Detecting small cell transformation in patients with advanced EGFR mutant lung adenocarcinoma through epigenomic cfDNA profiling.

PURPOSE: Histologic transformation to small cell lung cancer (SCLC) is a mechanism of treatment resistance in patients with advanced oncogene-driven lung adenocarcinoma (LUAD) that currently requires histologic review for diagnosis. Herein, we sought to develop an epigenomic cell-free (cf)DNA-based approach to non-invasively detect small cell transformation in patients with EGFR mutant (EGFRm) LUAD. EXPERIMENTAL DESIGN: To characterize the epigenomic landscape of transformed (t)SCLC relative to LUAD and de novo SCLC, we performed chromatin immunoprecipitation sequencing (ChIP-seq) to profile the histone modifications H3K27ac, H3K4me3, and H3K27me3, methylated DNA immunoprecipitation sequencing (MeDIP-seq), assay for transposase-accessible chromatin sequencing (ATAC-seq), and RNA sequencing on 26 lung cancer patient-derived xenograft (PDX) tumors. We then generated and analyzed H3K27ac ChIP-seq, MeDIP-seq, and whole genome sequencing cfDNA data from 1 ml aliquots of plasma from patients with EGFRm LUAD with or without tSCLC. RESULTS: Analysis of 126 epigenomic libraries from the lung cancer PDXs revealed widespread epigenomic reprogramming between LUAD and tSCLC, with a large number of differential H3K27ac (n=24,424), DNA methylation (n=3,298), and chromatin accessibility (n=16,352) sites between the two histologies. Tumor-informed analysis of each of these three epigenomic features in cfDNA resulted in accurate non-invasive discrimination between patients with EGFRm LUAD versus tSCLC (AUROC=0.82-0.87). A multi-analyte cfDNA-based classifier integrating these three epigenomic features discriminated between EGFRm LUAD versus tSCLC with an AUROC of 0.94. CONCLUSIONS: These data demonstrate the feasibility of detecting small cell transformation in patients with EGFRm LUAD through epigenomic cfDNA profiling of 1 ml of patient plasma.

Transcription factor dynamics, oscillation, and functions in human enteroendocrine cell differentiation.

Enteroendocrine cells (EECs) secrete serotonin (enterochromaffin [EC] cells) or specific peptide hormones (non-EC cells) that serve vital metabolic functions. The basis for terminal EEC diversity remains obscure. By forcing activity of the transcription factor (TF) NEUROG3 in 2D cultures of human intestinal stem cells, we replicated physiologic EEC differentiation and examined transcriptional and cis-regulatory dynamics that culminate in discrete cell types. Abundant EEC precursors expressed stage-specific genes and TFs. Before expressing pre-terminal NEUROD1, post-mitotic precursors oscillated between transcriptionally distinct ASCL1+ and HES6hi cell states. Loss of either factor accelerated EEC differentiation substantially and disrupted EEC individuality; ASCL1 or NEUROD1 deficiency had opposing consequences on EC and non-EC cell features. These TFs mainly bind cis-elements that are accessible in undifferentiated stem cells, and they tailor subsequent expression of TF combinations that underlie discrete EEC identities. Thus, early TF oscillations retard EEC maturation to enable accurate diversity within a medically important cell lineage.

A Lung Cancer Mouse Model Database.

Lung cancer, the leading cause of cancer mortality, exhibits diverse histological subtypes and genetic complexities. Numerous preclinical mouse models have been developed to study lung cancer, but data from these models are disparate, siloed, and difficult to compare in a centralized fashion. Here we established the Lung Cancer Mouse Model Database (LCMMDB), an extensive repository of 1,354 samples from 77 transcriptomic datasets covering 974 samples from genetically engineered mouse models (GEMMs), 368 samples from carcinogen-induced models, and 12 samples from a spontaneous model. Meticulous curation and collaboration with data depositors have produced a robust and comprehensive database, enhancing the fidelity of the genetic landscape it depicts. The LCMMDB aligns 859 tumors from GEMMs with human lung cancer mutations, enabling comparative analysis and revealing a pressing need to broaden the diversity of genetic aberrations modeled in GEMMs. Accompanying this resource, we developed a web application that offers researchers intuitive tools for in-depth gene expression analysis. With standardized reprocessing of gene expression data, the LCMMDB serves as a powerful platform for cross-study comparison and lays the groundwork for future research, aiming to bridge the gap between mouse models and human lung cancer for improved translational relevance.

