Quemliclustat and chemotherapy with or without zimberelimab in metastatic pancreatic adenocarcinoma: a randomized phase 1 trial.

Name: Verified Rigor Record — Quemliclustat and chemotherapy with or without zimberelimab in metastatic pancreatic adenocarcinoma: a randomized phase 1 trial.
Item: Quemliclustat and chemotherapy with or without zimberelimab in metastatic pancreatic adenocarcinoma: a randomized phase 1 trial.
Rating: 83
Author: Αlpha¹

Wainberg ZA, Manji GA, Bahary N, Ulahannan SV, Pant S, Spigel DR, Uboha NV, Oberstein PE, Saeed A, Beagle B, Kim JY, Wang N, Weeder B, Shitole S, Mrouj K, Scott JR, Ensign LG, DiRenzo DM, Walters MJ, Wu W, Kaplan A, Cho S, Kabbarah O, O'Reilly EM

Paper source

Open DOI

Αlpha¹ Rigor Review

Rigor Review Verified

Critical

High-priority

Action items to resolve · Critical / High

Confidence: moderateEngine: 7.6.0Source: full text

This Alpha1 Rigor Review uses AI Rigor Reviewers trained on a curated corpus of high-fidelity and retracted papers, with expert supervision and curation. It can still make mistakes; verify each finding against the source before relying on it.

The manuscript presents a phase 1b trial with a well-founded scientific premise, adequate demographic reporting, and transparent limitations. However, it fails to identify key resources (antibodies, cell line authentication, software) and has gaps in ethics committee naming, data/code deposition, and formal design elements like power analysis and outlier handling.

All eight dimensions were applicable and scored. The reviewers diverged on ethical approvals (pass vs. warn), key resources (pass vs. fail), and statistical analysis (pass vs. warn); evidence favoring the more conservative rating was adopted in each case. Verification components found no statistical errors, no retracted/not-found references, and no high-severity integrity concerns.

