The retracted paper, originally published by researchers Tang, Wang and Gu in 2018, reported that knocking down Sox2 — a transcription factor best known for its role in maintaining the pluripotency of embryonic stem cells — inhibited the invasion and migration of osteosarcoma cells by modulating the Wnt/β-catenin signaling pathway. That pathway is one of the most heavily studied signaling cascades in cancer research, acting as a molecular switch that can drive cell proliferation, survival and movement when improperly activated. The study appeared in volume 24 of the journal, pages 907 to 913, and was indexed under DOI 10.1007/s12253-018-0400-1.

Osteosarcoma, the malignancy at the center of the study, is the most common primary bone cancer in children and adolescents. Despite decades of research, treatment outcomes for patients with metastatic or relapsed disease have improved little, which explains why laboratory findings suggesting new molecular targets attract intense attention. Sox2, in particular, has been implicated in the behavior of cancer stem-like cells — a subpopulation of tumor cells thought to drive metastasis, therapy resistance and recurrence — making it a tempting therapeutic target for investigators seeking to blunt the aggressive spread of bone tumors.

The problem that led to the retraction centered on Figure 3A of the original article, specifically the control group images. According to the retraction notice, the authors themselves contacted the editorial office to report that these experimental images had been improperly cropped. Image cropping, when done to selectively present or omit data, can fundamentally distort what an experiment appears to show. In cell biology, where conclusions often rest on visual comparisons of stained cells, western blots or microscopy fields, the integrity of every panel in a figure is critical to the reliability of the entire dataset.

Following the authors’ disclosure, the journal launched an investigation in accordance with its own policies on publication ethics. That investigation confirmed the authors’ admission, concluding that the findings and conclusions reported in the article were no longer supported by the analyses and data. The retraction was subsequently approved by the Editor-in-Chief of Pathology & Oncology Research, and the authors agreed to the decision. The retraction notice was formally received on 28 July 2026, accepted the same day, and published on 13 August 2026 as part of volume 32 of the journal, under the new citation identifier 1612565 and DOI 10.3389/pore.2026.1612565.

Retractions initiated by the authors themselves are relatively uncommon in the scientific literature, and their occurrence is often viewed as a positive signal of self-correction within the research community. When scientists voluntarily flag problems in their own published work, it can prevent other laboratories from wasting time and resources attempting to replicate or build upon findings that cannot be trusted. In this case, the fact that the authors came forward to report the improper image cropping in their own figure suggests an acknowledgment that the study’s central claims — that Sox2 knockdown suppresses osteosarcoma cell invasion and migration through the Wnt/β-catenin pathway — could no longer be considered reliable.

The scientific premise of the retracted work, however, remains part of a broader and still-active research landscape. Sox2 belongs to the SOX family of transcription factors, which bind to DNA and regulate the expression of genes involved in cell fate decisions. In normal development, Sox2 helps maintain the undifferentiated state of stem cells. In cancer, aberrant Sox2 expression has been reported in numerous tumor types, where it is thought to contribute to stem-like properties, epithelial-mesenchymal transition and enhanced migratory capacity. The Wnt/β-catenin pathway, meanwhile, controls the stability and nuclear activity of β-catenin, a protein that, upon pathway activation, translocates to the nucleus and switches on genes promoting cell division and movement. Because both Sox2 and Wnt/β-catenin signaling are implicated in tumor aggressiveness, studies exploring crosstalk between the two have been attractive targets for osteosarcoma researchers.

For the osteosarcoma research community, the retraction serves as a reminder that individual findings, however plausible they may seem within an established biological framework, must be independently validated before they can inform therapeutic development. The claim that Sox2 knockdown modulates Wnt/β-catenin signaling to reduce invasion and migration in osteosarcoma cells is now formally withdrawn, and any future work building on that specific result will need to start from fresh, verifiable experimental evidence. Researchers citing the original 2018 paper are expected to reference the retraction notice going forward, and bibliographic databases will link the two records so that readers encountering the original study are alerted to its withdrawn status.

The case also highlights the ongoing challenge of image integrity in biomedical publishing. Improper cropping — whether intentional or the result of careless figure preparation — can alter the apparent relationship between experimental and control groups, and journals have increasingly adopted automated image-forensics tools to detect such manipulations before publication. In this instance, the problem surfaced after publication through the authors’ own admission, prompting a formal investigation and retraction under the journal’s established policies. The editorial office of Pathology & Oncology Research handled the process from receipt of the authors’ request through to the publication of the retraction notice within roughly two weeks, a relatively swift resolution by the standards of scientific self-correction.

The retracted article and the retraction notice are both openly accessible, with the retraction published under a Creative Commons Attribution License that permits unrestricted use and redistribution provided the original authors and copyright holders are credited. The editorial office has emphasized that all claims in the retraction notice are those of the authors and the journal, and that the publisher does not necessarily endorse any specific interpretation. For now, the scientific record reflects the correction: a once-promising link between a stem cell factor and a key cancer signaling pathway in osteosarcoma has been formally withdrawn, and the field moves forward with a clearer, if narrower, map of what the evidence actually supports.

Subject of Research: Retraction of a study on Sox2 knockdown and Wnt/β-catenin signaling in osteosarcoma cells due to improper image cropping.

Article Title: Retraction: Knockdown of Sox2 inhibits OS cells invasion and migration via modulating Wnt/β-catenin signaling pathway

Article References: Retraction: Knockdown of Sox2 inhibits OS cells invasion and migration via modulating Wnt/β-catenin signaling pathway. (n.d.). https://doi.org/10.3389/pore.2026.1612565

Image Credits: AI Generated

DOI: 10.3389/pore.2026.1612565

Keywords: Sox2, osteosarcoma, Wnt/beta-catenin signaling, retraction, research integrity, image manipulation, cancer stem cells, Pathology & Oncology Research, cell invasion, tumor metastasis, publication ethics, bone cancer