Nivolumab plus chemotherapy or ipilimumab in gastroesophageal cancer: exploratory biomarker analyses of a randomized phase 3 trial.
Shitara K, Janjigian YY, Ajani J, Moehler M, Yao J, Wang X, Chhibber A, Pandya D, Shen L, Garrido M, Gallardo C, Wyrwicz L, Yamaguchi K, Skoczylas T, Bragagnoli A, Liu T, Schenker M, Yañez P, Kowalyszyn R, Karamouzis M, Zander T, Feeney K, Elimova E, Doshi P, Li M, Lei M
- DOI
- 10.1038/s41591-025-03575-0
- Record issued
- 2026-08-15
- Engine
- 7.39.0
- 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/5a3dffc1-aa6a-40f2-b1a9-a0abb5da0b45 is authoritative.
How this rating was calculated
- IntegrityIntegrity concern ×2−1★
- CitationsUnresolved reference−0.25★
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 post hoc exploratory biomarker analysis of a randomized phase 3 trial, with clear scientific premise, rigorous methods, and transparent reporting. The main weaknesses are minor reporting gaps: exact p-values are often given as thresholds, code sharing is not mentioned, and the specific IRB is not named. The integrity check flagged a low-severity internal contradiction in patient counts that should be clarified.
Both reviewers classified the study as observational (post hoc analysis of a randomized trial). The evaluation covered the full text, with verification components for citations, statistics, reproducibility, preregistration, and integrity. The statistics verification covered only 1 test due to limited reporting of test statistics; most p-values are threshold-only and could not be machine-verified.
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 1 test: 1 consistent, 0 inconsistent; 1 via agent-written checks.
- CONSISTENTreported p < .001 · recomputed p = <.001Reviewer 1Check the p-value for the chi-squared test of genomic subtype vs histology.
“Exact P value for ( a ) was 6.673e-09.”
Taken as given: The p-value is from a two-sided chi-squared test.; The table is 2x2 with counts 41, 848, 533, 356.; The counts are derived from the WES-evaluable population (n=889).Method: Recomputed using Pearson's chi-squared test on the 2x2 table.How we recomputed it: pChi2x2(41, 848, 533, 356)
- lowinternal contradictionThe number of WES-evaluable patients in the nivolumab-plus-ipilimumab group is reported as 366 in the text, but the figure legend for Fig. 2c states TMB available in 447 patients (NIVO+IPI: n=226; Chemo: n=221). This discrepancy may be due to different denominators (WES-evaluable vs TMB-evaluable).
“Of 813 patients randomized to receive nivolumab-plus-ipilimumab versus chemotherapy, 366 (45%) were evaluable by WES and 402 (49%) were evaluable by RNA-seq (Supplementary Table ).”
Figure 2Find in source - lowinternal contradictionThe number of WES-evaluable patients in the nivolumab-plus-chemotherapy group is reported as 685, but the total WES-evaluable population is 889, and the nivolumab-plus-ipilimumab group has 366. The sum of 685 and 366 is 1051, which exceeds 889, suggesting overlapping or different denominators.
“Of 1,581 patients randomized to receive nivolumab-plus-chemotherapy versus chemotherapy, 685 (43%) were evaluable by whole-exome sequencing (WES) and 809 (51%) were evaluable by RNA sequencing (RNA-seq) (Supplementary Table ). Of 813 patients randomized to receive nivolumab-plus-ipilimumab versus chemotherapy, 366 (45%) were evaluable by WES and 402 (49%) were evaluable by RNA-seq (Supplementary Table ).”
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.
6 major claims checked against the paper's own evidence: all adequately supported.
- supportedReviewers 1, 2Nivolumab-based therapies demonstrated improved efficacy versus chemotherapy in hypermutated and, to a lesser degree, Epstein–Barr virus-positive tumors compared with chromosomally unstable and genomically stable tumors.The claim is supported by the reported HRs for hypermutated and EBV subtypes versus CIN and GS.Evidence: HRs for hypermutated (0.37 and 0.27) and EBV (0.61 and 0.76) are lower than for CIN (0.92 and 0.81) and GS (0.70 and 1.01).
“Nivolumab-based therapies demonstrated improved efficacy versus chemotherapy in hypermutated and, to a lesser degree, Epstein–Barr virus-positive tumors compared with chromosomally unstable and genomically stable tumors.”
AbstractFind in source - supportedReviewers 1, 2Within the KRAS-altered subgroup, only patients treated with nivolumab-plus-chemotherapy demonstrated improved OS benefit versus chemotherapy.The claim is supported by the reported HR for KRAS-altered patients treated with nivolumab-plus-chemotherapy (HR 0.53), while no notable impact was seen with nivolumab-plus-ipilimumab.Evidence: HR 0.53 for KRAS-altered with nivolumab-plus-chemotherapy; no notable impact with nivolumab-plus-ipilimumab.
“Within the KRAS-altered subgroup, only patients treated with nivolumab-plus-chemotherapy demonstrated improved OS benefit versus chemotherapy.”
AbstractFind in source - supportedReviewers 1, 2Low stroma gene expression signature scores were associated with OS benefit with nivolumab-based regimens; high regulatory T cell signatures were associated with OS benefit only with nivolumab-plus-ipilimumab.The claim is supported by the reported associations: low stroma GES with nivolumab-plus-chemotherapy and high Treg GES with nivolumab-plus-ipilimumab.Evidence: Low stroma GES associated with OS benefit with nivolumab-plus-chemotherapy; high Treg GES associated with OS benefit with nivolumab-plus-ipilimumab.
