Key takeaways Some immune cells are built to fight cancer, but a gene called Zfp148 can hold them back. When researchers turned off that gene in early studies, the immune cells were better at attacking cancer, and immunotherapy worked better too. The same pattern appeared in human tumor data, which may help explain why immunotherapy helps some people more than others. COLUMBUS, Ohio — Researchers at The Ohio State University Comprehensive Cancer Center – Arthur G. James Cancer Hospital and Richard J. Solove Research Institute (OSUCCC – James) report that a gene called Zfp148 makes a protein that acts like a “molecular brake”, limiting how strongly certain immune cells cancer attack cancer. In preclinical studies, switching off Zfp148 in CD8+ T cells helped them develop into a more powerful tumor-fighting state and improved tumor control—especially when combined with anti–PD-1 immunotherapy (a type of immune checkpoint inhibitor treatment designed to help T cells attack cancer). Why immunotherapy doesn’t always work Checkpoint inhibitor immunotherapies—such as drugs that target PD 1 or PD L1—are designed to release external brakes on the immune system. These treatments have transformed cancer care, but they don’t help everyone. This study highlights a key reason: even when checkpoint drugs remove those external brakes, immune cells may still be limited by internal genetic controls that prevent them from becoming powerful, long lasting cancer. Using laboratory models and advanced single cell techniques, the research team found that Zfp148 acts like a molecular off switch that prevents CD8⁺ T cells from fully switching into attack mode. When this brake was removed, the immune cells produced higher levels of cancer killing proteins and were better able to destroy tumor cells. In a preclinical colon cancer study, removing this internal brake also made anti–PD 1 immunotherapy more effective than the drug alone, leading to stronger tumor control and longer survival in laboratory models. To understand whether the findings might also apply to people, the researchers analyzed human tumor samples and patient data from immunotherapy studies. They found that tumors with lower levels of the human version of this genetic brake had more active cancer fighting immune cells and better responses to immunotherapy—supporting the relevance of the findings beyond the laboratory. “Zfp148 wasn’t on the short list of genes people talk about in T cell exhaustion or cancer immunity—and that’s what makes this so interesting,” said Tong Xiao, first author of the study, published in the April 2026 issue of Nature Immunology. Tong is a graduate student at the OSUCCC – James Pelotonia Institute for Immuno-Oncology (PIIO). “When we deleted it only in CD8+ T cells, they became better fighter cells, and they paired especially well with anti–PD-1 to improve tumor control.” Tong also noted that, in patient data, tumors with lower levels of the related human gene ZNF148 were associated with better responses to checkpoint therapy. “Now the big question is how to target this pathway in a way that safely strengthens the immune system’s ability to destroy cancer cells without pushing it too far, and whether this approach could also help in other diseases, such as chronic infections,” said Tong. CD8⁺ T cells come in distinct functional states—some attack cancer directly, while others modulate immunity. ‘We have identified the first specific genetic “off switch” controlling the development of cancer-killing T cells,’ said Zihai Li, MD, PhD, founding director of PIIO and senior author. Next steps in research The team is working to pinpoint when and where Zfp148 exerts its effects during T cell differentiation and to test whether targeting the Zfp148–Klf2 pathway can safely boost antitumor immunity in additional disease models. “These findings are early and come mainly from preclinical models, so they don’t mean a new treatment is ready yet,” said senior author Zihai Li, MD, PhD. “But they reveal a new way immune cells are held back inside tumors and suggest how future therapies might help the immune system fight cancer more effectively and for longer.” Funding and disclosures This research was supported by the OSUCCC and the National Institutes of Health (P01 CA278732 and R01 CA255334; Z.L.). This work was enabled by the PIIO and Pelotonia community. The authors declare no conflicts of interest related to this manuscript; Zihai Li serves on scientific advisory boards for HanchorBio. Additional co-authors in this study include Xingyu Chen; No-Joon Song; Ryan Joseph Brown; Anjun Ma; Jay K. Mandula; Amir Yousif; Yi Wang; Minh Quynh May Le; Jianying Li; Fengxia Gao; Bella Lee; Heng-Yi Chen; Fang-Yun Lay; Debasish Sundi; Maria Velegraki; Payton Weltge; Juanita L. Merchant; Mark P Rubinstein; Ken Oestreich; Chan-Wang Jerry Lio; Hazem E. Ghoneim; Xue Li; Dan Theodorescu; Gang Xin; Qin Ma and Weiguo Cui. ### MEDIA CONTACT: Amanda Harper amanda.harper2@osumc.edu | 614-685-5420