Pyrotinib or placebo in combination with trastuzumab and docetaxel for HER2 positive metastatic breast cancer: long term survival results from randomised phase 3 PHILA trial.
Ma F, Yan M, Li W, Ouyang Q, Tong Z, Teng Y, Wang Y, Wang S, Geng C, Luo T, Zhong J, Zhang Q, Liu Q, Zeng X, Sun T, Mo Q, Zhou S, Li P, Cheng J, Wang X, Nie J, Yang J, Wu X, Wang X, Li H, Yao G, Fan Y, Lin J, Zhu X, Xu B
- DOI
- 10.1136/bmj-2025-087259
- Record issued
- 2026-08-15
- Engine
- 7.39.0
- Exported
- 2026-09-20
Prepared by Alpha1. This document is confidential: it is intended for the recipient it was shared with and must not be redistributed. The live record at alpha1science.com/verify/bad43a1e-af51-4b4c-9dba-565bea1f3cc6 is authoritative.
How this rating was calculated
- IntegrityIntegrity concern ×3−1.5★
This Kaimen Rigor review uses Kaimen Rigor reviewers trained on a curated corpus of high-fidelity and retracted papers, with expert supervision and curation. It can still make mistakes; verify each finding against the source before relying on it.
This is a well-conducted phase 3 RCT with rigorous design, clear reporting of ethics, resources, and statistics, and a concrete data availability statement. Minor reporting gaps include lack of explicit CONSORT adherence and a small inconsistency in the safety narrative versus Table 2.
Both reviewers classified the study as interventional; no divergence. The statistics verification covered only 3 tests (those with test statistics/CI); the remaining analyses were not machine-verified but no errors were flagged. The citation check found no retracted or unresolved references.
Numerical inconsistencies
1 finding · worst lowValues that contradict each other or are impossible for the stated sample: recomputed p-values and test statistics, GRIM/GRIMMER checks on summary numbers, percentages against their own counts, totals against their parts, and estimates against their own confidence intervals.
- Internal contradictions in the reported numbersAssessed
Recomputed 3 tests: 3 consistent, 0 inconsistent; 3 via agent-written checks.
- CONSISTENTreported p = .004 · recomputed p = .004Reviewers 1, 2Overall survival hazard ratio at 30 April 2024: HR=0.64, 95% CI 0.46-0.89, one-sided P=0.004
“Overall survival was longer in the pyrotinib group (hazard ratio 0.64 (95% confidence interval (CI) 0.46 to 0.89); nominal one-sided P=0.004).”
Taken as given: The hazard ratio is 0.64 with 95% CI 0.46 to 0.89.; The CI is two-sided at 95%.; The reported P is one-sided, so the two-sided p from the CI is halved.Method: Compute two-sided p from HR and CI using normal approximation, then halve for one-sided.How we recomputed it: pCI(0.64, 0.46, 0.89, 1)/2 - CONSISTENTreported p < .001 · recomputed p = <.001Reviewers 1, 2Progression-free survival hazard ratio at 30 April 2024: HR=0.44, 95% CI 0.36-0.53, one-sided P<0.001
“The pyrotinib group showed a sustained benefit in investigator assessed progression-free survival compared with the placebo group, with a median of 22.1 months (95% CI 19.3 to 27.8) versus 10.5 months (95% CI 9.5 to 12.4), and a hazard ratio of 0.44 (95% CI 0.36 to 0.53; nominal one sided P<0.001; ).”
Taken as given: The hazard ratio is 0.44 with 95% CI 0.36 to 0.53.; The CI is two-sided at 95%.; The reported P is one-sided, so the two-sided p from the CI is halved.Method: Compute two-sided p from HR and CI using normal approximation, then halve for one-sided.How we recomputed it: pCI(0.44, 0.36, 0.53, 1)/2 - CONSISTENTreported p = .020 · recomputed p = .017Reviewers 1, 2Overall survival hazard ratio at 30 May 2025: HR=0.74, 95% CI 0.56-0.98, one-sided P=0.02
“Overall survival was longer in the pyrotinib group compared with placebo group (hazard ratio 0.74 (95% CI 0.56 to 0.98); nominal one sided P=0.02; ).”
