Faculty Bibliography

Image-based analysis as a method for mutation detection can be advantageous in settings when tumor tissue is limited or unavailable for direct testing. Here, we utilize two distinct and complementary machine learning methods of analyzing whole slide images (WSI) for predicting mutated BRAF. In the first method, WSI of melanomas from 256 patients were used to train a deep convolutional neural network (CNN) in order to develop a fully automated model that first selects for tumor-rich areas (Area Under the Curve AUC=0.96) then predicts for mutated BRAF (AUC=0.71). Saliency mapping was performed and revealed that pixels corresponding to nuclei were the most relevant to network learning. In the second method, WSI were analyzed using a pathomics pipeline that first annotates nuclei and then quantifies nuclear features, demonstrating that mutated BRAF nuclei were significantly larger and rounder nuclei compared to BRAF WT nuclei. Lastly, we developed a model that combines clinical information, deep learning, and pathomics that improves the predictive performance for mutated BRAF to AUC=0.89. Not only does this provide additional insights on how BRAF mutations affect tumor structural characteristics, machine learning-based analysis of WSI has the potential to be integrated into higher order models for understanding tumor biology.

SHP2 inhibitors (SHP2i) alone and in various combinations are being tested in multiple tumors with over-activation of the RAS/ERK pathway. SHP2 plays critical roles in normal cell signaling; hence, SHP2is could influence the tumor microenvironment. We found that SHP2i treatment depleted alveolar and M2-like macrophages, induced tumor-intrinsic CCL5/CXCL10 secretion and promoted B and T lymphocyte infiltration in Kras- and Egfr-mutant non-small cell lung cancer (NSCLC). However, treatment also increased intratumor gMDSCs via tumor-intrinsic, NF-kB-dependent production of CXCR2 ligands. Other RAS/ERK pathway inhibitors also induced CXCR2 ligands and gMDSC influx in mice, and CXCR2 ligands were induced in tumors from patients on KRASG12C-inhibitor trials. Combined SHP2(SHP099)/CXCR1/2(SX682) inhibition depleted a specific cluster of S100a8/9high gMDSCs, generated Klrg1+ CD8+ effector T cells with a strong cytotoxic phenotype but expressing the checkpoint receptor NKG2A, and enhanced survival in Kras- and Egfr-mutant models. Our results argue for testing RAS/ERK pathway/CXCR1/2/NKG2A inhibitor combinations in NSCLC patients.

Introduction: T-cell Acute Lymphoblastic Leukemia (T-ALL) is an aggressive leukemia with a high incidence in children, adolescents and young adults. Although multiple therapeutic options are available, almost one fifth of patients affected by T-ALL eventually succumb to the disease, suggesting an unrecognized biological complexity that might contribute to drug resistance. To better understand the differences between T-ALL subtypes and their clinical course, we systematically analyzed cohorts of patients with different risk profiles at the genetic and epigenetic level. We recently demonstrated that differences in three-dimensional (3D) chromatin architecture can influence the integrity of topologically associating domains (TADs) and rewire specific enhancer-promoter interactions, impacting gene expression and leading to disease. As an example, we focused in particular on the Myc family of oncogenes, revealing disease-specific patterns of enhancer-promoter interactions.
Method(s): We initially profiled a large cohort of T-ALL patients falling under different risk categories in order to identify differences in their genetic and epigenetic features. We then systematically integrated matched in situ Hi-C, RNA-seq and CTCF ChIP-seq datasets to reveal widespread differences in intra-TAD chromatin interactions and TAD boundary insulation in patients affected by TALL.
Result(s): Using primary acute leukemia patient samples, we have, for the first time, identified recurrent TAD disruptions in leukemia involving key oncogenes (e.g. NOTCH1, c-Myc) and their targets. For example, we identified a recurrent disruption of 3D chromatin topology in the c-Myc locus at a previously uncharacterized non-coding CTCF-bound region that insulates MYC from a downstream super-enhancer. This disruption enables chromatin interactions between the c-Myc oncogene and the downstream super-enhancer leading to an increase in c-Myc expression. In parallel, while focusing patients falling into a higher risk category, namely the early T cell progenitor acute lymphoblastic leukemia (ETP-ALL), we discovered a previously uncharacterized region of high activity encompassing a novel lncRNA interacting with the proto-oncogene N-Myc.
Conclusion(s):With the current study we demonstrated an inherent difference between subtypes of T-ALL based on their epigenetic profile, which in turn influences the expression of key oncogenes. By focusing on a new methods of regulating a known family of transcription factors, we provide a new mechanism which could open interesting ways for targeted therapy of patients at different risk levels

