Scientists studied one million immune cells and found clues to future disease risk

By Pooja Toshniwal PahariaReviewed by Susha Cheriyedath, M.Sc.Oct 22 2026

A sweeping analysis of the body's most abundant white blood cells examines how their molecular differences relate to serious illness, metabolic health, and the risk of future disease.

Study: A single-cell atlas of human neutrophils reveals neutrophil heterogeneity across health and disease. Image Credit: Arif biswas / Shutterstock

A recent study published in the journal Nature Immunology presents a single-cell atlas of human neutrophils (SCAHN), offering new insights into how these immune cells behave in health and disease. The researchers linked neutrophil gene activity patterns to cancer outcomes, metabolic health, and the risk of death over the long term. These patterns also helped identify mortality risk among individuals with acute illness at hospital presentation. The findings suggest that neutrophil profiling may eventually contribute to clinical risk assessment, pending prospective validation.

Neutrophils are the most abundant white blood cells (WBCs) in human blood. Their subsets perform distinct functions in health and disease, including antimicrobial defense, tissue repair, homeostasis, and modulation of adaptive immunity. Earlier single-cell RNA sequencing (scRNA-seq) studies have examined neutrophil diversity, but they generally covered limited conditions and differed in their laboratory procedures, data analysis methods, and classification systems.

About the study

In the present study, researchers created SCAHN by bringing together data from multiple cohorts to explore differences in neutrophil subtypes and gene activity across tissues, species, and health states. To build the atlas, they collected human scRNA-seq datasets from public repositories.

SCAHN brought together data on approximately one million human neutrophils, using 43 scRNA-seq datasets comprising 781 samples. The analysis included samples from cord blood, peripheral blood, lung, bone marrow, colorectum, liver, spleen, and pancreas. The datasets represented healthy participants and patients with infections, autoimmune conditions, cardiovascular disease (CVD), diabetes, or several types of cancer.

The team mapped neutrophils from mice and nonhuman primates to the human reference and validated human–mouse correspondence by projecting NeuMap data onto SCAHN. The researchers examined neutrophil subset signatures and their association with clinical outcomes across 8,205 tumor biopsies from The Cancer Genome Atlas (TCGA).

The team performed gene set enrichment analysis (GSEA) and used whole-blood transcriptomic data from healthy participants in the Framingham Heart Study (FHS) to assess whether body mass index (BMI) explained sex differences. The same analysis investigated links between neutrophil gene activity patterns, metabolic indicators such as glucose, glycated hemoglobin (HbA1c), triglycerides, and high-density lipoprotein (HDL) cholesterol, and the likelihood of developing diabetes, cancer, or cardiovascular disease.

The association between neutrophil gene signature scores and short-term mortality was examined across multiple independent critical illness cohorts. Associations between signature scores and all-cause mortality were assessed at five-, seven-, and 10-year follow-up in the Framingham cohort.

The team analyzed electronic health records from 430,588 eligible patients across two healthcare databases to investigate whether routine neutrophil counts were linked to chronic disease and death from any cause.

Results

The atlas covered more than 20 tissues and diseases across nine countries and seven sequencing platforms. Analysis of 17,023 bulk blood transcriptomes across 40 datasets linked neutrophil-specific signatures to acute mortality, cancer outcomes, metabolic dysfunction, and long-term all-cause mortality.

Neutrophil subsets reflected different stages of maturation and immune activity. Immature azurocidin 1 (AZU1)- and lactotransferrin (LTF)-expressing subsets were more abundant in bone marrow, cord blood, and spleen. Gene signatures associated with immature neutrophils were more evident in males, while females generally showed stronger interferon-related and mature neutrophil profiles. The differences in immature and interferon-related signatures remained after accounting for BMI, age, and smoking status. Cross-species comparisons identified shared neutrophil states alongside differences in granule-associated and inflammatory gene expression between humans and mice.

Matrix metalloproteinase-9 (MMP9), AZU1, and LTF subsets were more abundant in severe coronavirus disease 2019 (COVID-19) and bacterial sepsis, alongside increased expression of immune-suppression and inflammation-related genes. Several immature and degranulating neutrophil signatures were linked to greater disease severity and mortality, whereas cluster of differentiation 74 (CD74)-related signatures were linked to better outcomes. An exploratory subgroup analysis of the SAVE-MORE trial found that anakinra treatment was associated with significantly lower mortality within 28 or 30 days among hospitalized COVID-19 patients with low baseline IL1B gene expression.

Pretreatment tumors showed enrichment of several inflammatory neutrophil subsets, some of which were associated with exhausted T cells, regulatory T-cell enrichment, or reduced natural killer cell proportions. In separate analyses of bulk tumor gene-expression data, selected inflammatory signatures were associated with poorer survival, while CD74-related signatures were linked to better survival. The researchers also found that transcriptional immaturity did not necessarily indicate immunosuppressive function, with suppression-related gene expression concentrated in particular subsets.

Selected neutrophil gene signatures were linked to less favorable metabolic indicators, diabetes, and a greater risk of death from any cause. In the Framingham analysis, several signatures were associated with a cancer diagnosis within seven years, but none showed a significant association with a new cardiovascular disease diagnosis over the same period. Electronic health record analyses also linked higher neutrophil counts, even within the normal range and in people without baseline comorbidities, to increased risks of several chronic diseases and, in the Stanford cohort, higher mortality. These findings suggest that neutrophil profiles and routine counts may help identify long-term disease risk if validated prospectively.

Conclusion

Overall, SCAHN offers new insight into neutrophil diversity across tissues and disease states, including infection and cancer. The findings link neutrophil gene activity and cell counts to metabolic health, chronic disease risk, and survival, suggesting that these measures may help identify patients at increased risk. The retrospective analyses cannot establish causality or prove clinical predictive value. The authors acknowledge variation in sequencing methods, incomplete disease coverage, and potential biases in electronic health records. Prospective studies and combined single-cell transcriptomic and proteomic analyses are needed before clinical application.