Taken as given: The hazard ratio is 0.74 with 95% CI 0.56 to 0.98.; The CI is two-sided at 95%.; The reported P is one-sided, so the two-sided p from the CI is halved.Method: Compute two-sided p from HR and CI using normal approximation, then halve for one-sided.How we recomputed it: pCI(0.74, 0.56, 0.98, 1)/2
- lowinternal contradictionThe abstract reports 59 (20%) deaths in the pyrotinib group and 87 (30%) in the placebo group, but the results section reports 59 (20%) and 87 (30%) respectively, which is consistent. However, the long-term analysis reports 85 (29%) and 108 (37%) deaths, which is a different cut-off and not contradictory.
59 (20%) and 87 (30%) patients died, respectively. ... In the long term survival analysis, 85 deaths (29%) occurred in the pyrotinib group and 108 (37%) in the placebo group by 30 May 2025.
Abstractreviewer’s wording - lowinternal contradictionThe text reports diarrhoea of any grade in the placebo group as 55% (n=162) in the safety narrative, while Table 2 lists 54% (n=158). This is a minor inconsistency.
“Diarrhoea of any grade was reported in 99% (n=293) of patients in the pyrotinib group and 55% (n=162) in the placebo group”
ResultsFind in source
Overstated conclusions
None foundConclusions 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.
Checked — nothing surfaced.
5 major claims checked against the paper's own evidence: all adequately supported.
- supportedReviewers 1, 2Pyrotinib plus trastuzumab and docetaxel improves progression-free survival compared with placebo plus trastuzumab and docetaxel in HER2-positive metastatic breast cancer.The claim is supported by the reported hazard ratio of 0.44 (95% CI 0.36 to 0.53) with a nominal one-sided P<0.001, and consistent subgroup analyses.Evidence: Reported hazard ratio 0.44 (95% CI 0.36 to 0.53), median PFS 22.1 vs 10.5 months, and consistent subgroup analyses.
“Improvement in progression-free survival in the pyrotinib group was maintained (22.1 months (95% CI 19.3 to 27.8) v 10.5 months (9.5 to 12.4), hazard ratio 0.44 (95% CI 0.36 to 0.53); nominal one sided P<0.001).”
AbstractFind in source - supportedReviewers 1, 2Pyrotinib plus trastuzumab and docetaxel improves overall survival compared with placebo plus trastuzumab and docetaxel.The claim is supported by the reported hazard ratio of 0.64 (95% CI 0.46 to 0.89) with a nominal one-sided P=0.004, though median OS was not reached.Evidence: Reported hazard ratio 0.64 (95% CI 0.46 to 0.89), nominal one-sided P=0.004, and survival rate differences.
“Overall survival was longer in the pyrotinib group (hazard ratio 0.64 (95% confidence interval (CI) 0.46 to 0.89); nominal one-sided P=0.004).”
AbstractFind in source - supportedReviewer 1The safety profile of pyrotinib plus trastuzumab and docetaxel is consistent with interim findings, with no new safety signals.The claim is supported by the reported adverse event data and the statement that no new safety signals were identified.Evidence: Reported adverse event profiles and the statement 'No new safety signals were identified during the extended follow-up period.'
“The safety profile remained consistent with interim findings, with no new safety signals identified during extended follow-up.”
Results and discussionFind in source - supportedReviewers 1, 2The pyrotinib-based regimen is an effective treatment strategy for HER2-positive metastatic breast cancer.The claim is supported by the efficacy and safety data presented, though the lack of a pertuzumab comparator is a limitation.Evidence: Efficacy results (PFS, OS) and safety data.
“This analysis reinforces the efficacy of this dual anti-HER2 (pyrotinib plus trastuzumab) regimen as an effective treatment strategy for this patient population.”
ConclusionFind in source - supportedReviewer 2The safety profile of pyrotinib-based regimen is consistent with interim findings with no new safety signals.The claim is supported by the reported adverse event data showing similar types and frequencies as the interim analysis, and no new safety signals identified.Evidence: Adverse event profiles remained consistent with the interim analysis; no new safety signals identified.
Adverse event profiles remained consistent with the interim analysis for type, frequency, and severity. ... No new safety signals were identified during the extended follow-up period.
Abstractreviewer’s wording
Efficacy claim is anchored to an adequate endpoint and a meaningful effect.
- ADEQUATESurrogate endpointThe primary endpoint is progression-free survival (PFS), which is a clinical outcome, and the secondary endpoint includes overall survival (OS), a hard clinical outcome. The efficacy claim is based on these clinical endpoints, not a surrogate biomarker.
“The primary endpoint was investigator assessed progression-free survival. ... Overall survival was longer in the pyrotinib group (hazard ratio 0.64 (95% confidence interval (CI) 0.46 to 0.89); nominal one-sided P=0.004).”
