Interferon-Gamma-Producing CD8+ Tissue Resident Memory T Cells Are a Targetable Hallmark of Immune Checkpoint Inhibitor-Colitis

Abstract

Background & aims: The pathogenesis of immune checkpoint inhibitor (ICI)-colitis remains incompletely understood. We sought to identify key cellular drivers of ICI-colitis and their similarities to idiopathic ulcerative colitis, and to determine potential novel therapeutic targets.

Methods: We used a cross-sectional approach to study patients with ICI-colitis, those receiving ICI without the development of colitis, idiopathic ulcerative colitis, and healthy controls. A subset of patients with ICI-colitis were studied longitudinally. We applied a range of methods, including multiparameter and spectral flow cytometry, spectral immunofluorescence microscopy, targeted gene panels, and bulk and single-cell RNA sequencing.

Results: We demonstrate CD8⁺ tissue resident memory T (T_RM) cells are the dominant activated T cell subset in ICI-colitis. The pattern of gastrointestinal immunopathology is distinct from ulcerative colitis at both the immune and epithelial-signaling levels. CD8⁺ T_RM cell activation correlates with clinical and endoscopic ICI-colitis severity. Single-cell RNA sequencing analysis confirms activated CD8⁺ T_RM cells express high levels of transcripts for checkpoint inhibitors and interferon-gamma in ICI-colitis. We demonstrate similar findings in both anti-CTLA-4/PD-1 combination therapy and in anti-PD-1 inhibitor-associated colitis. On the basis of our data, we successfully targeted this pathway in a patient with refractory ICI-colitis, using the JAK inhibitor tofacitinib.

Conclusions: Interferon gamma-producing CD8⁺ T_RM cells are a pathological hallmark of ICI-colitis and a novel target for therapy.

Keywords: Checkpoint Colitis; Immunotherapy Colitis; Tofacitinib; Ulcerative Colitis.

Graphical abstract

Figure 1

Study design.

Figure 2

CD8⁺ T_RM cells predominate in ICI-colitis and their activation correlates with endoscopic and histologic findings. Mononuclear cells from colonic biopsies from HVs (n = 8), active UC (n = 7), DCC (n = 12), and DCNC (n = 8). Flow cytometry demonstrates (Ai–iii) DCC is associated with a CD3⁺ T cell lymphocytosis, and CD8⁺ T cells predominance. (iv–vi) The majority of CD8⁺ T cells express tissue-residency marker CD103, with higher activation in colitis than CD8⁺103^– counterparts. (vii–viii) The proportion of CD8⁺CD69⁺CD103⁺ T_RM cells does not significantly differ across disease states; however, activation is highest in the DCC group. ∗P < .02; ∗∗P < .01; ∗∗∗P < .001; and ∗∗∗∗P < .0001 by Mann-Whitney test with Bonferroni correction. (ix) Co-expression of activation markers CD38 and HLA-DR are highest in patients with UC and DCC with CD8⁺CD69⁺CD103⁺ T_RM cells (red) displaying higher expression of these markers than CD8⁺CD103^– nonresident T cells (blue). Live CD45⁺CD8⁺ T cells are shown. (B) Proportion of activated CD8⁺ T_RM cells positively correlates with anti–CTLA-4/PD-1 colitis severity and measured by UCEIS (Spearman correlation). (C) Multiplexed spectral microscopy of a patient with DCC. Colocalization of CD3, CD8, and CD103 is demonstrated in both gastrointestinal crypts and in the lamina propria. CK, cytokeratin; DAPI, 4′,6-diamidino-2-phenylindole nuclear stain. Data representative of 3 experiments. (D) Live, singlet CD45⁺CD3⁺ T cells are displayed. (I) Cellular activation of T cells (top right quadrant) in healthy stomach and patients with anti–CTLA-4/PD-1 gastritis. In both health and anti–CTLA-4/PD-1 gastritis the majority of T cells are (ii) CD8⁺ with (iii) T_RM cell phenotype. (iv) Increased cellular activation is present in anti–CTLA-4/PD-1 gastritis compared with healthy stomach.

