Concerns regarding data integrity and reporting inconsistencies in 'Compound kushen injection inhibits breast cancer lung metastasis through regulating MTSS1/ARPC3/F-actin' (J. Ethnopharmacol., 2025; DOI: 10.1016/j.jep.2025.120054)

Academic fraud report · Geng Detector

Summary

This report assesses a 2025 Journal of Ethnopharmacology paper on Compound kushen injection (CKI) and breast cancer lung metastasis. Overall verdict: highly suspicious. Five concerns are raised. (1) Figure 1A: reported MCF-7 IC50 values increase from 4.038 mg/mL at 12 h to 5.232 mg/mL at 24 h and 5.090 mg/mL at 48 h, contradicting the claim of time-dependent inhibition—biologically implausible. (2) Figure captions inconsistently alternate between SD and SEM across Figures 1, 2, 4, 5, 6, and 7, suggesting sloppy assembly. (3) Table 1 lists CKI quantitative-analysis values such as Matrine = 708586533338.32 and Oxymatrine = 322889676590.98 with no stated units, almost certainly raw or erroneous output rather than concentrations. (4) Animal ethics approval (R062020G005, Yunnan University of Chinese Medicine) expired 2024-12-31, only 5 days before the stated submission date of 2025-01-05, making the claimed experimental timeline implausible. (5) Image-level analysis was limited because the PDF states 'Low-resolution images were used to create this PDF.' Confidence in the textual findings is high; pixel-level image verification remains pending.

Verdict

🟠 Highly suspicious. Multiple independent concerns—illogical IC50 kinetics, absent units in Table 1, inconsistent SD/SEM usage, and an expired ethics approval dated only days before submission—together suggest serious problems in data handling, reporting, or both. Image-based verification could not be completed due to low-resolution figures.

Key findings

Implausible MCF-7 IC50 time course (Figure 1A). Reported IC50 at 12 h = 4.038 mg/mL; at 24 h = 5.232 mg/mL; at 48 h = 5.090 mg/mL. Higher concentrations should be required for *less* inhibition at earlier times, not more. Trend contradicts the authors' own 'time-dependent decrease' claim.
Inconsistent dispersion statistics across figures. Captions alternate between SEM and SD: Fig. 1 (SEMs), Fig. 2 (SDs), Fig. 4 (SEM), Fig. 5 (SEM), Fig. 6 (SD), Fig. 7 (SEM). A single study should report one metric consistently.
Table 1 quantitative-analysis column lacks units and contains implausible magnitudes. Matrine = 708586533338.32; Oxymatrine = 322889676590.98. No unit (e.g., ng/mL, μg/mg) is provided, making the values uninterpretable and indicative of raw/unprocessed or fabricated numbers.
Animal ethics approval effectively expired at submission. Approval R062020G005 (Yunnan University of Chinese Medicine) valid through 2024-12-31; manuscript submission stated as 2025-01-05, a 5-day gap that is inconsistent with the extensive in vivo work described (13-day treatment, histopathology, flow cytometry, proteomics).
Image analysis limited. The source PDF carries the notice 'Low-resolution images were used to create this PDF,' preventing pixel-level checks of Western blots, flow cytometry plots, and any potential splicing.

Evidence highlights

DOI: 10.1016/j.jep.2025.120054
Figure 1A MCF-7 IC50 values: 12 h = 4.038 mg/mL; 24 h = 5.232 mg/mL; 48 h = 5.090 mg/mL
Table 1 example values: Matrine = 708586533338.32; Oxymatrine = 322889676590.98 (no units)
Ethics approval: R062020G005, expiry 2024-12-31; submission 2025-01-05
Figure-caption SD/SEM pattern documented above (Figures 1, 2, 4, 5, 6, 7)

Notes

All numerical values are reproduced exactly as quoted in the source report; no new values have been inferred.
Image-level reuse or splicing could not be confirmed or ruled out due to the low-resolution PDF. This remains an open item.
The findings concern reporting integrity and internal consistency; a formal determination of misconduct requires institutional investigation and access to raw data (e.g., CCK-8 OD readings, LC-MS chromatograms and calibration, full ethics file).