Interferon Epsilon Loss Is Elusive 9p21 Link to Immune-Cold Tumors, Resistant to Immune Checkpoint Therapy, and Endogenous CXCL9/10 Induction

Abstract

Introduction

Copy number alterations of chromosome 9p, or parts thereof, impair immune response and confer immune-checkpoint therapy (ICT) resistance by direct elimination of immune-regulatory genes on this arm, notably interferon (IFN)-γ (at 9p24.1) and type I IFN (IFN-I) cluster (9p21.3) genes. Nevertheless, the primary 9p-loss human tumor immune readout is indirect (CXCL9/10 depletion at 4q21.1), and molecular alteration of chromosomes with engineered tandem elements–engineered 9p21.3-syntenic deletions in mice, Cdkn2a/b±Mtap (ΔS) versus larger Cdkn2a/b+Mtap+IFN-I (ΔL), revealed the causal link of IFN-I, primarily IFNϵ, to immune evasion.

Methods

This report updates and explicates the rapidly emerging body of clinical 9p ICT-cohort data and executes human (tumor, cell line) and mouse-model intrinsic 9p, IFN-I, and tumor-immune microenvironment CXCL9/10 deconvolution, mediation, and experimental studies. We analyzed CXCL9/10-CXCR3 cell sources and regulation by 9p deletion (size and depth) and mouse spatial single-cell RNA sequencing (scRNA-seq) syntenic chr4qC4 IFN-I (ΔS versus ΔL immune-evasive model) studies of immune-cell type, subtype, and subcluster Cxcl9/10⁺ numbers, fractions, and per-cell expression.

Results

Chr9p (9p, 9p21.3, 9p24.1) copy number loss is associated with immune-cold, programmed cell death protein 1 axis ICT-resistant human papillomavirus-negative head and neck squamous cancer, nonsquamous NSCLC, melanoma, urothelial cancer, and mesothelioma (13 reports, 36 ICT cohorts; <4 y). IFN-I has been associated with IFNα and ICT resistance. In human papillomavirus-negative head and neck squamous cancer, IFNE was the most highly expressed (and suppressed in 9p loss) IFN-I gene in tumors and cell lines, driven by 9p21.3 (q = 0.03; versus 9p24.1, q = 0.27); direct link to effector T-cell suppression (CD8 strongest, p = 0.006; mediation analysis), exhibited striking TP53 mutation co-occurrence and IFN-response pathway depletion. Progressively deep 9p21.3 loss (wild-type, shallow, deep) correlated with progressive IFNE and CXCL9/10-CXCR3 suppression; 9p21.3 ΔS (versus ΔL) IFN-I impact on CD8, NK (CD4, B, CD103) levels. Pan-tumor IFNE loss/tumor-immune microenvironment patterns were profoundly tissue-specific (Z ≤ 1.95 in 4/34 tumor types). IFN-intact ΔS (versus ΔL) KPC pancreatic model was linked to Cxcl9/10⁺ dendritic cell (DC), macrophage, and neutrophil number (confirmed in KPL-3M nonsquamous NSCLC), and macrophage per-cell expression. DC and macrophage subclustering revealed heterogeneity at the level of M1, Ccl5, and conventional type 1 DC (cDC1), particularly high in ΔS.

Conclusion

IFNϵ is the elusive, cell-intrinsic 9p21 IFN-I signal to human CD8 T-cell, myeloid DC, CXCL9/10, murine DC, and macrophage subtype and subcluster Cxcl9/10 expression. 9p-loss IFN-I and IFN-γ pathway (e.g., JAK2) genes at p21 and p24 lack the capacity of endogenous CXCL9/10 induction in an immune-desert, ICT-resistant state. These findings, 9p-loss/ICT-resistance data, and DC vaccine lung trials have led to a DC-CXCL9/10 vaccine, designed to bypass the severe chemokine deficit in 9p-loss tumors.