Deactivation of ligand-receptor interactions enhancing lymphocyte infiltration drives melanoma resistance to Immune Checkpoint Blockade.

Immune checkpoint blockade (ICB) is a promising cancer therapy; however, resistance often develops. To learn more about ICB resistance mechanisms, we developed IRIS (Immunotherapy Resistance cell-cell Interaction Scanner), a machine learning model aimed at identifying candidate ligand-receptor interactions (LRI) that are likely to mediate ICB resistance in the tumor microenvironment (TME). We developed and applied IRIS to identify resistance-mediating cell-type-specific ligand-receptor interactions by analyzing deconvolved transcriptomics data of the five largest melanoma ICB therapy cohorts. This analysis identifies a set of specific ligand-receptor pairs that are deactivated as tumors develop resistance, which we refer to as resistance deactivated interactions (RDI). Quite strikingly, the activity of these RDIs in pre-treatment samples offers a markedly stronger predictive signal for ICB therapy response compared to those that are activated as tumors develop resistance. Their predictive accuracy surpasses the state-of-the-art published transcriptomics biomarker signatures across an array of melanoma ICB datasets. Many of these RDIs are involved in chemokine signaling. Indeed, we further validate on an independent large melanoma patient cohort that their activity is associated with CD8+ T cell infiltration and enriched in hot/brisk tumors. Taken together, this study presents a new strongly predictive ICB response biomarker signature, showing that following ICB treatment resistant tumors turn inhibit lymphocyte infiltration by deactivating specific key ligand-receptor interactions.

Large Scale Comparative Deconvolution Analysis of the Canine and Human Osteosarcoma Tumor Microenvironment Uncovers Conserved Clinically Relevant Subtypes.

Osteosarcoma is a relatively rare but aggressive cancer of the bones with a shortage of effective biomarkers. Although less common in humans, Osteosarcomas are fairly common in adult pet dogs and have been shown to share many similarities with their human analogs. In this work, we analyze bulk transcriptomic data of 213 primary and 100 metastatic Osteosarcoma samples from 210 pet dogs enrolled in nation-wide clinical trials to uncover three Tumor Microenvironment (TME)-based subtypes: Immune Enriched (IE), Immune Enriched Dense Extra-Cellular Matrix-like (IE-ECM) and Immune Desert (ID) with distinct cell type compositions, oncogenic pathway activity and chromosomal instability. Furthermore, leveraging bulk transcriptomic data of canine primary tumors and their matched metastases from different sites, we characterize how the Osteosarcoma TME evolves from primary to metastatic disease in a standard of care clinical setting and assess its overall impact on clinical outcomes of canines. Most importantly, we find that TME-based subtypes of canine Osteosarcomas are conserved in humans and predictive of progression free survival outcomes of human patients, independently of known prognostic biomarkers such as presence of metastatic disease at diagnosis and percent necrosis following chemotherapy. In summary, these results demonstrate the power of using canines to model the human Osteosarcoma TME and discover novel biomarkers for clinical translation.

Deep Domain Adversarial Learning for Species-Agnostic Classification of Histologic Subtypes of Osteosarcoma.

Osteosarcomas (OSs) are aggressive bone tumors with many divergent histologic patterns. During pathology review, OSs are subtyped based on the predominant histologic pattern; however, tumors often demonstrate multiple patterns. This high tumor heterogeneity coupled with scarcity of samples compared with other tumor types render histology-based prognosis of OSs challenging. To combat lower case numbers in humans, dogs with spontaneous OSs have been suggested as a model species. Herein, a convolutional neural network was adversarially trained to classify distinct histologic patterns of OS in humans using mostly canine OS data during training. Adversarial training improved domain adaption of a histologic subtype classifier from canines to humans, achieving an average multiclass F1 score of 0.77 (95% CI, 0.74-0.79) and 0.80 (95% CI, 0.78-0.81) when compared with the ground truth in canines and humans, respectively. Finally, this trained model, when used to characterize the histologic landscape of 306 canine OSs, uncovered distinct clusters with markedly different clinical responses to standard-of-care therapy.

Deconvolving Clinically Relevant Cellular Immune Cross-talk from Bulk Gene Expression Using CODEFACS and LIRICS Stratifies Patients with Melanoma to Anti-PD-1 Therapy.