Introduction: Peripheral T-cell lymphomas (PTCLs) include heterogeneous entities of rare and aggressive neoplasms. The improved understanding of the biological/molecular mechanisms driving T-cell transformation and tumor maintenance has powerfully propelled new therapeutic programs. However, despite this progress, PTCLs remain an unmet medical need. Recurrent aberrations and the deregulated activation of distinct signaling pathways have been mapped and linked to selective subtypes. The JAK/STAT signaling pathway's deregulated activation plays a pathogenetic role in PTCL, including ALCL subtypes. STATs regulate the differentiation/phenotype, survival and cell-growth, metabolism, and drug resistance of T-cell lymphomas as well as host immunosuppressive microenvironments. Although many drugs' discovery programs were launched, a plethora of compounds has failed.
Method(s): We have discovered heterobifunctional molecules by an iterative medicinal chemistry SAR campaign that potently and selectively degrade STAT3 in a proteasome-dependent manner. Conventional PTCL cell lines and Patient Derived Tumor Xenograft (PDTX) and/or derived cell lines (PDTX-CL), carrying either WT- or mutated-STAT3, were exposed to increasing amounts (50nM5microM) of STAT3-degraders. Proteins and mRNA transcripts (2144hrs) were assessed by deep-proteomics and paired-end RNA sequencing, combined with WB/flow cytometry and qRT-PCR. Cell-titer-glo, cell titer blue, Annexin-V and S-cell cycle analyses were used as readouts. Chromatin accessibility, STAT3 DNA binding, 3D chromosomal architecture reorganization and 5-hmC profiling were assessed by ATACseq, CHIPseq and Hi-C and H3K27ac Hi-CHIP and mass-spectrometry. Drug testing/discovery combinations (96-well-plate) were performed using a semi-automated flow-cytometry. A battery of PTCL PDTX models were tested in pre-clinical trials.
Result(s): Treatment of ALK+ ALCL (SU-DHL1) led to the rapid (~6hrs.) and profound down-regulation of STAT3 followed by the loss of canonical STAT3-regulated proteins (SOCS3, MYC, Granzyme B, GAS1, and IL2RA), without appreciable changes for other STAT family members (STAT1, STAT5b). In vitro, cytoplasmic, nuclear, and mitochondrial STAT3 downregulation was maintained up to 144 hrs. Loss of STAT3 in ALK+/- ALCL and BIA-ALCL cells was associated with major transcriptional changes (7116-10615 and 15114 DEGs in ALK- and ALK+ ALCL, respectively), underscoring public/shared as well as private time-dependent signatures. Main down-regulated pathways included JAK-STAT, MAPK, NF-kB, PI3K, TGFb, and TNFa. Comparison of STAT3 shRNA (ALK+ ALCL) and STAT3 degrader (ALK-/ALK+ ALCL) signatures demonstrated a substantial and concordant gene modulation (24hrs) among all models with the highest overlaps between ALK+ ALCL (Figure 3). To identify direct STAT3 gene targets, we analyzed CHIPseq peaks and predicted bindings sites, demonstrating that canonical genes, i.e., SOCS3, Granzyme B, GAS1, IL2RA, STAT3, and CD30, were significantly downregulated. Conversely, CD58, CD274, and MCH-I/II were upregulated at late time points. By mapping the STAT3 binding sites in ALK+ and ALK- ALCL, we have identified 1077 and 2763 STAT3 peaks within promoter/5'-/3'- and distant intergenic regions, corresponding to both coding and non-coding genes. Therapeutically, in vitro treatments led to cell cycle arrest and profound growth inhibition, and over time increased cell death of PTCL cells, including ALCL. Accordingly, growth inhibition of ALCL xenograft and PDTX tumors was also achieved (Figure 2). To identify drugs that could synergize withSTAT3-degrader activity, we tested a compound library (40) targeting pro-tumorigenic PTCL pathways as well as FDA-approved compounds. Ongoing studies are in progress.
Conclusion(s): We have discovered selective STAT3 degraders which control PTCL growth. STAT3 degraders are powerful tools to define the STAT3 pathogenetic mechanisms and dissect genes/pathways to be targeted for T-cell lymphoma eradication. These data provide additional rationale for testing STAT3 degraders in the clinic for the treatment of aggressive malignancies including PTCL/ALCL. [Formula presented] Disclosures: Yang: Kymera Therapeutics: Current Employment, Current equity holder in publicly-traded company. Sharma: Kymera Therapeutics: Current Employment, Current equity holder in publicly-traded company. Dey: Kymera Therapeutics: Current Employment, Current equity holder in publicly-traded company. Karnik: Kymera Therapeutics: Current Employment, Current equity holder in publicly-traded company. Elemento: Owkin: Consultancy, Other: Current equity holder; Volastra Therapeutics: Consultancy, Other: Current equity holder, Research Funding; Johnson and Johnson: Research Funding; Eli Lilly: Research Funding; Janssen: Research Funding; Champions Oncology: Consultancy; Freenome: Consultancy, Other: Current equity holder in a privately-held company; One Three Biotech: Consultancy, Other: Current equity holder; AstraZeneca: Research Funding. Horwitz: Affimed: Research Funding; Aileron: Research Funding; ADC Therapeutics, Affimed, Aileron, Celgene, Daiichi Sankyo, Forty Seven, Inc., Kyowa Hakko Kirin, Millennium /Takeda, Seattle Genetics, Trillium Therapeutics, and Verastem/SecuraBio.: Consultancy, Research Funding; Acrotech Biopharma, Affimed, ADC Therapeutics, Astex, Merck, Portola Pharma, C4 Therapeutics, Celgene, Janssen, Kura Oncology, Kyowa Hakko Kirin, Myeloid Therapeutics, ONO Pharmaceuticals, Seattle Genetics, Shoreline Biosciences, Inc, Takeda, Trillium Th: Consultancy; Celgene: Research Funding; C4 Therapeutics: Consultancy; Crispr Therapeutics: Research Funding; Daiichi Sankyo: Research Funding; Forty Seven, Inc.: Research Funding; Kura Oncology: Consultancy; Kyowa Hakko Kirin: Consultancy, Research Funding; Millennium/Takeda: Research Funding; Myeloid Therapeutics: Consultancy; ONO Pharmaceuticals: Consultancy; Seattle Genetics: Consultancy, Research Funding; Secura Bio: Consultancy; Shoreline Biosciences, Inc.: Consultancy; Takeda: Consultancy; Trillium Therapeutics: Consultancy, Research Funding; Tubulis: Consultancy; Verastem/Securabio: Research Funding. DeSavi: Kymera Therapeutics: Current Employment, Current equity holder in publicly-traded company. Liu: Kymera Therapeutics: Current Employment, Current equity holder in publicly-traded company.
Copyright