Mammalian SWI/SNF complex activity regulates POU2F3 and constitutes a targetable dependency in small cell lung cancer.

Small cell lung cancers (SCLC) are comprised of heterogeneous subtypes marked by lineage-specific transcription factors, including ASCL1, NEUROD1, and POU2F3. POU2F3-positive SCLC, ∼12% of all cases, are uniquely dependent on POU2F3 itself; as such, approaches to attenuate POU2F3 expression may represent new therapeutic opportunities. Here using genome-scale screens for regulators of POU2F3 expression and SCLC proliferation, we define mSWI/SNF complexes, including non-canonical BAF (ncBAF) complexes, as top dependencies specific to POU2F3-positive SCLC. Notably, clinical-grade pharmacologic mSWI/SNF inhibition attenuates proliferation of all POU2F3-positive SCLCs, while disruption of ncBAF via BRD9 degradation is uniquely effective in pure non-neuroendocrine POU2F3-SCLCs. mSWI/SNF maintains accessibility over gene loci central to POU2F3-mediated gene regulatory networks. Finally, chemical targeting of SMARCA4/2 mSWI/SNF ATPases and BRD9 decrease POU2F3-SCLC tumor growth and increase survival in vivo . Taken together, these results characterize mSWI/SNF-mediated global governance of the POU2F3 oncogenic program and suggest mSWI/SNF inhibition as a therapeutic strategy for SCLC.

Genetically-engineered mouse models of small cell lung cancer: the next generation.

Small cell lung cancer (SCLC) remains the most fatal form of lung cancer, with patients in dire need of new and effective therapeutic approaches. Modeling SCLC in an immunocompetent host is essential for understanding SCLC pathogenesis and ultimately discovering and testing new experimental therapeutic strategies. Human SCLC is characterized by near universal genetic loss of the RB1 and TP53 tumor suppressor genes. Twenty years ago, the first genetically-engineered mouse model (GEMM) of SCLC was generated using conditional deletion of both Rb1 and Trp53 in the lungs of adult mice. Since then, several other GEMMs of SCLC have been developed coupling genomic alterations found in human SCLC with Rb1 and Trp53 deletion. Here we summarize how GEMMs of SCLC have contributed significantly to our understanding of the disease in the past two decades. We also review recent advances in modeling SCLC in mice that allow investigators to bypass limitations of the previous generation of GEMMs while studying new genes of interest in SCLC. In particular, CRISPR/Cas9-mediated somatic gene editing can accelerate how new genes of interest are functionally interrogated in SCLC tumorigenesis. Notably, the development of allograft models and precancerous precursor models from SCLC GEMMs provides complementary approaches to GEMMs to study tumor cell-immune microenvironment interactions and test new therapeutic strategies to enhance response to immunotherapy. Ultimately, the new generation of SCLC models can accelerate research and help develop new therapeutic strategies for SCLC.

Transcription factor dynamics, oscillation, and functions in human enteroendocrine cell differentiation.

Enteroendocrine cells (EECs), which secrete serotonin (enterochromaffin cells, EC) or a dominant peptide hormone, serve vital physiologic functions. As with any adult human lineage, the basis for terminal cell diversity remains obscure. We replicated human EEC differentiation in vitro , mapped transcriptional and chromatin dynamics that culminate in discrete cell types, and studied abundant EEC precursors expressing selected transcription factors (TFs) and gene programs. Before expressing the pre-terminal factor NEUROD1, non-replicating precursors oscillated between epigenetically similar but transcriptionally distinct ASCL1 + and HES6 hi cell states. Loss of either factor substantially accelerated EEC differentiation and disrupted EEC individuality; ASCL1 or NEUROD1 deficiency had opposing consequences on EC and hormone-producing cell features. Expressed late in EEC differentiation, the latter TFs mainly bind cis -elements that are accessible in undifferentiated stem cells and tailor the subsequent expression of TF combinations that specify EEC types. Thus, TF oscillations retard EEC maturation to enable accurate EEC diversification.