The tumor microenvironment (TME) is a complex mixture of cell types whose interactions affect tumor growth and clinical outcome. To discover such interactions, we developed CODEFACS (COnfident DEconvolution For All Cell Subsets), a tool deconvolving cell type-specific gene expression in each sample from bulk expression, and LIRICS (Ligand-Receptor Interactions between Cell Subsets), a statistical framework prioritizing clinically relevant ligand-receptor interactions between cell types from the deconvolved data. We first demonstrate the superiority of CODEFACS versus the state-of-the-art deconvolution method CIBERSORTx. Second, analyzing The Cancer Genome Atlas, we uncover cell type-specific ligand-receptor interactions uniquely associated with mismatch-repair deficiency across different cancer types, providing additional insights into their enhanced sensitivity to anti-programmed cell death protein 1 (PD-1) therapy compared with other tumors with high neoantigen burden. Finally, we identify a subset of cell type-specific ligand-receptor interactions in the melanoma TME that stratify survival of patients receiving anti-PD-1 therapy better than some recently published bulk transcriptomics-based methods. SIGNIFICANCE: This work presents two new computational methods that can deconvolve a large collection of bulk tumor gene expression profiles into their respective cell type-specific gene expression profiles and identify cell type-specific ligand-receptor interactions predictive of response to immune-checkpoint blockade therapy. This article is highlighted in the In This Issue feature, p. 873.

Hard wiring of normal tissue-specific chromosome-wide gene expression levels is an additional factor driving cancer type-specific aneuploidies.

BACKGROUND: Many carcinomas have recurrent chromosomal aneuploidies specific to the tissue of tumor origin. The reason for this specificity is not completely understood. METHODS: In this study, we looked at the frequency of chromosomal arm gains and losses in different cancer types from the The Cancer Genome Atlas (TCGA) and compared them to the mean gene expression of each chromosome arm in corresponding normal tissues of origin from the Genotype-Tissue Expression (GTEx) database, in addition to the distribution of tissue-specific oncogenes and tumor suppressors on different chromosome arms. RESULTS: This analysis revealed a complex picture of factors driving tumor karyotype evolution in which some recurrent chromosomal copy number reflect the chromosome arm-wide gene expression levels of the their normal tissue of tumor origin. CONCLUSIONS: We conclude that the cancer type-specific distribution of chromosomal arm gains and losses is potentially "hardwiring" gene expression levels characteristic of the normal tissue of tumor origin, in addition to broadly modulating the expression of tissue-specific tumor driver genes.

PD-L1 recruits phospholipase C and enhances tumorigenicity of lung tumors harboring mutant forms of EGFR.

Cancer immunotherapy focuses on inhibitors of checkpoint proteins, such as programmed death ligand 1 (PD-L1). Unlike RAS-mutated lung cancers, EGFR mutant tumors have a generally low response to immunotherapy. Because treatment outcomes vary by EGFR allele, intrinsic and microenvironmental factors may be involved. Among all non-immunological signaling pathways surveyed in patients' datasets, EGFR signaling is best associated with high PD-L1. Correspondingly, active EGFRs stabilize PD-L1 transcripts and depletion of PD-L1 severely inhibits EGFR-driven tumorigenicity and metastasis in mice. The underlying mechanisms involve the recruitment of phospholipase C-γ1 (PLC-γ1) to a cytoplasmic motif of PD-L1, which enhances PLC-γ1 activation by EGFR. Once stimulated, PLC-γ1 activates calcium flux, Rho GTPases, and protein kinase C, collectively promoting an aggressive phenotype. Anti-PD-L1 antibodies can inhibit these intrinsic functions of PD-L1. Our results portray PD-L1 as a molecular amplifier of EGFR signaling and improve the understanding of the resistance of EGFR+ tumors to immunotherapy.

Immunoproteasome expression is associated with better prognosis and response to checkpoint therapies in melanoma.

Predicting the outcome of immunotherapy treatment in melanoma patients is challenging. Alterations in genes involved in antigen presentation and the interferon gamma (IFNγ) pathway play an important role in the immune response to tumors. We describe here that the overexpression of PSMB8 and PSMB9, two major components of the immunoproteasome, is predictive of better survival and improved response to immune-checkpoint inhibitors of melanoma patients. We study the mechanism underlying this connection by analyzing the antigenic peptide repertoire of cells that overexpress these subunits using HLA peptidomics. We find a higher response of patient-matched tumor infiltrating lymphocytes against antigens diferentially presented after immunoproteasome overexpression. Importantly, we find that PSMB8 and PSMB9 expression levels are much stronger predictors of melanoma patients' immune response to checkpoint inhibitors than the tumors' mutational burden. These results suggest that PSMB8 and PSMB9 expression levels can serve as important biomarkers for stratifying melanoma patients for immune-checkpoint treatment.

UVB-Induced Tumor Heterogeneity Diminishes Immune Response in Melanoma.

