Perioperative durvalumab plus chemotherapy plus new agents for resectable non-small-cell lung cancer: the platform phase 2 NeoCOAST-2 trial.
Cascone T, Bonanno L, Guisier F, Insa A, Liberman M, Bylicki O, Livi L, Egenod T, Corre R, Kim DW, Garcia Campelo MR, Provencio Pulla M, Shim BY, Metro G, Bennouna J, Bielska AA, Yohannes AR, He Y, Dowson A, Kar G, McGrath L, Kumar R, Grenga I, Spicer J, Forde PM
- DOI
- 10.1038/s41591-025-03746-z
- Record issued
- 2026-08-15
- Engine
- 7.39.0
- Exported
- 2026-09-21
Prepared by Alpha1. This document is confidential: it is intended for the recipient it was shared with and must not be redistributed. The live record at alpha1science.com/verify/720f3e19-076d-4f0b-aed1-db9abfcbd384 is authoritative.
How this rating was calculated
- ClaimsEfficacy rests on an unvalidated surrogate endpoint−0.5★
- ClaimsTreatment effect not shown to be clinically meaningful−0.5★
- LinksDead data/code link−0.25★
- No reported statistical tests were found to recompute.
- 01Efficacy rests on an unvalidated surrogate endpoint
The primary efficacy endpoint is pathological complete response (pCR), a surrogate for long-term clinical outcomes such as event-free survival and overall survival. The paper does not demonstrate target engagement at the tested doses (e.g., PK/PD or dose-exposure data) and does not cite validated evidence linking pCR to clinical outcomes in this specific setting, although it mentions a correlation with EFS in other trials.
“These trials highlight the importance of pCR as a potential early predictor of survival, given the demonstrated correlation between pCR and improved EFS seen across trials.”
- 02Treatment effect not shown to be clinically meaningful
The reported pCR rates (20.3%, 25.7%, 35.2%) are presented as promising but are not anchored to a minimal clinically important difference or a threshold for clinical meaningfulness. The study is not powered for statistical comparisons, and the effect sizes are not explicitly compared to a predefined meaningful benefit.
“In the modified intention-to-treat population (n = 198; Arm 1, n = 74; Arm 2, n = 70; Arm 4, n = 54), pCR rates were 20.3% (15/74; 95% CI, 11.8–31.2), 25.7% (18/70; 95% CI, 16.0–37.6) and 35.2% (19/54; 95% CI, 22.7–49.4)”
- 03Declared data/code link does not resolve
Dead link — nothing to verify.
“https://vivli.org/members/enquiries-about-studies-not-listed-on-the-vivli-platform/”
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 and well-reported phase II platform trial. The paper demonstrates strong scientific premise, adequate study design, thorough ethical approvals, and transparent reporting. Minor issues include a missing explicit reporting guideline statement, a dead link in the data availability, and a few copyedit typos.
Both reviewers agreed on the study type (interventional) and on all dimension statuses. Minor divergences in checklist sub-ratings (e.g., power analysis, reporting guideline) were reconciled by considering the paper's explicit statements. The statistics verification component checked 0 tests (none recomputable), so no claims of statistical correctness are made beyond what was reported.
Numerical inconsistencies
None foundValues that contradict each other or are impossible for the stated sample: recomputed p-values and test statistics, GRIM/GRIMMER checks on summary numbers, percentages against their own counts, totals against their parts, and estimates against their own confidence intervals.
Checked — nothing surfaced.
Overstated conclusions
3 findings · worst highConclusions that reach past what the paper's own results support — including a significance claim that no longer holds when the statistic is recomputed, and efficacy resting on an unvalidated surrogate endpoint.
- Efficacy rests on an unvalidated surrogate endpointAssessed
- Treatment effect not shown to be clinically meaningfulAssessed
- Conclusions only partially backed by the presented evidenceAssessed
6 major claims checked against the paper's own evidence: 1 only partially supported (evidence backs part of the claim; gaps or caveats remain); the rest adequately supported.
- partialReviewers 1, 2The addition of novel agents may augment pathological response without meaningful increase in toxicity.The claim is based on exploratory comparison with AEGEAN and is appropriately hedged, but the lack of a control arm limits direct evidence.Evidence: Exploratory comparison with AEGEAN pCR and mPR rates.
