The Cocktail Report (sounds really smart around your friends):
A rare hybrid immune cell called a CD4 cytotoxic T lymphocyte, or CD4 CTL, normally makes up less than 5% of your T cells, but it climbs to a median of 17.6% in people past 110.
In each person, a single clone dominated, averaging 33.3% of all CD4 CTLs, and in one centenarian a single clone was 53.8%.
Those cells looked heavily battle-worn but showed no exhaustion markers, meaning they had been fighting for years without burning out.
The receptor sequences of the biggest clones matched T cells previously found expanding inside tumors, mostly non-small-cell lung cancer, yet none of these people had a lung cancer history.
The cohort was 28 people total, so this is a strong lead about how the longest-lived immune systems work, not a treatment.
Your immune system is one of the two organ systems most tightly tied to how long you live, which makes the immune profile of people who reach 110 unusually worth studying. A team from Keio University, RIKEN, and the University of Osaka just published the most detailed look yet, in Cell Reports.
They profiled 43,584 individual T cells from eight donors in their 70s to 90s, 10 centenarians, and 10 supercentenarians (people who reach 110 or beyond). For each cell, they read the genes being used, the proteins on the surface, and the T cell receptor, which is the molecular hand a T cell uses to grab and recognize its specific target.
The cell of interest is a strange one. CD4 T cells are supposed to be helpers that coordinate the immune response, while CD8 cells do the killing.
CD4 CTLs break that rule by carrying the CD4 helper badge while loading the killing machinery, including granzyme and perforin, the enzymes a killer cell injects to make a target cell self-destruct. In ordinary blood, they are under 5% of T cells.
Here is the age curve. The median share of CD4 CTLs was 4.0% in the 70s-to-90s group, 9.6% in centenarians, and 17.6% in supercentenarians.
The researchers then applied their model to more than five million cells from 1,512 public samples spanning ages 0 to 110-plus. The proportion stayed low across most of life, then rose sharply at extreme old age.
What makes this look like a targeted response rather than random drift is the clonal structure. A clone is a family of identical cells descended from one ancestor that got activated and copied itself, so a huge clone means one specific target was encountered over and over.
The largest clone averaged 33.3% of each person's CD4 CTLs. In one centenarian, the top clone was 126 of 234 cells, or 53.8%.
Critically, these cells were not exhausted. Exhausted T cells switch on shutdown markers such as PD-1, CTLA4, and LAG3, and these cells carried very little of that.
They had lost the co-stimulation proteins CD27 and then CD28, in that order, which is the signature of a fully matured fighter. What they did not show was any sign of a cell that has given up.
Then comes the part that matters most to you. The researchers took the receptor sequences of the dominant clones and searched public databases of T cells found in patient tissue.
Of the clonally expanded matches, 32 came from cancer patients: 17 non-small-cell lung cancer, nine breast cancer, six liver cancer. One sequence from a 70s-to-90s donor turned up in tumor tissue, adjacent tissue, and blood across five different lung cancer patients.
None of the centenarians or supercentenarians had any known history of lung, breast, or liver cancer. The authors' reading is that these clones may be responding to subclinical trouble, meaning abnormal cells too early to be detectable as disease, and clearing it before it becomes anything.
One more finding: when the team stimulated the cells, a single clone spread itself across multiple different cytokine profiles rather than staying locked into one job, while all of them kept producing the inflammatory signals IFN-gamma and TNF-alpha. That is versatility inside one family of cells.
A word of caution on how far this goes. The authors list their own limits plainly: no functional testing in a living body, blood cells only so the actual tissue targets are unidentified, receptor matches based on one chain only which they call exploratory overlap rather than proof of a shared target, and a cohort of 28 because 110-year-old blood donors are close to unobtainable.
Why Should You Care?
The people who dodge cancer into their second century may be running an active surveillance program, not simply enjoying good luck. If that program can be measured, it becomes a candidate for both an early-warning test and a therapy.
1. Hashimoto K, Kojima-Ishiyama M, Inokuchi H, et al. "CD4 CTLs in supercentenarians: Signs of adaptive expansion in healthy aging." Cell Reports, 117728, 2026. https://doi.org/10.1016/j.celrep.2026.117728
2. PubMed record for the study: https://pubmed.ncbi.nlm.nih.gov/42617599/
3. Hashimoto K, Kouno T, Ikawa T, et al. "Single-cell transcriptomics reveals expansion of cytotoxic CD4 T cells in supercentenarians." PNAS 116(48):24242-24251, 2019. https://doi.org/10.1073/pnas.1907883116
