Effect of methylcobalamin on capecitabine induced hand-foot syndrome in patients with HER2 negative early breast cancer: multicentre, double blind, randomised, placebo controlled, phase 3 trial.
Xia Y, Zhu Y, Ling L, Xu F, Yang Y, Ye J, Tan W, Chen Z, Liu Q, Wei W, Zhang J, Zhang A, Zhang L, Song E, Gong C
- DOI
- 10.1136/bmj-2025-084290
- Record issued
- 2026-08-10
- Engine
- 7.29.0
- Exported
- 2026-09-22
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/37f2c28b-e406-42f3-bdd2-828471a9d79a is authoritative.
How this rating was calculated
- ReportingBiological variables not met−0.5★
- ReportingEthical approvals partially met−0.25★
- ReportingKey resources partially met−0.25★
- ReportingData & code availability partially met−0.25★
- 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.
- 01Biological variables not reported
The paper fails to report key biological variables: weight is not reported, race/ethnicity is not explicitly reported in the baseline table, and no justification is given for the single-sex study.
“Eligible participants were women aged 18-75 years”
MethodsFind in source
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 phase 3 RCT is well-designed and well-reported on most dimensions: randomisation, blinding, power analysis, statistics, and transparency are strong, and the data are openly deposited. The main weaknesses are incomplete biological-variable reporting (no weight, no explicit race/ethnicity, no sex justification), unnamed ethics committees, missing drug-source identification, and code deposited only as a supplementary appendix. No statistical errors were found in the limited machine-verification, and the copyedit flagged a numeric inconsistency in a secondary-endpoint percentage.
Three independent reviewer runs were synthesised; they converged on premise, design, statistics, and transparency but diverged on biological variables, ethics, key resources, and data/code (Reviewer 1 was more lenient, Reviewers 2/3 more stringent). The statistics verification recomputed only 3 tests/effect estimates (all consistent); the rest of the analysis is unverified. Citation check found no retracted or not-found references; data links are live.
Numerical inconsistencies
None foundValues that contradict each other or are impossible for the stated sample: recomputed p-values and test statistics, GRIM/GRIMMER checks on summary numbers, percentages against their own counts, totals against their parts, and estimates against their own confidence intervals.
Checked — nothing surfaced.
Recomputed 3 tests: 3 consistent, 0 inconsistent; 1 recomputed directly from the reported test statistics, 2 via agent-written checks.
- CONSISTENTreported p = .007 · recomputed p = .005Recomputed risk ratio 0.47 (95% CI 0.28–0.81), reported p=0.007
“risk ratio 0.47, 95% confidence interval 0.28 to 0.81; P=0.007”
Taken as given: 0.28–0.81 is a two-sided 95% confidence interval for the risk ratio of 0.47, not a range, an IQR, or a different interval level; the risk ratio is a RATIO measure, so the interval is symmetric on the log scale; p=0.007 is the p for THIS estimate, not for another comparison in the same sentenceMethod: back the two-tailed p out of the log-scale CI width and compare it against the printed pHow we recomputed it: pCI(0.47, 0.28, 0.81, 1) - CONSISTENTreported p = .006 · recomputed p = .006Reviewer 3Two-sided p-value from the primary endpoint risk difference and its 95% CI
“unadjusted risk difference −14.5%, 95% confidence interval −24.9% to −4.1%; Cochran-Mantel-Haensel one sided P value=0.003”
Taken as given: The risk difference is normally distributed; The CI is two-sided at 95% confidenceMethod: pCI function from the estimate and CI, giving a two-sided p-valueHow we recomputed it: pCI(-14.5, -24.9, -4.1, 0) - CONSISTENTreported p = .360 · recomputed p = .358Reviewer 3Two-sided p-value from Pearson's chi-square test for any grade hand-foot syndrome between groups
“hand-foot syndrome of any grade occurred in 60 and 67 patients in the methylcobalamin and placebo groups, respectively (risk difference −6.0%, 95% confidence interval −18.7% to 6.8%; P=0.36)”
Taken as given: The two groups are independent; Expected cell counts are sufficient for chi-square approximationMethod: Pearson chi-square test from 2x2 contingency tableHow we recomputed it: pChi2x2(60, 57, 67, 50)
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 1Methylcobalamin has the potential to reduce the rate of dose modification of capecitabine.The reduction in capecitabine dose modification due to hand-foot syndrome was numerically lower in the methylcobalamin group (7.7% vs 13.7%), but the difference was not statistically significant (P=0.14).Evidence: Secondary endpoint: hand-foot syndrome related reduction or discontinuation of capecitabine: 7.7% vs 13.7%, risk difference −6.0% (95% CI −13.9% to 1.9%), P=0.14.