“Low stroma gene expression signature scores were associated with OS benefit with nivolumab-based regimens; high regulatory T cell signatures were associated with OS benefit only with nivolumab-plus-ipilimumab.”
AbstractFind in source - supportedReviewer 1Our analyses suggest that distinct and overlapping pathways contribute to the efficacy of nivolumab-based regimens in gastroesophageal adenocarcinoma.The claim is supported by the differential biomarker associations observed for the two regimens.Evidence: Distinct biomarkers (e.g., Treg for ipilimumab) and overlapping (e.g., hypermutated) are reported.
“Our analyses suggest that distinct and overlapping pathways contribute to the efficacy of nivolumab-based regimens in gastroesophageal adenocarcinoma.”
AbstractFind in source - supportedReviewer 2The biomarker-evaluable populations were representative of the entire randomized population.The paper states baseline characteristics and OS were comparable between all-randomized and biomarker-evaluable populations.Evidence: Supplementary Tables and text in Results.
Median OS and hazard ratios (HRs) were generally comparable between all-randomized, WES-evaluable and RNA-seq-evaluable patients in both the nivolumab-plus-chemotherapy and nivolumab-plus-ipilimumab versus chemotherapy groups, suggesting that the biomarker-evaluable populations were representative of the entire randomized population.
Resultsreviewer’s wording - supportedReviewer 2The study provides the largest WES and RNA-seq datasets with first-line immune checkpoint blockade in gastroesophageal adenocarcinoma.The paper claims this and provides no direct comparison, but the sample sizes are large.Evidence: Statement in Discussion.
“To our knowledge, this study provides the largest WES and RNA-seq datasets with first-line immune checkpoint blockade in gastroesophageal adenocarcinoma.”
Discussion ¶1Find in source
Efficacy claim is anchored to an adequate endpoint and a meaningful effect.
- ADEQUATESurrogate endpointThe primary efficacy claim is based on overall survival (OS), a hard clinical outcome, from the randomized phase 3 CheckMate 649 trial. The biomarker analyses are exploratory and use OS as the endpoint, not a surrogate.
“First-line nivolumab-plus-chemotherapy demonstrated superior overall survival (OS) and progression-free survival versus chemotherapy for advanced gastroesophageal adenocarcinoma with programmed death ligand 1 combined positive score ≥ 5, meeting both primary end points of the randomized phase 3 CheckMate 649 trial.”
- ADEQUATEEffect sizeThe effect sizes are reported as hazard ratios with confidence intervals, and the primary endpoint is overall survival, which is clinically meaningful. For example, the overall population HR for nivolumab+chemo vs chemo is 0.69 (95% CI 0.60–0.79), and subgroup HRs are provided. These are anchored to clinical outcomes.
“overall population, HR 0.69, 95% CI 0.60–0.79”
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
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.
The introduction cites prior studies on nivolumab-based regimens and molecular subtypes, and explicitly states that 'there has not been a comprehensive DNA or RNA analysis exploring the relationship of various pretreatment tumor characteristics with therapeutic outcomes in metastatic gastroesophageal adenocarcinoma.' The rationale linking the premise to the study objectives is logical, and the paper acknowledges limitations of prior work (e.g., limited studies on CIN/GS/EBV subtypes).
“there has not been a comprehensive DNA or RNA analysis exploring the relationship of various pretreatment tumor characteristics with therapeutic outcomes in metastatic gastroesophageal adenocarcinoma.”
“Identification of pretreatment biomarkers associated with outcomes with nivolumab-based regimens in gastroesophageal adenocarcinoma may help identify patient subgroups that are likely to derive the most clinical benefit.”
“there has not been a comprehensive DNA or RNA analysis exploring the relationship of various pretreatment tumor characteristics with therapeutic outcomes in metastatic gastroesophageal adenocarcinoma.”
“studies describing tumors with chromosomal instability (CIN) or that are genomically stable (GS) or Epstein–Barr virus (EBV)-positive are limited”
The parent trial was randomized (method not detailed here but referenced to prior publication), and the biomarker-evaluable populations are defined with inclusion criteria (e.g., 'evaluable baseline tumor tissue and matched whole-blood samples that passed the quality control (QC) criteria'). Blinding is not explicitly described but is standard for a phase 3 trial; the paper does not mention it, but the analysis is post hoc and exploratory. Power analysis is not applicable as this is an exploratory analysis. Outlier handling is addressed through QC criteria for sequencing data. Controls are inherent in the randomized comparator arms.
“adult patients with previously untreated, unresectable or advanced gastric, gastroesophageal and esophageal adenocarcinoma were randomized to receive nivolumab-plus-chemotherapy, nivolumab-plus-ipilimumab or chemotherapy alone.”
“Patients in the CheckMate 649 study who provided appropriate consent for the biomarker testing with evaluable baseline tumor tissue and matched whole-blood samples that passed the quality control (QC) criteria were eligible for the WES analyses”
“A WES QC report was generated, which marked tumor samples PASS if total reads 45 million, mean target coverage 50× and depth of coverage >20× at 80% of the targeted capture region or higher.”
“Patients in the CheckMate 649 study who provided appropriate consent for the biomarker testing with evaluable baseline tumor tissue and matched whole-blood samples that passed the quality control (QC) criteria were eligible for the WES analyses”
“All analyses were conducted using R v.4.0.3. Patient characteristics and clinical outcomes were compared between the biomarker-evaluable population and the intent-to-treat population using frequency statistics and descriptive statistics.”
