Verified Rigor RecordRigor Reviewed

Rigor-Reviewed Manuscript

Graft-versus-Host Disease Prophylaxis with Cyclophosphamide and Cyclosporin

Name: Verified Rigor Record — Graft-versus-Host Disease Prophylaxis with Cyclophosphamide and Cyclosporin
Item: Graft-versus-Host Disease Prophylaxis with Cyclophosphamide and Cyclosporin
Rating: 83
Author: Αlpha¹

Curtis DJ, Patil SS, Reynolds J, Purtill D, Lewis C, Ritchie DS, Gottlieb DJ, Yeung DT, Wong E, Tey SK, Perera T, Moore J, Koldej RM, De Abreu Lourenco R, Stubbs J, Morrissey CO, Munsef N, Arenas A, Hill GR, Australasian Leukaemia and Lymphoma Group.

Paper source

Open DOI

Αlpha¹ Editorial

This rigor review was examined and confirmed by Αlpha¹ Editorial · July 6, 2026

Αlpha¹ Rigor Review

Rigor Reviewed

WARN

Warnings to address · Interventional Study

Eight dimensions — click a segment for details

Critical

High-priority

Model-reported confidence: highReviewer agreement: moderateEngine: 7.6.0Source: full text

This Αlpha¹ Rigor Review uses AI 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 manuscript reports a well-conducted phase 3 randomized trial with a strong scientific premise, robust statistical methods, and transparent reporting. The main rigor gaps are the absence of a data availability statement (critical for a clinical trial), no regulatory compliance framework referenced in ethics, and inadequate identification of key resources (drug manufacturers and statistical software).

Three independent reviewers scored all eight dimensions, all had access to full-text. The statistical recomputation covered 1 test (consistent). Citation checking found 1 reference that could not be found in the registry; retracted references were 0. The copyedit flagged 3 minor issues. The reviewers broadly agreed, with minor divergences on ethical approvals and key resources resolved by examining specific checklist criteria.

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

Recomputed 1 test: 1 consistent, 0 inconsistent, 1 via agent-written checks.

Rigor

Eight Dimensions5 pass · 2 warn · 1 fail

9 major claims1 partial8 supported

The background cites numerous studies on GVHD and PTCy, explains why PTCy-cyclosporin is promising, and identifies gaps in prior trials (e.g., HOVON-96 limited to non-myeloablative conditioning, BMT CTN 1703 post-hoc analysis). All three sub-criteria are adequately reported.

Sub-criteria

prior work citedADEQUATE

premise rationaleADEQUATE

limitations addressedADEQUATE

Per reviewer (3) — the votes behind the verdict

PASSReviewer 1· 90% conf

The scientific premise is well-established, citing prior randomized and non-randomized studies, and clearly outlining the rationale for the trial and how it addresses limitations of prior work.

Evidence

direct quote[Background, paragraphs 4-5]

“The BMT CTN 1703 trial assessed triple GVHD prophylaxis in recipients of PBSCs from a matched related or unrelated donor receiving reduced-intensity or non-myeloablative conditioning.”

direct quote[Background, paragraph 6]

“As a result, the benefit of PTCy for PBSCT using a MRD remains uncertain, and unknown in the context of myeloablative conditioning, which was excluded from both the BMT CTN 1703 and HOVON-96 trials.”

direct quote[Background, final paragraph]

“Here, we present the results of the phase 3 randomized ALLG BM12 CAST trial, which aimed to establish the benefit of the dual GVHD prophylaxis regimen (PTCy-cyclosporin) for PBSCT using a MRD in the context of either reduced-intensity or myeloablative conditioning.”

PASSReviewer 2· 95% conf

The paper establishes a clear scientific foundation by citing prior work, providing a logical rationale, and addressing limitations of previous studies.

Evidence

direct quote[Background, paragraphs 2-4]

“The BMT CTN 1703 trial assessed triple GVHD prophylaxis in recipients of PBSCs from a matched related or unrelated donor...”

direct quote[Background, final paragraph]

paraphrase[Discussion, paragraph 2]

“The authors note that HOVON-96 was limited by small MRD number and exclusively non-myeloablative conditioning.”

PASSReviewer 3· 95% conf

The paper clearly establishes the scientific premise by citing relevant prior work, providing a logical rationale, and discussing limitations of prior research.

