Germline modifiers of the tumor immune microenvironment implicate drivers of cancer risk and immunotherapy response - PubMed (original) (raw)

. 2023 May 12;14(1):2744.

doi: 10.1038/s41467-023-38271-5.

Timothy J Sears 2, Victoria H Wu 3, Eva Pérez-Guijarro 4, Hyo Kim 5, Andrea Castro 2, James V Talwar 2, Cristian Gonzalez-Colin 6, Steven Cao 7, Benjamin J Schmiedel 6, Shervin Goudarzi 8, Divya Kirani 9, Jessica Au 2, Tongwu Zhang 10, Teresa Landi 10, Rany M Salem 7, Gerald P Morris 11, Olivier Harismendy 2 12, Sandip Pravin Patel 13, Ludmil B Alexandrov 14 15, Jill P Mesirov 16 17, Maurizio Zanetti 16 18, Chi-Ping Day 4, Chun Chieh Fan 19 20, Wesley K Thompson 21, Glenn Merlino 4, J Silvio Gutkind 3, Pandurangan Vijayanand 6, Hannah Carter 22 23

Affiliations

Germline modifiers of the tumor immune microenvironment implicate drivers of cancer risk and immunotherapy response

Meghana Pagadala et al. Nat Commun. 2023.

Abstract

With the continued promise of immunotherapy for treating cancer, understanding how host genetics contributes to the tumor immune microenvironment (TIME) is essential to tailoring cancer screening and treatment strategies. Here, we study 1084 eQTLs affecting the TIME found through analysis of The Cancer Genome Atlas and literature curation. These TIME eQTLs are enriched in areas of active transcription, and associate with gene expression in specific immune cell subsets, such as macrophages and dendritic cells. Polygenic score models built with TIME eQTLs reproducibly stratify cancer risk, survival and immune checkpoint blockade (ICB) response across independent cohorts. To assess whether an eQTL-informed approach could reveal potential cancer immunotherapy targets, we inhibit CTSS, a gene implicated by cancer risk and ICB response-associated polygenic models; CTSS inhibition results in slowed tumor growth and extended survival in vivo. These results validate the potential of integrating germline variation and TIME characteristics for uncovering potential targets for immunotherapy.

© 2023. The Author(s).

PubMed Disclaimer

Conflict of interest statement

S.P.P. receives scientific advisory income from: Amgen, AstraZeneca, Bristol-Myers Squibb, Certis, Eli Lilly, Jazz, Genentech, Illumina, Merck, Pfizer, Rakuten, and Tempus. S.P.P.’s university receives research funding from: Amgen, AstraZeneca/MedImmune, Bristol-Myers Squibb, Eli Lilly, Fate Therapeutics, Gilead, Iovance, Merck, Pfizer, Roche/Genentech, and SQZ Biotechnologies. R.M.S. has a service contract with Travere Theraputics. L.B.A. is a compensated consultant and has equity interest in io9, LLC. His spouse is an employee of Biotheranostics, Inc. L.B.A. is also an inventor of a US Patent 10,776,718 for source identification by non-negative matrix factorization. L.B.A. declares U.S. provisional applications with serial numbers: 63/289,601; 63/269,033; 63/366,392; 63/367,846; 63/412,835. J.S.G. reports scientific advisory income from Domain Pharmaceuticals, Pangea Therapeutics, and io9, and is founder of Kadima Pharmaceuticals, all unrelated to the current study. O.H. is a current employee and stockholder of Zentalis pharmaceuticals Inc. M.Z. is a board member of Invectys Inc. All other authors declare that they have no competing interests.

Figures

Fig. 1

Fig. 1. Identifying heritable characteristics of the tumor immune microenvironment.

