

No major wording or formatting errors. 7 minor suggestions below.

7 copyedit issues flagged: mostly typo, consistency, clarity.

MINORtypoFig 3 footnote
“Thrombopoietin receptor agonists include eltrombopag, herombopag, or avatrombopag.”
→ Change 'herombopag' to 'hetrombopag'.
Elsewhere in the paper the drug is spelled 'hetrombopag'.
MINORconsistencyResults, Efficacy (subgroup paragraph)
“10 out of 12 patients (83%) who tested positive for anti-GP Ib/IX autoantibodies (either alone or in combination with anti-GP Ib/IX autoantibodies) at baseline achieved the primary endpoint.”
→ The second 'anti-GP Ib/IX autoantibodies' should likely read 'anti-GP IIb/IIIa autoantibodies'.
The combination group refers to positivity for both glycoproteins IIb/IIIa and Ib/IX per Table 1.
MINORclarityData availability statement
“Data availability statement The code used to analyse”
→ Complete the statement with a concrete data/code access route.
The statement appears truncated in the provided text.
MINORotherEnd of manuscript
“Data availability statement The code used to analyse”
→ Complete the data availability statement, e.g., 'The code used to analyse the data is available at [repository URL]. The de-identified participant data can be requested from the corresponding author.'
The statement is cut off and incomplete.
MINORclarityResults, Table 4
“Increased platelet count*”
→ Define the asterisk footnote in the table legend, as it is currently defined in the text but not in the table.
The footnote '* At least one platelet count ≥450×10^9/L' is in the text but not in the table caption.
MINORtypoTable 3 footnote
“herombopag”
→ hetrombopag
Possible typo in the drug name
MINORconsistencyData availability statement
“Data availability statement The code used to analyse”
→ Complete the sentence with a repository link or contact information
Statement is truncated and incomplete

Kaimen Rigor Action ItemsFinal pass before journal submission

As a post-publication audit, the published work is robust in design and reporting, but an informed reader should weigh two gaps: the truncated data availability statement (no data/code access route) and the threshold-only primary p-value. These warrant a correction or a brief erratum/letter providing the exact p-value and a concrete data-access mechanism, rather than re-analysis; the trial registration, ethics, and statistical results appear sound.

1.
HIGHdata code
Complete the truncated data availability statement (currently 'The code used to analyse') with a concrete access route: name a repository (e.g., figshare/Zenodo) with a DOI for the de-identified dataset, or a managed-access platform/committee with request conditions and a timeframe.
A data-driven clinical trial with an incomplete data availability statement is a serious reporting gap that undermines reproducibility and transparency.
2.
HIGHdata code
Share the analysis code in a public repository (e.g., GitHub/Zenodo) with a persistent identifier and reference it in the code availability statement.
The statement references analysis code but gives no repository, so the code is currently inaccessible to readers.
3.
HIGHcopyedit
Fix the antibody-consistency error in the Results subgroup paragraph: the second 'anti-GP Ib/IX autoantibodies' should read 'anti-GP IIb/IIIa autoantibodies'.
The sentence as written is internally contradictory and misreports the combination-group definition given in Table 1.
4.
MEDIUMstatistics
Report the exact p-value for the primary outcome instead of the threshold 'P<0.001'.
Printing a threshold for a computed p-value is imprecise reporting and was flagged independently by two reviewers; an exact value (e.g., P=0.000...) improves transparency.
5.
MEDIUMstatistics
Name the group-comparison test used for time-to-event endpoints (e.g., log-rank test) in the Methods.
The paper reports P values for durations/median times but only describes Kaplan-Meier estimation, leaving the group-comparison method unspecified.
6.
MEDIUMreporting
Add an explicit statement of adherence to the CONSORT 2010 reporting guideline and reference the completed CONSORT flow diagram/checklist.
A randomised trial should reference its reporting guideline; its absence was flagged by all three reviewers.
7.
MEDIUMcopyedit
Correct the drug-name typo 'herombopag' to 'hetrombopag' in the Fig 3 footnote and Table 3 footnote.
The drug is spelled 'hetrombopag' elsewhere in the paper; the inconsistent spelling is a copyedit defect.
8.
MEDIUMcopyedit
Define the asterisk footnote ('At least one platelet count ≥450×10^9/L') in the Table 4 legend, not only in the text.
The table caption lacks the footnote definition, making the 'Increased platelet count*' row ambiguous to readers.
9.
LOWother
Add vendor, catalog number, and RRID for the platelet autoantibody detection kit (and any flow-cytometry antibody clones).
Full reagent identification improves reproducibility; the kit is currently named without source identifiers.
10.
LOWreporting
Add a statement on patient and public involvement in the design or conduct of the study.
The paper notes this is not standard practice in China, but a brief statement would strengthen reporting transparency.
11.
LOWother
Tabulate race/ethnicity explicitly in Table 1 or state clearly that all participants were Chinese.
Reviewer 3 flagged demographics as inadequate because ethnicity is only mentioned in the limitations, not presented in the baseline table.
12.
LOWstatistics
Consider reporting a normality/assumption check for the mixed-model residuals.
Assumptions are handled adequately by the clinical idiom, but an explicit check would strengthen the statistical reporting.

Full text (46,561 chars)Analyzed: Aug 10, 2026

RigorReviewer 1

RigorReviewer 2

RigorReviewer 3

RigorPre-Submission Reviewer

CopyeditAgent Scribe

CitationsCitation Integrity

StatisticsStatistics Consistency

ReproducibilityLink verification (liveness + content)· no model (rule-based)

PreregistrationDeterministic check· no model (rule-based)

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.”

How stars are deducted

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.

The eight reporting dimensions

Scientific PremisePrior work, rationale, and how it addresses gaps.

Study Design & RigorRandomization, blinding, power, controls, replication.

Biological VariablesSex, age, species/strain, housing, demographics.

Ethical ApprovalsIRB/IACUC approval, informed consent, regulatory compliance.

Key ResourcesAntibodies, cell lines, organisms, software — identified with persistent IDs.

Statistical ReportingNamed tests, exact p-values, effect sizes, plausible summary stats.

Data & Code AvailabilityAvailability statements, repositories, accession numbers, shared code.

Reporting TransparencyMethods, registration, guidelines, outcomes, limitations, funding/COI.

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.

Anti-CD38 monoclonal antibody CM313 for primary immune thrombocytopenia: multicentre, randomised, placebo controlled, phase 2 trial.

Chen Y, Xu Y, Dai J, Sun T, Li H, Hua Z, Zhou Z, Zhou H, Yan Z, Zhao X, Xue F, Liu W, Liu X, Fu R, Wang W, Chi Y, Dong H, Ju M, Dai X, Gu W, Pei X, Yang R, Zhang L

DOI

10.1136/bmj-2025-084314

Record issued

2026-08-10

Engine

7.29.0

Exported

2026-09-21