Immune Cell Markers: A Closer Look at Their Importance

Immune cell markers are molecules, mostly proteins, found on the surface of white blood cells that allow researchers, clinicians, and laboratory instruments to identify which cell type they are looking at, what that cell is doing, and even whether it is healthy or diseased. The best-known system for cataloging these markers assigns each one a “CD” number, and over 400 such molecules have been formally designated through international workshops spanning three decades. Far from being a niche laboratory concern, these markers now underpin cancer diagnostics, guide billion-dollar immunotherapies, and are reshaping how we understand the immune system at the single-cell level.

Where the CD Numbers Come From

If you have ever seen references to “CD4 cells” or “CD8 cells” and wondered what the letters stand for, CD means “cluster of differentiation.” The naming system was created by the Human Leukocyte Differentiation Antigens Workshops, a series of collaborative meetings that began in the early 1980s. Over ten workshops spanning roughly thirty years, researchers characterized and formally assigned numbers to more than 400 surface molecules found on human white blood cells.¹ Before this system existed, different labs used different names for the same protein, which made comparing results almost impossible. The CD system solved that by giving each confirmed molecule a single, universal number.

Not every immune marker carries a CD number. Transcription factors like FoxP3, which sits inside cells rather than on the surface, are identified differently. Cytokines, the signaling molecules cells release, are another category entirely. But for surface proteins, the CD system remains the common language that laboratories, hospitals, and pharmaceutical companies use worldwide.

CD4 and CD8, the Markers That Split T Cells Into Two Armies

The most familiar immune markers are CD4 and CD8, the proteins that divide T cells into two major branches. CD4 sits on “helper” T cells, which coordinate the immune response by directing other cells. CD8 marks “killer” T cells, which destroy virus-infected or cancerous cells directly. Both proteins are glycoproteins that interact with major histocompatibility complex molecules, the cellular ID tags that let the immune system distinguish self from non-self.²

These markers are not just passive labels. They actively participate in the signaling that occurs when a T cell recognizes a threat. In immature T cells developing in the thymus, both CD4 and CD8 are expressed simultaneously. At that stage, the molecules help amplify weak signals from the T cell receptor, which is critical for the selection process that determines which T cells survive and mature.³ Eventually, each cell commits to expressing only one, becoming either a CD4+ helper or a CD8+ killer. This is why clinicians track CD4 counts in HIV patients: the virus specifically infects CD4+ T cells, so a dropping CD4 count signals a deteriorating immune defense.

Markers for B Cells, NK Cells, and Myeloid Cells

T cells get most of the public attention, but every other immune lineage relies on surface markers for identification too. B cells, the antibody-producing arm of the immune system, are commonly identified by CD19 and CD20 on their surface. As B cells mature into plasma cells, which are essentially antibody factories, their marker profile shifts. Circulating plasma cells split into populations based on whether they express CD138: roughly 57% lack it and about 43% carry it, with the CD138-positive fraction showing a more mature profile, including higher levels of CD38 and CD27 and lower CD45.⁴

Natural killer (NK) cells are often identified by CD56, but this marker is less exclusive than many people assume. CD56 can also appear on certain T cells, dendritic cells, and monocytes.⁵ This is an important lesson about immune markers in general: very few are truly unique to one cell type. Most are shared across several, and identification depends on combinations rather than any single marker.

Myeloid cells, the branch that includes monocytes, macrophages, and dendritic cells, illustrate this combinatorial approach well. Monocytes are separated from other cells using a cocktail of lineage markers (CD3, CD14, CD19, CD56) along with CD33, and then further subdivided into classical, intermediate, and non-classical populations based on their relative expression of CD14 and CD16.⁶ Dendritic cells, which serve as sentinels that present threats to T cells, are identified by lacking those lineage markers while expressing HLA-DR. Getting these panels right matters because different monocyte subsets behave differently in infection and inflammation.

Markers That Change With Activation

A cell’s surface markers are not fixed. When an immune cell encounters a pathogen or receives a stimulation signal, it rapidly changes which markers it displays, almost like flipping switches in a specific order. CD69, sometimes called the earliest activation marker, appears on human T cells within hours of stimulation and can reach over 90% of cells by 24 hours. CD25, the receptor for a key growth factor, rises on a similar timeline.⁷ Other markers follow later: CD38 expression on mouse CD4+ T cells, for instance, does not peak until around 72 hours after stimulation.