A Overview of the TIME germline analysis. Sample sizes are shown in gray. B Clustermap depicting 733 IP components and their pairwise correlation across 30 tumor types in the TCGA. C Horizontal barplot of variance in phenotype explained by variance in genotype (Vg/Vp) for 235 immune genes estimated separately genome-wide excluding the HLA locus (left panel, blue) and using only the HLA locus (right panel, orange). Source data are provided in the Source Data file.

Fig. 2

Fig. 2. Detecting putative germline modifiers of the tumor immune microenvironment.

A Locuszoom plot summarizing 890 associations between TIME eQTLs and 93 immune genes. Outer ring represents locations of all 157 tested IP components. Links are colored if implicated in cancer risk (orange), survival (green), or immunotherapy response (blue). B Significant associations between TIME eQTLs and 17 genes in the HLA region detected through conditional GWAS analysis for effects on gene expression using either a basic alignment to the reference genome (conditional), or allele-specific expression obtained by aligning to a patient-specific HLA reference allele set. C Ideogram plot of TIME eQTLs implicated by our discovery analysis (red) and literature curation (blue). Source data are provided in the Source Data file.

Fig. 3

Fig. 3. TIME eQTLs underlying antigen presentation stratify melanoma and prostate cancer risk.

A Violinplot of melanoma PRS trained on UK Biobank and validated in 1317 melanoma cases and 382 controls in High Density Melanoma cohort,. B Logistic regression odds ratio of melanoma risk ±SE among individuals in the top and bottom 10th quantile of PRS in High Density Melanoma cohort. C Top 15 TIME eQTL features most important in melanoma PRS. D Violinplot of prostate cancer PRS trained on UK Biobank and validated in 54,283 prostate cancer cases and 37,361 controls in ELLIPSE Consortium. E Logistic regression odds ratio of prostate cancer risk ±SE among individuals in the top and bottom 10th quantile of PRS in ELLIPSE consortium. F Top 15 TIME eQTL features most important in prostate cancer PRS. G Boxplot of M1 and M2 macrophage infiltration in primary TCGA SKCM (melanoma) in top (n = 10) and bottom (n = 10) 10th quantile of melanoma PRS. H Boxplot of CD8+ T cell and CD4+ T regulatory cell infiltration in TCGA SKCM (melanoma) in top and bottom 10th quantile of melanoma PRS. Boxplots show median (line), 25th and 75th percentiles (box), and 1.5× the interquartile range (IQR, whiskers). Two-sided Mann–Whitney U _p_-values were used for comparisons, adjusted if >2 comparisons were being made.

Fig. 4

Fig. 4. TIME eQTLs associated with survival implicate immune evasion.

A Cox Proportional Hazards ratios ±SE for cancer type-specific polygenic survival score (PSS) with overall survival separated by TCGA cancer type. Sample sizes used for PSS evaluation are indicated in parentheses. Red color indicates that the PSS hazard ratio was significant with an FDR < 0.05 after correcting for the number of tumor types modeled. B Cox Proportional Hazards odds ratios ±SE for cancer type-specific PSS with progression-free survival separated by TCGA cancer type. C Overall survival Kaplan–Meier curve based on LUAD PSS in TCGA LUAD (n = 121). D Overall survival Kaplan–Meier curve based on LUAD PSS in SHERLOCK (n = 166). E Cox Proportional Hazards ratio ±SE for LUAD PSS in TCGA LUAD and SHERLOCK. F Top 15 TIME eQTL features most important in the LUAD PSS. For C and D, High indicates top 25%, Med indicates middle 50% and Low indicates lowest 25% of PSS. Error bars represent standard error of Cox Proportional Hazards model. Significance is marked as: *p < 0.05, **p < 0.01, ***p < 0.001. Source data are provided in the Source Data file.

Fig. 5

Fig. 5. TIME eQTLs implicate targets for modulating immune responses.