Figure 3

Targeted gene panel analysis of ICI-colitis includes unique and common features compared with UC and high expression of the IFNG signaling pathway. A 780-gene Nanostring analysis of colonic biopsy RNA from HVs (n = 8), patients with active UC (n = 5), DCC (n = 9), and DCNC (n = 8). (A) Heatmap of top 50 differentially expressed genes. (B) Number of genes up-regulated ≥2-fold compared with HVs demonstrates 173 of up-regulated genes in DCC, only 12 of 173 are also up-regulated in the DCNC group (limited on-treatment effect); 144 of 173 genes are common between DCC and UC, 28 of 173 genes are unique to DCC. (C) Manhattan plot indicating significantly up-regulated pathways, including response to IFNG. The complete list is provided in Supplementary Table 4. (D) RNA expression of canonical markers of IFNG signaling JAK1, JAK2, STAT1, and STAT2 are higher in DCC and UC groups compared with healthy controls and DCNC groups. ∗∗P < .01 and ∗∗∗P < .001 by 1-way analysis of variance.

Figure 4

Bulk RNASeq analysis confirms anti–CTLA-4/PD-1 colitis has a transcriptome distinct from UC with IFNG signaling stronger than TNFα signaling. Bulk RNASeq data generated from total RNA extracted from patients with DCC, those with active UC and HVs are shown. (A) Partial least squares-discriminate analysis (PLS-DA) demonstrate the divergent transcriptome of DCC and UC. (B) Module enrichment analysis demonstrated overexpression of hallmark gene “modules” 3 and 4 in DCC. (C) Over-representation analysis demonstrates the over-represented genesets in modules 1, 3, and 4. Over-expressed pathways include IFNG signaling (box),which was more highly expressed than TNFα signaling. (D) Co-expression and interaction of genes in modules 1, 3, and 4 as determined by Biological General Repository using Interaction Datasets (BioGRID). Blue indicates co-expressed genes; brown indicates gene interaction; and green indicates gene co-expression and interaction.

Figure 5

CD8⁺ T_RM cells in ICI-colitis express high proportions of checkpoint inhibitors, cellular activation/cytotoxicity markers and IFNG. scRNASeq analysis of 5876 cells from HVs (n = 3), active UC (n = 2), DCC (n = 3), DCNC (n = 3), PD-1 colitis (PDC; n = 2), and PD-1 treated with no colitis (PDNC; n = 3). (A) t-stochastic neighbor embedding (t-SNE) projection of live CD45⁺ lymphocytes formed 8 transcriptionally distinct clusters. (B) Proportion of clusters formed from cells from each disease state, with cluster 4 (box) most common in DCC. (C) Canonical gene markers of each clusters used to define annotation with cluster 4 (box) expressing CD3, CD8, CD69, and CD103 consistent with T_RM cells. (D) High expression of immune checkpoint molecules on (cluster 4) CD8⁺ T_RM cells (box). (E) t-SNE projection of T cells, highlighted by patient group. (F) Distribution of CD8⁺ T_RM cells as shown by cells co-expressing CD8, CD69, and ITGAE(CD103) in pink (low expression) and red (high expression). (G) Histogram showing cells that express a canonical gene-set list for T cells (dark blue) were selected from the total data for analysis in E, F, G, H, I, and K. (H) Expression of activation markers HLADR, GZMB, PRF1, and CD38 (to a lesser extent) overlap with the CD8⁺ T_RM cell zone. (I) Expression of IFNG overlaps with the CD8⁺ T_RM cell zone with IFNG being detected in UC, DCC, and PDC groups. (J) Heatmap based on all CD45⁺ cells showing the most differentially expressed genes in each patient group. (K) Heatmap based on T cells only showing differential expression between ITGAE(CD103)⁺ and ITGAE^– cells.