Evidence

direct quote[Background, paragraphs 2-4]

“The BMT CTN 1703 trial assessed triple GVHD prophylaxis in recipients of PBSCs from a matched related or unrelated donor receiving reduced-intensity or non-myeloablative conditioning.”

direct quote[Background, paragraph 5]

“the benefit of PTCy for PBSCT using a MRD remains uncertain, and unknown in the context of myeloablative conditioning, which was excluded from both the BMT CTN 1703 and HOVON-96 trials.”

direct quote[Background, final sentence]

The trial is a phase 3 open-label RCT. Randomization was stratified by age and conditioning intensity, but the specific method (e.g., computer-generated sequence) is not stated. Sample size and power are provided. Inclusion/exclusion criteria are listed, and ITT analysis is pre-specified. Open-label design is acknowledged. Six applicable criteria: 5 adequate, 1 inadequate (randomization method).

Sub-criteria

controlsN/A

power analysisADEQUATE

blinding levelsADEQUATE

outlier handlingADEQUATE

randomization unitADEQUATE

inclusion exclusionADEQUATE

randomization methodINADEQUATE

replicate distinctionN/A

independent replicationN/A

Per reviewer (3) — the votes behind the verdict

PASSReviewer 1· 85% conf

The study design is robust: a prospective randomized trial with pre-specified stratification, sample-size calculation, and intention-to-treat analysis. Applicable sub-criteria (randomization, power analysis, inclusion/exclusion, controls) are adequately reported.

Evidence

direct quote[Methods, Statistical Analysis paragraph 1]

“A total of 134 patients were randomized to the two treatment arms using a 1:1 allocation ratio”

direct quote[Methods, Statistical Analysis paragraph 1]

“A sample size of 134 patients, guided by results of the single arm PTCy-cyclosporin study, was selected to achieve an 80% power for a two-sided log-rank test (a=0.05) to detect a hazard ratio of 0.54, corresponding to 86 total primary events.”

direct quote[Methods, Study Design paragraph 1]

“prospective, open-label, two-arm randomized phase 3 trial”

PASSReviewer 2· 90% conf

The study design is robust with a clear randomization scheme, power analysis, and ITT analysis, though the randomization method is not fully described.

Evidence

direct quote[Methods, Statistical Analysis paragraph]

“A total of 134 patients were randomized to the two treatment arms using a 1:1 allocation ratio”

direct quote[Methods, Statistical Analysis paragraph]

“A sample size of 134 patients... was selected to achieve an 80% power for a two-sided log-rank test (a=0.05) to detect a hazard ratio of 0.54”

direct quote[Methods, Study Design paragraph]

“open-label... randomized phase 3 trial”

PASSReviewer 3· 80% conf

The study design is adequately reported for a human RCT, including power analysis, inclusion/exclusion criteria, and analysis population, though the randomization method and blinding are not fully detailed.

Evidence

direct quote[Methods, Statistical Analysis]

“A total of 134 patients were randomized to the two treatment arms using a 1:1 allocation ratio, following the decision to transplant but not more than 28 days before transplant admission to minimize dropout before PBSCT. Randomization was stratified by age (dichotomised at 50 years) and intensity of the conditioning regimen (myeloablative or reduced intensity).”

direct quote[Methods, Statistical Analysis]

direct quote[Methods, Study Design]

“The Australasian Leukaemia & Lymphoma Group (ALLG) BM12 CAST trial is a prospective, open-label, two-arm randomized phase 3 trial comparing cyclosporin and methotrexate (control arm) with PTCy and cyclosporin (investigational arm) for GVHD prophylaxis.”

Table 1 reports age, sex, CMV status, disease, comorbidity index, and donor sex. Age and health status (comorbidity-age index, disease risk index) are provided. Demographics are comprehensive. Both sexes are included, so sex justification is not applicable.

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

Key human biological variables (age, sex, health status/conditioning intensity, disease type) are well-reported in Table 1. Species/strain and housing conditions are not applicable. Demographics are adequately reported.

Evidence

direct quote[Table 1]

“Male sex – n (%) 37 (54.4) 37 (56.1) 74 (55.2)”

direct quote[Table 1]

“Median, years (range) 56.2 (22-70) 55.3 (21-71) 55.7 (21-71)”

direct quote[Table 1]

“Primary disease – n (%) | Acute myeloid leukemia 44 (64.7) 38 (57.6) 82 (61.2)”

PASSReviewer 2· 95% conf

Biological variables are adequately reported including sex, age, health status, and demographics.

Evidence

direct quote[Table 1]

“Male sex – n (%) 37 (54.4) 37 (56.1) 74 (55.2)”

direct quote[Table 1, Results paragraph 1]

“The median age was 55.7 years... 44% received myeloablative conditioning.”

PASSReviewer 3· 85% conf

Sex, age/weight/health status, and most demographics are reported, but race/ethnicity is not included, which is a minor reporting gap.