“A WES QC report was generated, which marked tumor samples PASS if total reads 45 million, mean target coverage 50× and depth of coverage >20× at 80% of the targeted capture region or higher.”
The paper reports that baseline characteristics were 'generally balanced between the treatment groups' and consistent across populations, implying demographics are reported (likely in supplementary tables). Sex is reported as a covariate in the GSEA analysis. Age and health status are not explicitly stated in the main text but are standard in the parent trial. Demographics are mentioned as being representative.
“Covariates included prognostic factors such as age, sex, Eastern Cooperative Oncology Group performance status, region, primary tumor location, disease status, Lauren classification, peritoneal metastases, previous surgery related to current cancer or radiotherapy before randomization, number of organs with baseline lesion, signet ring cell and planned chemotherapy regimen as well as PD-L1 CPS level and RNA-seq batch ID.”
“Baseline characteristics were generally balanced between the treatment groups and were consistent for all-randomized, WES-evaluable and RNA-seq-evaluable patients (Supplementary Tables and ).”
“Although the biomarker-evaluable populations were generally representative of the all-randomized study population with respect to demographics, baseline clinical characteristics and outcomes”
“Baseline characteristics were generally balanced between the treatment groups and were consistent for all-randomized, WES-evaluable and RNA-seq-evaluable patients”
“Covariates included prognostic factors such as age, sex, Eastern Cooperative Oncology Group performance status, region, primary tumor location, disease status, Lauren classification, peritoneal metastases, previous surgery related to current cancer or radiotherapy before randomization, number of organs with baseline lesion, signet ring cell and planned chemotherapy regimen as well as PD-L1 CPS level and RNA-seq batch ID.”
The paper states that the study was conducted in accordance with Good Clinical Practice and the Declaration of Helsinki, and that written informed consent was obtained. The trial is registered at ClinicalTrials.gov (NCT02872116). However, the specific ethics committee that approved the study is not named in the text, which is a minor gap but not a failure given the explicit consent and registration statements.
“Written informed consent was obtained from all patients per the Declaration of Helsinki principles.”
“The study is registered at ClinicalTrials.gov ( NCT02872116 (https://clinicaltrials.gov/ct2/show/NCT02872116) ).”
“This study was conducted in accordance with the trial protocol and with Good Clinical Practice guidelines developed by the International Council for Harmonisation.”
“This study was conducted in accordance with the trial protocol and with Good Clinical Practice guidelines developed by the International Council for Harmonisation. Written informed consent was obtained from all patients per the Declaration of Helsinki principles.”
“The study is registered at ClinicalTrials.gov ( NCT02872116 (https://clinicaltrials.gov/ct2/show/NCT02872116) ).”
The investigational products (nivolumab, ipilimumab, chemotherapy) are named but not with manufacturer details in the text (likely in the protocol). Sequencing platforms (Illumina NovaSeq), capture panels (Agilent SureSelect, Illumina TruSeq RNA Access), and software (Sentieon, GATK, R packages) are identified with versions. The PD-L1 assay is identified (Dako PD-L1 IHC 28-8 pharmDx). Antibodies are not used in this analysis, so that criterion is not applicable.
“Baseline tumor tissue and matched whole-blood samples were processed using the Agilent SureSelect Human All Exon V6 in-solution hybrid capture panel and underwent subsequent next-generation sequencing on the Illumina NovaSeq platform.”
“All analyses were conducted using R v.4.0.3.”
“PD-L1 CPS was determined using a validated immunohistochemistry assay (Dako PD-L1 IHC 28-8 pharmDx assay; Agilent Technologies Inc.)”
“PD-L1 CPS was determined using a validated immunohistochemistry assay (Dako PD-L1 IHC 28-8 pharmDx assay; Agilent Technologies Inc.)”
“All analyses were conducted using R v.4.0.3.”
“patients with previously untreated, unresectable or advanced gastric, gastroesophageal and esophageal adenocarcinoma were randomized to receive nivolumab-plus-chemotherapy, nivolumab-plus-ipilimumab or chemotherapy alone.”
The paper names statistical tests (Cox PH models, LRT, Mann-Whitney U, Kruskal-Wallis, chi-squared) and software (R v.4.0.3). Effect sizes are reported as HRs with 95% CIs, which is the standard for survival analysis. Exact p-values are sometimes reported (e.g., '6.673e-09') but often only thresholds (e.g., 'P < 0.1'), which is acceptable for exploratory analyses. Data presentation includes Kaplan-Meier curves and forest plots. Mathematical plausibility is not applicable for large-N survival data.
“Comparisons among groups used Mann–Whitney U -test (two groups) and Kruskal–Wallis test (three or more groups) for continuous variables and chi-squared test for categorical variables.”
“The hypermutated subtype benefited most from both nivolumab-based regiments versus chemotherapy (nivolumab-plus-chemotherapy, HR 0.37, 95% CI 0.15–0.90; Fig. ; nivolumab-plus-ipilimumab, HR 0.27, 95% CI 0.07–1.06; Fig. ).”
“Exact P value for ( a ) was 6.673e-09.”
“Comparisons among groups used Mann–Whitney U -test (two groups) and Kruskal–Wallis test (three or more groups) for continuous variables and chi-squared test for categorical variables.”
