Article Text

Download PDFPDF

1502 Tumor heterogeneity impedes immunotherapy through driving spatially organized pockets of immunosuppression
  1. Miho Tanaka1,
  2. Lotus Lum1,
  3. Kenneth Hu1,
  4. Cecilia Ledezma-Soto1,
  5. Bushra Samad1,
  6. Daphne Superville1,
  7. Kenneth Ng1,
  8. Zoe Adams1,
  9. Kelly Kersten1,
  10. Lawrence Fong1,
  11. Alexis Combes1,
  12. Max Krummel1 and
  13. Melissa Q Reeves2
  1. 1University of California, San Francisco, San Francisco, CA, USA
  2. 2University of Utah Huntsman Cancer Institute, Salt Lake City, UT, USA
  • Journal for ImmunoTherapy of Cancer (JITC) preprint. The copyright holder for this preprint are the authors/funders, who have granted JITC permission to display the preprint. All rights reserved. No reuse allowed without permission.


Background Intratumoral heterogeneity—defined as genetic and cellular diversity within a tumor—is linked to the failure of immunotherapy in multiple cancer types.1 2 The reasons for this are not well understood. Recent multi-region studies have found that as many as two-thirds of patient tumors contain both distinct ‘hot’ and ‘cold’ tumor regions, defined by high and low T cell infiltrates, respectively.3 We developed a novel system to reproducibly model tumor heterogeneity in mice, and employed this system to ask how ITH shapes the efficacy of immunotherapy.

Methods We introduced fluorescent tags into each ‘immune hot’ and ‘immune cold’ squamous cell skin carcinoma cell lines, and mixed them together to establish heterogeneous tumors in which tumor populations could be precisely spatially tracked. We analyzed immune cells in the vicinity of each hot and cold tumor cell populations by microdissection and flow cytometry as well as by spatial single cell RNAseq. We subsequently treated mice with a combination of anti-PD-1 blockade + CD40 agonist or control antibodies. We analyzed the response to therapy in each hot and cold regions of heterogeneous tumors, as well as overall tumor growth.

Results In untreated heterogeneous tumors, we find local tumor cells direct the formation distinct immune microenvironments. Cold tumor cells establish neighborhoods of immunosuppression, characterized by high macrophage infiltration, few inflammatory monocytes and neutrophils, and limited T cell abundance and function. We identified CX3CL1 as a candidate mediator of immunosuppression in cold tumor regions. When CX3CL1 was overexpressed in hot tumor cells, it drove an increase in suppressive CD206hi macrophages and reduction of monocytes and neutrophils. When heterogeneous tumors were treated with a combination of anti-PD-1 and CD40 agonist antibodies, we observed an influx of T cells post-treatment but only a modest slowing of tumor growth. Microdissection of tumors 6 days after treatment revealed that dysfunctional immune spatial organization persisted following therapy: T cells continued to exhibit limited abundance and function near cold tumor cells.

Conclusions We conclude that tumor cells create an architectural blueprint for the abundance and functional activity of tumor-infiltrating immune cells, which shapes both the pre-treatment immune microenvironment and responses to immunotherapy. Treatment with a combinatorial checkpoint inhibition regimen, despite leading to an influx in T cells, failed to eliminate spatial pockets of immunosuppression near cold tumor cells, and was ultimately ineffective at inducing tumor regression. Mechanistically, we identify CX3CL1 as a mediator of intratumoral accumulation of immunosuppressive macrophages.


  1. McGranahan N, Furness AJS, Rosenthal R, et al. Clonal neoantigens elicit T cell immunoreactivity and sensitivity to immune checkpoint blockade. Science. 2016;351:1463–1469.

  2. Liu D, Schilling B, Liu D, et al. Integrative molecular and clinical modeling of clinical outcomes to PD1 blockade in patients with metastatic melanoma. Nat. Med. 2019;25:1916–1927.

  3. Abduljabbar K, Ahmed Raza SE, Rosenthal R, et al. Geospatial immune variability illuminates differential evolution of lung adenocarcinoma. Nat. Med. 2020;26:1054–1062.

This is an open access article distributed in accordance with the Creative Commons Attribution Non Commercial (CC BY-NC 4.0) license, which permits others to distribute, remix, adapt, build upon this work non-commercially, and license their derivative works on different terms, provided the original work is properly cited, appropriate credit is given, any changes made indicated, and the use is non-commercial. See

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.