Researchers have classified these into “early” activation markers like CD69 and CD71, and “late” markers like CD25 and HLA-DR, because their timing and response to inhibition differ.⁸ The practical value here is that a clinician or researcher can look at a blood sample and infer not just which cells are responding but roughly how far along the response is. A surge of CD69 without much CD25 suggests a very recent activation event. A sample where HLA-DR is elevated on T cells points to a response that has been building for days.⁹

Exhaustion Markers and Why They Matter for Cancer

If activation markers tell you the immune system has switched on, exhaustion markers tell you it has been running too long and is burning out. T cells fighting chronic infections or living inside tumors gradually lose their ability to kill and produce protective cytokines. The hallmark of this exhaustion is the accumulation of inhibitory receptors on the cell surface, most famously PD-1. But PD-1 alone does not actually mean a cell is exhausted. In follicular lymphoma, researchers found that PD-1-positive T cells that lacked a second marker called LAG-3 were simply activated and functionally healthy. Only the T cells expressing both PD-1 and LAG-3 showed the truly impaired, exhausted phenotype with reduced cytokine and granule production.¹⁰

A similar finding emerged with Tim-3, another inhibitory receptor. In tumor-infiltrating T cells, the population co-expressing both Tim-3 and PD-1 showed the most severe exhaustion, failing to proliferate and unable to produce the key cytokines needed to fight cancer.¹¹ This matters enormously for immunotherapy. Checkpoint inhibitor drugs like nivolumab and pembrolizumab work by blocking PD-1, essentially removing the brakes from exhausted T cells. Understanding which combination of markers actually defines true exhaustion helps researchers design better drugs and better predict who will respond to treatment.

The Tricky Business of Identifying Regulatory T Cells

Regulatory T cells, or Tregs, are the immune system’s peacekeepers. They suppress overactive immune responses and prevent autoimmune disease. Identifying them has been a persistent headache because their most definitive marker, the transcription factor FoxP3, sits inside the cell. You have to break the cell open to detect it, which means you cannot isolate living Tregs for experiments or clinical use using FoxP3 alone.

The workaround came from discovering that Tregs have low expression of CD127, the receptor for the signaling molecule IL-7. Combining CD4, CD25, and low CD127 produces a highly enriched population of FoxP3-positive cells, capturing even those Tregs that express little or no CD25.¹² Head-to-head comparisons of various Treg staining profiles confirmed that CD4+CD25+CD127low/− cells showed the highest FoxP3 levels and the strongest ability to suppress other T cells, outperforming alternatives like CD39-based or CD73-based identification.¹³ This combination has since been used to track Tregs in autoimmune diseases like lupus, where Treg numbers and function are often disrupted.

Stem Cells and the Meaning of CD34

CD34 is one of the most clinically consequential markers because it identifies hematopoietic stem cells, the progenitors in bone marrow that give rise to every type of blood cell. Transplant teams use CD34 counts to assess whether a stem cell collection is adequate for a bone marrow transplant. Within the CD34-positive pool, the most primitive stem cells can be further distinguished by low expression of CD90 and absence of CD38, HLA-DR, and mature lineage markers.¹⁴ These cells sit at the top of the hierarchy and have the greatest capacity to replenish the entire blood system. Lose the ability to identify them reliably, and transplant medicine would face a serious setback.

Markers in Cancer Diagnostics

One of the highest-stakes applications of immune markers is tracking minimal residual disease (MRD), the tiny number of cancer cells that survive treatment and can cause relapse. In acute lymphoblastic leukemia, flow cytometry panels using carefully chosen marker combinations can detect a single leukemic cell among 100,000 normal bone marrow cells.¹⁵ The leukemic cells display an abnormal pattern of surface markers that distinguishes them from healthy cells at the same stage of development.

In acute myeloid leukemia, MRD status detected by flow cytometry carries strong prognostic weight. Patients who test MRD-positive after treatment have substantially shorter overall survival: one multicenter study found a two-year survival rate of about 63% for MRD-positive patients compared with 92% for MRD-negative patients, with MRD status retaining its prognostic significance even after accounting for other risk factors.¹⁶ These numbers are not abstractions for patients. They directly influence whether doctors recommend additional chemotherapy, a stem cell transplant, or ongoing surveillance.

Markers as Drug Targets

Identifying a reliable surface marker on cancer cells opens the door to therapies that attack those cells with precision. CD38, a marker highly expressed on myeloma cells, is the target of antibodies like daratumumab and isatuximab. These drugs engage several killing mechanisms at once, including recruiting immune cells to destroy the cancer and triggering direct cell death, while maintaining a tolerable side-effect profile.¹⁷ All three approved CD38-targeting antibodies have shown meaningful activity even in heavily pretreated myeloma patients who have run out of other options.¹⁸

CAR-T cell therapy takes this one step further. Instead of an antibody drug, a patient’s own T cells are genetically engineered to recognize a specific surface marker on cancer cells. The remarkable early success of CAR-T in blood cancers owes a great deal to the choice of target: CD19, a protein uniformly and strongly expressed on malignant B cells.¹⁹ Extending CAR-T to solid tumors has proven far harder, partly because finding a marker that is abundant on the tumor but absent from normal tissue is much more difficult outside of blood cancers.²⁰ The safety requirement is strict: if a target marker also appears on heart or lung tissue, an engineered T cell will attack those organs too.