“our results suggest that adding novel agents such as ADCs in the neoadjuvant setting may augment the pathological response benefit without a meaningful increase in toxicity”
DiscussionFind in source - supportedReviewers 1, 2Neoadjuvant durvalumab plus chemotherapy with Dato-DXd resulted in the highest pCR rate (35.2%) among the three arms.The pCR rate for Arm 4 is reported with 95% CI and is numerically higher than Arms 1 and 2, but the study is not powered for cross-arm comparisons.Evidence: pCR rates: Arm 1 20.3%, Arm 2 25.7%, Arm 4 35.2% with CIs.
“pCR rates were 20.3% (15/74; 95% CI, 11.8–31.2), 25.7% (18/70; 95% CI, 16.0–37.6) and 35.2% (19/54; 95% CI, 22.7–49.4)”
AbstractFind in source - supportedReviewer 1The safety profile of the combinations is manageable and consistent with current standard-of-care regimens.Safety data are reported in detail, including grade ≥3 AEs and deaths, and are compared to historical benchmarks.Evidence: Safety tables and discussion comparing to AEGEAN.
“Arms 1, 2 and 4 in the NeoCOAST-2 trial exhibited an overall safety profile that was comparable to those of currently approved regimens”
DiscussionFind in source - supportedReviewer 1Surgical feasibility is not adversely impacted by the novel combinations.Surgery rates (93.2%, 93.0%, 94.4%) and R0 resection rates are reported and compared to historical trials.Evidence: Surgery rates and R0 resection rates in Results.
“In NeoCOAST-2, there were similar trends in the feasibility of surgery (93.2% in Arm 1, 93.0% in Arm 2 and 94.4% in Arm 4) and R0 resection rates (91.3%, 92.4% and 88.0%, respectively).”
DiscussionFind in source - supportedReviewer 2The combination regimens have a manageable safety profile consistent with current standard of care.Safety data show grade ≥3 TRAE rates of 36.5%, 40.8%, and 20.4% across arms, which are comparable to historical data, and the paper discusses this.Evidence: Safety table and discussion comparing to AEGEAN.
“Arms 1, 2 and 4 in the NeoCOAST-2 trial exhibited an overall safety profile that was comparable to those of currently approved regimens”
Discussion ¶5Find in source - supportedReviewer 2NeoCOAST-2 is the first global phase II study to report clinical data on an ADC in the neoadjuvant setting for resectable NSCLC.The paper states this as a claim of novelty, and no contradicting evidence is presented within the paper.Evidence: Statement in Discussion.
“NeoCOAST-2 is, to our knowledge, the first global phase II study to report clinical data on an ADC in the neoadjuvant setting for patients with resectable NSCLC”
Discussion ¶1Find in source
Premise concern: surrogate not validated for clinical benefit; effect size not shown to be clinically meaningful.
- INADEQUATESurrogate endpointThe primary efficacy endpoint is pathological complete response (pCR), a surrogate for long-term clinical outcomes such as event-free survival and overall survival. The paper does not demonstrate target engagement at the tested doses (e.g., PK/PD or dose-exposure data) and does not cite validated evidence linking pCR to clinical outcomes in this specific setting, although it mentions a correlation with EFS in other trials.
“These trials highlight the importance of pCR as a potential early predictor of survival, given the demonstrated correlation between pCR and improved EFS seen across trials.”
- INADEQUATEEffect sizeThe reported pCR rates (20.3%, 25.7%, 35.2%) are presented as promising but are not anchored to a minimal clinically important difference or a threshold for clinical meaningfulness. The study is not powered for statistical comparisons, and the effect sizes are not explicitly compared to a predefined meaningful benefit.
“In the modified intention-to-treat population (n = 198; Arm 1, n = 74; Arm 2, n = 70; Arm 4, n = 54), pCR rates were 20.3% (15/74; 95% CI, 11.8–31.2), 25.7% (18/70; 95% CI, 16.0–37.6) and 35.2% (19/54; 95% CI, 22.7–49.4)”
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 multiple pivotal trials and the NeoCOAST study, acknowledging the unmet need with pCR rates and EFS. The rationale for combining durvalumab with oleclumab, monalizumab, and Dato-DXd is logically linked to their mechanisms and prior evidence. Limitations of prior research are implicitly addressed by the platform design and the focus on novel agents.