“The rate of reduction or discontinuation of capecitabine treatment because of hand-foot syndrome was 7.7% (9 of 117) in the methylcobalamin group and 13.7% (16 of 117) in the placebo group (risk difference −6.0%, 95% confidence interval −13.9% to 1.9%).”
AbstractFind in source - partialReviewer 1Methylcobalamin does not increase cancer risk and does not appear to influence long-term survival.Survival outcomes (disease-free and overall survival) showed no significant differences, but the follow-up was short (median ~24 months) with few events, limiting the ability to draw definitive conclusions.Evidence: Survival analysis: disease-free survival (3 vs 5 events, P=0.49) and overall survival (1 vs 3 events, P=0.32) showed no significant differences.
“However, because of the short follow-up period with few events in this study, we did not observe a significant difference in survival outcomes between the two groups. This might suggest that methylcobalamin does not increase cancer risk, a finding supported by previous studies, and its application during chemotherapy does not appear to influence long term survival outcomes for patients with breast cancer.”
DiscussionFind in source - partialReviewer 2Methylcobalamin may reduce the rate of dose modification of capecitabine.The secondary endpoint showed a trend (7.7% vs 13.7%) but was not statistically significant (P=0.14), so the claim is only partially supported.Evidence: Table 2: total reduction/discontinuation related to HFS: 9 (7.7%) vs 16 (13.7%), risk difference -6.0% (-13.9% to 1.9%), P=0.14.
Eight patients (6.8%) in the methylcobalamin group and 13 (11.1%) in the placebo group reduced capecitabine because of hand-foot syndrome ...
Resultsreviewer’s wording - partialReviewer 3Methylcobalamin could potentially maintain the efficacy of capecitabine treatment.The reduction in dose modifications suggests potential maintenance of dose intensity, but survival data show no significant difference, and the study is not powered for survival.Evidence: Discussion: 'our results showed that the need to reduce the dose or discontinue capecitabine treatment was relatively lower in the methylcobalamin group than the placebo group, although without statistical significance, and so methylcobalamin could potentially maintain the efficacy of capecitabine treatment.'
“methylcobalamin could potentially maintain the efficacy of capecitabine treatment in women with HER2 negative early breast cancer.”
DiscussionFind in source - supportedReviewers 1, 2Oral methylcobalamin significantly lowered the severity of hand-foot syndrome by reducing the incidence of grade ≥2 symptoms in women with HER2 negative early breast cancer receiving adjuvant capecitabine.The primary endpoint shows a statistically significant reduction in grade ≥2 hand-foot syndrome (14.5% vs 29.1%, one-sided P=0.003), with consistent results in subgroup and sensitivity analyses.Evidence: Primary endpoint results: risk difference −14.5% (95% CI −24.9% to −4.1%), one-sided P=0.003.
“Grade ≥2 hand-foot syndrome occurred in 17 (14.5%) of 117 patients in the methylcobalamin group and 34 (29.1%) of 117 patients in the placebo group (risk difference −14.5%, 95% confidence interval −24.9% to −4.1%; one sided P value=0.003).”
