Heterogeneous Tumor-Immune Microenvironments among Differentially Growing Metastases in an Ovarian Cancer Patient - PubMed (original) (raw)

Case Reports

. 2017 Aug 24;170(5):927-938.e20.

doi: 10.1016/j.cell.2017.07.025.

Danish Memon 2, Stephane Pourpe 3, Harini Veeraraghavan 4, Yanyun Li 5, Hebert Alberto Vargas 6, Michael B Gill 1, Kay J Park 7, Oliver Zivanovic 8, Jason Konner 9, Jacob Ricca 5, Dmitriy Zamarin 10, Tyler Walther 3, Carol Aghajanian 9, Jedd D Wolchok 11, Evis Sala 6, Taha Merghoub 5, Alexandra Snyder 12, Martin L Miller 13

Affiliations

Case Reports

Heterogeneous Tumor-Immune Microenvironments among Differentially Growing Metastases in an Ovarian Cancer Patient

Alejandro Jiménez-Sánchez et al. Cell. 2017.

Abstract

We present an exceptional case of a patient with high-grade serous ovarian cancer, treated with multiple chemotherapy regimens, who exhibited regression of some metastatic lesions with concomitant progression of other lesions during a treatment-free period. Using immunogenomic approaches, we found that progressing metastases were characterized by immune cell exclusion, whereas regressing and stable metastases were infiltrated by CD8+ and CD4+ T cells and exhibited oligoclonal expansion of specific T cell subsets. We also detected CD8+ T cell reactivity against predicted neoepitopes after isolation of cells from a blood sample taken almost 3 years after the tumors were resected. These findings suggest that multiple distinct tumor immune microenvironments co-exist within a single individual and may explain in part the heterogeneous fates of metastatic lesions often observed in the clinic post-therapy. VIDEO ABSTRACT.

Copyright © 2017 The Authors. Published by Elsevier Inc. All rights reserved.

PubMed Disclaimer

Figures

None

Graphical abstract

Figure 1

Figure 1

Metastatic Tumors Exhibit Heterogeneous Growth and Somatic Mutation Patterns after Multi-line Chemotherapy (A) Representative CT scans showing concomitant progression/regression of the different resected metastatic tumors. RUQ = right upper quadrant. “Spleen” refers to the tumor deposit adjacent to the spleen. (B) CT-based volume of metastatic lesions represented with the solid vertical lines and dynamics of quantified CA125 levels with the red line indicating the CA125 upper limit of normal (35 units/ml). The x axis at the bottom shows a timeline of therapeutic interventions and clinical follow up. (C) Number of missense, silent, and nonsense mutations. (D) The phylogenetic tree represents the relationship of the samples based on binary calls of non-silent point mutations (Table S1A). Length of the branches is proportional to the number of mutations. Potential driver mutations are indicated. (E) Hierarchical cluster analysis (Euclidean distance metric and “average” linkage method) of the cellular prevalence of point mutations (n = 299) estimated with PyClone (Roth et al., 2014) (Table S1B).

Figure S1

Figure S1

Non-Silent Somatic Mutations and Copy-Number Alterations, Related to Figures 1 and S2 and Table S1 (A) Binary matrix of present/absent non-silent point mutations (n = 188) used for the phylogeny tree reconstruction in Figure 1D (Table S1A). (B) Relative copy-number alterations inferred from WES data of the primary and metastatic samples using CopywriteR (Kuilman et al., 2015). (C) Relative copy number profiles and tumor purity inferred after ABSOLUTE (Carter et al., 2012) analysis. Amplified and deep deleted DNA segments were defined as copy number alterations with at least ± 2 median absolute deviations for each sample. MAD = median absolute deviation.

Figure 2

Figure 2

Differential Expression of Immune-Related Pathways in Heterogeneously Growing Tumors (A) Expression levels and genetic alterations of genes associated with chemotherapy resistance in HGSOC (Patch et al., 2015) and multidrug resistance. Amplification and deep deletion were defined as at least ± 2 median absolute deviations of copy-number alterations for each sample (Figure S1C). (B) Single-sample gene set enrichment analysis (Barbie et al., 2009, Subramanian et al., 2005) of upregulated pathways using the KEGG (Kanehisa and Goto, 2000, Kanehisa et al., 2016) and REACTOME (Fabregat et al., 2016) databases (Tables S2D and S2E). Significantly enriched pathways (q < 0.05) with at least ± 1 log2 change relative to the median of the other samples are colored (Table S2G). False-discovery rate adjusted p value (q value) was calculated using the Benjamini-Hochberg method.