Measuring Dozens of Markers at Once

Traditional flow cytometry uses fluorescent dyes attached to antibodies, each lighting up a different marker. The problem is that fluorescent signals bleed into each other when too many are used simultaneously, limiting most conventional instruments to somewhere around 15–20 parameters. Designing panels that avoid this interference requires meticulous planning around antigen expression levels, fluorochrome brightness, and spillover spreading.²¹

Mass cytometry sidesteps the fluorescence problem entirely by replacing fluorescent labels with rare metal isotopes. Because metals do not have overlapping emission spectra the way fluorescent dyes do, the technique routinely measures around 60 parameters per cell simultaneously.²² This leap in capacity means researchers can characterize far more cell populations in a single experiment, uncovering rare subsets that would be invisible with fewer markers.

Going even further, a technology called CITE-seq pairs antibody-based marker detection with single-cell RNA sequencing. Each cell is simultaneously profiled for its surface proteins and its entire gene expression program.²³ This means researchers can link what a cell looks like on the outside to what it is doing on the inside, revealing genomic programs that explain how progenitor cells decide their fate.²⁴ The technology has been especially useful for T cell biology, clarifying how different surface markers relate to functional subsets that RNA alone could not distinguish.²⁵

Seeing Markers in Place With Spatial Imaging

A limitation shared by flow cytometry, mass cytometry, and CITE-seq is that they all require breaking tissue apart into individual cells. You learn what markers each cell carries, but you lose all information about where those cells were sitting in the tissue and which neighbors they were touching. Spatial imaging technologies are solving this. Methods like Imaging Mass Cytometry, Multiplexed Ion Beam Imaging, and Cyclic Immunofluorescence can measure dozens of protein markers on intact tissue slices while preserving the exact location of every cell.²⁶ This is transformative for cancer research, where the spatial arrangement of immune cells around and within a tumor can predict whether immunotherapy will work.²⁷ A tumor densely infiltrated by CD8+ T cells at its margins is biologically different from one where the same cells are excluded to the periphery, even though a dissociated sample would show identical total counts.

When Markers Lie

Relying on surface markers comes with pitfalls that even experienced researchers can overlook. One particularly tricky phenomenon is trogocytosis, in which a cell physically rips a patch of membrane from a neighbor and pastes it onto its own surface. The transferred proteins remain functional, which creates genuine confusion. T cells, for example, can acquire CD20, a protein normally specific to B cells, through trogocytosis.²⁸ A flow cytometry readout would then label those T cells as though they are B cells, or flag them as some bizarre hybrid. Some studies report remarkably high percentages of trogocytic cells, raising questions about how much this process could distort standard immunophenotyping. Experts have recommended pairing flow cytometry with microscopy to catch such artifacts.²⁹

Another interpretive challenge involves markers whose meaning depends entirely on context. CD69 is the classic example: on a T cell in blood, it signals recent activation, but on a T cell embedded in tissue like the lung, spleen, or brain, it marks a tissue-resident memory T cell, a long-lived sentinel that stays put in the organ rather than recirculating. In the human brain, CD8+ T cells express high levels of CD69 without showing other activation markers like Ki-67 or HLA-DR, confirming that the CD69 reflects a residency state rather than an active immune response.³⁰ Researchers identified a core transcriptional profile of these resident cells, defined by CD69 positivity, that is distinct from circulating memory cells in both lung and spleen.³¹ Reading CD69 as “activation” in a tissue biopsy would be a misinterpretation.

Soluble Markers in Blood

Some CD molecules can be shed or cleaved from the cell surface and detected as soluble proteins in blood. Soluble CD24, for instance, has emerged as a potential biomarker for rheumatoid arthritis. In one study, blood levels of soluble CD24 were roughly nine times higher in RA patients than in healthy individuals, and the marker was elevated in about two-thirds of early RA cases and a similar proportion of patients who tested negative on standard antibody tests.³² This is relevant because seronegative RA patients, who lack the classic antibodies used for diagnosis, are often harder to identify early. A soluble marker detectable with a simple blood draw could fill that gap.

Why Mouse Markers Do Not Always Translate to Humans

A persistent challenge in immunology is that much of what we know about immune markers comes from mouse experiments, and mice are not miniature humans. Global transcription profiles of corresponding immune cell types are largely conserved between species, but several hundred genes show clearly divergent expression patterns, and 169 genes meet even the most stringent criteria for genuine species differences.³³ In practical terms, this means a marker that reliably identifies a cell type in mice may be unreliable in humans, and vice versa. The rise of targeted biological drugs makes understanding these molecular differences increasingly important, because a therapy designed around a mouse-validated target could behave unexpectedly in human patients.³⁴

Isoform Switching and CD45

Not all marker changes involve gaining or losing a protein. Sometimes the same gene produces different versions of a marker through alternative splicing, where the cell edits its RNA to generate a slightly different protein. CD45 is the best-known example. Naive T cells express a larger form called CD45RA, while memory T cells switch to a shorter form called CD45RO. This switch is not cosmetic; it is driven by a specific RNA-processing protein, and mutations that disrupt this process alter the ratio of naive to memory T cells in mice.³⁵ Clinically, the CD45RA/RO distinction is used to assess whether a patient’s T cell compartment is dominated by inexperienced or battle-tested cells, information relevant to transplant medicine, vaccine responses, and aging research.