Ovarian Cancer Study Retracted After Image Duplication Scandal

In a decision that underscores the growing scrutiny of cancer biology research, the editors of BMC Cancer have formally retracted a 2021 study that claimed a long non-coding RNA called LINC00852 drives the proliferation and invasion of ovarian cancer cells. The retraction notice, published on 14 September 2026, states bluntly that the editors have lost confidence in both the data and the conclusions of the article, which originally appeared on 8 September 2021 under the title describing how LINC00852 competitively binds miR-140-3p to regulate AGTR1 expression.

The original study proposed an elegant molecular mechanism. Long non-coding RNAs are transcripts that do not code for proteins but can act as molecular sponges, soaking up microRNAs and preventing them from suppressing their downstream targets. According to the paper, LINC00852 functioned as a competing endogenous RNA that sequestered miR-140-3p, thereby relieving inhibition of AGTR1, the angiotensin II receptor type 1 gene. This axis, the authors argued, promoted ovarian cancer cell growth and invasion, positioning LINC00852 as a potential therapeutic target in a disease that remains among the deadliest gynecological malignancies.

That mechanism now stands retracted. The retraction notice catalogs a disturbing pattern of image similarities spanning at least eight separate instances, connecting panels from the ovarian cancer paper to figures in five unrelated publications covering colon cancer, lung cancer, and even hyperoxia-induced lung injury in newborn rats. The editors identified duplicated images between Fig. 2C of the retracted article, showing SKOV-3 ovarian cancer cells with LINC00852 knockdown, and Fig. 2C of a 2021 BMC Cancer paper on LINC01287 in colon cancer cells. Additional overlaps linked Fig. 2E and Fig. 6B of the ovarian cancer study to the same colon cancer publication, while other panels matched figures from papers on Ganoderan in non-small cell lung cancer, HNF1B overexpression in A549 cells, tetrandrine in lung injury, and GANT61 in osteosarcoma.

Perhaps most tellingly, the editors also flagged internal duplication within the retracted paper itself. Fig. 2E row 1 panel 1, labeled as a SKOV-3 control, and Fig. 2E row 2 panel 1, labeled as an OV-90 control, were found to be partially similar to each other. In experimental terms, these panels should represent distinct cell lines serving as untreated baselines for comparison. Their partial overlap suggests that images intended to demonstrate independent experimental conditions may not have been generated independently at all, a hallmark of image manipulation that automated screening tools and vigilant readers increasingly detect across the scientific literature.

The technical implications of such duplication are severe. Western blot panels, microscopy images, and flow cytometry plots form the evidentiary backbone of studies like this one. When the same image appears under different experimental labels across multiple papers, the underlying data cannot support the biological conclusions drawn from them. In this case, the duplicated panels were central to demonstrating that LINC00852 knockdown reduced proliferation and invasion in SKOV-3 and OV-90 cells, and that miR-140-3p mimics could rescue the effects of LINC00852 overexpression. Without trustworthy primary data, the proposed ceRNA mechanism collapses entirely.

The retraction notice states that the authors, including Zhi-wei Qiao, Ying Jiang, Ling Wang, Lei Wang, Jing Jiang, Jing-ru Zhang, and Peng Mu, all affiliated with the Department of Gynaecology at the Cancer Hospital of China Medical University in Shenyang, did not respond to correspondence from the publisher about the retraction. This silence extends to the concerns raised about the data itself. When authors fail to engage with editors during a retraction investigation, the outcome typically reflects the severity of the evidence rather than any mitigating explanation.

This case fits within a broader pattern of concern surrounding image integrity in cancer research. The five papers cited in the retraction notice as sources of the duplicated images include one from Onco Targets and Therapy on Ganoderan signaling in lung cancer, one from Translational Cancer Research on PI3K/AKT phosphorylation, one from Annals of Translational Medicine on tetrandrine and NF-κB p65 inhibition, and one from Cell Biology International on GANT61 in osteosarcoma that itself carries a retraction notice. The web of connections suggests that images may have circulated across multiple manuscripts, potentially from shared laboratory resources or systematic reuse, rather than isolated error.

For the ovarian cancer research community, the retraction removes a claimed therapeutic target from the literature. LINC00852 had been positioned alongside other long non-coding RNAs as part of a growing class of transcripts implicated in gynecological malignancies. The competing endogenous RNA hypothesis, which underpinned the paper’s mechanism, remains an active area of investigation, with researchers exploring how lncRNA-miRNA-mRNA networks regulate tumor behavior. However, each retracted study in this space erodes confidence in the field’s accumulated findings and complicates meta-analyses that aggregate results across publications.

The retraction also highlights the evolving role of post-publication scrutiny. Modern image forensics tools can detect duplicated panels across thousands of papers by comparing pixel patterns, background textures, and band shapes. Journals increasingly rely on such tools during routine screening, and readers can flag suspicious figures through platforms like PubPeer. The BMC Cancer editors acted on concerns raised after publication, systematically documenting each similarity with precise figure and panel references before reaching the decision to retract. This methodical approach, while slower than some would like, builds a transparent record that other journals and institutions can act upon.

The retracted article remains accessible online with a watermarked retraction notice, preserving the scientific record while ensuring that readers encountering the work understand its status. The canonical DOI for the retraction note is 10.1186/s12885-026-16963-7, published as volume 26, article number 1092 of BMC Cancer. The original article, published under DOI 10.1186/s12885-021-08730-7 as volume 21, article number 1004, now carries the retraction as part of its permanent record. For researchers citing LINC00852 in ovarian cancer, the message is clear: the data supporting this mechanism can no longer be trusted, and the field must look elsewhere for reliable insights into the molecular drivers of this devastating disease.

Subject of Research: Scientific misconduct and research integrity

Article Title: Retraction Note: LINC00852 promotes the proliferation and invasion of ovarian cancer cells by competitively binding with miR-140-3p to regulate AGTR1 expression

Article References: Qiao, Z.-W., Jiang, Y., Wang, L., Wang, L., Jiang, J., Zhang, J.-R., & Mu, P. (2026). Retraction Note: LINC00852 promotes the proliferation and invasion of ovarian cancer cells by competitively binding with miR-140-3p to regulate AGTR1 expression. BMC Cancer, 26(1), Article 1092. https://doi.org/10.1186/s12885-026-16963-7

Image Credits: AI Generated

DOI: 10.1186/s12885-026-16963-7

Keywords: retraction, ovarian cancer, LINC00852, image duplication, BMC Cancer, research integrity, microRNA, ceRNA mechanism, Note, promotes, proliferation, invasion