“nivolumab-plus-chemotherapy, HR 0.37, 95% CI 0.15–0.90”
“Several GES were associated with OS benefit from nivolumab-plus-chemotherapy versus chemotherapy ( P < 0.1; likelihood ratio test (LRT))”
WES and RNA-seq data are deposited in the EGA under study ID EGAS50000000747. Clinical trial data access is described via a managed-access process through Vivli, which is a concrete route. Code sharing is not explicitly mentioned, but the analysis uses standard software and no custom code is described as being shared.
“WES and RNA-seq datasets have been deposited in the European Genome–phenome Archive under study ID EGAS50000000747 (https://ega-archive.org/studies/EGAS50000000747) .”
“Bristol Myers Squibb will honor legitimate requests for our clinical trial data from qualified researchers with a clearly defined scientific objective.”
“WES and RNA-seq datasets have been deposited in the European Genome–phenome Archive under study ID EGAS50000000747 (https://ega-archive.org/studies/EGAS50000000747) .”
“Bristol Myers Squibb will honor legitimate requests for our clinical trial data from qualified researchers with a clearly defined scientific objective.”
The trial is registered (NCT02872116). Limitations are discussed in the Discussion section. Conclusions are appropriately cautious, noting the exploratory nature and need for prospective validation. Funding and competing interests are disclosed. A reporting guideline (e.g., CONSORT) is not explicitly mentioned, but the paper references a 'Reporting Summary' linked to the article.
“The study is registered at ClinicalTrials.gov ( NCT02872116 (https://clinicaltrials.gov/ct2/show/NCT02872116) ).”
“The current analyses have some limitations. Although the biomarker-evaluable populations were generally representative of the all-randomized study population with respect to demographics, baseline clinical characteristics and outcomes, the small sample sizes of each biomarker subgroup may have resulted in imbalances that are not captured by these measurements.”
“This study was supported by Bristol Myers Squibb and Ono Pharmaceutical Co.”
“The study is registered at ClinicalTrials.gov ( NCT02872116 (https://clinicaltrials.gov/ct2/show/NCT02872116) ).”
“The current analyses have some limitations. Although the biomarker-evaluable populations were generally representative of the all-randomized study population with respect to demographics, baseline clinical characteristics and outcomes, the small sample sizes of each biomarker subgroup may have resulted in imbalances that are not captured by these measurements.”
“This study was supported by Bristol Myers Squibb and Ono Pharmaceutical Co.”
Registered (1 ID: ClinicalTrials.gov). No reporting guideline cited.
Broken references and links
1 finding · worst lowReferences 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.
- References not resolvable to a published paperRecomputed
Checked 96 references by DOI: 2 verified — 1 DOI unresolved, 93 no DOI (shown, not verified).
- UNRESOLVED10.1016/s1470-2045(13OPDIVO (nivolumab prescribing information)Cited DOI does not resolve to any Crossref record.
- NO DOICapecitabine and cisplatin with or without cetuximab for patients with previously untreated advanced gastric cancer (EXPAND): a randomised, open-label phase 3 trialNo DOI in the reference — shown for manual review; not independently verifiable (not a fabrication signal).
- NO DOIRilotumumab plus epirubicin, cisplatin, and capecitabine as first-line therapy in advanced MET-positive gastric or gastro-oesophageal junction cancer (RILOMET-1): a randomised, double-blind, placebo-controlled, phase 3 trialNo DOI in the reference — shown for manual review; not independently verifiable (not a fabrication signal).
- NO DOIEffect of fluorouracil, leucovorin, and oxaliplatin with or without onartuzumab in HER2-negative, MET-positive gastroesophageal adenocarcinoma: the METGastric randomized clinical trialNo DOI in the reference — shown for manual review; not independently verifiable (not a fabrication signal).
- NO DOIRamucirumab with cisplatin and fluoropyrimidine as first-line therapy in patients with metastatic gastric or junctional adenocarcinoma (RAINFALL): a double-blind, randomised, placebo-controlled, phase 3 trialNo DOI in the reference — shown for manual review; not independently verifiable (not a fabrication signal).
- NO DOIFirst-line nivolumab plus chemotherapy versus chemotherapy alone for advanced gastric, gastro-oesophageal junction, and oesophageal adenocarcinoma (CheckMate 649): a randomised, open-label, phase 3 trialNo DOI in the reference — shown for manual review; not independently verifiable (not a fabrication signal).
- NO DOINivolumab plus chemotherapy or ipilimumab in gastro-oesophageal cancerNo DOI in the reference — shown for manual review; not independently verifiable (not a fabrication signal).
- NO DOIPD-1 and CTLA-4 combination blockade expands infiltrating T cells and reduces regulatory T and myeloid cells within B16 melanoma tumorsNo DOI in the reference — shown for manual review; not independently verifiable (not a fabrication signal).
- NO DOICombination therapy with anti-CTLA-4 and anti-PD-1 leads to distinct immunologic changes in vivoNo DOI in the reference — shown for manual review; not independently verifiable (not a fabrication signal).
- NO DOIIn vitro characterization of the anti-PD-1 antibody nivolumab, BMS-936558, and in vivo toxicology in non-human primatesNo DOI in the reference — shown for manual review; not independently verifiable (not a fabrication signal).
- NO DOIThe blockade of immune checkpoints in cancer immunotherapyNo DOI in the reference — shown for manual review; not independently verifiable (not a fabrication signal).
- NO DOIFundamental mechanisms of immune checkpoint blockade therapyNo DOI in the reference — shown for manual review; not independently verifiable (not a fabrication signal).