AbstractFind in source - supportedReviewer 1Methylcobalamin has an excellent safety profile with no unexpected adverse events.The safety analysis shows similar rates of adverse events between groups, with no serious adverse events or deaths related to treatment.Evidence: Safety results: 88 (75.2%) in methylcobalamin group vs 95 (81.2%) in placebo group had any adverse event; no serious adverse events or deaths.
“The two groups showed similar incidence of any other adverse events (88 (75.2%) in the methylcobalamin group and 95 (81.2%) in the placebo group). No methylcobalamin specific adverse events were observed.”
AbstractFind in source - supportedReviewer 2Methylcobalamin is safe with no unexpected safety concerns.The safety analysis shows similar adverse event rates between groups and no methylcobalamin-specific events.Evidence: Table 3: any adverse event 75.2% vs 81.2% (P=0.27); grade 3-4 events 10.3% vs 15.4%; no SAEs.
Overall, 183 (78.2%) patients experienced other treatment related adverse events of any grade: 88 (75.2%) in the methylcobalamin group and 95 (81.2%) in the placebo group. ... No patient experienced serious adverse events, and no deaths caused by adverse events were reported.
Resultsreviewer’s wording - supportedReviewer 2The findings support the use of methylcobalamin to prevent capecitabine associated severe hand-foot syndrome in this patient population.The primary efficacy endpoint supports this claim, and safety data show no added risks. The conclusion is proportional to the evidence from this single trial.Evidence: Primary endpoint result and safety comparison.
“Oral methylcobalamin, compared with placebo, significantly lowered the severity of hand-foot syndrome by reducing the incidence of grade ≥2 hand-foot syndrome with excellent safety among women with HER2 negative early breast cancer who received adjuvant capecitabine treatment. These findings support the use of methylcobalamin to prevent capecitabine associated severe hand-foot syndrome in women with HER2 negative early breast cancer receiving adjuvant capecitabine treatment.”
ConclusionFind in source - supportedReviewers 2, 3Oral methylcobalamin significantly lowered the severity of hand-foot syndrome by reducing the incidence of grade ≥2 symptoms.The primary endpoint analysis shows a statistically significant reduction in grade ≥2 hand-foot syndrome with methylcobalamin vs placebo (14.5% vs 29.1%, one-sided P=0.003).Evidence: Primary endpoint in Results: 'The incidence of grade ≥2 hand-foot syndrome was lower in the methylcobalamin group (17 of 117, 14.5%) than the placebo group (34 of 117, 29.1%; unadjusted risk difference −14.5%, 95% confidence interval −24.9% to −4.1%; Cochran-Mantel-Haensel one sided P value=0.003).'
The incidence of grade ≥2 hand-foot syndrome was lower in the methylcobalamin group (17 of 117, 14.5%) than the placebo group (34 of 117, 29.1%)
Resultsreviewer’s wording - supportedReviewer 3Methylcobalamin did not increase safety concerns.Adverse events were similar between groups, and no methylcobalamin-specific events were observed.Evidence: Safety results: 'Overall, 183 (78.2%) patients experienced other treatment related adverse events of any grade: 88 (75.2%) in the methylcobalamin group and 95 (81.2%) in the placebo group... No methylcobalamin specific adverse events were observed.'
“No methylcobalamin specific adverse events were observed.”
ResultsFind in source - supportedReviewer 3The findings support the use of methylcobalamin to prevent capecitabine associated severe hand-foot syndrome.The primary endpoint demonstrates efficacy, and safety data are reassuring. The claim is proportional to the evidence.Evidence: Conclusion: 'These findings support the use of methylcobalamin to prevent capecitabine associated severe hand-foot syndrome in women with HER2 negative early breast cancer receiving adjuvant capecitabine treatment.'
These findings support the use of methylcobalamin to prevent capecitabine associated severe hand-foot syndrome in this patient population.
Conclusionreviewer’s wording
Efficacy claim is anchored to an adequate endpoint and a meaningful effect.