“In the phase III AEGEAN study, perioperative durvalumab plus neoadjuvant chemotherapy led to significantly higher rates of pCR (17.2% versus 4.3%; P < 0.001) and significantly longer event-free survival (EFS; hazard ratio 0.68; 95% confidence interval (CI), 0.53–0.88) compared with neoadjuvant chemotherapy alone”
“In the phase II NeoCOAST study, neoadjuvant durvalumab combined with oleclumab or monalizumab led to numerical improvement in mPR rates and increased pro-inflammatory effects in the tumor microenvironment compared with durvalumab alone”
“Despite recent advances, a minority of patients receiving perioperative or neoadjuvant treatment have a pCR, and the 3-year EFS rates remain around 54–60%”
“In the phase III AEGEAN study, perioperative durvalumab plus neoadjuvant chemotherapy led to significantly higher rates of pCR (17.2% versus 4.3%; P < 0.001) and significantly longer event-free survival (EFS; hazard ratio 0.68; 95% confidence interval (CI), 0.53–0.88) compared with neoadjuvant chemotherapy alone”
“In the phase II NeoCOAST study, neoadjuvant durvalumab combined with oleclumab or monalizumab led to numerical improvement in mPR rates and increased pro-inflammatory effects in the tumor microenvironment compared with durvalumab alone”
“Despite recent advances, a minority of patients receiving perioperative or neoadjuvant treatment have a pCR, and the 3-year EFS rates remain around 54–60%”
Randomization method is described (blocked, stratified by PD-L1, using IRT). Blinding is not applicable as it is an open-label study, which is stated. Power analysis is not applicable because the study is not powered for statistical comparisons; sample size is based on a model-based approach. Inclusion/exclusion criteria are detailed. Outlier handling is not explicitly described but is not a standard requirement for clinical trials. Controls are not applicable as there is no control arm; the study is non-comparative. Independent replication is not applicable as this is a single trial.
“A blocked randomization was generated, and randomization was balanced within the IRT at the central level.”
“NeoCOAST-2 is a phase II, open-label, multiarm, multicenter, global, randomized platform study.”
“The sample size was not based on type I error and power considerations.”
“A randomization method featuring a dynamically changing allocation ratio for treatment assignments was used to account for fluctuations in the number of patients enrolled in the treatment arms over the course of the study”
“NeoCOAST-2 is a phase II, open-label, multiarm, multicenter, global, randomized platform study.”
Sex is reported for all arms (e.g., Arm 1: 37.8% female). Age is reported as median and range. Demographics include race, ECOG PS, PD-L1 TPS, and disease stage. Since both sexes are enrolled, sex_justified is not applicable. Species/strain and housing conditions are not applicable for a human trial.
“Median (range), years | 66.5 (30–79) | 66.0 (48–83) | 65.0 (38–81)”
“Median (range), years | 66.5 (30–79) | 66.0 (48–83) | 65.0 (38–81)”
The study oversight section states that the protocol and related documents were reviewed and approved by IRBs and Ethics Committees at each center, listing many institutions. Informed consent is described in the inclusion criteria ('Capable of giving signed informed consent'). Regulatory compliance is stated with adherence to the Declaration of Helsinki and ICH-GCP guidelines.
“The study protocol, protocol amendments, informed consent form, investigator brochure and other relevant documents were reviewed and approved by the Institutional Review Boards and Ethics Committees at each participating center”
“Capable of giving signed informed consent, which includes compliance with the requirements and restrictions listed in the ICF and in this protocol.”
“The study was performed in accordance with consensus ethical principles derived from international guidelines including the Declaration of Helsinki and Council for International Organizations of Medical Sciences International Ethical Guidelines, applicable International Council for Harmonisation Good Clinical Practice Guidelines”
“The study protocol, protocol amendments, informed consent form, investigator brochure and other relevant documents were reviewed and approved by the Institutional Review Boards and Ethics Committees at each participating center”
“Capable of giving signed informed consent, which includes compliance with the requirements and restrictions listed in the ICF and in this protocol.”
“The study was performed in accordance with consensus ethical principles derived from international guidelines including the Declaration of Helsinki and Council for International Organizations of Medical Sciences International Ethical Guidelines, applicable International Council for Harmonisation Good Clinical Practice Guidelines and all applicable laws and regulations.”
