CTLA-4 immune checkpoint blockade is clinically effective in a subset of patients with metastatic melanoma. We identify a subcluster of MAGE-A cancer-germline antigens, located within a narrow 75 kb region of chromosome Xq28, that predicts resistance uniquely to blockade of CTLA-4, but not PD-1. We validate this gene expression signature in an independent anti-CTLA-4-treated cohort and show its specificity to the CTLA-4 pathway with two independent anti-PD-1-treated cohorts. Autophagy, a process critical for optimal anti-cancer immunity, has previously been shown to be suppressed by the MAGE-TRIM28 ubiquitin ligase in vitro. We now show that the expression of the key autophagosome component LC3B and other activators of autophagy are negatively associated with MAGE-A protein levels in human melanomas, including samples from patients with resistance to CTLA-4 blockade. Our findings implicate autophagy suppression in resistance to CTLA-4 blockade in melanoma, suggesting exploitation of autophagy induction for potential therapeutic synergy with CTLA-4 inhibitors.

MicroRNA (miRNA) dysregulation frequently occurs in cancer. Analysis of whole blood miRNA in tumor models has not been widely reported, but could potentially lead to novel assays for early detection and monitoring of cancer. To determine whether miRNAs associated with malignancy could be detected in the peripheral blood, we used real-time reverse transcriptase-PCR to determine miRNA profiles in whole blood obtained from transgenic mice with c-MYC-induced lymphoma, hepatocellular carcinoma and osteosarcoma. The PCR-based assays used in our studies require only 10 nanograms of total RNA, allowing serial mini-profiles (20 - 30 miRNAs) to be carried out on individual animals over time. Blood miRNAs were measured from mice at different stages of MYC-induced lymphomagenesis and regression. Unsupervised hierarchical clustering of the data identified specific miRNA expression profiles that correlated with tumor type and stage. The miRNAs found to be altered in the blood of mice with tumors frequently reverted to normal levels upon tumor regression. Our results suggest that specific changes in blood miRNA can be detected during tumorigenesis and tumor regression.

To elucidate mechanisms by which T cells eliminate leukemia, we study donor lymphocyte infusion (DLI), an established immunotherapy for relapsed leukemia. We model T cell dynamics by integrating longitudinal, multimodal data from 94,517 bone marrow-derived single T cell transcriptomes in addition to chromatin accessibility and single T cell receptor sequencing from patients undergoing DLI. We find that responsive tumors are defined by enrichment of late-differentiated T cells before DLI and rapid, durable expansion of early differentiated T cells after treatment, highly similar to "terminal" and "precursor" exhausted subsets, respectively. Resistance, in contrast, is defined by heterogeneous T cell dysfunction. Surprisingly, early differentiated T cells in responders mainly originate from pre-existing and novel clonotypes recruited to the leukemic microenvironment, rather than the infusion. Our work provides a paradigm for analyzing longitudinal single-cell profiling of scenarios beyond adoptive cell therapy and introduces Symphony, a Bayesian approach to infer regulatory circuitry underlying T cell subsets, with broad relevance to exhaustion antagonists across cancers.

Prediction of HLA epitopes is important for the development of cancer immunotherapies and vaccines. However, current prediction algorithms have limited predictive power, in part because they were not trained on high-quality epitope datasets covering a broad range of HLA alleles. To enable prediction of endogenous HLA class I-associated peptides across a large fraction of the human population, we used mass spectrometry to profile >185,000 peptides eluted from 95 HLA-A, -B, -C and -G mono-allelic cell lines. We identified canonical peptide motifs per HLA allele, unique and shared binding submotifs across alleles and distinct motifs associated with different peptide lengths. By integrating these data with transcript abundance and peptide processing, we developed HLAthena, providing allele-and-length-specific and pan-allele-pan-length prediction models for endogenous peptide presentation. These models predicted endogenous HLA class I-associated ligands with 1.5-fold improvement in positive predictive value compared with existing tools and correctly identified >75% of HLA-bound peptides that were observed experimentally in 11 patient-derived tumor cell lines.

