Why Cancer-Fighting Immune Cells Wear Out

Give an immune cell a cancer target and we naturally want it to keep attacking. Tomorrow, next month, whenever the tumor tries again. A cell that never quits sounds like excellent engineering. Unfortunately, the body also contains healthy organs that would prefer not to become collateral damage. They tend to be rather insistent about this.

Research involving St. Jude Children’s Research Hospital is exploring how gene regulation helps determine whether T cells keep functioning under persistent stimulation. ASXL1, DNMT3A, and TOX have emerged as important players. Their names resemble forgotten passwords, but their biology raises a practical question: how can cancer treatments preserve a useful immune response without removing the controls that make it survivable? [1] [2] [3]

Exhaustion is more than a tired cell

The phrase “T-cell exhaustion” invites a familiar picture: a hardworking cell needs a nap. The actual process involves changes in gene activity, cell function, and the packaging of DNA. During persistent stimulation, certain T cells become less effective at killing targets and producing immune signals. They also display more inhibitory receptors, which can restrain their activity. [4]

DNA is organized around proteins called histones. Together they form chromatin, and its arrangement affects which genetic instructions a cell can use. Changes to that arrangement are part of epigenetic regulation. The sequence of DNA letters need not change for the cell’s behavior to change substantially. Think of having the same instruction manual while access to particular pages has become restricted. [4]

This helps explain why exhaustion can become a durable biological state. Giving a cell more fuel will not necessarily reopen the relevant pages. Nor does a cell remaining alive establish that it is still doing the job we hired it for. A head count and a performance review answer different questions.

Keeping tomorrow’s response available

Even the category “exhausted T cell” contains important differences. In a 2016 mouse study of chronic viral infection, researchers identified a subset that could renew itself and produce more differentiated descendants. Most of the expansion after blocking the inhibitory PD-1 pathway came from this subset. A protein called TCF1 was essential to its development. [5]

These cells are often called stem-like because they retain renewal capacity. They remain T cells; nobody has discovered a tiny collection of replacement kidneys hiding in the immune system. The point is that sustained immunity may require a continuing source of responsive cells, alongside cells carrying out immediate attacks. [5]

Checkpoint drugs can help by blocking inhibitory interactions, including PD-1 binding to PD-L1. They have restored antitumor responses in patients, but can also permit immune attacks on healthy organs. Already, “make the response stronger” requires a footnote. The footnote happens to include the patient’s liver, lungs, and other useful equipment. [6]

ASXL1 and the cell’s future options

The October 2024 Science study examined three epigenetic regulators: ASXL1, DNMT3A, and TET2. In mice facing chronic viral stimulation, T cells lacking any one of these regulators maintained a renewable population for more than a year, without observed malignant transformation. These were experiments on cellular persistence. They did not demonstrate a year of cancer cures in people. [1]

ASXL1 works through machinery that modifies histones. Disrupting it changed chromatin regulation and preserved properties associated with self-renewal. In separate mouse tumor experiments, ASXL1-deficient T cells improved tumor control when combined with PD-L1 blockade. This suggests that changing a cell’s developmental trajectory could improve treatment responsiveness. It does not mean the researchers created cells immune to every form of dysfunction. [1]

Notice the change of setting: chronic infection in one experiment, tumors in another. Both can provide persistent stimulation, but a tumor has additional ways to obstruct immunity. Biology does not issue transferable warranties.

DNMT3A and engineered cancer treatments

CAR T-cell therapy equips T cells with a manufactured receptor that recognizes a chosen cancer target. The cells are expanded outside the body and infused into the patient. Such treatments have produced lasting remissions in some blood cancers, although many patients still do not achieve durable benefit. [7]

DNMT3A adds methyl groups to DNA, helping regulate gene activity. In a 2021 study, St. Jude investigators deleted DNMT3A from human CAR T cells. With repeated exposure to tumors in experimental systems, the edited cells retained greater ability to multiply and mount antitumor responses. The study connected exhaustion with epigenetic silencing of programs that preserve developmental flexibility. [2]

The improved function was also associated with increased interleukin-10, an immune signaling molecule. One gene edit, several downstream changes. The experiment reminds us that a cell is an interacting system, however convenient a single-switch diagram might look in a presentation. These findings supported further engineering work; they were not a clinical trial proving that DNMT3A-edited cells cured patients. [2]

Timing matters too. Preventing a restrictive program from becoming established and reversing an already established state are different tasks. A successful experiment on the first does not automatically settle the second. By the time we measure an exhausted cell, its earlier encounters may already have narrowed its options.

