Concerns Regarding Misclassification and Arithmetic Errors in MSI2 Cancer Prognosis Meta-Analysis

Academic fraud report · Geng Detector

Summary

This report evaluates the systematic review and meta-analysis titled "Prognostic value of Musashi 2 (MSI2) in cancer patients: A systematic review and meta-analysis" published in Frontiers in Oncology (DOI: 10.3389/fonc.2022.969632). Three major concerns were flagged and verified. First, reference 31 (Aly RM 2015), which studies adult B-acute lymphoblastic leukemia (ALL), was systematically misclassified as acute myeloid leukemia (AML) throughout the manuscript and used to conduct an AML subgroup analysis. Second, multiple arithmetic inconsistencies were identified in Table 3: the reported totals for Age subgroups (1414 vs. verified 1404), Gender (1650 vs. verified 1603), and Lymph node metastasis (1166 vs. verified 1159) do not match the sum of the constituent study sample sizes. Third, Table 1 contains formatting errors suggesting an unrefined draft was published, including a "Median" entry for a follow-up time and a quality score of "8/6" for one study. The misclassification of cancer types is a fundamental factual error in a meta-analysis, and the repeated summation mistakes suggest serious data extraction or statistical processing problems. These findings collectively warrant editorial investigation.

Verdict

🟠 Highly suspicious. The paper contains a fundamental cancer-type misclassification affecting a subgroup analysis, multiple arithmetic errors in summary tables, and evidence of an unrefined data-extraction draft being published. These issues raise serious questions about the reliability of the underlying data extraction and synthesis.

Key findings

Cancer-type misclassification (Finding 1): Reference 31 (Aly RM 2015), titled *Prognostic significance of MSI2 predicts unfavorable outcome in adult b-acute lymphoblastic leukemia*, was incorrectly labeled as acute myeloid leukemia (AML) in the main text, Table 1, and Table 2, and was used to construct an AML subgroup meta-analysis.
Arithmetic inconsistencies in Table 3 (Finding 2): Reported totals do not match the sum of constituent study sample sizes:
Age subgroup: reported 1414 vs. verified total of 1404 (studies 16, 19, 21, 22, 24, 25, 27, 28, 30, 32, 33, 34).
Gender subgroup: reported 1650 vs. verified total of 1603 (studies 16, 21, 27, 28, 30, 31, 32, 34, 35, 36, 37).
Lymph node metastasis: reported 1166 vs. verified total of 1159 (studies 16, 19, 21, 22, 24, 25, 26, 27, 28, 30, 34).
Draft-state Table 1 (Finding 3): Zhao HZ 2016 records "Median" in the follow-up time column instead of a numeric value; Thol F 2013 records "8/6" as a quality score, inconsistent with other single-number entries.

Evidence highlights

Misclassification severity: ALL and AML are distinct hematological malignancies; conflating them invalidates the AML subgroup pooled estimate and indicates the authors may not have consulted the original source.
Magnitude of arithmetic drift: Errors range from 7 to 47 patients, inconsistent with rounding and indicative of extraction/aggregation errors.
Formatting anomalies: "Median" and "8/6" entries strongly suggest the published Table 1 is an unedited working draft.

Notes

DOI: 10.3389/fonc.2022.969632
All three findings were marked as verified (✅ 成立) in the original report.
This automated AI-assisted assessment is for academic discussion only; final determination of misconduct requires investigation by the journal or relevant institution.
Confidence is high for the misclassification and arithmetic findings, which can be checked against the published tables and the cited primary sources.