- ADEQUATESurrogate endpointThe primary endpoint is the incidence of grade ≥2 hand-foot syndrome, a directly assessed clinical symptom (graded per NCI-CTCAE), not a biomarker or surrogate. It reflects a clinically meaningful outcome that affects treatment adherence and quality of life.
“The primary endpoint was the incidence of grade ≥2 hand-foot syndrome occurring for the first time during capecitabine treatment.”
- ADEQUATEEffect sizeThe incidence of grade ≥2 hand-foot syndrome was reduced from 29.1% in placebo to 14.5% in methylcobalamin (absolute risk difference −14.5%, 95% CI −24.9% to −4.1%, P=0.003). This is a substantial and clinically meaningful reduction, as grade ≥2 hand-foot syndrome can lead to dose reduction or discontinuation of capecitabine and impairs quality of life.
“Grade ≥2 hand-foot syndrome occurred in 17 (14.5%) of 117 patients in the methylcobalamin group and 34 (29.1%) of 117 patients in the placebo group (risk difference −14.5%, 95% confidence interval −24.9% to −4.1%; one sided P value=0.003).”
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
4 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
- Data/code availability incompleteAssessed
- Ethics/consent reporting incompleteAssessed
- Key resources under-identified (antibodies, cell lines, RRIDs)Assessed
The introduction cites prior studies on the incidence and impact of hand-foot syndrome, the ineffectiveness of pyridoxine, the safety concerns with celecoxib, and the uncertainty of urea-based cream. It also discusses the potential role of small fibre neuropathy and the neuroprotective effects of methylcobalamin, providing a logical rationale for the hypothesis. The limitations of prior research (e.g., safety concerns, uncertain efficacy) are explicitly acknowledged and addressed by the study design.
“Several interventions have been investigated to prevent capecitabine associated hand-foot syndrome, including pyridoxine (vitamin B6), the cyclooxygenase 2 enzyme inhibitor celecoxib, and urea based cream. Unfortunately, pyridoxine did not significantly reduce the risk of hand-foot syndrome, and the effect of urea based cream in preventing grade 2 or 3 disease remains uncertain. Although celecoxib has been reported as effective in preventing hand-foot syndrome, its long term use is limited because of safety concerns about cardiovascular events, resulting in a relatively low grade of recommendation.”
“Methylcobalamin (vitamin B12) has been found to promote neurite outgrowth and neuronal survival by enhancing Erk 1/2 and Akt activities through the methylation cycle in vitro and in vivo, alleviating neuropathic pain and dysesthesia, and making it a common treatment for neuropathy in clinical settings. We hypothesised that methylcobalamin might decrease the risk of hand-foot syndrome among patients with HER2 negative early breast cancer receiving adjuvant capecitabine treatment.”
“Several interventions have been investigated to prevent capecitabine associated hand-foot syndrome, including pyridoxine (vitamin B6), the cyclooxygenase 2 enzyme inhibitor celecoxib, and urea based cream. Unfortunately, pyridoxine did not significantly reduce the risk of hand-foot syndrome, and the effect of urea based cream in preventing grade 2 or 3 disease remains uncertain. Although celecoxib has been reported as effective in preventing hand-foot syndrome, its long term use is limited because of safety concerns about cardiovascular events, resulting in a relatively low grade of recommendation.”
“Previous studies identified that small fibre neuropathy might be a potential cause of the pain and palmoplantar dysesthesia in the palms and soles in capecitabine induced hand-foot syndrome. Methylcobalamin (vitamin B12) has been found to promote neurite outgrowth and neuronal survival by enhancing Erk 1/2 and Akt activities through the methylation cycle in vitro and in vivo, alleviating neuropathic pain and dysesthesia, and making it a common treatment for neuropathy in clinical settings.”
“We hypothesised that methylcobalamin might decrease the risk of hand-foot syndrome among patients with HER2 negative early breast cancer receiving adjuvant capecitabine treatment.”