Figure S2

Figure S2

Gene Set Analysis of Transcript Abundance and Somatic Alteration Patterns across Samples, Related to Figure 2 and Table S2 (A–C) Gene-expression levels and genetic alterations of the DNA damage, apoptosis pathways, and caspases. (D) Expression levels of the 50 most variant genes according to their coefficient of variation (Table S2A). (E) Differential enrichment scores and enrichment q values of downregulated pathways between tumor samples (Tables S2D and S2E). No significantly enriched pathways (q < 0.05) with at least ± 1 log2 change relative to the median of the other samples were detected (Table S2G). False-discovery rate adjusted p value (q value) was calculated using the Benjamini-Hochberg method.

Figure 3

Figure 3

Immune Infiltration Status Shows Heterogeneous Microenvironments across Tumor Samples (A) Tumor purity and immune component estimated by analyzing Affymetrix-based transcriptomics (Table S3A) (Yoshihara et al., 2013). (B) Fractions of immune cell subsets in tumor samples inferred from gene-expression data using CIBERSORT (Newman et al., 2015). Width of bars is proportional to the −log10 p value of the deconvolution (Table S3B). CIBERSORT empirical p value, ∗p < 0.05. (C) Representative images of hematoxylin and eosin staining of tumor samples and immunofluorescence staining for DAPI, cytotoxic T cells (CD8+), helper T cells (CD4+FOXP3−), T cells (CD3+), T-regs (CD4+FOXP3+), macrophages (CD68+), and immune-checkpoint PD-L1. Complete slides are shown in Figure S3. (D) Image-based cell quantification of whole slides (Table S3C).

Figure S3

Figure S3

Complete Slide Hematoxylin and Eosin and Immunofluorescent Staining, Related to Figure 3 and Table S3 Hematoxylin and eosin staining of tumor samples. Immunofluorescence staining for cytotoxic T cells (CD8+), helper T cells (CD4+FOXP3−), and regulatory T cells (CD4+FOXP3+).

Figure 4

Figure 4

Higher HLA Expression and T Cell Oligoclonal Expansion Detected in Regressing Tumors (A) HLA-I and II gene differential expression across samples (Table S2A). (B) Number of predicted neoepitopes per sample (Tables S4B–S4D). (C) TCR sequencing of FFPE tumor samples and blood. The most prevalent TCR clonotypes (top 5 for each sample and blood) are shown (Table S5A). The blood sample was collected from the patient 550 days after secondary debulking (Figure S6A). Inset shows detection of the most frequent TCR rearrangement (CASSNDEYRGPTYEQYF) and its abundance comparison between samples (two-sided binomial tests with Benjamini-Hochberg multiple test correction, ∗∗∗ q < 0.001).

Figure S4

Figure S4

Neoepitope Distributions and HLA-I Neoepitope Depletion Analysis, Related to Figure 4 and Table S4 (A) Number of unique and overlapping expressed missense mutations, HLA-I and II neoepitopes between samples (Table S4D). (B) Correlations between expressed missense mutations and predicted HLA-I neoepitopes using NetMHC applied to TCGA ovarian samples (n = 150) and the primary and metastatic tumors (Tables S4E–S4G). KDE = kernel density estimate. (C) Top: Estimated neoepitope deviation from expected in the five tumor samples compared to TCGA ovarian cancer samples (n = 150). The expected number of neoepitopes was calculated by taking into account the expected number of missense mutations and the number of silent mutations according to Rooney et al., 2015 (see STAR Methods). Bottom: Neoepitope depletion analysis of 150 random unique permutations of the patient’s tumors (primary, spleen, RUQ, liver, and vaginal cuff) and their mutations. Each sample was compared against its own 150 unique permutations to control for the number of mutations (Tables S4H and S4I). Two-sided empirical p values were calculated from each distribution.

Figure S5

Figure S5

Predicted Immunogenicity of HLA Class I Neoepitopes, Related to Figures 1 and 4 and Tables S1 and S4 (A) Predicted immunogenic properties of trunk (clonal) and private HLA-I neoepitopes. Positive immunogenicity scores have biochemical properties associated with higher immunogenicity that outweigh properties associated with lower immunogenicity, and vice versa for negative scores (Calis et al., 2013). Horizontal lines within violin plots show the median and interquartile range of the data distribution. (B and C) Comparison between clonal and sub-clonal (including shared between two or more samples but not all) predicted immunogenicity of predicted binders and non-binders (two-sided Mann-Whitney rank test). Horizontal lines within violin plots show the median and interquartile range of the data distribution. (D–F) Probability of an HLA-I neoepitope having immunogenic properties considering its clonality and HLA-I binding affinity using the neoepitope data in (A), (B), and (C), respectively. Clonal neoepitopes have a lower probability of having immunogenic properties than sub-clonal predicted binders (chi-square test, p = 0.02). For non-binders (NetMHC score > HLA-I specific cutoff), clonal neoepitopes have a higher probability of having immunogenic properties (chi-square test, p = 0.003), as well as peptides with higher HLA-I affinities (chi-square test, p = 0.0001), although the absolute differences are minor. No significant interaction between clonality and predicted HLA-I binding affinity was detected for either binders or non-binders. GLM = generalized linear model.