- NO DOIAnti-CTLA-4 antibodies drive myeloid activation and reprogram the tumor microenvironment through FcγR engagement and type I interferon signalingNo DOI in the reference — shown for manual review; not independently verifiable (not a fabrication signal).
- NO DOICTLA-4 blockade drives loss of Treg stability in glycolysis-low tumoursNo DOI in the reference — shown for manual review; not independently verifiable (not a fabrication signal).
- NO DOICombination anti-CTLA-4 plus anti-PD-1 checkpoint blockade utilizes cellular mechanisms partially distinct from monotherapiesNo DOI in the reference — shown for manual review; not independently verifiable (not a fabrication signal).
- NO DOIParadigms on immunotherapy combinations with chemotherapyNo DOI in the reference — shown for manual review; not independently verifiable (not a fabrication signal).
- NO DOIComprehensive molecular characterization of gastric adenocarcinomaNo DOI in the reference — shown for manual review; not independently verifiable (not a fabrication signal).
- NO DOIPositive status of Epstein–Barr virus as a biomarker for gastric cancer immunotherapy: a prospective observational studyNo DOI in the reference — shown for manual review; not independently verifiable (not a fabrication signal).
- NO DOIPembrolizumab in microsatellite instability high or mismatch repair deficient cancers: updated analysis from the phase II KEYNOTE-158 studyNo DOI in the reference — shown for manual review; not independently verifiable (not a fabrication signal).
- NO DOISafety and efficacy of pembrolizumab monotherapy in patients with previously treated advanced gastric and gastroesophageal junction cancer: phase 2 clinical KEYNOTE-059 trialNo DOI in the reference — shown for manual review; not independently verifiable (not a fabrication signal).
- NO DOIMolecular determinants of clinical outcomes with pembrolizumab versus paclitaxel in a randomized, open-label, phase III trial in patients with gastroesophageal adenocarcinomaNo DOI in the reference — shown for manual review; not independently verifiable (not a fabrication signal).
- NO DOIGenomic features of response to combination immunotherapy in patients with advanced non-small-cell lung cancerNo DOI in the reference — shown for manual review; not independently verifiable (not a fabrication signal).
- NO DOIAtezolizumab in patients with locally advanced and metastatic urothelial carcinoma who have progressed following treatment with platinum-based chemotherapy: a single-arm, multicentre, phase 2 trialNo DOI in the reference — shown for manual review; not independently verifiable (not a fabrication signal).
- NO DOIGenetic basis for clinical response to CTLA-4 blockade in melanomaNo DOI in the reference — shown for manual review; not independently verifiable (not a fabrication signal).
- NO DOIGenomic correlates of response to CTLA-4 blockade in metastatic melanomaNo DOI in the reference — shown for manual review; not independently verifiable (not a fabrication signal).
- NO DOITargeted next generation sequencing identifies markers of response to PD-1 blockadeNo DOI in the reference — shown for manual review; not independently verifiable (not a fabrication signal).
- NO DOIAssociation of tumour mutational burden with outcomes in patients with advanced solid tumours treated with pembrolizumab: prospective biomarker analysis of the multicohort, open-label, phase 2 KEYNOTE-158 studyNo DOI in the reference — shown for manual review; not independently verifiable (not a fabrication signal).
- NO DOITumor mutational burden predicts the efficacy of pembrolizumab monotherapy: a pan-tumor retrospective analysis of participants with advanced solid tumorsNo DOI in the reference — shown for manual review; not independently verifiable (not a fabrication signal).
- NO DOIAnalyses of PD-L1 and inflammatory gene expression association with efficacy of nivolumab ± ipilimumab in gastric cancer/gastroesophageal junction cancerNo DOI in the reference — shown for manual review; not independently verifiable (not a fabrication signal).
- NO DOIClinical features and multiplatform molecular analysis assist in understanding patient response to anti-PD-1/PD-L1 in renal cell carcinomaNo DOI in the reference — shown for manual review; not independently verifiable (not a fabrication signal).
- NO DOIEMT- and stroma-related gene expression and resistance to PD-1 blockade in urothelial cancerNo DOI in the reference — shown for manual review; not independently verifiable (not a fabrication signal).
- NO DOIPGC-1α-responsive genes involved in oxidative phosphorylation are coordinately downregulated in human diabetesNo DOI in the reference — shown for manual review; not independently verifiable (not a fabrication signal).
- NO DOIGene set enrichment analysis: a knowledge-based approach for interpreting genome-wide expression profilesNo DOI in the reference — shown for manual review; not independently verifiable (not a fabrication signal).
- NO DOITumor aneuploidy correlates with markers of immune evasion and with reduced response to immunotherapyNo DOI in the reference — shown for manual review; not independently verifiable (not a fabrication signal).
- NO DOIChromosomal instability drives metastasis through a cytosolic DNA responseNo DOI in the reference — shown for manual review; not independently verifiable (not a fabrication signal).
- NO DOINon-cell-autonomous cancer progression from chromosomal instabilityNo DOI in the reference — shown for manual review; not independently verifiable (not a fabrication signal).
- NO DOIMetastasis and immune evasion from extracellular cGAMP hydrolysisNo DOI in the reference — shown for manual review; not independently verifiable (not a fabrication signal).
- NO DOIKRAS mutation: from undruggable to druggable in cancerNo DOI in the reference — shown for manual review; not independently verifiable (not a fabrication signal).
- NO DOIAssociation between KRAS variant status and outcomes with first-line immune checkpoint inhibitor-based therapy in patients with advanced non-small-cell lung cancerNo DOI in the reference — shown for manual review; not independently verifiable (not a fabrication signal).