- ADEQUATEEffect sizeThe effect sizes are clinically meaningful: PFS improved from 10.5 to 22.1 months (HR 0.44), and OS showed a 36% reduction in risk of death (HR 0.64) with survival rates at 3, 4, and 5 years favoring pyrotinib. These are substantial improvements in hard clinical outcomes.
“Improvement in progression-free survival in the pyrotinib group was maintained (22.1 months (95% CI 19.3 to 27.8) v 10.5 months (9.5 to 12.4), hazard ratio 0.44 (95% CI 0.36 to 0.53); nominal one sided P<0.001). ... Overall survival was longer in the pyrotinib group (hazard ratio 0.64 (95% CI 0.46 to 0.89); nominal one-sided P=0.004).”
Data authenticity concerns
1 finding · worst lowAn 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.
- Data look implausibly cleanAssessed
3 integrity concerns flagged (0 high).
- lowdata too cleanThe reported progression-free survival benefit (HR 0.44) is large and consistent across subgroups, but this is plausible given the mechanism and prior interim results.
hazard ratio of 0.44 (95% CI 0.36 to 0.53; nominal one sided P<0.001)
Resultsreviewer’s wording
Reporting gaps
None foundRequired detail the manuscript never states — study design, biological variables, ethics approval and consent, key resources, statistical reporting, data and code availability, and overall transparency.
Checked — nothing surfaced.
Prior work is cited (CLEOPATRA, PHILA interim analysis) and the rationale for the study is clearly linked to the need for long-term survival data. The limitations of prior work (preliminary overall survival data) are explicitly addressed as the motivation for this updated analysis.
“Among the dual anti-HER2 regimens, the combination of pertuzumab and trastuzumab along with docetaxel is the established standard of care for the initial treatment of patients with HER2 positive metastatic breast cancer, based on the landmark Clinical Evaluation of Pertuzumab and Trastuzumab (CLEOPATRA) study.”
“Among the dual anti-HER2 regimens, the combination of pertuzumab and trastuzumab along with docetaxel is the established standard of care for the initial treatment of patients with HER2 positive metastatic breast cancer, based on the landmark Clinical Evaluation of Pertuzumab and Trastuzumab (CLEOPATRA) study.”
“Here, we present the prespecified final analysis of progression-free survival results, including overall survival, duration of response, and safety results after a longer follow-up period, as well as long term survival outcomes.”
Randomization was centralized via an interactive web response system with stratification factors. Blinding of patients, investigators, and sponsor is stated. A priori power analysis is provided. Inclusion/exclusion criteria are detailed. The analysis populations (full analysis set and safety set) are defined, addressing missing data and outliers.
“Patients, investigators, and the sponsor were blinded to treatment allocation.”
“Based on an assumption of a median progression-free survival of 12.5 months with trastuzumab and docetaxel, the study aimed to have 410 investigator assessed progression-free survival events to provide 80% power to detect a four month increase with the pyrotinib plus trastuzumab and docetaxel regimen (16.5 months, hazard ratio 0.76), at a one sided significance level of 0.025 for superiority.”
“Randomisation was done through a centralised interactive web response system.”
“Patients, investigators, and the sponsor were blinded to treatment allocation.”
“the study aimed to have 410 investigator assessed progression-free survival events to provide 80% power to detect a four month increase with the pyrotinib plus trastuzumab and docetaxel regimen (16.5 months, hazard ratio 0.76), at a one sided significance level of 0.025 for superiority.”
Sex is reported (all female) and justified by the disease context (HER2-positive breast cancer predominantly in women). Age range (18-75) is given. Demographics such as hormone receptor status and prior trastuzumab use are reported as stratification factors. Detailed baseline characteristics are said to be similar and previously reported.
“590 female patients with untreated HER2 positive metastatic breast cancer.”
“Previous reports have indicated that the baseline characteristics of participants were generally similar between the two groups.”
“590 female patients with untreated HER2 positive metastatic breast cancer.”
“Participants aged 18 to 75 years with histologically confirmed HER2 positive recurrent or metastatic breast cancer, who had not received the treatment for this stage of the disease were eligible.”
“Previous reports have indicated that the baseline characteristics of participants were generally similar between the two groups.”
The paper states approval by the ethics committee of each study centre and that the study was conducted in accordance with the Declaration of Helsinki and Good Clinical Practice Guidelines. Written informed consent was obtained from all patients. The approval statement is adequate, though specific protocol numbers are not provided.
“This study was approved by the ethics committee of each study centre (see supplementary table S9) and was conducted in accordance with the Declaration of Helsinki and Good Clinical Practice Guidelines.”