Evidence

direct quote[Table 1]

“Male sex – n (%): 37 (54.4), 37 (56.1), 74 (55.2)”

direct quote[Table 1]

“Comorbidity-age index – n (%): 0, 1-2, 3-4, 5+”

absence[Table 1]

Race/ethnicity is not reported in the demographics.

The paper states approval by the Human Research Ethics Committee at each site and that all patients provided informed consent. However, there is no statement of adherence to the Declaration of Helsinki, ICH-GCP, or other regulatory standards. For a human clinical trial, such a statement is expected.

Sub-criteria

iacuc statementN/A

informed consentADEQUATE

irb ethics statementADEQUATE

regulatory complianceNOT REPORTED

Per reviewer (3) — the votes behind the verdict

PASSReviewer 1· 90% conf

Ethical approvals, informed consent, and regulatory compliance are explicitly reported with institutional and registry details.

Evidence

direct quote[Methods, Study Design paragraph 1]

“approved by the Human Research Ethics Committee at each participating site”

direct quote[Methods, Study Design paragraph 1]

“All patients provided informed consent to participate.”

direct quote[Methods, Study Design paragraph 1]

“The trial was registered with the Australian New Zealand Clinical Trials Registry (ACTRN12618000505202)”

WARNReviewer 2· 80% conf

Ethics approval and informed consent are reported but regulatory compliance (e.g., Declaration of Helsinki) is not explicitly stated.

Evidence

direct quote[Methods, Study Design paragraph]

“approved by the Human Research Ethics Committee at each participating site”

direct quote[Methods, Study Design paragraph]

“All patients provided informed consent to participate.”

absence[Ethics section]

No statement of regulatory compliance.

WARNReviewer 3· 90% conf

Ethics approval and informed consent are reported, but compliance with a recognized regulatory framework (e.g., Declaration of Helsinki) is not explicitly stated, which is a reporting gap.

Evidence

direct quote[Methods, Study Design]

“The trial was registered with the Australian New Zealand Clinical Trials Registry (ACTRN12618000505202) and approved by the Human Research Ethics Committee at each participating site.”

direct quote[Methods, Study Design]

“All patients provided informed consent to participate.”

absence[Methods]

No regulatory compliance framework is named.

The drugs cyclophosphamide, cyclosporin, and methotrexate are described with doses and regimen, which is adequate for a clinical trial. However, the statistical software used for analysis is not mentioned, which is a common reporting gap.

Sub-criteria

mycoplasma testingN/A

reagents identifiedADEQUATE

organisms identifiedN/A

antibodies identifiedN/A

cell line authenticationN/A

software tools identifiedINADEQUATE

Per reviewer (3) — the votes behind the verdict

WARNReviewer 1· 70% conf

The investigational product (cyclophosphamide, cyclosporin, methotrexate) is named but not identified with manufacturer/lot numbers — this is reported_but_inadequate for reagents_identified. Software tools are identified (SAS? not stated). Other resource types are not applicable.

Evidence

direct quote[Methods, Trial procedures]

“methotrexate (intravenous bolus of 15 mg per meter squared on day 1 and 10 mg per meter squared on days 3, 6 and 11) and cyclosporin”

absence[Methods, Statistical Analysis]

Statistical software is not identified.

paraphrase[Methods, Trial procedures]

“The drugs cyclophosphamide, cyclosporin, methotrexate are described only by dose and schedule, without manufacturer or supplier.”

WARNReviewer 2· 85% conf

Investigational drugs are named with dosing schedules, but statistical software is not identified.

Evidence

direct quote[Methods, Trial procedures]

“cyclophosphamide (50 mg per kilogram on days 3 and 4 after transplant”

absence[Methods, Statistical Analysis]

Statistical software not identified.

FAILReviewer 3· 85% conf

The investigational drugs are named with doses, but manufacturers/sources are not provided, and statistical software is not identified, leading to inadequate reporting of key resources.

Evidence

direct quote[Methods, Trial procedures]

“The investigational GVHD prophylaxis consisted of cyclophosphamide (50 mg per kilogram on days 3 and 4 after transplant followed by cyclosporin (dosing and monitoring levels as per institutional guidelines)”

direct quote[Methods, Statistical Analysis]

“Statistical analysis was performed, unless otherwise specified, using the Intention to Treat (ITT) principle with all randomized subjects analyzed according to the assigned treatment arm.”

direct quote[Methods, Trial procedures]

“The control GVHD prophylaxis consisted of methotrexate (intravenous bolus of 15 mg per meter squared on day 1 and 10 mg per meter squared on days 3, 6 and 11)”

Tests include log-rank, Cox regression, cumulative incidence. HR and 95% CIs are reported. P-values for primary endpoint is given as '<0.0001' but not an exact value. Secondary endpoints report CIs only. Software is not specified. Data presentation includes KM curves and tables. Mathematical plausibility checks reveal no discrepancies.