Aurora A kinase inhibition induces accumulation of SCLC tumor cells in mitosis with restored interferon signaling to increase response to PD-L1.

Despite small cell lung cancers (SCLCs) having a high mutational burden, programmed death-ligand 1 (PD-L1) immunotherapy only modestly increases survival. A subset of SCLCs that lose their ASCL1 neuroendocrine phenotype and restore innate immune signaling (termed the "inflammatory" subtype) have durable responses to PD-L1. Some SCLCs are highly sensitive to Aurora kinase inhibitors, but early-phase trials show short-lived responses, suggesting effective therapeutic combinations are needed to increase their durability. Using immunocompetent SCLC genetically engineered mouse models (GEMMs) and syngeneic xenografts, we show durable efficacy with the combination of a highly specific Aurora A kinase inhibitor (LSN3321213) and PD-L1. LSN3321213 causes accumulation of tumor cells in mitosis with lower ASCL1 expression and higher expression of interferon target genes and antigen-presentation genes mimicking the inflammatory subtype in a cell-cycle-dependent manner. These data demonstrate that inflammatory gene expression is restored in mitosis in SCLC, which can be exploited by Aurora A kinase inhibition.

Liquid biopsy epigenomic profiling for cancer subtyping.

Although circulating tumor DNA (ctDNA) assays are increasingly used to inform clinical decisions in cancer care, they have limited ability to identify the transcriptional programs that govern cancer phenotypes and their dynamic changes during the course of disease. To address these limitations, we developed a method for comprehensive epigenomic profiling of cancer from 1 ml of patient plasma. Using an immunoprecipitation-based approach targeting histone modifications and DNA methylation, we measured 1,268 epigenomic profiles in plasma from 433 individuals with one of 15 cancers. Our assay provided a robust proxy for transcriptional activity, allowing us to infer the expression levels of diagnostic markers and drug targets, measure the activity of therapeutically targetable transcription factors and detect epigenetic mechanisms of resistance. This proof-of-concept study in advanced cancers shows how plasma epigenomic profiling has the potential to unlock clinically actionable information that is currently accessible only via direct tissue sampling.

KDM6A epigenetically regulates subtype plasticity in small cell lung cancer.

Small cell lung cancer (SCLC) exists broadly in four molecular subtypes: ASCL1, NEUROD1, POU2F3 and Inflammatory. Initially, SCLC subtypes were thought to be mutually exclusive, but recent evidence shows intra-tumoural subtype heterogeneity and plasticity between subtypes. Here, using a CRISPR-based autochthonous SCLC genetically engineered mouse model to study the consequences of KDM6A/UTX inactivation, we show that KDM6A inactivation induced plasticity from ASCL1 to NEUROD1 resulting in SCLC tumours that express both ASCL1 and NEUROD1. Mechanistically, KDM6A normally maintains an active chromatin state that favours the ASCL1 subtype with its loss decreasing H3K4me1 and increasing H3K27me3 at enhancers of neuroendocrine genes leading to a cell state that is primed for ASCL1-to-NEUROD1 subtype switching. This work identifies KDM6A as an epigenetic regulator that controls ASCL1 to NEUROD1 subtype plasticity and provides an autochthonous SCLC genetically engineered mouse model to model ASCL1 and NEUROD1 subtype heterogeneity and plasticity, which is found in 35-40% of human SCLCs.

Regulation of neuroendocrine plasticity by the RNA-binding protein ZFP36L1.

Some small cell lung cancers (SCLCs) are highly sensitive to inhibitors of the histone demethylase LSD1. LSD1 inhibitors are thought to induce their anti-proliferative effects by blocking neuroendocrine differentiation, but the mechanisms by which LSD1 controls the SCLC neuroendocrine phenotype are not well understood. To identify genes required for LSD1 inhibitor sensitivity in SCLC, we performed a positive selection genome-wide CRISPR/Cas9 loss of function screen and found that ZFP36L1, an mRNA-binding protein that destabilizes mRNAs, is required for LSD1 inhibitor sensitivity. LSD1 binds and represses ZFP36L1 and upon LSD1 inhibition, ZFP36L1 expression is restored, which is sufficient to block the SCLC neuroendocrine differentiation phenotype and induce a non-neuroendocrine "inflammatory" phenotype. Mechanistically, ZFP36L1 binds and destabilizes SOX2 and INSM1 mRNAs, two transcription factors that are required for SCLC neuroendocrine differentiation. This work identifies ZFP36L1 as an LSD1 target gene that controls the SCLC neuroendocrine phenotype and demonstrates that modulating mRNA stability of lineage transcription factors controls neuroendocrine to non-neuroendocrine plasticity.