“All patients provided written informed consent.”
“This study was approved by the ethics committee of each study centre (see supplementary table S9) and was conducted in accordance with the Declaration of Helsinki and Good Clinical Practice Guidelines.”
“All patients provided written informed consent.”
The trial uses a drug, biologic, and chemotherapy; the investigational products are named with manufacturer (Jiangsu Hengrui Pharmaceuticals for pyrotinib) and dosing regimen. Trastuzumab and docetaxel are standard agents, and their doses are specified. Software used for statistical analysis (SAS version 9.4) is identified. No bench reagents or cell lines are used, so those criteria are not applicable.
“Patients were randomly assigned in a 1:1 ratio to receive either oral pyrotinib (400 mg once daily) or placebo, both combined with intravenous trastuzumab (8 mg/kg in treatment cycle 1 and 6 mg/kg in subsequent cycles) and docetaxel on day 1 of each 21 day treatment cycle.”
“All statistical analyses were performed using SAS (SAS Institute), version 9.4 or higher.”
“Eligible patients were randomly assigned in a 1:1 ratio to receive either the irreversible pan-HER inhibitor pyrotinib (400 mg orally once daily) or placebo, both in combination with intravenous trastuzumab (8 mg/kg for the first cycle, then 6 mg/kg in subsequent cycles) and docetaxel (75 mg/m 2 ) on day 1 of each 21 day treatment cycle.”
“All statistical analyses were performed using SAS (SAS Institute), version 9.4 or higher.”
The statistical analysis plan is detailed, naming tests (Kaplan-Meier, log-rank, Cox, Cochran-Mantel-Haenszel). Effect sizes (hazard ratios) with 95% CIs are reported for all primary and secondary endpoints. P-values are reported as nominal one-sided, which is appropriate given the prespecified alpha spending. The paper reports by estimation with CIs, so exact p-values are not required. Data presentation includes Kaplan-Meier curves and forest plots.
“Overall survival was longer in the pyrotinib group (hazard ratio 0.64 (95% confidence interval (CI) 0.46 to 0.89); nominal one-sided P=0.004).”
“Overall survival was longer in the pyrotinib group (hazard ratio 0.64 (95% confidence interval (CI) 0.46 to 0.89); nominal one-sided P=0.004).”
“no formal hypothesis testing on overall survival was prespecified, so the results reported in this paper, for both progression-free survival and overall survival, are with nominal one sided P values.”
The data availability statement provides a specific repository (Mendeley Data) with a DOI and URL for the anonymised individual patient data. This is a concrete access route, satisfying the requirement. No bespoke code was written for analysis (SAS was used), so code sharing is not applicable.
“The anonymised individual patient data supporting the findings of this study are available in the Mendeley Data ( https://data.mendeley.com/datasets/62224ppy2z/1 ).”
“The anonymised individual patient data supporting the findings of this study are available in the Mendeley Data ( https://data.mendeley.com/datasets/62224ppy2z/1 ).”
Trial registration number is provided (NCT03863223). Methods are comprehensive. Limitations are explicitly discussed (absence of pertuzumab control, post-discontinuation treatments). Conclusions are proportional to the evidence. Funding sources and competing interests are declared.
“Trial registration ClinicalTrials.gov NCT03863223 (https://clinicaltrials.gov/ct2/show/NCT03863223) .”
“One of the major limitations of this study was the absence of a control group using pertuzumab-trastuzumab combination treatment.”
“Trial registration ClinicalTrials.gov NCT03863223 (https://clinicaltrials.gov/ct2/show/NCT03863223) .”
“Funding: This study was funded by Jiangsu Hengrui Pharmaceuticals, with partial support from the National Natural Science Foundation of China (92459304), CAMS Innovation Fund for Medical Sciences (CIFMS; 2021-I2M-1-014), and National High Level Hospital Clinical Research Funding (2025-LYZX-D-A02).”
“One of the major limitations of this study was the absence of a control group using pertuzumab-trastuzumab combination treatment.”
Registered (1 ID: ClinicalTrials.gov). No reporting guideline cited.
Broken references and links
None foundReferences checked against Crossref, OpenAlex and Retraction Watch for retractions and resolvability, plus declared data and code links probed for whether they resolve to content matching the paper.
Checked — nothing surfaced.
Checked 25 references by DOI: 25 verified.
Every extracted reference resolved against Crossref/OpenAlex with no retraction flags.
1 data/code link checked; 1 live.