Figure 6

CD8⁺ T_RM cells express high levels of IFNG in PD-1–associated colitis. Data from a single cell protein and RNASeq analysis of 23,265 gut-derived T cells from HVs (n = 4), patients with active UC (n = 3), PD-1 colitis (PDC; n = 5) and PD-1 treated with no colitis (PDNC; n = 2). (A) t-stochastic neighbor embedding (t-SNE) projection of live T cells formed 7 distinct clusters. t-SNE plots of all groups showing expression of (B) CD4, (C) CD8A, and (D) ITGAE(CD103). (E) Distribution of CD8⁺ T_RM cells as shown by cells co-expressing CD8, CD69, and ITGAE(CD103) in pink (low expression) and red (high expression). (F) Expression of IFNG is shown in all groups, displaying overlap with CD8⁺ T_RM cell zones. (G) Distribution of cells based on patient groups demonstrates T cells from patients with PDC are found predominantly in the CD8⁺ T_RM cell zones (clusters 1 and 6). (H) Heatmap based on CD8⁺ T_RM cell populations 1–3 only displaying top differentially expressed genes.

Figure 7

Tofacitinib results in rapid resolution of treatment-refractory ICI-colitis, and correlates with resolution of CD8⁺ T_RM cell activation and down-regulation of JAK/STAT signaling. (A) Clinical time course of a 61-year-old man with non–small-cell lung cancer treated with carboplatin, pemetrexed, and pembrolizumab. The anti–PD-1 colitis was refractory to multiple therapies. Tofacitinib resulted in prompt resolution of clinical symptoms, and endoscopic and histopathology inflammation. Tofacitinib was continued for 6 weeks. Star, crypt abscess; thick arrow, attenuated crypt; thin arrow, crypt architectural distortion; triangle, erosion. (B) FMT response in a previous patient with ICI-colitis, where clinical resolution was associated with normalization of CD8⁺ T_RM cell activation. Flow cytometry gated on live CD3⁺CD8⁺CD69⁺CD103⁺ T_RM cells. (C) Using the same donor stool, FMT did not result in resolution of clinical symptoms or resolution of CD8⁺ T_RM cell activation in this 61-year-old man who subsequently received tofacitinib. (D) Flow cytometry plots gated on live single CD45⁺CD3⁺ T cells are shown. Before tofacitinib, widespread activation of CD4⁺ and CD8⁺ T cells is evident, with the highest level of activation in CD8⁺CD103⁺ T_RM cell subset (61%). After 6 weeks of tofacitinib, there is resolution of T cell activation, including in the CD8⁺CD103⁺ T_RM cell subset (7%). (E) Gene set enrichment analysis of bulk RNASeq data demonstrates IFNG signaling pathway enrichment in ICI-colitis. (F) Data from Nanostring RNAplex assay from the tofacitinib-treated patient and 3 HVs. Tofacitinib results in significant down-regulation of JAK1, JAK3, STAT1, STAT2, STAT3, STAT4, and STAT5A. ∗P < .05; ∗∗P < .01; ∗∗∗P < .001; ∗∗∗∗P < .0001; and ∗∗∗∗∗P < .00001 by Mann-Whitney test. (G) Volcano plot depicting pre and post tofacitinib.

Supplementary Figure 1

UCEIS and Nancy Index system for the scoring of UC.

Supplementary Figure 2

(A) Flow cytometric gating strategy. (B) CD8⁺CD69⁺CD103⁺ T_RM cells in anti-CTLA-4/PD-1 colitis (red) displayed high levels of checkpoint molecules LAG-3, Tim-3, TIGIT, and chemokine receptor CXCR6. Intracellularly they had high Ki-67 and granzyme-B and low levels of anti-apoptotic protein Bcl-2. Anti-CTLA-4/PD-1–associated colitis non-T_RM cells (orange), CTLA-4/PD-1 treatment with no colitis CD8⁺ T_RM cells (dark blue) and non-T_RM cells (light blue). Histograms based on live CD45⁺CD8⁺ T cells. Data are representative of 5 experiments.

Supplementary Figure 3

Data generated from a 780-gene Nanostring experiment analyzing RNA extracted from the gastrointestinal tract of HVs (n = 8), patients with active UC (n = 5), anti–CTLA-4/PD-1 colitis (DCC; n = 9) and anti–CTLA-4/PD-1 treated with no colitis (DCNC; n = 8) are shown. Plots show differentially expressed genes in (i) UC vs HV and (ii) DCC vs DCNC and DCC vs UC.