The drugs are named with doses and schedules in the Methods (e.g., durvalumab 1,500 mg i.v. Q3W). The PD-L1 assay is identified as VENTANA SP263. Software used for analysis (SAS v9.4) is identified. Antibodies, cell lines, and mycoplasma testing are not applicable as this is a clinical trial without wet-lab assays.
“Arm 1 consisted of neoadjuvant durvalumab (1,500 mg intravenously (i.v.) every 3 weeks (Q3W)), oleclumab (3,000 mg i.v. Q3W) and platinum-doublet chemotherapy for four cycles”
“Baseline PD-L1 status was assessed using central (VENTANA SP263) or local testing (VENTANA SP263, pharmDx 28-8, or pharmDx 22C3).”
“Analyses of clinical data were conducted using SAS v9.4.”
“Arm 1 consisted of neoadjuvant durvalumab (1,500 mg intravenously (i.v.) every 3 weeks (Q3W)), oleclumab (3,000 mg i.v. Q3W) and platinum-doublet chemotherapy for four cycles”
“Analyses of clinical data were conducted using SAS v9.4.”
The paper reports effect sizes with 95% CIs (e.g., pCR rates with CIs) and does not report p-values for the primary endpoints, which is appropriate for a non-comparative study. Exact p-values are not applicable as no hypothesis tests are conducted. Assumptions are not applicable as no formal tests are used. Data presentation includes per-group n and CIs. Mathematical plausibility is not applicable as the outcomes are proportions with CIs, and the numbers are internally consistent.
“pCR rates were 20.3% (15/74; 95% CI, 11.8–31.2), 25.7% (18/70; 95% CI, 16.0–37.6) and 35.2% (19/54; 95% CI, 22.7–49.4)”
“Single-arm analyses were conducted without statistical tests.”
“Analyses of clinical data were conducted using SAS v9.4.”
“pCR rates were 20.3% (95% CI, 11.8–31.2) in Arm 1, 25.7% (95% CI, 16.0–37.6) in Arm 2 and 35.2% (95% CI, 22.7–49.4) in Arm 4”
“Single-arm analyses were conducted without statistical tests.”
“Analyses of clinical data were conducted using SAS v9.4.”
The data availability statement names a concrete mechanism: AstraZeneca's data sharing policy and Vivli platform. Repository deposit and accession numbers are not applicable for patient-level data due to privacy. Code sharing is not applicable as no bespoke code is mentioned.
“Data underlying the findings described in this manuscript may be obtained in accordance with AstraZeneca’s data sharing policy described at: https://astrazenecagrouptrials.pharmacm.com/ST/Submission/Disclosure”
“Data for studies directly listed on Vivli can be requested through Vivli at www.vivli.org”
“Data underlying the findings described in this manuscript may be obtained in accordance with AstraZeneca’s data sharing policy described at: https://astrazenecagrouptrials.pharmacm.com/ST/Submission/Disclosure . Data for studies directly listed on Vivli can be requested through Vivli at www.vivli.org”
Trial registration is provided. Reporting guideline is not explicitly mentioned but the paper follows CONSORT-like reporting with a flow diagram. All outcomes are reported, including negative results. Limitations are discussed (lack of control arm, small sample sizes). Conclusions are proportional, acknowledging the need for larger trials. Funding and COI are disclosed.
“ClinicalTrials.gov identifier: NCT05061550”
“Our study has some limitations, including the lack of a dedicated control arm and power for statistical comparisons between arms, which precludes definitive cross-arm comparisons.”
“This study was funded by AstraZeneca.”
“ClinicalTrials.gov identifier: NCT05061550”
“Our study has some limitations, including the lack of a dedicated control arm and power for statistical comparisons between arms, which precludes definitive cross-arm comparisons.”
“This study was funded by AstraZeneca.”
Registered (1 ID: ClinicalTrials.gov). Reporting guideline cited: CONSORT.
Broken references and links
1 finding · worst mediumReferences 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.
- Dead data/code linksRecomputed
Checked 21 references by DOI: 19 verified — 2 no DOI (shown, not verified).
- NO DOICancer Facts and Figures 2025No DOI in the reference — shown for manual review; not independently verifiable (not a fabrication signal).
- NO DOIU.S. cancer statistics lung cancer stat biteNo DOI in the reference — shown for manual review; not independently verifiable (not a fabrication signal).