- dataMendeley DataLIVEHTTP 200https://data.mendeley.com/datasets/62224ppy2z/1Resolves to Mendeley Data (data repository).
Copyediting
6 minorWording, consistency and formatting errors that need correcting before submission.
No major wording or formatting errors. 6 minor suggestions below.
6 copyedit issues flagged: mostly typo, consistency, clarity.
- MINORtypoAbstract, Conclusions“initial treatment ofHER2 positive metastatic breast cancer”→ Add a space: 'treatment of HER2'Missing space between 'of' and 'HER2'.
- MINORconsistencyResults, Safety“Diarrhoea of any grade was reported in 99% (n=293) of patients in the pyrotinib group and 55% (n=162) in the placebo group”→ Ensure consistency with Table 2 where diarrhoea is 99% (293) and 54% (158).The text says 55% (n=162) but Table 2 says 54% (n=158).
- MINORclarityDiscussion, Comparison with other studies“The hazard ratio for overall survival in the pyrotinib group (0.74) was comparable to the hazard ratio reported for the pertuzumab arm (0.68) in the CLEOPATRA study.”→ Clarify that the comparison is with the pertuzumab arm of CLEOPATRA, not the overall study.The sentence is clear but could be more precise.
- MINORtypoConclusions“initial treatment ofHER2 positive metastatic breast cancer”→ initial treatment of HER2 positive metastatic breast cancerMissing space after 'of'.
- MINORconsistencyAbstract, Results“59 (20%) and 87 (30%) patients died”→ 59 (20%) and 87 (30%) patients died, respectivelyAdd 'respectively' for clarity.
- MINORclarityDiscussion, Comparison with other studies“The hazard ratio for overall survival in the pyrotinib group (0.74) was comparable to the hazard ratio reported for the pertuzumab arm (0.68) in the CLEOPATRA study.”→ The hazard ratio for overall survival in the pyrotinib group (0.74) was comparable to that reported for the pertuzumab arm (0.68) in the CLEOPATRA study.Rephrase for clarity.
The published work is robust; an informed reader should weigh the minor internal inconsistency in the diarrhoea safety numbers and the lack of explicit CONSORT adherence, but neither undermines the main conclusions. No erratum is warranted for the minor issues, though the authors may consider a correction for the diarrhoea percentage discrepancy.
- 1.HIGHcopyeditIn Results, Safety, reconcile the diarrhoea percentage in the placebo group: text says 55% (n=162) but Table 2 lists 54% (n=158). Correct the text or the table to match.An internal inconsistency between the narrative and the table undermines data credibility and could be flagged by readers or reviewers.
- 2.MEDIUMreportingAdd an explicit statement of adherence to the CONSORT reporting guideline in the Methods or a dedicated section.Both reviewers noted the absence of an explicit reporting guideline statement, which is a minor transparency gap for a clinical trial.
- 3.MEDIUMreportingProvide more detailed demographic data (e.g., race/ethnicity, comorbidities) in the main text or supplementary tables, rather than only stating baseline characteristics were similar.Reviewer 2 flagged demographics as 'reported_but_inadequate'; fuller reporting would strengthen the paper's transparency.
- 4.MEDIUMstatisticsClarify the handling of missing data and any sensitivity analyses in the statistical analysis section.Reviewer 2 noted this is not explicitly described; adding it would improve methodological transparency.
- 5.MEDIUMstatisticsConsider reporting exact p-values for all secondary endpoints, as some are only reported as thresholds (e.g., P<0.001).Exact p-values would allow readers to assess the strength of evidence more precisely.
- 6.MEDIUMdata codeInclude a statement on whether the Mendeley Data repository is under a managed-access or open-access model, to clarify access conditions.Reviewer 2 suggested this clarification to help readers understand how to access the data.
- 7.LOWcopyeditFix the missing space in the Abstract/Conclusions: 'treatment ofHER2' should be 'treatment of HER2'.Typographical error that should be corrected for professionalism.
- 8.LOWcopyeditAdd 'respectively' to the Abstract Results sentence: '59 (20%) and 87 (30%) patients died' → '59 (20%) and 87 (30%) patients died, respectively'.Improves clarity by indicating the order of the groups.
- 9.LOWcopyeditRephrase the Discussion sentence comparing hazard ratios for clarity: 'The hazard ratio for overall survival in the pyrotinib group (0.74) was comparable to that reported for the pertuzumab arm (0.68) in the CLEOPATRA study.'The original phrasing is slightly ambiguous; the suggested rephrase is clearer.
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.