Sub-criteria

tests namedADEQUATE

exact p valuesINADEQUATE

effect sizes ciADEQUATE

data presentationADEQUATE

software identifiedINADEQUATE

assumptions verifiedADEQUATE

mathematical plausibilityPASS

Per reviewer (3) — the votes behind the verdict

PASSReviewer 1· 85% conf

Statistical methods are appropriate and well-described: named tests (log-rank, Cox PH, Kaplan-Meier, cumulative incidence), exact p-values reported, effect sizes with 95% CIs, software not named but analysis package referenced. Data presentation includes KM curves, cumulative incidence plots, and baseline tables with n per group.

Evidence

direct quote[Results, Primary Endpoint]

“GRFS was significantly improved in the PTCy group ... Log-Rank test p-value <0.0001”

direct quote[Results, Primary Endpoint]

“estimated hazard ratio (HR) was 0.419 (95% CI:0.267-0.656)”

direct quote[Methods, Statistical Analysis]

“Statistical analysis followed the pre-specified statistical analysis package (available online)”

PASSReviewer 2· 90% conf

Statistical methods are appropriate, effect sizes with CIs are provided, and no arithmetic errors are detected; however, exact p-values are not reported for secondary endpoints and software is not identified.

Evidence

direct quote[Results, PRIMARY ENDPOINT]

“Log-Rank test p<0.0001”

direct quote[Results, PRIMARY ENDPOINT]

“The estimated hazard ratio (HR) was 0.419 (95% CI:0.267-0.656)”

absence[Methods, Statistical Analysis]

Statistical software not identified.

PASSReviewer 3· 80% conf

Statistical methods are adequately described and results are presented with effect sizes and confidence intervals, though exact p-values are not reported (only threshold) and software is not named.

Evidence

direct quote[Methods, Statistical Analysis]

“There was a significant difference in GRFS (Log-Rank test p<0.0001)”

direct quote[Results, Primary Endpoint]

“Log-Rank test p-value <0.0001”

absence[Methods]

Statistical software is not identified anywhere in the manuscript.

For a clinical trial, a data availability statement is expected. The paper does not include any statement about data access. Although the protocol and disclosure forms are available at NEJM.org, no mention is made of sharing individual patient data. Code sharing is not applicable as no custom code is described.

Sub-criteria

code sharingN/A

accession numbersN/A

repository depositN/A

data availability statementNOT REPORTED

Per reviewer (3) — the votes behind the verdict

WARNReviewer 1· 75% conf

A data availability statement is present (protocol online, registry), but no individual patient data or code access is described beyond a vague reference. The statement is reported_but_inadequate.

Evidence

direct quote[Methods, Study Design paragraph 1]

“The full protocol is available at NEJM.org (http://NEJM.org) .”

direct quote[Methods, Study Design paragraph 1]

“The trial was registered with the Australian New Zealand Clinical Trials Registry (ACTRN12618000505202)”

absence[entire paper]

Code availability not reported.

FAILReviewer 2· 95% conf

No data availability statement is provided, and no repository deposition or code sharing is mentioned.

Evidence

absence[Entire paper]

No data availability statement.

FAILReviewer 3· 90% conf

No data availability statement is provided in the manuscript, nor are data deposited or code shared, which is a critical gap for a clinical trial.

Evidence

absence[Entire manuscript]

No data availability statement found in the manuscript.

absence[Entire manuscript]

No repository deposit mentioned.

absence[Entire manuscript]

No code sharing mentioned.

Methods are detailed enough for replication. Trial registration number is provided. All pre-specified outcomes appear to be reported (with some post-hoc analyses noted). Limitations are discussed. Conclusions are proportional. Funding and COI are stated. CONSORT checklist is not mentioned.

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· 85% conf

Reporting is generally transparent: trial is registered, methods are detailed enough for replication, limitations are discussed, conclusions are proportional, and funding/COI are reported. The open-label design is a limitation acknowledged by authors.

Evidence

direct quote[Abstract, Methods]

“ACTRN12618000505202”

direct quote[Discussion, paragraph 4]

“Some limitations of CAST should be noted including the non-blinded nature of the study that could bias GVHD assessments”

direct quote[Funding Sources and Disclosure forms statement]

“The trial was funded by grants from the Australian Government Medical Research Future Fund (MRF1152188) ... Disclosure forms provided by the authors are available with the full text of this article at NEJM.org”

PASSReviewer 2· 90% conf

The paper is well-reported: methods are complete, trial is registered, outcomes are reported, limitations discussed, and funding/COI disclosed, though no reporting guideline is referenced.

