Macrophage-Derived CXCL9 and CXCL10 Are Required for Antitumor Immune Responses Following Immune Checkpoint Blockade

Imran G. House(The University of Melbourne), Peter Savas(The University of Melbourne), Junyun Lai(The University of Melbourne), Amanda X. Y. Chen(The University of Melbourne), Amanda J. Oliver(The University of Melbourne), Zhi L. Teo(The University of Melbourne), Kirsten L. Todd(The University of Melbourne), Melissa A. Henderson(The University of Melbourne), Lauren Giuffrida(The University of Melbourne), Emma V. Petley(The University of Melbourne), Kevin Sek(The University of Melbourne), Sherly Mardiana(The University of Melbourne), Tuba N. Gide(Melanoma Institute Australia), Camelia Quek(Melanoma Institute Australia), Richard A. Scolyer(Royal Prince Alfred Hospital), Georgina V. Long(Royal North Shore Hospital), James S. Wilmott(Melanoma Institute Australia), Sherene Loi(The University of Melbourne), Phillip K. Darcy(The University of Melbourne), Paul A. Beavis(The University of Melbourne)
Clinical Cancer Research
October 21, 2019
Cited by 749

Abstract

Abstract Purpose: Response rates to immune checkpoint blockade (ICB; anti-PD-1/anti-CTLA-4) correlate with the extent of tumor immune infiltrate, but the mechanisms underlying the recruitment of T cells following therapy are poorly characterized. A greater understanding of these processes may see the development of therapeutic interventions that enhance T-cell recruitment and, consequently, improved patient outcomes. We therefore investigated the chemokines essential for immune cell recruitment and subsequent therapeutic efficacy of these immunotherapies. Experimental Design: The chemokines upregulated by dual PD-1/CTLA-4 blockade were assessed using NanoString-based analysis with results confirmed at the protein level by flow cytometry and cytometric bead array. Blocking/neutralizing antibodies confirmed the requirement for key chemokines/cytokines and immune effector cells. Results were confirmed in patients treated with immune checkpoint inhibitors using single-cell RNA-sequencing (RNA-seq) and paired survival analyses. Results: The CXCR3 ligands, CXCL9 and CXCL10, were significantly upregulated following dual PD-1/CTLA-4 blockade and both CD8+ T-cell infiltration and therapeutic efficacy were CXCR3 dependent. In both murine models and patients undergoing immunotherapy, macrophages were the predominant source of CXCL9 and their depletion abrogated CD8+ T-cell infiltration and the therapeutic efficacy of dual ICB. Single-cell RNA-seq analysis of patient tumor-infiltrating lymphocytes (TIL) revealed that CXCL9/10/11 was predominantly expressed by macrophages following ICB and we identified a distinct macrophage signature that was associated with positive responses to ICB. Conclusions: These data underline the fundamental importance of macrophage-derived CXCR3 ligands for the therapeutic efficacy of ICB and highlight the potential of manipulating this axis to enhance patient responses.


Related Papers

No related papers found

Powered by citation graph analysis