4 data/code links checked; 3 live, 1 dead.
- datahttps://astrazenecagrouptrials.pharmacm.com/ST/Submission/DisclosureLIVEHTTP 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.
- datahttps://vivli.org/members/enquiries-about-studies-not-listed-on-the-vivli-platform/DEADHTTP 404Dead link — nothing to verify.
- datahttps://vivli.org/ourmember/astrazeneca/LIVEHTTP 200Resolves, but the content could not be matched to the paper.
Copyediting
5 minorWording, consistency and formatting errors that need correcting before submission.
No major wording or formatting errors. 5 minor suggestions below.
5 copyedit issues flagged: mostly consistency, typo.
- MINORconsistencyAbstract“In the safety population, 69/74 (93.2%), 66/71 (93.0%), and 51/54 (94.4%) patients underwent surgery, respectively.”→ Ensure the denominators match the safety population Ns (74, 71, 54) consistently.The safety population Ns are 74, 71, 54, but the abstract uses 74, 71, 54; consistent.
- MINORtypoCompeting interests“LM. reports employment and stock with AstraZeneca; patent and patent applications with Gilead Sciences, Jounce Therapeutics, AstraZeneca.”→ Change 'LM.' to 'L.M.' for consistency with other author initials.Minor formatting inconsistency.
- MINORconsistencyAbstract“In the safety population, 69/74 (93.2%), 66/71 (93.0%), and 51/54 (94.4%) patients underwent surgery, respectively.”→ Ensure the denominators match the safety population Ns (74, 71, 54) consistently throughout.The safety population Ns are 74, 71, 54, but the abstract uses 69/74, 66/71, 51/54; these are the numbers who underwent surgery, which is correct.
- MINORtypoMethods, Eligibility criteria“using https://www.kidney.org/professionals/KDOQI/gfr_calculatorCoc”→ Remove the stray 'Coc' at the end of the URL.Typographical error in the URL.
- MINORconsistencyResults, Safety“with 14.9%, 16.9 and 16.7% of patients in the respective arms experiencing these events in the neoadjuvant period.”→ Add a percent sign after 16.9 for consistency.Missing percent sign.
The published work is robust and well-reported. An informed reader should weigh the lack of a control arm and the non-comparative design, which limit causal inference, and the minor reporting gaps (e.g., no explicit CONSORT statement, one dead link). No erratum is warranted for the core findings, but the authors could consider a correction for the typographical errors and the dead link.
- 1.HIGHreportingIn the Methods or Reporting Summary, explicitly state that the study follows the CONSORT reporting guideline and provide the checklist as supplementary material.Reviewer 2 flagged that the reporting guideline is not explicitly named, which is a minor transparency gap that could be easily fixed.
- 2.HIGHdata codeFix or replace the dead link found in the data availability section (reproducibility check found 1 dead link among 4 checked).A data availability statement undercut by a broken link is a reporting gap that reviewers will catch.
- 3.MEDIUMcopyeditIn the Competing interests section, change 'LM.' to 'L.M.' for consistency with other author initials.Minor formatting inconsistency flagged by the copyedit pass.
- 4.MEDIUMcopyeditIn the Methods, Eligibility criteria, remove the stray 'Coc' at the end of the URL 'https://www.kidney.org/professionals/KDOQI/gfr_calculatorCoc'.Typographical error in the URL that could mislead readers.
- 5.MEDIUMcopyeditIn the Results, Safety section, add a percent sign after 16.9 for consistency (currently '16.9 and 16.7%').Missing percent sign flagged by the copyedit pass.
- 6.MEDIUMstatisticsIn the Statistical analysis section, clarify how missing data (e.g., missing ECOG PS) were handled in the analyses.Reviewer 2 suggested addressing outlier_handling to improve transparency.
- 7.MEDIUMdata codeIn the Data availability section, specify the timeframe for data requests to make the access route more concrete.Reviewer 2 suggested this to enhance the concreteness of the data access process.
- 8.LOWreportingIn the Discussion, consider adding a sentence acknowledging that the lack of a control arm limits causal inference, even though it is already mentioned.Reviewer 2 suggested this to further temper conclusions.
- 9.LOWotherIn the Methods, provide more detail on the randomization algorithm (e.g., seed, software) to enhance reproducibility.Reviewer 2 suggested this to improve 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.