Substantial progress has been made in understanding how tumors escape immune surveillance. However, few measures to counteract tumor immune evasion have been developed. Suppression of tumor antigen expression is a common adaptive mechanism that cancers use to evade detection and destruction by the immune system. Epigenetic modifications play a critical role in various aspects of immune invasion, including the regulation of tumor antigen expression. To identify epigenetic regulators of tumor antigen expression, we established a transplantable syngeneic tumor model of immune escape with silenced antigen expression and used this system as a platform for a CRISPR-Cas9 suppressor screen for genes encoding epigenetic modifiers. We found that disruption of the genes encoding either of the chromatin modifiers activating transcription factor 7-interacting protein (Atf7ip) or its interacting partner SET domain bifurcated histone lysine methyltransferase 1 (Setdb1) in tumor cells restored tumor antigen expression. This resulted in augmented tumor immunogenicity concomitant with elevated endogenous retroviral (ERV) antigens and mRNA intron retention. ERV disinhibition was associated with a robust type I interferon response and increased T-cell infiltration, leading to rejection of cells lacking intact Atf7ip or Setdb1. ATF7IP or SETDB1 expression inversely correlated with antigen processing and presentation pathways, interferon signaling, and T-cell infiltration and cytotoxicity in human cancers. Our results provide a rationale for targeting Atf7ip or Setdb1 in cancer immunotherapy.

Oncogene-induced senescence (OIS) is an inherent and important tumor suppressor mechanism. However, if not removed timely via immune surveillance, senescent cells also have detrimental effects. Although this has mostly been attributed to the senescence-associated secretory phenotype (SASP) of these cells, we recently proposed that "escape" from the senescent state is another unfavorable outcome. The mechanism underlying this phenomenon remains elusive. Here, we exploit genomic and functional data from a prototypical human epithelial cell model carrying an inducible CDC6 oncogene to identify an early-acquired recurrent chromosomal inversion that harbors a locus encoding the circadian transcription factor BHLHE40. This inversion alone suffices for BHLHE40 activation upon CDC6 induction and driving cell cycle re-entry of senescent cells, and malignant transformation. Ectopic overexpression of BHLHE40 prevented induction of CDC6-triggered senescence. We provide strong evidence in support of replication stress-induced genomic instability being a causative factor underlying "escape" from oncogene-induced senescence.

Loss of imprinting (LOI) results in severe developmental defects, but the mechanisms preventing LOI remain incompletely understood. Here, we dissect the functional components of the imprinting control region of the essential Dlk1-Dio3 locus (called IG-DMR) in pluripotent stem cells. We demonstrate that the IG-DMR consists of two antagonistic elements: a paternally methylated CpG island that prevents recruitment of TET dioxygenases and a maternally unmethylated non-canonical enhancer that ensures expression of the Gtl2 lncRNA by counteracting de novo DNA methyltransferases. Genetic or epigenetic editing of these elements leads to distinct LOI phenotypes with characteristic alternations of allele-specific gene expression, DNA methylation, and 3D chromatin topology. Although repression of the Gtl2 promoter results in dysregulated imprinting, the stability of LOI phenotypes depends on the IG-DMR, suggesting a functional hierarchy. These findings establish the IG-DMR as a bipartite control element that maintains imprinting by allele-specific restriction of the DNA (de)methylation machinery.