- NO DOIEfficacy of immunotherapy in KRAS-mutant advanced NSCLC: a real-world study in a Chinese populationNo DOI in the reference — shown for manual review; not independently verifiable (not a fabrication signal).
- NO DOITransforming growth factor-β signaling in immunity and cancerNo DOI in the reference — shown for manual review; not independently verifiable (not a fabrication signal).
- NO DOIExploiting the therapeutic implications of KRAS inhibition on tumor immunityNo DOI in the reference — shown for manual review; not independently verifiable (not a fabrication signal).
- NO DOIMutant KRAS regulates transposable element RNA and innate immunity via KRAB zinc-finger genesNo DOI in the reference — shown for manual review; not independently verifiable (not a fabrication signal).
- NO DOICancer immunology. Mutational landscape determines sensitivity to PD-1 blockade in non-small cell lung cancerNo DOI in the reference — shown for manual review; not independently verifiable (not a fabrication signal).
- NO DOITumor hypoxia in cancer therapyNo DOI in the reference — shown for manual review; not independently verifiable (not a fabrication signal).
- NO DOIMicrovascular permeability and interstitial penetration of sterically stabilized (stealth) liposomes in a human tumor xenograftNo DOI in the reference — shown for manual review; not independently verifiable (not a fabrication signal).
- NO DOIThe intersection between tumor angiogenesis and immune suppressionNo DOI in the reference — shown for manual review; not independently verifiable (not a fabrication signal).
- NO DOINormalizing tumor microenvironment to treat cancer: bench to bedside to biomarkersNo DOI in the reference — shown for manual review; not independently verifiable (not a fabrication signal).
- NO DOIImmunosuppressive effects of vascular endothelial growth factor (review)No DOI in the reference — shown for manual review; not independently verifiable (not a fabrication signal).
- NO DOIDual angiopoietin-2 and VEGFA inhibition elicits antitumor immunity that is enhanced by PD-1 checkpoint blockadeNo DOI in the reference — shown for manual review; not independently verifiable (not a fabrication signal).
- NO DOISimultaneous blockade of programmed death 1 and vascular endothelial growth factor receptor 2 (VEGFR2) induces synergistic anti-tumour effect in vivoNo DOI in the reference — shown for manual review; not independently verifiable (not a fabrication signal).
- NO DOIVEGF-A modulates expression of inhibitory checkpoints on CD8 + T cells in tumorsNo DOI in the reference — shown for manual review; not independently verifiable (not a fabrication signal).
- NO DOICombined antiangiogenic and anti-PD-L1 therapy stimulates tumor immunity through HEV formationNo DOI in the reference — shown for manual review; not independently verifiable (not a fabrication signal).
- NO DOIEfficacy of cotargeting angiopoietin-2 and the VEGF pathway in the adjuvant postsurgical setting for early breast, colorectal, and renal cancersNo DOI in the reference — shown for manual review; not independently verifiable (not a fabrication signal).
- NO DOITumor induction of VEGF promoter activity in stromal cellsNo DOI in the reference — shown for manual review; not independently verifiable (not a fabrication signal).
- NO DOIFirst-line regorafenib with nivolumab and chemotherapy in advanced oesophageal, gastric, or gastro-oesophageal junction cancer in the USA: a single-arm, single-centre, phase 2 trialNo DOI in the reference — shown for manual review; not independently verifiable (not a fabrication signal).
- NO DOIRegorafenib plus nivolumab in patients with advanced gastric or colorectal cancer: an open-label, dose-escalation, and dose-expansion phase Ib trial (REGONIVO, EPOC1603)No DOI in the reference — shown for manual review; not independently verifiable (not a fabrication signal).
- NO DOILenvatinib plus pembrolizumab in patients with advanced gastric cancer in the first-line or second-line setting (EPOC1706): an open-label, single-arm, phase 2 trialNo DOI in the reference — shown for manual review; not independently verifiable (not a fabrication signal).
- NO DOITMB and inflammatory gene expression associated with clinical outcomes following immunotherapy in advanced melanomaNo DOI in the reference — shown for manual review; not independently verifiable (not a fabrication signal).
- NO DOIAssociation of inflammatory biomarkers with clinical outcomes in nivolumab-treated patients with advanced hepatocellular carcinomaNo DOI in the reference — shown for manual review; not independently verifiable (not a fabrication signal).
- NO DOITargeting immunogenic cell death in cancerNo DOI in the reference — shown for manual review; not independently verifiable (not a fabrication signal).
- NO DOIQuantification of regulatory T cells enables the identification of high-risk breast cancer patients and those at risk of late relapseNo DOI in the reference — shown for manual review; not independently verifiable (not a fabrication signal).
- NO DOICD4 + CD25 + regulatory T cells in patients with gastrointestinal malignancies: possible involvement of regulatory T cells in disease progressionNo DOI in the reference — shown for manual review; not independently verifiable (not a fabrication signal).
- NO DOISpecific recruitment of regulatory T cells in ovarian carcinoma fosters immune privilege and predicts reduced survivalNo DOI in the reference — shown for manual review; not independently verifiable (not a fabrication signal).
- NO DOIIntraepithelial CD8 + tumor-infiltrating lymphocytes and a high CD8 + /regulatory T cell ratio are associated with favorable prognosis in ovarian cancerNo DOI in the reference — shown for manual review; not independently verifiable (not a fabrication signal).