Evidence

direct quote[Abstract, Methods]

“ACTRN12618000505202”

direct quote[Discussion, paragraph 4]

“Some limitations of CAST should be noted including the non-blinded nature of the study...”

absence[Methods]

No reporting guideline mentioned.

PASSReviewer 3· 85% conf

The manuscript is generally well-reported with registration, funding, limitations, and proportional conclusions, but does not reference a reporting guideline (e.g., CONSORT).

Evidence

direct quote[Methods, Study Design]

“The trial was registered with the Australian New Zealand Clinical Trials Registry (ACTRN12618000505202)”

direct quote[Discussion, paragraph 3]

“Some limitations of CAST should be noted including the non-blinded nature of the study that could bias GVHD assessments and the lack of adverse events collected beyond the first 100 days after transplant”

absence[Methods]

No reporting guideline (e.g., CONSORT) is referenced.

3 copyedit issues flagged: mostly consistency, clarity.

Checked 35 references: 33 verified — 1 unresolved, 1 not checked.

Reproducibility

1 data/code link checked; 0 live.

Registered (1 ID: ANZCTR). No reporting guideline cited.

Αlpha¹ Rigor Action ItemsFinal pass before journal submission

The manuscript is conceptually and methodologically strong but requires a few critical fixes before submission: add a data availability statement, include a regulatory compliance statement, and identify statistical software. These are straightforward additions that should be made before submitting to a journal.

1.
HIGHdata code
Add a data availability statement in the Methods section specifying how de-identified patient-level data can be accessed (e.g., by reasonable request to the ALLG, or through a repository like Vivli or clinicalstudydatarequest.com).
ICMJE and most journals require a data availability statement for clinical trials; its absence is likely to result in a mandatory revision request.
2.
HIGHethics
Add a sentence to the Methods (Ethics section) stating that the trial was conducted in accordance with the Declaration of Helsinki and/or ICH-GCP guidelines.
This is a standard requirement for human clinical trials and is missing from the current manuscript.
3.
HIGHreporting
Identify the statistical software (e.g., SAS version 9.4, R version 4.2, Stata 17) used for all analyses in the Statistical Analysis section.
Reproducibility requires naming the software and version used; this is a common and easy fix.
4.
MEDIUMreporting
Reference the CONSORT reporting checklist in the Methods and ensure a CONSORT flow diagram is included (Figure 1 appears to show patient flow but should be explicitly labeled as CONSORT).
Journals endorse CONSORT for RCTs; its absence is a reporting transparency gap.
5.
MEDIUMstatistics
Report exact p-values for the primary endpoint (e.g., p=0.00003 rather than p<0.0001) and for secondary endpoints where appropriate, noting any multiplicity adjustment.
Threshold-only p-values ('p<0.0001') are less informative than exact values; this is a reporting best practice.
6.
MEDIUMcopyedit
Revise the Table 2 footnote to clarify that the 'five patients' who did not proceed to transplant are accounted for: 1 death and 4 event-free patients not transplanted.
The copyedit flagged this as a consistency issue that could confuse readers.
7.
MEDIUMcopyedit
Specify a target range or typical dose for cyclosporin (instead of 'as per institutional practice') or state that the dose was per institutional guidelines and provide a reference if the guidelines are published.
Reproducibility is enhanced by a more precise dosing description; the current phrase is vague.
8.
MEDIUMcopyedit
Clarify whether the neutrophil engraftment percentages (94.1% vs 89.4%) refer to all randomized patients or only those who received a transplant.
The denominator is ambiguous and could affect interpretation of these rates.
9.
LOWreporting
Consider adding a brief justification for the open-label design in the Methods, noting any measures taken to minimize bias (e.g., blinded outcome assessment for GVHD).
Addressing this known limitation strengthens transparency and reviewer confidence.
10.
LOWreporting
Describe the randomization method in more detail, e.g., 'computer-generated random sequence with permuted blocks, stratified by age and conditioning intensity'.
A fully described randomization method is expected for a phase 3 trial to allow assessment of allocation concealment.
11.
LOWother
Verify the reference flagged as 'not_found_in_registry' (Phase III study comparing methotrexate and tacrolimus...) and correct or replace it if it cannot be found.
A reference that cannot be located in any registry may indicate a fabrication signal; ensure it exists and is correctly cited.

Full text (45,574 chars)Analyzed: Jul 6, 2026

Αlpha¹ assesses methodological rigor, not the importance of the findings. See how we calibrate →