Randomization was performed using a computer-generated code with stratification by hormone receptor status and study centre, and a block size of six. Blinding was double-blind with participants, investigators, and site staff masked. An a priori power analysis was conducted with 80% power to detect a reduction from 30% to 14% in grade ≥2 hand-foot syndrome, with a one-sided alpha of 0.025 and 10% dropout. Inclusion and exclusion criteria are pre-specified and detailed in the supplementary appendix. The use of opaque sealed envelopes is noted as a potential limitation, but the paper acknowledges this and describes measures to ensure allocation concealment. Outlier handling is not explicitly discussed, but the analysis population (ITT, per-protocol, safety) is defined. Controls are adequate (placebo). Independent replication is not applicable for a single pivotal trial.
“Randomisation was performed using a computer generated code created with R studio software (version 4.2.2), stratified by hormone receptor status (positive—oestrogen receptor or progesterone receptor ≥1% v negative—oestrogen receptor and progesterone receptor <1%) and study centres with a block size of six.”
“The participants, investigators, and site staff were masked to treatment allocation.”
“A target sample size of 234 participants was determined based on a grade ≥2 hand-foot syndrome rate of 30% for the placebo group, derived from the results of a trial evaluating celecoxib. This trial had 80% power to detect a reduction in hand-foot syndrome rate to 14% in the methylcobalamin group, with a one sided level of significance of 0.025 and a dropout rate of 10%.”
“Randomisation was performed using a computer generated code created with R studio software (version 4.2.2), stratified by hormone receptor status (positive—oestrogen receptor or progesterone receptor ≥1% v negative—oestrogen receptor and progesterone receptor <1%) and study centres with a block size of six.”
“The participants, investigators, and site staff were masked to treatment allocation.”
“A target sample size of 234 participants was determined based on a grade ≥2 hand-foot syndrome rate of 30% for the placebo group, derived from the results of a trial evaluating celecoxib. This trial had 80% power to detect a reduction in hand-foot syndrome rate to 14% in the methylcobalamin group, with a one sided level of significance of 0.025 and a dropout rate of 10%.”
“The participants, investigators, and site staff were masked to treatment allocation.”
Sex is reported (all women), but no justification for the single-sex study is provided. Age and health status (ECOG, organ function) are reported, but weight is omitted. Demographics include age, menopausal status, TNM stage, hormone receptor status, and previous chemotherapy, but race/ethnicity is not explicitly listed (only stated as 'Chinese participants' in limitations). The baseline table does not include weight or race. These omissions lead to a fail status under the scoring rule (only 1 of 4 applicable sub-criteria adequate).
“Table 1 Baseline characteristics of patients with HER2 negative early breast cancer”
“Eligible participants were women aged 18-75 years”
“The median age of the 234 patients at randomisation was 50 years (interquartile range 41-59), and 53.0% were premenopausal. Most patients had TNM stage II tumours (69.7%). A total of 66.7% of patients had triple negative breast cancer, and 94.0% had previously received chemotherapy.”
The paper states 'Ethics approval was obtained from the ethics committee of each hospital' but does not name the specific committees or provide protocol numbers. Informed consent is adequately described (written informed consent). Regulatory compliance is stated (Declaration of Helsinki). The absence of named ethics committees and protocol numbers means the irb_ethics_statement is reported_but_inadequate, triggering a warn status.
“Ethics approval was obtained from the ethics committee of each hospital and all participants gave written informed consent.”
“The study was conducted in accordance with the Declaration of Helsinki.”
“all participants gave written informed consent.”
“Ethics approval was obtained from the ethics committee of each hospital”
“all participants gave written informed consent”
“The study was conducted in accordance with the Declaration of Helsinki.”
“Ethics approval was obtained from the ethics committee of each hospital and all participants gave written informed consent.”
“The study was conducted in accordance with the Declaration of Helsinki.”