Although clonal neo-antigen burden is associated with improved response to immune therapy, the functional basis for this remains unclear. Here we study this question in a novel controlled mouse melanoma model that enables us to explore the effects of intra-tumor heterogeneity (ITH) on tumor aggressiveness and immunity independent of tumor mutational burden. Induction of UVB-derived mutations yields highly aggressive tumors with decreased anti-tumor activity. However, single-cell-derived tumors with reduced ITH are swiftly rejected. Their rejection is accompanied by increased T cell reactivity and a less suppressive microenvironment. Using phylogenetic analyses and mixing experiments of single-cell clones, we dissect two characteristics of ITH: the number of clones forming the tumor and their clonal diversity. Our analysis of melanoma patient tumor data recapitulates our results in terms of overall survival and response to immune checkpoint therapy. These findings highlight the importance of clonal mutations in robust immune surveillance and the need to quantify patient ITH to determine the response to checkpoint blockade.

Combined Analysis of Antigen Presentation and T-cell Recognition Reveals Restricted Immune Responses in Melanoma.

The quest for tumor-associated antigens (TAA) and neoantigens is a major focus of cancer immunotherapy. Here, we combine a neoantigen prediction pipeline and human leukocyte antigen (HLA) peptidomics to identify TAAs and neoantigens in 16 tumors derived from seven patients with melanoma and characterize their interactions with their tumor-infiltrating lymphocytes (TIL). Our investigation of the antigenic and T-cell landscapes encompassing the TAA and neoantigen signatures, their immune reactivity, and their corresponding T-cell identities provides the first comprehensive analysis of cancer cell T-cell cosignatures, allowing us to discover remarkable antigenic and TIL similarities between metastases from the same patient. Furthermore, we reveal that two neoantigen-specific clonotypes killed 90% of autologous melanoma cells, both in vitro and in vivo, showing that a limited set of neoantigen-specific T cells may play a central role in melanoma tumor rejection. Our findings indicate that combining HLA peptidomics with neoantigen predictions allows robust identification of targetable neoantigens, which could successfully guide personalized cancer immunotherapies.Significance: As neoantigen targeting is becoming more established as a powerful therapeutic approach, investigating these molecules has taken center stage. Here, we show that a limited set of neoantigen-specific T cells mediates tumor rejection, suggesting that identifying just a few antigens and their corresponding T-cell clones could guide personalized immunotherapy. Cancer Discov; 8(11); 1366-75. ©2018 AACR. This article is highlighted in the In This Issue feature, p. 1333.

An optimization framework for network annotation.

Motivation: A chief goal of systems biology is the reconstruction of large-scale executable models of cellular processes of interest. While accurate continuous models are still beyond reach, a powerful alternative is to learn a logical model of the processes under study, which predicts the logical state of any node of the model as a Boolean function of its incoming nodes. Key to learning such models is the functional annotation of the underlying physical interactions with activation/repression (sign) effects. Such annotations are pretty common for a few well-studied biological pathways. Results: Here we present a novel optimization framework for large-scale sign annotation that employs different plausible models of signaling and combines them in a rigorous manner. We apply our framework to two large-scale knockout datasets in yeast and evaluate its different components as well as the combined model to predict signs of different subsets of physical interactions. Overall, we obtain an accurate predictor that outperforms previous work by a considerable margin. Availability and implementation: The code is publicly available at https://github.com/spatkar94/NetworkAnnotation.git.

A network diffusion approach to inferring sample-specific function reveals functional changes associated with breast cancer.

Guilt-by-association codifies the empirical observation that a gene's function is informed by its neighborhood in a biological network. This would imply that when a gene's network context is altered, for instance in disease condition, so could be the gene's function. Although context-specific changes in biological networks have been explored, the potential changes they may induce on the functional roles of genes are yet to be characterized. Here we analyze, for the first time, the network-induced potential functional changes in breast cancer. Using transcriptomic samples for 1047 breast tumors and 110 healthy breast tissues from TCGA, we derive sample-specific protein interaction networks and assign sample-specific functions to genes via a diffusion strategy. Testing for significant changes in the inferred functions between normal and cancer samples, we find several functions to have significantly gained or lost genes in cancer, not due to differential expression of genes known to perform the function, but rather due to changes in the network topology. Our predicted functional changes are supported by mutational and copy number profiles in breast cancers. Our diffusion-based functional assignment provides a novel characterization of a tumor that is complementary to the standard approach based on functional annotation alone. Importantly, this characterization is effective in predicting patient survival, as well as in predicting several known histopathological subtypes of breast cancer.