Plasticity in the Absence of NOTCH Uncovers a RUNX2-Dependent Pathway in Small Cell Lung Cancer.

Neuroendocrine to nonneuroendocrine plasticity supports small cell lung cancer (SCLC) tumorigenesis and promotes immunogenicity. Approximately 20% to 25% of SCLCs harbor loss-of-function (LOF) NOTCH mutations. Previous studies demonstrated that NOTCH functions as a SCLC tumor suppressor, but can also drive nonneuroendocrine plasticity to support SCLC growth. Given the dual functionality of NOTCH, it is not understood why SCLCs select for LOF NOTCH mutations and how these mutations affect SCLC tumorigenesis. In a CRISPR-based genetically engineered mouse model of SCLC, genetic loss of Notch1 or Notch2 modestly accelerated SCLC tumorigenesis. Interestingly, Notch-mutant SCLCs still formed nonneuroendocrine subpopulations, and these Notch-independent, nonneuroendocrine subpopulations were driven by Runx2-mediated regulation of Rest. Notch2-mutant nonneuroendocrine cells highly express innate immune signaling genes including stimulator of interferon genes (STING) and were sensitive to STING agonists. This work identifies a Notch-independent mechanism to promote nonneuroendocrine plasticity and suggests that therapeutic approaches to activate STING could be selectively beneficial for SCLCs with NOTCH2 mutations. SIGNIFICANCE: A genetically engineered mouse model of NOTCH-mutant SCLC reveals that nonneuroendocrine plasticity persists in the absence of NOTCH, driven by a RUNX2-REST-dependent pathway and innate immune signaling.

Intrinsic Immunogenicity of Small Cell Lung Carcinoma Revealed by Its Cellular Plasticity.

Small cell lung carcinoma (SCLC) is highly mutated, yet durable response to immune checkpoint blockade (ICB) is rare. SCLC also exhibits cellular plasticity, which could influence its immunobiology. Here we discover that a distinct subset of SCLC uniquely upregulates MHC I, enriching for durable ICB benefit. In vitro modeling confirms epigenetic recovery of MHC I in SCLC following loss of neuroendocrine differentiation, which tracks with derepression of STING. Transient EZH2 inhibition expands these nonneuroendocrine cells, which display intrinsic innate immune signaling and basally restored antigen presentation. Consistent with these findings, murine nonneuroendocrine SCLC tumors are rejected in a syngeneic model, with clonal expansion of immunodominant effector CD8 T cells. Therapeutically, EZH2 inhibition followed by STING agonism enhances T-cell recognition and rejection of SCLC in mice. Together, these data identify MHC I as a novel biomarker of SCLC immune responsiveness and suggest novel immunotherapeutic approaches to co-opt SCLC's intrinsic immunogenicity. SIGNIFICANCE: SCLC is poorly immunogenic, displaying modest ICB responsiveness with rare durable activity. In profiling its plasticity, we uncover intrinsically immunogenic MHC Ihi subpopulations of nonneuroendocrine SCLC associated with durable ICB benefit. We also find that combined EZH2 inhibition and STING agonism uncovers this cell state, priming cells for immune rejection.This article is highlighted in the In This Issue feature, p. 1861.

Targeting oncoproteins with a positive selection assay for protein degraders.

Most intracellular proteins lack hydrophobic pockets suitable for altering their function with drug-like small molecules. Recent studies indicate that some undruggable proteins can be targeted by compounds that can degrade them. For example, thalidomide-like drugs (IMiDs) degrade the critical multiple myeloma transcription factors IKZF1 and IKZF3 by recruiting them to the cereblon E3 ubiquitin ligase. Current loss of signal ("down") assays for identifying degraders often exhibit poor signal-to-noise ratios, narrow dynamic ranges, and false positives from compounds that nonspecifically suppress transcription or translation. Here, we describe a gain of signal ("up") assay for degraders. In arrayed chemical screens, we identified novel IMiD-like IKZF1 degraders and Spautin-1, which, unlike the IMiDs, degrades IKZF1 in a cereblon-independent manner. In a pooled CRISPR-Cas9-based screen, we found that CDK2 regulates the abundance of the ASCL1 oncogenic transcription factor. This methodology should facilitate the identification of drugs that directly or indirectly degrade undruggable proteins.