The paper reports the dose and regimen of methylcobalamin (0.5 mg three times daily) and capecitabine (2000 mg/m2/day), but does not provide the manufacturer or source for these drugs. The placebo supplier is acknowledged (Yangtze River Pharmaceutical Group). Software (R studio version 4.2.2) is adequately identified. No other biological resources (antibodies, cell lines, etc.) are used. With 1 of 2 applicable sub-criteria adequate (software_tools_identified), the dimension receives a warn.
“Patients were given 0.5 mg of methylcobalamin orally three times daily (a dosage used in neuropathy) or a placebo to be taken for a maximum of 24 weeks. The physical properties of the methylcobalamin and placebo treatment, such as appearance, size, colour, dosage form, weight, taste, and odour, were the same.”
“All statistical analyses were performed using R studio software (version 4.2.2).”
“The support provided by Yangtze River Pharmaceutical Group for supplying the placebo used in this study is appreciated.”
“Patients were given 0.5 mg of methylcobalamin orally three times daily (a dosage used in neuropathy) or a placebo”
“All patients received oral capecitabine 2000 mg/m 2 /day, which was divided into two daily doses”
“All statistical analyses were performed using R studio software (version 4.2.2).”
“All statistical analyses were performed using R studio software (version 4.2.2).”
The primary analysis uses the Cochran-Mantel-Haenszel test with one-sided p-value, and exact p-values are reported (e.g., P=0.003). Effect sizes are reported with 95% confidence intervals. The statistical software (R studio version 4.2.2) is identified. Data presentation includes per-group n, risk differences, and confidence intervals. Assumptions are not explicitly verified for normality/equal variance, but for a large-sample trial with a binary outcome, the CMH test is appropriate and assumptions are handled by design. Mathematical plausibility checks are not applicable for continuous outcomes and large N. The paper reports both ITT and per-protocol analyses.
“All statistical analyses were performed using R studio software (version 4.2.2).”
“The difference between the groups (P value) was evaluated using the Cochran-Mantel-Haensel test, adjusting for the stratification factors (hormone receptor status and study centre).”
“−14.5 (−24.9 to−4.1); −14.1 (−25.0 to−3.2)* | 0.003†”
“Fig 1 CONSORT (consolidated standards of reporting trials) diagram of trial”
“Cochran-Mantel-Haensel one sided P value=0.003”
“unadjusted risk difference −14.5%, 95% confidence interval −24.9% to −4.1%”
“All statistical analyses were performed using R studio software (version 4.2.2).”
The paper includes a data availability statement that names a repository (Dryad with DOI: https://doi.org/10.5061/dryad.gmsbcc31k) and states that the code is in supplementary appendix 3. The data repository deposit is adequate (Dryad is a recognised repository). However, the code is not deposited in a version-controlled public repository (e.g., GitHub with Zenodo DOI), making code_sharing reported_but_inadequate. This results in a warn status (2 of 3 applicable sub-criteria adequate).
“The raw database underlying the study is openly and publicly available ( https://doi.org/10.5061/dryad.gmsbcc31k ).”
“The code used to analyse the data in the paper can be found in supplementary appendix 3.”
“The raw database underlying the study is openly and publicly available ( https://doi.org/10.5061/dryad.gmsbcc31k ).”
“The code used to analyse the data in the paper can be found in supplementary appendix 3.”
“The raw database underlying the study is openly and publicly available ( https://doi.org/10.5061/dryad.gmsbcc31k ).”
“The code used to analyse the data in the paper can be found in supplementary appendix 3.”
Methods are detailed enough for replication, including randomization, blinding, intervention, endpoints, and statistical analysis. The trial is registered at ClinicalTrials.gov (NCT05165069). A CONSORT diagram is provided (Figure 1). Limitations are discussed in the Discussion section, including the single-country population, lack of patient involvement, potential selection bias from sealed envelopes, and lack of biomarker data. Conclusions are proportional to the evidence, noting the need for further confirmation. Funding sources and conflicts of interest are disclosed. The paper does not explicitly reference a reporting guideline (e.g., CONSORT checklist), but the CONSORT diagram is provided.
