{
  "abstract": "Background The terms ‘hot’ or ‘cold,’ pioneered by Galon et al., classify tumors based on their high or low levels of immune infiltration. 1 These descriptions influence the available treatment options for different cancer types,2–4 yet they do not inform patient-specific treatments that could harness each patient’s unique tumor microenvironment. Moreover, the high heterogeneity in tumor-immune contexture within and across various cancer types5–7 highlights the need for a quantitative immune temperature, beyond a qualitative hot or cold, to fully inform therapy options. We hypothesized that we could use pan-cancer pretreatment immune compositions to generate patient-specific immune temperatures that predict treatment response and outcome. Our objective was to generate data-driven immune temperatures validated on separate clinical cohorts.Methods We analyzed untreated tissue samples of 10,469 patients across 31 tumor types. We used ESTIMATE 8 to calculate tumor purity and CIBERSORT9 to estimate anti- and pro-tumor immune phenotypes.10 For each patient, we mapped the proportions of tumor, anti- and pro-tumor cells on a triangular domain, which represents every possible combination of the relative proportions of the three populations. Next, we generated anti- and pro-tumor scores and combined them in an optimal way to generate a normalized immune temperature, 0-100°. We then validated the immune temperature in a separate cohort of 59 non-small cell lung cancer (NSCLC) patients who underwent conventional fractionated radiotherapy (RT) (n = 36, locoregional control (LRC); n = 23, locoregional failure (LRF)).Results The derived quantitative immune temperature broadly aligns with the qualitative hot/cold cancer types. Furthermore, the immune temperature correlates with LRC after RT in NSCLC. Patients who achieved LRC had significantly higher immune temperatures compared to those with LRF (mean 51.66° vs. 34.77°, p = 0.004), with a classification accuracy of 0.71. Additionally, patients with immune temperatures above the median (42.3°) had significantly improved overall survival compared to those with lower values (p = 0.031). We found a similar correlation with survival outcomes in eleven different cancer types downloaded from TIMEDB, 11 although specific treatment information for those cases was unavailable. The immune temperature also correlated with tumor control (p = 0.012, week 3), with a high prediction accuracy (AUC=0.805), in an independent cervical cancer cohort treated with chemoradiotherapy.Conclusions The novel concept of a quantitative immune temperature to predict patient-specific treatment response and outcome demonstrated clinically actionable accuracy and is poised to help guide clinical decision-making towards improved precision medicine that harnesses an individual’s immune contexture.References Galon J, et al. Type, density, and location of immune cells within human colorectal tumors predict clinical outcome. Science. 2006;313(5795):1960–1964.Galon J, Bruni D. Approaches to treat immune hot, altered and cold tumours with combination immunotherapies. Nature Reviews Drug Discovery. 2019;18(3):197–218.Maleki Vareki S. High and low mutational burden tumors versus immunologically hot and cold tumors and response to immune checkpoint inhibitors. J Immunother Cancer. 2018;6(1):157.Wu B, et al. Cold and hot tumors: from molecular mechanisms to targeted therapy. Signal Transduction and Targeted Therapy. 2024;9(1):274.Genomic map of the cellular composition of the tumor microenvironment. Nature Immunology. 2024;25(10):1789–1790.Flemming A. Tumour heterogeneity determines immune response. Nature Reviews Immunology. 2019;19(11):662–663.Jia Q, et al. Heterogeneity of the tumor immune microenvironment and its clinical relevance. Exp Hematol Oncol. 2022;11(1):24.Yoshihara K, et al. Inferring tumour purity and stromal and immune cell admixture from expression data. Nature Communications. 2013;4(1):2612.Newman AM, et al. Robust enumeration of cell subsets from tissue expression profiles. Nat Methods. 2015;12(5):453–7.Alfonso JCL, et al. Tumor-immune ecosystem dynamics define an individual radiation immune score to predict pan-cancer radiocurability. Neoplasia. 2021;23(11):1110–1122.Wang X, et al. TIMEDB: tumor immune micro-environment cell composition database with automatic analysis and interactive visualization. Nucleic Acids Res. 2023;51(D1):D1417-d1424.",
  "authors": [
    {
      "affiliations": [
        "The University of Texas MD Anderson Cancer Center, Houston, TX, USA"
      ],
      "name": "Awino Maureiq E Ojwang"
    },
    {
      "affiliations": [
        "The University of Texas MD Anderson Cancer Center, Houston, TX, USA"
      ],
      "name": "Mohammad U Zahid"
    },
    {
      "affiliations": [
        "The University of Texas MD Anderson Cancer Center, Houston, TX, USA"
      ],
      "name": "Aleksandra Aiurova"
    },
    {
      "affiliations": [
        "The University of Texas MD Anderson Cancer Center, Houston, TX, USA"
      ],
      "name": "Tiffany Nguyen"
    },
    {
      "affiliations": [
        "The University of Texas MD Anderson Cancer Center, Houston, TX, USA"
      ],
      "name": "Molly El Alam"
    },
    {
      "affiliations": [
        "The University of Texas MD Anderson Cancer Center, Houston, TX, USA"
      ],
      "name": "Tim Harris"
    },
    {
      "affiliations": [
        "Comprehensive Cancer Center, San Juan, PR, USA"
      ],
      "name": "Stephanie Dorta-Estremera"
    },
    {
      "affiliations": [
        "The University of Texas MD Anderson Cancer Center, Houston, TX, USA"
      ],
      "name": "Jagan Sastry"
    },
    {
      "affiliations": [
        "The University of Texas MD Anderson Cancer Center, Houston, TX, USA"
      ],
      "name": "Tatiana Cisneros Napravnik"
    },
    {
      "affiliations": [
        "The University of Texas MD Anderson Cancer Center, Houston, TX, USA"
      ],
      "name": "Lauren E Colbert"
    },
    {
      "affiliations": [
        "The University of Texas MD Anderson Cancer Center, Houston, TX, USA"
      ],
      "name": "Heiko Enderling"
    }
  ],
  "title": "110 Quantifying a pan-cancer immune temperature to predict patient-specific treatment response and outcome",
  "uid": "04048452-0e36-502d-ad4f-57899eda7d96"
}