When your immune system encounters a virus, a tumor, or a flare of inflammation, your CD8 T cells, a type of white blood cell that hunts and kills infected or abnormal cells, switch on. CD38 is a surface protein that appears when these cells are actively working. The percentage of your CD8 T cells carrying CD38 is one of the most studied measurements of immune activation in blood.
A standard complete blood count can tell you how many immune cells you have, but not how many are mobilized and fighting. This test fills that gap. A high reading can flag a chronic infection your body is quietly battling, signal whether an antiviral or anti-inflammatory treatment is working, or help explain why someone with apparently normal labs feels persistently unwell.
CD38 (cluster of differentiation 38) is a surface molecule that CD8 T cells switch on when they are activated. The lab uses flow cytometry, a technique that counts individual cells based on the markers on their surface, and reports the result as the percentage of CD8 T cells that carry CD38. A higher percentage means a larger share of your CD8 army is currently engaged.
Because the marker reflects systemic T cell activation, the same elevation can show up in very different settings: a chronic viral infection, a critical illness like sepsis, a cancer that the immune system is trying to fight, or an autoimmune disease where the immune system is overactivated against the body's own tissues. The reading is not disease-specific. It tells you the immune system is working hard. The cause has to come from context.
The strongest evidence for this marker comes from HIV. In untreated adults, a baseline CD8+CD38+ percentage above 25% predicts faster loss of CD4 helper cells and faster progression to AIDS. Each 10 percentage-point rise in CD8+CD38+ has been linked to roughly 37 to 88 percent higher AIDS risk, independently of CD4 count and other markers.
On effective antiretroviral therapy (ART, the standard combination of drugs that suppress HIV), CD8+CD38+ falls along with viral load. In primary HIV infection, very high pre-treatment activated CD8 counts (median around 461 per cubic millimeter) drop rapidly in parallel with viral suppression, and some people normalize to below 20 cells per cubic millimeter within seven months. Persistent elevation despite suppressed virus signals ongoing immune activation that is increasingly recognized as a driver of long-term complications.
In HIV-positive adults on suppressive therapy, persistently high CD8+CD38+ has also been linked to about three times the odds of HPV-related anal or cervical dysplasia compared to those with lower activation, independently of HPV status itself.
In chronic hepatitis B, CD8+CD38+ averages around 62 percent (with a typical spread of plus or minus about 15 points), far higher than in inactive carriers or healthy controls. With adefovir treatment that suppresses the virus, levels drop from roughly 54 percent down toward 20 percent as viral load falls.
In hepatitis C, activated CD8 T cells (CD38+HLA-DR+) decline from about 18 percent down to 12 percent in people coinfected with HIV, and from around 14 percent to 12 percent in HCV-monoinfected adults, between pre-treatment and one year after a sustained virologic response with direct-acting antivirals (DAAs, modern oral pills that cure HCV).
In severe COVID-19, patients who died within 28 days had nearly double the percentage of activated CD8+CD38+HLA-DR+ T cells compared to survivors (13.5 percent versus 7.6 percent). Severe and critical Omicron-era cases also showed persistently high CD38 expression on both CD8 T cells and natural killer cells (a type of innate immune cell).
In fatal H7N9 influenza, CD38+HLA-DR+CD8+ cells stayed high over time (mean around 32 percent) compared to survivors (around 19 percent). In sepsis (a life-threatening response to infection that lands people in the ICU), a pattern of rising CD8+CD38+ combined with falling CD8+CD28+ identified patients with the highest 28-day mortality, roughly 50 percent.
In several solid cancers, the ratio of CD8+CD38+ cells to total CD8 cells is elevated compared to healthy controls and rises with tumor stage, lymph node involvement, and distant spread, suggesting it tracks with disease progression. In hepatocellular carcinoma, peripheral CD8+CD38+ has emerged as a candidate prognostic marker for response to certain liver-directed treatments.
In systemic lupus erythematosus (SLE, an autoimmune disease where the body attacks its own tissues), expanded CD8+CD38+ T cells in peripheral blood correlate with disease activity, organ damage, and treatment resistance, and have been linked to a higher rate of infections. In rapidly progressive interstitial lung disease tied to anti-MDA5 dermatomyositis (a rare autoimmune muscle and lung disease), elevated CD38+CD8+ T cells appear to track with the most aggressive lung involvement.
There is no universally accepted clinical cutpoint for CD8+CD38+. The numbers below come from disease-specific research cohorts using different flow cytometry protocols, gating strategies, and lab equipment. Treat them as orientation, not as targets. Your lab will likely use its own reference range, and absolute numbers should be compared within the same lab over time rather than across studies.
| Setting | Reported CD8+CD38+ Pattern | What It Suggests |
|---|---|---|
| Healthy adults | Lower percentages, varies widely by lab and assay | Baseline immune activation |
| Untreated HIV | Above 25% baseline | Higher risk of CD4 decline and AIDS progression |
| Chronic hepatitis B | Around 62% | Active HBV-driven immune activation |
| Severe COVID-19 (non-survivors) | Around 13.5% | Worse 28-day outcome compared to survivors at 7.6% |
| HIV + HCV coinfected women | Upper tertile above 43% | Roughly 3x higher AIDS risk than below 26% |
Some of these percentages cannot be compared directly to each other because studies used different antibody combinations (CD38 alone versus CD38 with HLA-DR), different gating strategies, and different patient populations. The take-home is directional: higher than your own personal baseline in the context of new symptoms or a known condition is worth investigating.
A single CD8+CD38+ reading is far less useful than a trend. The marker rises with any acute immune challenge, including infections you may not even notice, and falls again as the trigger resolves. Within-person change over time is what separates a transient bump from a chronic process. Studies showing that activated CD8 cells fall steadily over months on antiviral therapy demonstrate this principle: the trajectory tells you whether treatment is working.
If you are managing a known chronic condition (HIV, chronic hepatitis, an autoimmune disease, post-cancer surveillance), a reasonable approach is to establish a baseline now, retest in 3 to 6 months, then at least annually. If you are starting a new treatment expected to modulate immune activation, retest at 3 months to confirm the direction of change. People without a specific clinical reason can still use a baseline reading as a personal reference point, but should not over-interpret a single result.
CD8+CD38+ is a non-specific marker. Many things can shift the number without any underlying disease worth pursuing. The most important confounders to know about:
An elevated CD8+CD38+ on its own is a starting point, not a diagnosis. The decision pathway depends on context. If you have a known chronic viral infection, the next step is checking the relevant viral load and treatment adherence. If you have a known autoimmune disease, pair this with disease activity markers and discuss with your rheumatologist. If the result is unexpectedly high and you have no known condition, the workup should include screening for chronic viral infections (HIV, hepatitis B and C, EBV, CMV), basic inflammatory markers like hs-CRP (high-sensitivity C-reactive protein, a measure of body-wide inflammation), and a CD4:CD8 ratio for a fuller picture of immune balance.
Persistently elevated activation despite a thorough workup is worth discussing with an immunologist or infectious disease specialist, especially if accompanied by unexplained symptoms. A normal result does not rule out localized immune problems, since the test measures circulating cells, not tissue-resident activation.
Evidence-backed interventions that affect your % CD8+ CD38+ level
% CD8+ CD38+ is best interpreted alongside these tests.