A multi-institutional study led by researchers at UNC Lineberger Comprehensive Cancer Center, Salk Institute for Biological Studies, and UC San Diego has uncovered new genetic rules that determine how powerful immune cells - known as CD8 killer T cells - choose between becoming long-lasting, protective defenders or slipping into exhausted, dysfunctional states. The findings reveal new strategies for sustaining immune memory while preserving the ability to fight cancer and infections, with broad implications for immunotherapy and infectious disease research.
CD8 killer T cells play a central role in immune defense by seeking out and destroying virus-infected cells and cancer cells. However, during chronic infections or within tumors, these cells can gradually lose their killing ability and enter a state known as T cell exhaustion, where they become ineffective.
"Because protective and dysfunctional CD8 T cell states can look very similar, we designed this study to ask whether protective immune memory and dysfunction could be genetically separated. We flipped specific genetic switches in the T cells to see if we could restore their tumor-killing function without damaging their ability to provide long-term immune protection," said H. Kay Chung, PhD, lead and co-corresponding author and assistant professor in the Department of Cell Biology & Physiology at UNC School of Medicine and UNC Lineberger Comprehensive Cancer Center member. Chung began this research at the Salk Institute before joining UNC. "We found that it was indeed possible to separate these two outcomes."
A key advance in the study was the creation of a detailed atlas of CD8 T cell states, capturing how these immune cells change across a spectrum from highly protective to deeply dysfunctional.
"From the outset, our goal was to move beyond describing T cell states and instead build a predictive framework that tells us how to intentionally program them," Chung said.
"Our long-term goal is to make immune therapies work better by creating clear 'recipes' for designing T cells," said Susan M. Kaech, PhD, co-corresponding author and professor at the Salk Institute for Biological Studies. "To do that, we first needed to identify which molecular ingredients are uniquely active in one T cell state but not others. By building a comprehensive atlas of CD8 T cell states, we were able to pinpoint the key factors that define protective versus dysfunctional programs-information that is essential for precisely engineering effective immune responses."
"This is a challenging task," said Wei Wang, PhD, co-corresponding author and professor at UC San Diego. "Because genes work together in complex regulatory networks that are difficult to decipher, powerful computational tools are essential to pinpoint which regulators drive specific cell states."
Using advanced laboratory techniques, genetic tools, mouse models, and computational approaches, the researchers analyzed nine distinct CD8 T cell states. They identified specific transcription factors-proteins that control gene activity-that act like molecular switches, steering T cells toward either long-term function or exhaustion.
Among these, the team discovered two transcription factors, ZSCAN20 and JDP2, that had not previously been linked to T cell exhaustion. When these factors were turned off, exhausted T cells regained their ability to kill tumors without losing their capacity for long-term immune memory.
"Once we had this map, we could start giving T cells much clearer instructions-helping them keep the traits that allow them to fight cancer or infection over the long term, while avoiding the pathways that cause them to burn out," Kaech said.
"Our findings should be relevant for both solid tumors and blood-borne cancers," Chung added. "But this work may be especially important for solid tumors, where separating protective immune responses from exhaustion is critical for effective therapy."
More broadly, the study challenges the long-standing belief that immune exhaustion is an unavoidable consequence of prolonged immune activity.
T cell states are actively programmed by transcription factors, highlighting their predictability and potential controllability for immunotherapy.
"By mapping how killer T cells decide between staying strong or burning out, we've shown that immune exhaustion isn't inevitable," Kaech said. "By separating these two programs, we can begin to design immune cells that are both durable and effective in cancer and chronic infection."
"This study shows that we can begin to precisely manipulate immune cell fates and unlock new possibilities for enhancing immune therapies." Wang said.
Looking ahead, the team will combine advanced laboratory methods with AI-guided computational modeling to develop a larger number of precise genetic recipes for programming T cells, enabling greater precision for cellular therapies such as adoptive cell transfer (ACT) and CAR T cell therapy.
"It is important to emphasize that this work was truly a team effort," Chung said. "It began with the synergy between Dr. Kaech's immunology lab at Salk and Dr. Wang's computational platform at UC San Diego, and after my move to UNC, collaborations here allowed us to strengthen and extend the findings."
The study was published January 28, 2026, in Nature. In addition to Chung, co-corresponding authors include Susan M. Kaech, PhD, at the Salk Institute for Biological Studies, and Wei Wang, PhD, at UC San Diego.
By revealing how killer T cells choose between resilience and burnout, this research brings scientists closer to guiding the immune system with intention-rather than watching it fail under pressure.
The study was supported by supported by NIH grants R37AI066232, R01AI123864, R21AI151986, R01CA240909, R01AI150282, R01HG009626, K01EB034321, R01AI177864, R01CA248359, R01CA244361, AI151123, EB029122 and GM140929. Chung is a Damon Runyon Fellow supported by Damon Runyon Cancer Research Foundation grant DRG-2374-19.
Authors and disclosures: A comprehensive listing of the study authors and conflicts of interest can be found at the journal website.