“Trial registration ClinicalTrials.gov NCT05165069”
“Trial registration ClinicalTrials.gov NCT05165069 (https://clinicaltrials.gov/ct2/show/NCT05165069) .”
“Trial registration ClinicalTrials.gov NCT05165069”
“Limitations of this study This study has several limitations.”
Registered (1 ID: ClinicalTrials.gov). Reporting guideline cited: CONSORT.
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 37 references by DOI: 35 verified — 2 no DOI (shown, not verified).
- NO DOIThe hand-foot syndrome—a frequent secondary manifestation in antineoplastic chemotherapyNo DOI in the reference — shown for manual review; not independently verifiable (not a fabrication signal).
- NO DOIEffect of celecoxib on capecitabine-induced hand-foot syndrome and antitumor activityNo DOI in the reference — shown for manual review; not independently verifiable (not a fabrication signal).
2 data/code links checked; 2 live.
- dataDryadLIVEHTTP 200https://doi.org/10.5061/dryad.gmsbcc31kResolves to Dryad (data repository).
- datahttps://clinicaltrials.gov/ct2/show/NCT05165069LIVEHTTP 200Resolves, but the content could not be matched to the paper.
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 consistency, typo, clarity.
- MINORtypoMethods, Statistical analysis“Cochran-Mantel-Haensel test”→ Cochran-Mantel-Haenszel testMissing 'z' in 'Haenszel'.
- MINORconsistencyResults, Secondary endpoints“two patients (6.8%) in the methylcobalamin group and three (11.1%) in the placebo group reduced capecitabine owing to other adverse events”→ The percentages 6.8% and 11.1% appear to be calculated from denominators of 117, but the text says 'two patients' and 'three patients', which would be 1.7% and 2.6%, respectively. Verify the numbers.Potential inconsistency: 2/117 = 1.7%, not 6.8%; 3/117 = 2.6%, not 11.1%. The percentages may refer to a different subgroup.
- MINORclarityResults, Secondary endpoints“Additionally, two patients (6.8%) in the methylcobalamin group and three (11.1%) in the placebo group reduced capecitabine owing to other adverse events”→ Clarify the denominator for these percentages, as they do not match the total group size.The percentages are inconsistent with the stated numbers of patients.
- MINORconsistencyThroughout“P value”→ Use consistent capitalization for 'P value' (e.g., 'P value' vs 'P'). The paper uses both 'P value' and 'P'.Minor inconsistency, does not affect clarity.
- MINORotherTable 2“−14.5 (−24.9 to−4.1); −14.1 (−25.0 to−3.2)*”→ Add a space after 'to' for consistency: '−14.5 (−24.9 to −4.1); −14.1 (−25.0 to −3.2)*'Missing space before the minus sign.
- MINORtypoTable 2 footnote“From log-binomial regression adjusting for stratification factors (hormone receptor status and study centre).”→ Change 'log-binomial' to 'log binomial' (elsewhere they use 'log binomial' without hyphen).Minor inconsistency.
- MINORconsistencyAbstract, authors list“Xia Yuan doctoral student Zhu Yingying associate research professor Ling Li professor Xu Fei professor Yang Yaping associate professor Ye Jinhui professor Tan Weige associate professor Chen Zouxiang statistician Liu Qiang professor Wei Wei professor Zhang Jun professor Zhang Ailing professor Zhang Lehong professor Song Erwei professor Gong Chang professor”→ Reformat as a standard author list with titles separated by commas or on separate lines.The author list in the abstract appears as a continuous string without clear separation.
- MINORpunctuationResults, Secondary endpoints“Eight patients (6.8%) in the methylcobalamin group and 13 (11.1%) in the placebo group reduced capecitabine because of hand-foot syndrome ()”→ Remove the empty parentheses or add a citation.The parentheses are empty, likely a formatting artifact.