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

partialDiscussion, paragraph 2Reviewer 1
The median OS of 15.7 months in the Quemli100 cohort compares favorably with historical benchmarks.
Supported by SCA comparison showing median OS 15.7 vs 9.8 months, but the comparison is post hoc and not from a randomized concurrent control.
Evidence: Extended Data Fig. 2 shows OS Kaplan-Meier for Quemli100 vs SCA (HR=0.634, P=0.003). The paper cites historical benchmarks (NAPOLI 3, MPACT) with median OS ~9 months.
“The median OS of 15.7 months in the Quemli100 cohort compares favorably with the median OS of historical benchmarks of G/nP alone, including the phase 3 NAPOLI 3 study (median OS, 9.2 months) and the registrational phase 3 MPACT study (median OS, 8.7 months)”
partialDiscussionReviewer 2
An NR4A expression signature may represent a reasonable surrogate for assessing adenosine levels in the TME.
The claim is supported by strong in vitro and correlative clinical data showing that NR4A expression is regulated by adenosine and that downregulation by quemliclustat is linked to benefit. However, a direct measurement of adenosine levels in patient tumors is not presented, so 'surrogate' is supported but not definitively proven.
Evidence: In vitro mechanistic studies and correlative clinical analyses.
“Thus, an NR4A expression signature may represent a reasonable surrogate for assessing adenosine levels in the TME.”
partialResults, Efficacy; Post hoc SCA analysisReviewer 3
Clinical response rates and survival outcomes were encouraging in patients treated with quemliclustat-containing regimens.
ORR, PFS, and OS are reported in a phase 1b single-arm or small randomized arms with no concurrent control. The synthetic control arm analysis shows a significant OS improvement (HR=0.634, P=0.003), but this is post hoc and subject to residual confounding.
Evidence: Results: ORR 38% (Q+G/nP) and 25% (Q+G/nP+Z); median OS 19.4 and 14.6 months; Quemli100 vs SCA OS HR=0.634 (0.471-0.854), P=0.003.
“Median OS was significantly longer in the Quemli100 arm (15.7 months (95% CI: 12.4–20.9)) versus the SCA (9.8 months (95% CI: 7.8–11.4)) (P = 0.003).”
supportedAbstract, Results, DiscussionReviewer 1
Quemliclustat combined with G/nP with or without zimberelimab showed encouraging clinical response rates and survival.
Supported by reported ORR, DCR, and median OS in the Quemli100 cohort and SCA comparison.
Evidence: Table 3 reports confirmed ORR 38% (95% CI 21-58) in Q+G/nP arm and 25% (95% CI 15-37) in Q+G/nP+Z arm. Median OS 15.7 months in Quemli100 cohort. SCA comparison shows improved OS vs historical control (P=0.003).
“Quemliclustat and chemotherapy with or without zimberelimab in metastatic pancreatic adenocarcinoma: a randomized phase 1 trial”
supportedResults, Baseline tumor NR4A family expression, paragraph 1Reviewer 1
High tumor NR4A expression was associated with improved OS in ARC-8 but not in two external cohorts.
Supported by forest plots and Kaplan-Meier analyses showing significant association in ARC-8 BEP but not in PRINCE or MORPHEUS cohorts.
Evidence: Figure 2e shows HR for OS 0.678 (95% CI 0.49-0.95, P=0.0239) in ARC-8, with non-significant HRs in external cohorts. Figure 2g shows Kaplan-Meier curves for OS (P=0.015).
“The NR4A family expression significantly correlated with survival benefit in the BEP (PFS: hazard ratio = 0.732 (95% CI: 0.56–0.96); P = 0.0238; OS: hazard ratio = 0.678 (95% CI: 0.49–0.95); P = 0.0239) but not in the PRINCE G/nP + nivo or MORPHEUS G/nP clinical cohorts.”
supportedResults, Greater magnitude of reduction in tumor NR4A expression, paragraph 1Reviewer 1
Maximal downregulation of NR4A expression was associated with T cell activation and improved OS.
Supported by analysis of paired biopsies showing T cell activation signatures in the maximal decrease subgroup and significant OS improvement (HR=0.24, P=0.0035).
Evidence: Figure 3c,d show T cell activation signatures upregulated in maximal decrease subgroup. Figure 3f shows OS Kaplan-Meier with HR=0.24 (95% CI 0.08-0.67, P=0.0035).
“Maximal decrease in NR4A family expression after treatment was associated with a positive trend toward improved PFS (hazard ratio = 0.49 (95% CI: 0.22–1.08); P = 0.073) and was significantly associated with improved OS (hazard ratio = 0.24 (95% CI: 0.08–0.67); P = 0.0035)”
supportedResults, NR4A family expression is upregulated by adenosine, paragraph 1-2Reviewers 1, 2
NR4A family gene expression is upregulated by adenosine in vitro and by chemotherapy in human PDACs.
Supported by in vitro experiments showing NR4A upregulation by AMP/NECA and inhibition by quemliclustat/etrumadenant, and by snRNA-seq data from treated patients.
Evidence: Figure 2a,b show upregulation by AMP and inhibition by quemliclustat. Figure 2c shows upregulation in various cell types from chemotherapy-treated patients.
“Upregulation of NR4A family expression after addition of AMP was confirmed and extended to additional cell types using reverse transcription polymerase chain reaction (RT−PCR)”
supportedResults, A scarcity of activated T cells, paragraph 2Reviewer 1
Spatial tissue analyses revealed a scarcity of activated T cells near regions with high NR4A1 expression.
Supported by spatial analysis showing fewer IFNγ+ T cells within 50 μm of NR4A1 high cells and greater distance between CD3+ and PanCK+ cells in NR4A high tumors.
Evidence: Extended Data Fig. 9c,d show significantly lower IFNγ expression in NR4A1 high regions and higher proportion of IFNγ+ cells >50 μm away.
“There were significantly fewer IFNγ transcripts per cell in NR4A1 high versus NR4A1 low ROIs”
supportedResults, Safety, paragraph 1Reviewer 1
The safety profile was consistent with that of G/nP.
Supported by safety data showing most grade ≥3 TEAEs were related to G/nP rather than quemliclustat or zimberelimab.
Evidence: Table 2 shows grade ≥3 study drug-related TEAEs: gemcitabine related 69%, nab-paclitaxel related 69%, quemliclustat related 22%, zimberelimab related 28%.
“Most reported grade 3 or higher TEAEs were related to G/nP rather than to quemliclustat or zimberelimab.”
supportedAbstractReviewer 2
Quemliclustat combined with chemotherapy is safe and shows encouraging clinical activity in first-line mPDAC.
The safety profile is well-characterized and consistent with chemotherapy alone. Response rates (ORR 25-41%) and median OS (14.6-19.4 months) are presented across arms, which are favorable compared to historical benchmarks.
Evidence: Tables 2 and 3, Figure 1c-d
“In all treatment arms, the safety profile was consistent with that of G/nP. Clinical response rates and survival outcomes were encouraging.”
supportedAbstractReviewer 2
High tumor NR4A expression is associated with improved OS in ARC-8 but not in external cohorts.
The hazard ratios and p-values for NR4A expression are statistically significant in ARC-8 BEP but not in the PRINCE G/nP+nivo or MORPHEUS G/nP cohorts, supporting the predictive rather than prognostic claim.
Evidence: Figure 2d-g
“High tumor NR4A expression was associated with improved overall survival (OS) in ARC-8 but not in two external cohorts from the PRINCE (G/nP + nivolumab (nivo)) or Morpheus-PDAC (G/nP) trials.”
supportedAbstractReviewer 2
Spatial analyses reveal a scarcity of activated T cells near high NR4A1-expressing regions, consistent with an immunosuppressed TME.
Dual ISH and proximity analyses quantitively show fewer IFNγ+ T cells within 50 μm of NR4A1-high regions vs NR4A1-low regions.
Evidence: Extended Data Fig. 9, Extended Data Fig. 10
“Spatial tissue analyses revealed a scarcity of activated T cells near regions with high NR4A1 expression, consistent with an immunosuppressed tumor microenvironment.”

Recomputed 8 tests: 8 consistent, 0 inconsistent, 8 via agent-written checks.

