Integrity review of 'Endothelial-Derived CCL7 Promotes Macrophage Polarization and Aggravates Septic Acute Lung Injury via CCR1-Mediated STAT1 Succinylation' (Adv. Sci., 2025)

Academic fraud report · Geng Detector

Summary

This report presents a 'questionable' (doubtful) verdict on the above paper by Li, Long, Zhu, Gu, Zhou, and Miao, published in Advanced Science in 2025. Two issues are flagged. First, the TUNEL assay description in the Experimental Section refers to 'paraffin-embedded intestinal tissue sections,' whereas the study investigates septic acute lung injury and the surrounding Methods and Results consistently describe lung tissue, indicating a copy-paste error from a previous intestinal-tissue protocol and raising concerns about whether the described experiments were actually performed on lung tissue. Second, the authors claim that CCL7 'could serve as a prognostic marker' on the basis of a ROC AUC of 0.67 (Figure S2H), which is generally regarded as poor-to-unre discriminatory accuracy in clinical prediction, suggesting overstatement of clinical significance. Image-duplication, image-tampering, and digit-randomness analyses could not be performed because raw images and numeric source tables were unavailable. The verdict is moderate and based on textual evidence only; definitive determination of misconduct requires institutional investigation.

Verdict

🟡 Questionable. Two notable concerns were identified in the manuscript text, but image-level and statistical-digit analyses could not be conducted due to lack of source files. Final determination requires institutional review.

Key findings

Organ-name inconsistency in TUNEL Methods: The Experimental Section describing the TUNEL assay refers to 'paraffin-embedded intestinal tissue sections,' while the study's title, abstract, Results, and the rest of the Methods concern septic acute lung injury using lung tissue. This is consistent with template/copy-paste carryover from a prior intestinal study rather than deliberate fabrication, but it reflects serious lapses in methodological rigor.
Overstated clinical conclusion based on weak ROC: In Results 2.1 / Figure S2H, the authors state that ROC analysis (AUC = 0.67) indicates CCL7 'could serve as a prognostic marker.' An AUC of 0.67 is generally considered to have poor-to-marginal discriminative ability (AUC > 0.8 is typically regarded as clinically useful), so the framing of CCL7 as a prognostic biomarker is not well supported by the presented data.

Evidence highlights

Finding 1 location: Experimental Section → 'Terminal Deoxynucleotidyl Transferase dUTP Nick End Labeling (TUNEL) Assay' paragraph. Verbatim phrasing: '*Paraﬃn-embedded intestinal tissue sections were subjected to TUNEL assays...*' — a direct contradiction with the lung-injury focus of the manuscript.
Finding 2 location: Results 2.1 and Figure S2H. The reported AUC is 0.67, used to support a claim of prognostic-marker potential for CCL7 in septic patients.
Detections not performed:
Image-reuse / image-splicing checks (Western blot backgrounds, histology staining patterns, flow cytometry plots) — no raw image files were available.
Last-digit / Benford-type randomness tests — the paper presents summary graphs with small n (n = 3, 5, or 6) and does not provide unmodified numeric tables.

Notes

DOI: 10.1002/advs.202506209
Corresponding author affiliation identified as the Department of Anesthesiology, Zhongshan Hospital, Fudan University (Changhong Miao group). Given the group's high publication output, reviewers may wish to cross-check other papers from the same group for similar template-misuse or organ-naming errors.
The findings here are textual and methodological; they do not by themselves constitute proof of data fabrication. The TUNEL paragraph error may reflect careless writing rather than a fabricated experiment, but the mismatch is conspicuous and the authors should be asked to clarify which tissue was actually sectioned and to supply original images.
The clinical-significance claim is a matter of interpretation and statistical reporting standards; a more conservative conclusion would be appropriate given the AUC of 0.67.
This report was AI-assisted and is intended for academic discussion only. Final judgments of academic misconduct must come from the relevant institutional or publisher investigation.