A Boxplot of polygenic ICB score (PICS) constructed in melanoma ICB cohort,,, validated in Miao et al. cohort (n = 70). B Boxplot of PICS constructed in melanoma ICB cohort validated in Rizvi et al. cohort (n = 34). C ROC-AUC analysis of PICS in Miao et al. and Rizvi et al. validation cohorts. D Top 15 TIME eQTL features most important in PICS. E Grid plot of log odds ratio of variants with responder status in 6 ICB cohorts with beta coefficients of classic ICB biomarkers (TMB, PD-L1, PD-1, CTLA-4) association with responder status. Data are presented as mean values ±SE. Sample sizes for each cohort are indicated in parentheses. F Tumor growth curve for C57BL/6 mice implanted with MC38 treated with anti-PD-1, anti-CTSS, and combination of anti-PD-1 and anti-CTSS. Data are presented as mean values ±SE; n = 10 mice per group. G Survival curve for C57BL/6 mice implanted with MC38 treated with anti-PD-1, anti-CTSS, and combination of anti-PD-1 and anti-CTSS. H Barplot of the proportion of F4/80 Macrophages that are Arginase+ M2 macrophages and MHCII+ M1 macrophages respectively for MC38 tumors treated with anti-CTSS compared to control. Data are presented as mean values ±SE. Boxplots show median (line), 25th and 75th percentiles (box), and 1.5× the interquartile range (IQR, whiskers). Two-sided Mann–Whitney U _p_-values were used for PICS comparisons, adjusted if >2 comparisons were being made. Two-sample _t_-test was used for MC38 tumor growth comparisons, Bonferroni-adjusted for multiple tests. Logrank test was used for MC38 survival comparisons, Bonferroni-adjusted for multiple tests. Source data are provided in the Source Data file.

Fig. 6

Fig. 6. Characterization of TIME eQTLs using in genetic models.

A Cancer relevant associations by category with barplot showing the total number of genes implicated by polygenic risk scores (PRSs), polygenic survival score (PSS) and polygenic ICB score (PICS). B Mean enrichment ratio of genetic model immune microenvironment variants in 11 histone marks with corresponding enrichment ratios in 14 specific cell types. Number of cell types analyzed for each histone mark are given. C Barplot of cell-type specific TIME eQTLs implicated by DICE and ieQTL analysis. Source data are provided in the Source Data file.

Fig. 7

Fig. 7. Characterization of genes implicated by PICS model TIME eQTLs.

A map of TIME eQTL biological functions, immune functions and cancer associations for 15 genes implicated as modifiers of immune checkpoint blockade response. Innate immune function indicates that TIME eQTLs are also DICE eQTLs for macrophages, monocytes or dendritic cells. Adaptive immune function indicates that TIME eQTLs are also DICE eQTLs for CD8+ T cells, CD4+ T cells, or B cells. Risk indicates whether a gene was also implicated in PRS models. Survival indicates whether a gene was also implicated in PSS models. Asterisks indicate that a small molecule inhibitor has been reported for a gene. Source data are provided in the Source Data file.

References

    1. Stratton, M. R., Campbell, P. J. & Andrew Futreal, P. The cancer genome. Nature458 719–724 (2009). - PMC - PubMed
    1. Vogelstein B, et al. Cancer genome landscapes. Science. 2013;339:1546–1558. doi: 10.1126/science.1235122. - DOI - PMC - PubMed
    1. Hanahan D, Weinberg RA. The hallmarks of cancer. Cell. 2000;100:57–70. doi: 10.1016/S0092-8674(00)81683-9. - DOI - PubMed
    1. Hanahan D, Weinberg RA. Hallmarks of cancer: the next generation. Cell. 2011;144:646–674. doi: 10.1016/j.cell.2011.02.013. - DOI - PubMed
    1. Grivennikov SI, Greten FR, Karin M. Immunity, inflammation, and cancer. Cell. 2010;140:883–899. doi: 10.1016/j.cell.2010.01.025. - DOI - PMC - PubMed

Publication types

MeSH terms

Grants and funding

LinkOut - more resources