Immune signatures of checkpoint inhibitor-induced autoimmunity-A focus on neurotoxicity

Abstract

Background: Neurologic immune-related adverse events (irAE-n) are rare but severe toxicities of immune checkpoint inhibitor (ICI) treatment. To overcome diagnostic and therapeutic challenges, a better mechanistic understanding of irAE-n is paramount.

Methods: In this observational cohort study, we collected serum and peripheral blood samples from 34 consecutive cancer patients with irAE-n (during acute illness) and 49 cancer control patients without irAE-n (pre- and on-ICI treatment, n = 44 without high-grade irAEs, n = 5 with high-grade nonneurologic irAEs). Patients received either anti-programmed cell death protein (PD)-1 or anti-PD ligand-1 monotherapy or anti-PD-1/anti-cytotoxic T-lymphocyte-associated protein-4 combination therapy. Most common cancers were melanoma, lung cancer, and hepatocellular carcinoma. Peripheral blood immune profiling was performed using 48-marker single-cell mass cytometry and a multiplex cytokine assay.

Results: During acute illness, patients with irAE-n presented higher frequencies of cluster of differentiation (CD)8+ effector memory type (EM-)1 and central memory (CM) T cells compared to controls without irAEs. Multiorgan immunotoxicities (neurologic + nonneurologic) were associated with higher CD8+ EM1 T cell counts. While there were no B cell changes in the overall cohort, we detected a marked decrease of IgD- CD11c+ CD21low and IgD- CD24+ CD21high B cells in a subgroup of patients with autoantibody-positive irAE-n. We further identified signatures indicative of enhanced chemotaxis and inflammation in irAE-n patients and discovered C-X-C motif chemokine ligand (CXCL)10 as a promising marker to diagnose high-grade immunotoxicities such as irAE-n.

Conclusions: We demonstrate profound and partly subgroup-specific immune cell dysregulation in irAE-n patients, which may guide future biomarker development and targeted treatment approaches.

Keywords: immune checkpoint inhibitors; immune-related adverse events; immunotherapy; mass cytometry; neurotoxicity.

Graphical Abstract

Figure 1.

Unsupervised mass cytometry analysis of peripheral blood T cell signatures in patients with irAE-n and controls. (A) Overlay t-distributed stochastic neighbor embedding (t-SNE) maps of all patients with irAE-n and controls, each plot represents the expression of the indicated marker. One dot represents one cell. Heat colors show overall expression levels (red, high expression; dark blue, low or no expression). (B) 2D single-cell t-SNE maps with 50 metaclusters in patients with irAE-n and controls. Analysis for differential cluster abundance using significance analysis of microarrays (SAM) reveals differences in 3 metaclusters (cluster 2, cluster 18, cluster 21; printed in bold). (C) Heatmap of 37 different T cell markers illustrates individual expression levels for the 3 differential clusters 2, 18, and 21. Colour intensity shows overall expression levels (dark = high expression, bright = low or no expression). (D) Inspection of marker expression and plotted frequencies demonstrate a decrease of CD4⁺ CD127⁺ T cells (cluster 2) and CD4⁺ CXCR⁺ CD127⁺ EM1 T cells (cluster 18) as well as an increase of CD8⁺ effector memory type 1 (EM1) T cells (cluster 21) in patients with irAE-n compared to controls. Tukey’s box plots depict median (horizontal bar), mean (cross), interquartile range (hinges), and whiskers (fences). Outliers are represented as dots. * = P value ≤ .05, *** = P ≤ .001. CTLA-4, cytotoxic T-lymphocyte-associated protein 4; FoxP3, forkhead-Box-Protein P3; ICOS, inducible T-cell costimulatory; KLRG-1, killer cell lectin-like receptor subfamily G member 1; PD-1, programmed cell death protein 1; PD-L1, programmed death-ligand 1; PD-L2, programmed death-ligand 2; Tbet, T-box expressed in T cells; TIGIT, T cell immunoreceptor with Ig and ITIM domains; TIM-3, T-cell immunoglobulin and mucin-domain containing-3.

Figure 2.