- NO DOIThe PD-1 expression balance between effector and regulatory T cells predicts the clinical efficacy of PD-1 blockade therapiesNo DOI in the reference — shown for manual review; not independently verifiable (not a fabrication signal).
- NO DOIPD-1(+) regulatory T cells amplified by PD-1 blockade promote hyperprogression of cancerNo DOI in the reference — shown for manual review; not independently verifiable (not a fabrication signal).
- NO DOIIpilimumab-dependent cell-mediated cytotoxicity of regulatory T cells ex vivo by nonclassical monocytes in melanoma patientsNo DOI in the reference — shown for manual review; not independently verifiable (not a fabrication signal).
- NO DOIIntegration of tumor inflammation, cell proliferation, and traditional biomarkers improves prediction of immunotherapy resistance and responseNo DOI in the reference — shown for manual review; not independently verifiable (not a fabrication signal).
- NO DOIProliferative potential and resistance to immune checkpoint blockade in lung cancer patientsNo DOI in the reference — shown for manual review; not independently verifiable (not a fabrication signal).
- NO DOIIntegrated genomic characterization of oesophageal carcinomaNo DOI in the reference — shown for manual review; not independently verifiable (not a fabrication signal).
- NO DOIAccelerating discovery of functional mutant alleles in cancerNo DOI in the reference — shown for manual review; not independently verifiable (not a fabrication signal).
- NO DOISentieon DNA pipeline for variant detection - software-only solution, over 20× faster than GATK 3.3 with identical resultsNo DOI in the reference — shown for manual review; not independently verifiable (not a fabrication signal).
- NO DOIThe Genome Analysis Toolkit: a MapReduce framework for analyzing next-generation DNA sequencing dataNo DOI in the reference — shown for manual review; not independently verifiable (not a fabrication signal).
- NO DOIFast and accurate long-read alignment with Burrows–Wheeler transformNo DOI in the reference — shown for manual review; not independently verifiable (not a fabrication signal).
- NO DOILong read alignment based on maximal exact match seedsNo DOI in the reference — shown for manual review; not independently verifiable (not a fabrication signal).
- NO DOITumor mutational burden and efficacy of nivolumab monotherapy and in combination with ipilimumab in small-cell lung cancerNo DOI in the reference — shown for manual review; not independently verifiable (not a fabrication signal).
- NO DOISensitive detection of somatic point mutations in impure and heterogeneous cancer samplesNo DOI in the reference — shown for manual review; not independently verifiable (not a fabrication signal).
- NO DOIStrelka2: fast and accurate calling of germline and somatic variantsNo DOI in the reference — shown for manual review; not independently verifiable (not a fabrication signal).
- NO DOIIntegrating human sequence data sets provides a resource of benchmark SNP and indel genotype callsNo DOI in the reference — shown for manual review; not independently verifiable (not a fabrication signal).
- NO DOIdbSNP: the NCBI database of genetic variationNo DOI in the reference — shown for manual review; not independently verifiable (not a fabrication signal).
- NO DOIAnalysis of protein-coding genetic variation in 60,706 humansNo DOI in the reference — shown for manual review; not independently verifiable (not a fabrication signal).
- NO DOIThe mutational constraint spectrum quantified from variation in 141,456 humansNo DOI in the reference — shown for manual review; not independently verifiable (not a fabrication signal).
- NO DOIPerformance comparison of exome DNA sequencing technologiesNo DOI in the reference — shown for manual review; not independently verifiable (not a fabrication signal).
- NO DOIBioinformatic methods and bridging of assay results for reliable tumor mutational burden assessment in non-small-cell lung cancerNo DOI in the reference — shown for manual review; not independently verifiable (not a fabrication signal).
- NO DOIPathSeq: software to identify or discover microbes by deep sequencing of human tissueNo DOI in the reference — shown for manual review; not independently verifiable (not a fabrication signal).
- NO DOImclust 5: clustering, classification and density estimation using Gaussian finite mixture modelsNo DOI in the reference — shown for manual review; not independently verifiable (not a fabrication signal).
- NO DOIGISTIC2.0 facilitates sensitive and confident localization of the targets of focal somatic copy-number alteration in human cancersNo DOI in the reference — shown for manual review; not independently verifiable (not a fabrication signal).
- NO DOIGenomic landscape associated with potential response to anti-CTLA-4 treatment in cancersNo DOI in the reference — shown for manual review; not independently verifiable (not a fabrication signal).
- NO DOIControl-FREEC: a tool for assessing copy number and allelic content using next-generation sequencing dataNo DOI in the reference — shown for manual review; not independently verifiable (not a fabrication signal).
- NO DOISTAR: ultrafast universal RNA-seq alignerNo DOI in the reference — shown for manual review; not independently verifiable (not a fabrication signal).
- NO DOIdupRadar: a Bioconductor package for the assessment of PCR artifacts in RNA-seq dataNo DOI in the reference — shown for manual review; not independently verifiable (not a fabrication signal).
- NO DOIA scaling normalization method for differential expression analysis of RNA-seq dataNo DOI in the reference — shown for manual review; not independently verifiable (not a fabrication signal).
- NO DOIThe Molecular Signatures Database (MSigDB) hallmark gene set collectionNo DOI in the reference — shown for manual review; not independently verifiable (not a fabrication signal).
4 data/code links checked; 4 live.
- dataEGALIVEHTTP 200https://ega-archive.org/studies/EGAS50000000747Resolves to EGA (data repository).