- MINORclarityDiscussion, comparison with other studies“However, the increased risk of myocardial infarction associated with celecoxib (hazard ratio 2.26, 95% confidence interval 1.0 to 5.1) might limit its clinical use.”→ Consider adding a reference for the hazard ratio.The hazard ratio is cited without a reference number; the paper should include a citation.
As a published work, the trial is fundamentally sound and the headline result is supported; an informed reader should weigh the reporting gaps (missing baseline weight/race-ethnicity and sex justification, unnamed ethics committees, unstated drug source, code not in a versioned repository) as transparency issues rather than validity threats. The numeric inconsistency in the secondary-endpoint percentages (2/117=1.7% and 3/117=2.6% reported as 6.8% and 11.1%) warrants verification and likely a correction/erratum.
- 1.HIGHstatisticsVerify and correct the percentages in 'Results, Secondary endpoints': 'two patients (6.8%)... three (11.1%)... reduced capecitabine owing to other adverse events' — 2/117=1.7% and 3/117=2.6%, so the reported percentages are inconsistent unless they refer to a different subgroup whose denominator is stated.An internally inconsistent/impossible number in a published paper warrants correction via erratum; as written it is a data transcription error.
- 2.HIGHreportingAdd baseline weight/BMI and an explicit race/ethnicity row to Table 1, and add a sentence in Methods justifying enrolment of women only (e.g., 'Male breast cancer is extremely rare; only women were enrolled').These are standard demographic variables whose absence (and the unaddressed single-sex design) is the reason the biological_variables dimension fails.
- 3.HIGHethicsName the ethics committees of each of the seven hospitals and provide approval/protocol numbers in the Methods 'Study design and patients' section.An ethics statement that does not identify the approving bodies is a reporting gap an informed reader or journal may require to be closed.
- 4.HIGHdata codeDeposit the analysis code in a version-controlled public repository (e.g., GitHub with a Zenodo DOI) and reference it in the Data availability statement, instead of only supplementary appendix 3.Code in a supplementary appendix lacks versioning and a persistent identifier, which weakens reproducibility claims.
- 5.HIGHrigorState the manufacturer/source (and lot numbers/expiry if available) of methylcobalamin and capecitabine in the Methods 'Intervention' section.Identifying the investigational products' source is required for full reagent/resource identification and reproducibility.
- 6.MEDIUMreportingAdd an explicit statement in Methods that the trial is reported in accordance with the CONSORT 2010 statement (the CONSORT diagram is already provided).Explicit adherence to a reporting guideline closes the one transparency gap flagged by all reviewers.
- 7.MEDIUMcopyeditCorrect the typo 'Cochran-Mantel-Haensel' to 'Cochran-Mantel-Haenszel' in Methods, Statistical analysis.Missing 'z' in the test name is a factual typo in a key methodological term.
- 8.MEDIUMcopyeditRemove the empty parentheses after 'hand-foot syndrome' in Results, Secondary endpoints (the sentence about 'Eight patients (6.8%)... 13 (11.1%) reduced capecitabine because of hand-foot syndrome ()').Empty parentheses are a formatting artifact that should be cleaned up.
- 9.MEDIUMcopyeditAdd a space in Table 2 risk differences: '−24.9 to−4.1' → '−24.9 to −4.1' (and the analogous value).Missing spaces before the minus signs are minor readability/presentation errors.
- 10.MEDIUMcopyeditReformat the abstract author list (currently a continuous string of names and titles) into a standard, clearly separated list.The author list as printed is unreadable and would be corrected during production.
- 11.LOWcopyeditAdd a citation/reference for the celecoxib hazard ratio (2.26, 95% CI 1.0 to 5.1) in the Discussion.The hazard ratio is currently given without a supporting reference.
- 12.LOWcopyeditStandardize the capitalization of 'P value' throughout the manuscript.Inconsistent capitalization of a statistical term is a minor style issue.
- 13.LOWcopyeditUse consistent 'log binomial' (no hyphen) in the Table 2 footnote, matching usage elsewhere.Hyphenation of the model name is inconsistent between sections.
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.