Association of peripheral blood CD8+ effector memory type 1 (EM1) T cell frequency, number of irAEs, and outcome. (A) Spearman’s rank correlation reveals a positive correlation between the frequency of CD8⁺ EM1 T cells and the total number of irAEs (neurologic and nonneurologic). (B) Elevation of CD8⁺ EM1 T cells in patients with irAE-n is not associated with better tumor response. * = P value ≤ .05. CR, complete remission; EM1, effector memory type 1; irAE-n, neurologic immune-related adverse event; no., number; PD, progressive disease; PR, partial remission; SD, stable disease; V1, visit 1 (after ICI treatment initiation).

Figure 3.

Unsupervised mass cytometry analysis of peripheral blood non-T cell signatures in patients with irAE-n and controls. (A) 2D single-cell t-distributed stochastic neighbor embedding (t-SNE) maps with 30 metaclusters in patients with irAE-n and controls. Cluster analysis using significance analysis of microarrays (SAM) reveals between-group differences in 5 metaclusters (cluster 2, cluster 4, cluster 8, cluster 14, cluster 20; printed in bold). (B) Overlay t-SNE maps of all patients with irAE-n and controls, each plot represents the expression of the indicated marker. One dot represents one cell. Heat colors show overall expression levels (red = high expression; dark blue = low or no expression). (C) Heatmap shows the individual expression of 21 different markers of B cells, NK cells, monocytes, and DCs for the 5 differential metaclusters (cluster 2, cluster 4, cluster 8, cluster 14, cluster 20). Colour intensity shows overall expression levels (dark = high expression, bright = low or no expression). (D) Inspection of marker expression and plotted frequencies demonstrate (1) a decrease of CD11c⁺ conventional DCs (cDCs; cluster 14), (2) a decrease of CD161⁺ NK cells (cluster 2) and CD161⁻ NK cells (cluster 8), and (3) a decrease of CD16⁺ CD14⁻ monocytes (cluster 20) with a concurrent increase of CD14⁺ CD11b⁺ monocytes (cluster 4) in patients with irAE-n compared to controls. Tukey’s box plots depict median (horizontal bar), mean (cross), interquartile range (hinges), and whiskers (fences). Outliers are represented as dots. * = P value ≤ .05, *** = P ≤ .001. KLRG1, killer cell lectin-like receptor subfamily G member 1; PD-L1/2, programmed death-ligand 1/2.

Figure 4.

Serum cytokine signatures in patients with irAE-n and controls. Comparison of serum levels of 45 different cytokines in patients with irAE-n and controls using Mann–Whitney U test reveals significant differences for 10 cytokines. Scatter dot plots depict individual log2 transformed cytokine levels (dots), log2 transformed median cytokine concentration in pg/mL (middle bar), and interquartile range (upper and lower error bar). * = P ≤ .05, ** = P ≤ .01, *** = P ≤ .001; irAE-n, neurologic immune-related adverse events; IL-1RA, interleukin-1 receptor antagonist; ns, not significant; PIGF-1, placental growth factor 1; V0, baseline (before ICI treatment initiation); V1, after ICI therapy initiation.

Figure 5.

Potential peripheral blood markers of irAE-n. (A, B) In a logistic regression model, a probability of more than .5 for the diagnosis of irAE-n was predicted at (A) CXCL10 levels >49.5 pg/mL and (B) CD8⁺ EM1 T cell frequencies of >7.2%. Blue area filling indicates 95% confidence intervals. (C) ROC analysis of CD8⁺ EM1 T cell frequency, CXCL10 levels, and a combination of both parameters revealed an area under the curve (AUC) of 67.1%, 74.5%, and 86.8%, respectively, when all patients with irAE-n were included. (D) When only patients with irAE-n of the PNS were included, ROC analysis revealed an AUC of 65.9%, 71.8%, and 85.9% for CD8⁺ EM1 T cell frequency, CXCL10 concentration, and the combination of both parameters, respectively. (E) CD8⁺ EM1 T cell frequency and serum concentration of CXCL10 are comparable between patients with PNS-irAE-n and patients with CNS-irAE-n. (F) Correlation matrix of peripheral blood immune markers shows a low correlation between CD8⁺ EM1 T cell frequency and CXCL10 levels. cDCs, conventional dendritic cells; CNS, central nervous system; EM1, effector memory type 1; IL-1RA, interleukin 1 receptor antagonist; ncMono, nonconventional monocytes; PIGF-1, placental growth factor 1; PNS, peripheral nervous system.