TOX complicates the easy answer

Several groups established TOX’s importance in 2019. This protein helps control gene activity and connects prolonged stimulation with an exhaustion-associated program. Its sustained expression helps cells acquire that distinct state. Removing it might therefore sound like the obvious next step. [4]

A study involving St. Jude tested a more specific intervention: disabling TOX’s DNA-binding region. In chronically infected mice, the altered T cells initially produced a stronger response. They also caused more immune-mediated tissue damage. Later, their numbers fell sharply, including the self-renewing population. The restraint that limited their immediate activity was also helping maintain the response over time. [3]

This is where the slogan “unstoppable immune cells” starts to lose its charm. Early enthusiasm is understandable; later cell counts are less impressed. A treatment must preserve useful, targeted function over time, and its effects on healthy tissue belong in the same assessment.

Where people and aging enter the story

Human observations helped motivate the research, but they need careful labeling. A related phase I/II trial treated 33 people, median age 73, with guadecitabine and the checkpoint drug atezolizumab. Participants had difficult-to-treat myelodysplastic syndromes or chronic myelomonocytic leukemia, both blood disorders. One third responded, including two complete remissions, and important toxicities occurred. This was a drug-combination trial, not treatment with ASXL1-edited T cells. [8]

Finding a mutation in a patient whose cells persist can suggest a mechanism worth investigating. It cannot isolate that mechanism from the patient’s disease, other mutations, or treatment history. The ASXL1 research connects observations in people with experiments designed to examine the underlying biology. Those forms of evidence strengthen one another while retaining their different limitations. [1]

Aging adds another complication. In a study of more than 17,000 people, acquired mutations in blood cells became more common with age; DNMT3A, TET2, and ASXL1 accounted for most detected variants. This expansion of particular blood-cell lineages, called clonal hematopoiesis, was associated with greater risks of blood cancer and death. These findings offer no justification for broadly disabling the genes as a rejuvenation strategy. [9]

Changing selected mature T cells for cancer treatment differs from a mutation spreading through blood-forming stem cells and their descendants. The same gene name does not make the biological situations equivalent. And a longer-lived T-cell population is not evidence of a longer-lived person, tempting though that shortcut may be on a longevity website.

What progress would actually look like

Exhaustion is only one obstacle. Solid tumors can lack a uniform target, exclude incoming cells, or suppress immune activity locally. CAR T-cell treatments can themselves cause serious inflammatory and neurological complications. Preserving a cell’s persistence cannot automatically fix its targeting or establish its safety. [7]

The next steps must show reproducible benefits, controlled growth, and acceptable risks in patients. Researchers need to learn which cancers benefit and whether better laboratory performance becomes durable clinical improvement.

The promise is substantial: immune cells that retain more useful options when cancer keeps demanding a response. Getting there requires understanding why those options close, which restrictions can be changed, and which safeguards still matter. The immune system has to live in the body it is trying to protect. That detail deserves to survive every exciting headline.

References

[1] Kang TG, Lan X, Mi T, et al. Epigenetic regulators of clonal hematopoiesis control CD8 T cell stemness during immunotherapy. Science. 2024;386(6718):eadl4492. doi:10.1126/science.adl4492.

[2] Prinzing B, Zebley CC, Petersen CT, et al. Deleting DNMT3A in CAR T cells prevents exhaustion and enhances antitumor activity. Sci Transl Med. 2021;13(620):eabh0272. doi:10.1126/scitranslmed.abh0272.

[3] Alfei F, Kanev K, Hofmann M, et al. TOX reinforces the phenotype and longevity of exhausted T cells in chronic viral infection. Nature. 2019;571(7764):265-269. doi:10.1038/s41586-019-1326-9.

[4] Khan O, Giles JR, McDonald S, et al. TOX transcriptionally and epigenetically programs CD8+ T cell exhaustion. Nature. 2019;571(7764):211-218. doi:10.1038/s41586-019-1325-x.

[5] Im SJ, Hashimoto M, Gerner MY, et al. Defining CD8+ T cells that provide the proliferative burst after PD-1 therapy. Nature. 2016;537(7620):417-421. doi:10.1038/nature19330.

[6] National Cancer Institute. Immune Checkpoint Inhibitors. Accessed September 24, 2026.

[7] National Cancer Institute. CAR T Cells: Engineering Patients’ Immune Cells to Treat Their Cancers. Accessed September 24, 2026.

[8] O'Connell CL, Baer MR, Ørskov AD, et al. Safety, Outcomes, and T-Cell Characteristics in Patients with Relapsed or Refractory MDS or CMML Treated with Atezolizumab in Combination with Guadecitabine. Clin Cancer Res. 2022;28(24):5306-5316. doi:10.1158/1078-0432.ccr-22-1810.

[9] Jaiswal S, Fontanillas P, Flannick J, et al. Age-related clonal hematopoiesis associated with adverse outcomes. N Engl J Med. 2014;371(26):2488-2498. doi:10.1056/nejmoa1408617.