Summary
This document serves as a formal retraction of the study titled "Polyphyllin I suppresses the formation of vasculogenic mimicry via Twist1/VE-cadherin pathway," originally published in Cell Death and Disease in 2018. Following an extensive editorial investigation, the journal’s Editors-in-Chief have withdrawn the paper due to significant concerns regarding data integrity and image manipulation. The findings regarding the anti-cancer properties of Polyphyllin I are no longer considered reliable by the scientific community.
Background: The Original Research
The 2018 study investigated the therapeutic potential of Polyphyllin I, a natural compound derived from the Paris polyphylla plant, in treating aggressive cancers. The authors focused on vasculogenic mimicry (VM)—a process where tumor cells form fluid-conducting channels to create their own blood supply, independent of traditional angiogenesis. This phenomenon often makes tumors more resilient to conventional treatments.
The original findings suggested that Polyphyllin I could inhibit this process by targeting the Twist1/VE-cadherin signaling pathway in various liver cancer cell lines, including Hep3B, HepG2, and Bel7402. The study utilized standard molecular biology techniques such as tube formation assays, Western blotting, and immunofluorescence to track protein expression and channel formation.
Editorial Investigation and Findings
Post-publication analysis and an internal audit by the editorial board revealed systemic issues with the visual data presented in the manuscript. The investigation focused on "frameshift" duplications—a sophisticated form of data manipulation where overlapping or slightly shifted microscopic fields are used to represent distinct experimental conditions. This technique can be difficult to detect without specialized software or highly detailed scrutiny.
Specific discrepancies were identified in the following areas:
- Figure 2A: Duplicated panels were found representing different experimental groups.
- Supplementary Figures 3B and 3C: Extensive duplications were noted across multiple cell lines (PLC, Hep3B, Bel7402, and HepG2) and treatment groups (comparing Control vs. VEGF-a+PPI).
- Figure 6C: Duplications were discovered within the original raw data files provided to the journal during the inquiry.
The Editors-in-Chief concluded that these overlapping images across different experimental conditions undermined the reliability of the entire dataset, leading to a total loss of confidence in the study's conclusions.
Conclusion and Current Status
As of November 2025, the paper is officially retracted. While two authors (Cheng Yang and Tao Sun) expressed disagreement with the decision, the remaining authors did not respond to the journal's inquiries.
The retraction highlights a critical challenge in oncology research: the detection of sophisticated image manipulation. The "frameshift" duplications identified here serve as a red flag for peer reviewers and editors. Consequently, the biological claims regarding Polyphyllin I’s effect on the Twist1/VE-cadherin pathway are now considered unsupported.
For the broader scientific community, this case underscores the importance of data transparency and the necessity of robust, AI-driven image screening tools during the initial review process. Future research in this area will require independent replication using rigorous, transparent methodologies to determine if Polyphyllin I holds any genuine clinical promise for suppressing vasculogenic mimicry.
Summary updated