3 integrity concerns flagged (0 high).

Rigor

Eight Dimensions4 pass · 4 warn

12 major claims3 partial9 supported

8 copyedit issues flagged: mostly consistency, clarity, typo.

Checked 65 references: 62 verified — 3 not checked.

Reproducibility

6 data/code links checked; 6 live.

datahttps://trials.arcusbio.com/our-transparency-policyLIVEHTTP 200
Resolves, but the content could not be matched to the paper.
datahttp://clinicaltrials.gov/study/NCT04104672LIVEHTTP 200
Resolves, but the content could not be matched to the paper.
datahttps://clinicaltrials.gov/study/NCT04104672LIVEHTTP 200
Resolves, but the content could not be matched to the paper.
datahttps://clinicaltrials.gov/ct2/show/NCT06608927LIVEHTTP 200
Resolves, but the content could not be matched to the paper.
dataGEOLIVEHTTP 200
https://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GSE202051
Resolves to GEO (data repository).
codeGitHubLIVEHTTP 200
https://github.com/ParkerICI/prince-trial-data
Resolves to GitHub (code repository).

Registered (2 IDs: ClinicalTrials.gov). No reporting guideline cited.

Alpha¹ Rigor Action ItemsFinal pass before journal submission

The manuscript requires substantive revision before submission, primarily to address the key resources fail (provider missing cell line authentication, antibody identifiers, and software version) and the ethics committee naming gap. Data/code deposition and study design details (blinding rationale, outlier handling) should also be addressed. The core clinical results appear sound based on the verification checks.

1.
HIGHrigor
Report manufacturer, catalog numbers, and lot numbers for all investigational drugs (quemliclustat, gemcitabine, nab-paclitaxel, zimberelimab) in the Methods (Drug preparation section).
Current text only gives drug names and doses; reviewers require source identification for reproducibility.
2.
HIGHrigor
Provide authentication details (e.g., STR profiling) and mycoplasma testing results for all cell lines used (MIA PaCa-2, PANC-1, HCT-116, NCI-H650) in the Methods (Cell culture section).
Lack of authentication is a common reproducibility issue; journals increasingly require this information.
3.
HIGHrigor
Identify all antibodies used in ISH and mIF assays with vendor, catalog number, clone, and dilution in the Methods (Spatial analysis section).
Without antibody identifiers, the biomarker analyses cannot be reproduced.
4.
HIGHdata code
Deposit RNA-seq and spatial transcriptomics data in a public repository (e.g., GEO) and provide accession numbers in the Data Availability section.
The paper's biomarker conclusions rely on these data; deposition is required for transparency and reproducibility.
5.
HIGHethics
List the approving ethics committee(s)/IRB name(s) and protocol number(s) in the main text (Methods, Study design) rather than only in a supplementary table.
A cross-reference to a supplement for the ethics committee name is insufficient per standard reporting guidelines.
6.
HIGHstatistics
Name the statistical software (e.g., R version 4.x, SAS 9.4) and describe the procedure for verifying proportional hazards assumptions for Cox models in the Statistical Analysis section.
Software version and assumption checks are standard expectations for statistical reporting.
7.
HIGHreporting
Reference the CONSORT reporting guideline for clinical trials and provide the completed checklist as supplementary material.
Adherence to CONSORT is expected for trial reports; its absence could delay or complicate review.
8.
HIGHreporting
Share custom analysis code (e.g., R scripts for NR4A signature calculation, propensity score matching) in a public repository (GitHub/Zenodo) with a persistent DOI in the Data Availability section.
Custom code is necessary to evaluate the biomarker and SCA analyses; its absence reduces reproducibility.
9.
HIGHrigor
Add a statement in the Study Design section explaining the rationale for the open-label design and discuss its potential impact on investigator-assessed endpoints such as response rate.
Open-label design is acceptable but the rationale and potential bias should be transparent.
10.
HIGHrigor
Describe the handling of outliers and missing data for safety and efficacy analyses in the Statistical Analysis section.
Outliers and missing data can influence results; the current Methods do not address this.
11.
MEDIUMcopyedit
Standardize the abbreviation for the 100 mg quemliclustat arm throughout the manuscript (e.g., consistently use 'Quemli100' or 'quemliclustat 100 mg').
Inconsistent use (e.g., 'Quemli100' vs 'Quemli 100') may confuse readers.
12.
MEDIUMcopyedit
Ensure figure references are consistently formatted (e.g., 'Fig. 1' in text and 'Figure 1' in captions, not both).
Minor formatting inconsistency but can be flagged during peer review.
13.
LOWreporting
Explicitly state in the Statistical Analysis section that no multiplicity correction was applied to the numerous post hoc exploratory analyses (e.g., NR4A best cut, subgroup analyses), or clarify that these are purely descriptive.
Multiplicity is a common concern; transparency about handling reduces potential criticism.
14.
LOWreporting
Ensure the CONSORT flow diagram (Figure 1) legend defines all abbreviations (RP2D, PD, etc.) and explains the patient flow succinctly.
Legends should be self-contained for accessibility.

Full text (92,946 chars)Analyzed: Jul 6, 2026

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