CDK7 Inhibition Potentiates Genome Instability Triggering Anti-tumor Immunity in Small Cell Lung Cancer.

Cyclin-dependent kinase 7 (CDK7) is a central regulator of the cell cycle and gene transcription. However, little is known about its impact on genomic instability and cancer immunity. Using a selective CDK7 inhibitor, YKL-5-124, we demonstrated that CDK7 inhibition predominately disrupts cell-cycle progression and induces DNA replication stress and genome instability in small cell lung cancer (SCLC) while simultaneously triggering immune-response signaling. These tumor-intrinsic events provoke a robust immune surveillance program elicited by T cells, which is further enhanced by the addition of immune-checkpoint blockade. Combining YKL-5-124 with anti-PD-1 offers significant survival benefit in multiple highly aggressive murine models of SCLC, providing a rationale for new combination regimens consisting of CDK7 inhibitors and immunotherapies.

The KDM5A/RBP2 histone demethylase represses NOTCH signaling to sustain neuroendocrine differentiation and promote small cell lung cancer tumorigenesis.

More than 90% of small cell lung cancers (SCLCs) harbor loss-of-function mutations in the tumor suppressor gene RB1 The canonical function of the RB1 gene product, pRB, is to repress the E2F transcription factor family, but pRB also functions to regulate cellular differentiation in part through its binding to the histone demethylase KDM5A (also known as RBP2 or JARID1A). We show that KDM5A promotes SCLC proliferation and SCLC's neuroendocrine differentiation phenotype in part by sustaining expression of the neuroendocrine transcription factor ASCL1. Mechanistically, we found that KDM5A sustains ASCL1 levels and neuroendocrine differentiation by repressing NOTCH2 and NOTCH target genes. To test the role of KDM5A in SCLC tumorigenesis in vivo, we developed a CRISPR/Cas9-based mouse model of SCLC by delivering an adenovirus (or an adeno-associated virus [AAV]) that expresses Cre recombinase and sgRNAs targeting Rb1, Tp53, and Rbl2 into the lungs of Lox-Stop-Lox Cas9 mice. Coinclusion of a KDM5A sgRNA decreased SCLC tumorigenesis and metastasis, and the SCLCs that formed despite the absence of KDM5A had higher NOTCH activity compared to KDM5A+/+ SCLCs. This work establishes a role for KDM5A in SCLC tumorigenesis and suggests that KDM5 inhibitors should be explored as treatments for SCLC.

Cyclin D-CDK4 relieves cooperative repression of proliferation and cell cycle gene expression by DREAM and RB.

The retinoblastoma protein (RB) restricts cell cycle gene expression and entry into the cell cycle. The RB-related protein p130 forms the DREAM (DP, RB-like, E2F, and MuvB) complex and contributes to repression of cell cycle-dependent genes during quiescence. Although both RB and DREAM bind and repress an overlapping set of E2F-dependent gene promoters, it remains unclear whether they cooperate to restrict cell cycle entry. To test the specific contributions of RB and DREAM, we generated RB and p130 knockout cells in primary human fibroblasts. Knockout of both p130 and RB yielded higher levels of cell cycle gene expression in G0 and G1 cells compared to cells with knockout of RB alone, indicating a role for DREAM and RB in repression of cell cycle genes. We observed that RB had a dominant role in E2F-dependent gene repression during mid to late G1 while DREAM activity was more prominent during G0 and early G1. Cyclin D-Cyclin-Dependent Kinase 4 (CDK4)-dependent phosphorylation of p130 occurred during early G1, and led to the release of p130 and MuvB from E2F4 and decreased p130 and MuvB binding to cell cycle promoters. Specific inhibition of CDK4 activity by palbociclib blocked DREAM complex disassembly during cell cycle entry. In addition, sensitivity to CDK4 inhibition was dependent on RB and an intact DREAM complex in both normal cells as well as in palbociclib-sensitive cancer cell lines. Although RB knockout cells were partially resistant to CDK4 inhibition, RB and p130 double knockout cells were significantly more resistant to palbociclib treatment. These results indicate that DREAM cooperates with RB in repressing E2F-dependent gene expression and cell cycle entry and supports a role for DREAM as a therapeutic target in cancer.

