The pivotal role of immune functional assays in deciphering immune function alterations

Abstract

Growing evidence suggests that conventional immunomonitoring alone may not be sufficient to fully capture the complexity of immune dysfunctions. Immune functional assays (IFAs) have therefore emerged as valuable complementary tools, offering functional insights that extend beyond traditional phenotypic or quantitative approaches. Nevertheless, although in vitro stimulation represents a central component of IFAs, its specific contribution has never been rigorously evaluated, raising the critical question of whether this step is truly essential for detecting clinically relevant immune dysfunctions. To address this question, the present study compared gene expression levels (Nanostring) obtained from samples stimulated (TruCulture) or unstimulated (PaxGene) using the same analytical pipeline, in two distinct clinical settings: immune reconstitution following allogeneic hematopoietic stem cell transplantation (allo-HSCT) and sepsis progression. In allo-HSCT patients, post-stimulation data revealed immune heterogeneity and alterations related to ongoing immunosuppressive treatment or infectious event, not detected using unstimulated transcriptomic or cellular profiles alone. Similarly, post-stimulation transcriptomic profiles in patients with sepsis revealed immune clusters linked to disease severity and outcomes, surpassing traditional markers like mHLA-DR, while analyses from the unstimulated datasets failed to generate clinically relevant stratification. These findings emphasize the value of IFAs in uncovering immune function alterations that unstimulated assessments may miss, which could offer deeper insights into immune dysfunction. This study supports the use of IFAs as complementary tools to current clinical practices to enhance patient management by offering a functional view of immune system dynamics.

Keywords: immune dysfunction; immune functional assay; immune monitoring tool; method comparison; transcriptomic analysis.

Graphical Abstract

Figure 1.

Comparison of the capacity of gene expressions obtained from stimulated and unstimulated datasets to discriminate allo-HSCT patients according to ongoing Herpesviridae infectious episode or immunosuppressive treatment. Volcano plots of gene expression after SEB or LPS stimulation (A and D) or unstimulated (B and E) differing between allo-HSCT patients with or without ongoing Herpesviridae infectious episode (A and B) or ongoing immunosuppressive treatment (D and E). Fold change (x-axis) is plotted against statistical significance (y axis for each gene). Red circles represent differentially expressed genes. ROC curves obtained from the comparison between allo-HSCT recipients with and without ongoing infectious episodes (n = 19 and n = 40, respectively) (C) or with and without immunosuppressive treatment (n = 19 and n = 40, respectively) (F) based on the genes with the highest AUC values in each condition (stimulated or unstimulated), as well as quantitative cell counts with the highest AUC. CCL4 and STAT2 gene expression levels were obtained following LPS and SEB stimulation, respectively. AUC [95% CI] is indicated for each parameter. P < .05 is considered significant. Abbreviations: CI, confidence interval; HSCT, hematopoietic stem cell transplantation; LPS, lipopolysaccharide; PNN, polynuclear neutrophils; SEB, staphylococcal enterotoxin B .

Figure 2.

PAM clustering analysis of transcriptomic data from patients with sepsis in stimulated and unstimulated conditions. Hierarchical clustering was performed on 10 healthy volunteers (HVs) and 28 patients with sepsis. Using the PAM method with correlation distance, 3 clusters of individuals were revealed using (left) gene expression fold change in the post-SEB stimulation condition (Healthier; n = 11, Severe; n = 13, Intermediate; n = 14) or (right) normalized gene expression in the unstimulated condition (Cluster 1; n = 16, Cluster 2; n = 14, Cluster 3; n = 8). For both heatmaps, darker purple colors indicate upregulated genes while darker orange colors indicate downregulated genes. Between the two heatmaps, an alluvial plot illustrates the repartition (in percentage) of HVs (n = 10, gray), survivor patients with (n = 7, yellow) or without (n = 18, blue) hospital-acquired infection (HAI), and non-survivors (n = 3, orange) within the different clusters generated post-stimulation (left) or under unstimulated conditions (right). The percentages displayed in each box represent the proportion of the respective population included within each cluster. Abbreviations: HAI, hospital-acquired infection; HVs, healthy volunteers.