Figure S6

Figure S6

PBMCs Sample Timeline and T Cell-Neoepitope Recognition Assay, Related to Figures 4 and 5 and Table S5 (A) Blood samples obtained from the patient 550 and 978 days after resection were used for TCR sequencing and T cell – neoepitope recognition assays respectively. (B) Experimental setup and flow cytometry gating strategy for the T cell –neoepitope recognition assays (intracellular cytokine staining assay) with surface staining of CD3, CD4, CD8, CD45, and intracellular staining of IL-4, IFN-γ, TNF-α. PBMC = peripheral blood mononuclear cells.

Figure 5

Figure 5

Predicted Neoepitopes with Higher Mutant than Wild-Type HLA-I Binding Affinity Elicit a T Cell Response (A) Representative scatterplots of TNF-α and IFN-γ intracellular cytokine staining of CD8+ T cells after 21 days of culture with CEF peptides or DMSO as positive and negative controls or the predicted mutant peptides (Figure S6B). CEF = Cytomegalovirus, Epstein-Barr virus, Influenza virus. (B) Percentage of CD8+ T cells with double-positive intracellular staining (TNF-α and IFN-γ) after incubation with each of the 43 predicted HLA-I neoepitopes, and HLA-I predicted binding affinity wild-type to mutant ratio (Table S5B). Mutation in gene FLG2 E1608K (P12) was found to be clonal after manual inspection in IGV (Table S1A).

Figure S7

Figure S7

Overall Associations between Tumor Fates and Tumor-Immune Microenvironmental Features, Related to Figures 1, 2, 3, and 4 and Tables S1, S2, S3, and S4 Cellular and molecular associations with change in tumor growth. Change in tumor growth (y axis) was calculated by dividing the tumor volume at CT scan 11 by the tumor volume at CT scan 10 (Figure 1B). Fitted curves are 2nd order polynomial regression lines plotted for trend visualization rather than prediction purposes. Capase 1 and 4 are considered inflammatory caspases involved in a type of apoptosis related to immune response called pyroptosis. The enrichment score x axis and the _q_-values come from the ssGSEA analysis. HLA-I genes include HLA-A, B, C, E, and F. HLA-II genes include HLA-DPA1, DMA, DRA, DQA1, DMB, DPB1, DQB2, DRB5, DRB1, DQB1, and DOA.

Comment in

Similar articles

Cited by

References

    1. Ascierto M.L., Makohon-Moore A., Lipson E.J., Taube J.M., McMiller T.L., Berger A.E., Fan J., Kaunitz G., Cottrell T., Kohutek Z. Transcriptional mechanisms of resistance to anti-PD-1 therapy. Clin. Cancer Res. 2017;23 clincanres.0270.2016. - PMC - PubMed
    1. Bahram S., Arnold D., Bresnahan M., Strominger J.L., Spies T. Two putative subunits of a peptide pump encoded in the human major histocompatibility complex class II region. Proc. Natl. Acad. Sci. USA. 1991;88:10094–10098. - PMC - PubMed
    1. Barbie D.A., Tamayo P., Boehm J.S., Kim S.Y., Moody S.E., Dunn I.F., Schinzel A.C., Sandy P., Meylan E., Scholl C. Systematic RNA interference reveals that oncogenic KRAS-driven cancers require TBK1. Nature. 2009;462:108–112. - PMC - PubMed
    1. Bashashati A., Ha G., Tone A., Ding J., Prentice L.M., Roth A., Rosner J., Shumansky K., Kalloger S., Senz J. Distinct evolutionary trajectories of primary high-grade serous ovarian cancers revealed through spatial mutational profiling. J. Pathol. 2013;231:21–34. - PMC - PubMed
    1. Bauer S., Groh V., Wu J., Steinle A., Phillips J.H., Lanier L.L., Spies T. Activation of NK cells and T cells by NKG2D, a receptor for stress-inducible MICA. Science. 1999;285:727–729. - PubMed

Publication types

MeSH terms

Substances

Grants and funding

LinkOut - more resources