Deubiquitinases Maintain Protein Homeostasis and Survival of Cancer Cells upon Glutathione Depletion.

Cells are subjected to oxidative stress during the initiation and progression of tumors, and this imposes selective pressure for cancer cells to adapt mechanisms to tolerate these conditions. Here, we examined the dependency of cancer cells on glutathione (GSH), the most abundant cellular antioxidant. While cancer cell lines displayed a broad range of sensitivities to inhibition of GSH synthesis, the majority were resistant to GSH depletion. To identify cellular pathways required for this resistance, we carried out genetic and pharmacologic screens. Both approaches revealed that inhibition of deubiquitinating enzymes (DUBs) sensitizes cancer cells to GSH depletion. Inhibition of GSH synthesis, in combination with DUB inhibition, led to an accumulation of polyubiquitinated proteins, induction of proteotoxic stress, and cell death. These results indicate that depletion of GSH renders cancer cells dependent on DUB activity to maintain protein homeostasis and cell viability and reveal a potentially exploitable vulnerability for cancer therapy.

Cells Lacking the RB1 Tumor Suppressor Gene Are Hyperdependent on Aurora B Kinase for Survival.

Small cell lung cancer (SCLC) accounts for 15% of lung cancers and is almost always linked to inactivating RB1 and TP53 mutations. SCLC frequently responds, albeit briefly, to chemotherapy. The canonical function of the RB1 gene product RB1 is to repress the E2F transcription factor family. RB1 also plays both E2F-dependent and E2F-independent mitotic roles. We performed a synthetic lethal CRISPR/Cas9 screen in an RB1 -/- SCLC cell line that conditionally expresses RB1 to identify dependencies that are caused by RB1 loss and discovered that RB1 -/- SCLC cell lines are hyperdependent on multiple proteins linked to chromosomal segregation, including Aurora B kinase. Moreover, we show that an Aurora B kinase inhibitor is efficacious in multiple preclinical SCLC models at concentrations that are well tolerated in mice. These results suggest that RB1 loss is a predictive biomarker for sensitivity to Aurora B kinase inhibitors in SCLC and perhaps other RB1 -/- cancers. SIGNIFICANCE: SCLC is rarely associated with actionable protooncogene mutations. We did a CRISPR/Cas9-based screen that showed that RB1 -/- SCLC are hyperdependent on AURKB, likely because both genes control mitotic fidelity, and confirmed that Aurora B kinase inhibitors are efficacious against RB1 -/- SCLC tumors in mice at nontoxic doses.See related commentary by Dick and Li, p. 169.This article is highlighted in the In This Issue feature, p. 151.

Autochthonous tumors driven by Rb1 loss have an ongoing requirement for the RBP2 histone demethylase.

Inactivation of the retinoblastoma gene (RB1) product, pRB, is common in many human cancers. Targeting downstream effectors of pRB that are central to tumorigenesis is a promising strategy to block the growth of tumors harboring loss-of-function RB1 mutations. One such effector is retinoblastoma-binding protein 2 (RBP2, also called JARID1A or KDM5A), which encodes an H3K4 demethylase. Binding of pRB to RBP2 has been linked to the ability of pRB to promote senescence and differentiation. Importantly, genetic ablation of RBP2 is sufficient to phenocopy pRB's ability to induce these cellular changes in cell culture experiments. Moreover, germline Rbp2 deletion significantly impedes tumorigenesis in Rb1+/- mice. The value of RBP2 as a therapeutic target in cancer, however, hinges on whether loss of RBP2 could block the growth of established tumors as opposed to simply delaying their onset. Here we show that conditional, systemic ablation of RBP2 in tumor-bearing Rb1+/- mice is sufficient to slow tumor growth and significantly extend survival without causing obvious toxicity to the host. These findings show that established Rb1-null tumors require RBP2 for growth and further credential RBP2 as a therapeutic target in human cancers driven by RB1 inactivation.