Data integrity in scientific research is crucial, yet sometimes, misconduct threatens to undermine trust in published findings. Recently, a prominent case emerged when Nature, one of the world’s most prestigious scientific journals, retracted a paper after discovering evidence of manipulated data. This case involves a doctoral student whose actions have not only called into question the validity of the research but also highlighted significant issues regarding research oversight and accountability.
The paper in question, published in April 2023, focused on the responsiveness of lung cancers to immunotherapy treatments. It gained considerable attention, being cited 192 times according to Clarivate’s Web of Science. However, recent investigations revealed that the data supporting several of the study's conclusions had been intentionally altered.
According to the official retraction notice, the first author, Kevin Ng, was responsible for the data manipulation, which included falsified information in multiple figures within the publication. At the time of conducting the research, Ng was enrolled as a Ph.D. student at the renowned Francis Crick Institute in London, working under the supervision of co-corresponding author George Kassiotis.
Currently, Ng is listed as a postdoctoral fellow at Rockefeller University in New York, according to his ORCID profile. Attempts to contact him directly have proved unsuccessful; he has not responded to messages via social media platforms like Instagram, nor could a current email address be found.
The allegations of misconduct surfaced after concerns were raised following the paper’s publication, prompting the Francis Crick Institute’s research integrity team to launch an investigation. The institute’s spokesperson, Kathryn Ingham, confirmed that the team was made aware of these concerns but declined to specify their exact origin. The authors, including Kassiotis, Charles Swanton, and Julian Downward—each affiliated with the Crick—agreed that the paper should be retracted. This consensus indicates acknowledgment of the serious issues uncovered.
The investigation identified specific misconduct by Ng, notably the manipulation of cell-binding data presented in one of the figures. Additionally, experiments associated with three other figures could not be verified at all. Significantly, the altered data supported the main conclusions of the paper, which suggests that the entire study’s findings could be compromised. As a result, the institute recommended an immediate retraction of the publication.
The Crick Institute informed the journal about the findings, leading Nature to attach an editor’s note on November 21st, warning readers about concerns related to data reliability. Interestingly, the investigation found no evidence of misconduct by the other authors, raising questions about oversight and the review process that failed to detect the manipulation before publication.
Julian Downward, one of the co-corresponding authors, has an extensive publication record with 27 papers featuring comments on PubPeer, many of which have been flagged for image concerns. Notably, Downward experienced a similar issue in 2015, when two of his papers were retracted due to data problems, including one from Nature. This pattern sparks debate about the challenges researchers face in maintaining data integrity amidst intense publication pressures.
The case raises critical questions: How did manipulated data make it through the peer review process, and what responsibilities do senior authors have to verify raw data? It also underscores the importance of institutional oversight and the potential vulnerabilities in the current scientific publishing system.
For those interested in supporting investigative work on scientific misconduct, resources such as Retraction Watch offer ways to contribute and stay informed about similar cases. As the scientific community continues to grapple with these issues, the main question remains: Are sufficient measures in place to prevent such misconduct, or is the system fundamentally flawed? We encourage discussion—what are your thoughts on how the research community should address data manipulation and uphold integrity in published science?