- datahttps://clinicaltrials.gov/ct2/show/NCT02872116LIVEHTTP 200Resolves, but the content could not be matched to the paper.
- datahttp://www.vivli.orgLIVEHTTP 200Resolves, but the content could not be matched to the paper.
- codeGitHubLIVEHTTP 200https://github.com/kassambara/survminerResolves to GitHub (code 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.
- MINORtypoResults, OS by genomic subtypes“nivolumab-based regiments”→ nivolumab-based regimensTypo: 'regiments' should be 'regimens'.
- MINORconsistencyResults, Patient population“Of 1,581 patients randomized to receive nivolumab-plus-chemotherapy versus chemotherapy, 685 (43%) were evaluable by whole-exome sequencing (WES)”→ Ensure the denominator is consistent with the parent trial's total randomized population.The 1,581 number may refer to the combined arms; verify consistency with the parent trial.
- MINORclarityMethods, Statistical methods“Calculated P values were nominal (not adjusted for multiple testing unless otherwise stated), descriptive and not intended to demonstrate statistical significance.”→ Clarify that these are exploratory and not confirmatory.This is clear but could be emphasized in the abstract.
- MINORtypoResults, OS by genomic subtypes“regiments”→ regimensTypographical error.
- MINORconsistencyResults, OS by TMB and MSI status“TMB-high denotes ≥199 mutations/exome”→ Ensure consistent use of 'mutations/exome' vs 'mutations per exome' throughout.Minor inconsistency in terminology.
- MINORclarityMethods, Statistical methods“Calculated P values were nominal (not adjusted for multiple testing unless otherwise stated), descriptive and not intended to demonstrate statistical significance.”→ Clarify which analyses were adjusted for multiple testing, if any.Could be clearer.
The published work is robust and generally well-reported. An informed reader should weigh the exploratory nature of the biomarker analyses, the lack of multiple-testing correction, and the minor internal contradiction in patient counts. No erratum is urgently required, but the authors should consider clarifying the patient-count discrepancy and adding code availability to enhance reproducibility.
- 1.HIGHreportingIn the Results section, reconcile the discrepancy in WES-evaluable patient counts: the text reports 685 for nivolumab-plus-chemotherapy and 366 for nivolumab-plus-ipilimumab, but the total WES-evaluable population is 889, and the sum exceeds this. Clarify the denominators (e.g., WES-evaluable vs TMB-evaluable) in the text and figure legends.The internal contradiction in patient counts could confuse readers and undermine confidence in the reported analyses.
- 2.HIGHstatisticsIn the Results and Methods, report exact p-values for all key comparisons instead of thresholds like 'P < 0.1', or provide a supplementary table with exact p-values for the GES interaction analyses.Threshold-only p-values limit the reader's ability to assess the strength of evidence and are a common reviewer request.
- 3.HIGHdata codeIn the Data Availability section, add a statement about code availability, including a link to a public repository (e.g., GitHub) for custom analysis scripts, or state that code is available upon request.Code sharing is not mentioned, which reduces reproducibility and transparency.
- 4.MEDIUMethicsIn the Methods, specify the IRB/ethics committee that approved the study and provide the protocol approval number, even if it is a central IRB.Naming the specific ethics committee strengthens the ethics statement and is standard for clinical trials.
- 5.MEDIUMstatisticsIn the Methods, clarify the statistical handling of multiple testing in the GES analyses; state whether any correction was applied and, if not, acknowledge this as a limitation.The paper reports nominal p-values without adjustment; clarifying this helps readers interpret the findings appropriately.
- 6.MEDIUMreportingIn the Methods, explicitly state whether the biomarker analysis plan was pre-specified or post hoc, and if post hoc, acknowledge this limitation in the Discussion.Clarifying the exploratory nature of the analysis is important for interpreting the results.
- 7.MEDIUMreportingIn the Results, provide a CONSORT-style flow diagram for the biomarker-evaluable populations to clarify patient selection and the denominators for each analysis.A flow diagram would help resolve the patient-count discrepancies and improve transparency.
- 8.MEDIUMreportingIn the Methods, explicitly describe outlier handling for sequencing data, such as criteria for excluding samples beyond QC metrics.The QC criteria are described, but explicit outlier handling would be clearer.
- 9.MEDIUMreportingIn the Results, report the number of patients with missing data for each biomarker analysis to improve transparency.Missing data can affect the generalizability of the findings; reporting it is good practice.
- 10.MEDIUMcopyeditIn the Results, fix the typo 'regiments' to 'regimens' in the OS by genomic subtypes section.Typographical errors detract from the professionalism of the manuscript.
- 11.MEDIUMcopyeditIn the Results, ensure consistent use of 'mutations/exome' vs 'mutations per exome' throughout the TMB and MSI status section.Consistent terminology improves clarity.
- 12.LOWreportingIn the Discussion, expand on the limitations regarding the lack of independent replication and the exploratory nature of the analyses, and suggest specific validation studies.Strengthening the limitations section helps temper overgeneralization and guides future research.
- 13.LOWreportingIn the Discussion, avoid overstating the predictive value of biomarkers; emphasize the exploratory nature and need for prospective validation more strongly.Proportional conclusions are important for clinical interpretation.
- 14.LOWdata codeIn the Data Availability, include a data availability statement for the clinical data access process in the main text, not just in the supplementary.Making the data access process more prominent improves transparency.
- 15.LOWotherIn the Methods, specify the version of the Dako PD-L1 assay and any validation data, to fully identify the reagent.Complete reagent identification is important for reproducibility.
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.