Unlike many other hematologic malignancies, Richter syndrome (RS), an aggressive B cell lymphoma originating from indolent chronic lymphocytic leukemia, is responsive to PD-1 blockade. To discover the determinants of response, we analyze single-cell transcriptome data generated from 17 bone marrow samples longitudinally collected from 6 patients with RS. Response is associated with intermediate exhausted CD8 effector/effector memory T cells marked by high expression of the transcription factor ZNF683, determined to be evolving from stem-like memory cells and divergent from terminally exhausted cells. This signature overlaps with that of tumor-infiltrating populations from anti-PD-1 responsive solid tumors. ZNF683 is found to directly target key T cell genes (TCF7, LMO2, CD69) and impact pathways of T cell cytotoxicity and activation. Analysis of pre-treatment peripheral blood from 10 independent patients with RS treated with anti-PD-1, as well as patients with solid tumors treated with anti-PD-1, supports an association of ZNF683high T cells with response.

The recent successes of cancer immunotherapies have stimulated interest in the potential widespread application of these approaches; haematological malignancies have provided both initial proofs of concept and an informative testing ground for various immune-based therapeutics. The immune-cell origin of many of the blood malignancies provides a unique opportunity both to understand the mechanisms of cancer immune responsiveness and immune evasion, and to exploit these mechanisms for therapeutic purposes.

Oncogene addiction is thought to occur cell autonomously. Immune effectors are implicated in the initiation and restraint of tumorigenesis, but their role in oncogene inactivation-mediated tumor regression is unclear. Here, we show that an intact immune system, specifically CD4(+) T cells, is required for the induction of cellular senescence, shutdown of angiogenesis, and chemokine expression resulting in sustained tumor regression upon inactivation of the MYC or BCR-ABL oncogenes in mouse models of T cell acute lymphoblastic lymphoma and pro-B cell leukemia, respectively. Moreover, immune effectors knocked out for thrombospondins failed to induce sustained tumor regression. Hence, CD4(+) T cells are required for the remodeling of the tumor microenvironment through the expression of chemokines, such as thrombospondins, in order to elicit oncogene addiction.

Relapse of myeloproliferative neoplasms (MPNs) after allogeneic hematopoietic stem cell transplantation (HSCT) is associated with poor outcomes, as therapeutic approaches to reinstate effective graft-versus-leukemia (GVL) responses remain suboptimal. Immune escape through overexpression of PD-L1 in JAK2V617F-mutated MPN provides a rationale for therapeutic PD-1 blockade, and indeed, clinical activity of nivolumab in relapsed MPN post-HSCT has been observed. Elucidation of the features of response following PD-1 blockade in such patients could inform novel therapeutic concepts that enhance GVL. Here, we report an integrated high-dimensional analysis using single-cell RNA sequencing, T-cell receptor sequencing, cellular indexing of transcriptomes and epitopes by sequencing (CITE-seq), and assay for transposase-accessible chromatin using sequencing (scATAC-seq), together with mass cytometry, in peripheral blood mononuclear cells collected at 6 timepoints before, during, and after transient response to PD-1 blockade from an index case of relapsed MPN following HSCT. Before nivolumab infusion, acute myeloid leukemia (AML) blasts demonstrated high expression of chemokines, and T cells were characterized by expression of interferon-response genes. This baseline inflammatory signature disappeared after nivolumab infusion. Clinical response was characterized by transient expansion of a polyclonal CD4+ T-cell population and contraction of an AML subpopulation that exhibited megakaryocytic features and elevated PD-L1 expression. At relapse, the proportion of the AML subpopulation with progenitor-like features progressively increased, suggesting coevolution of AML blasts and donor-derived T cells. We thus demonstrate how single-cell technologies can provide complementary insight into cellular mechanisms underlying response to PD-1 blockade, motivating future longitudinal high-dimensional single-cell studies of GVL responses in relapsed myeloid disease.