Abstract Lung cancer encompasses multiple histological entities with substantial molecular heterogeneity that remain incompletely resolved at population scale. Here, we constructed a unified reference landscape of lung cancer by analyzing raw RNA sequencing data from 1,824 tumors spanning adenocarcinoma (n=966), squamous cell carcinoma (n=628), small cell lung cancer (n=150), and unclassified non–small cell lung cancer (n=80). Following batch correction, samples were analyzed using consensus clustering and visualized with PaCMAP to generate a molecular atlas annotated with clinical and biological metadata. Rather than segregating by pathological diagnosis, tumors organized along conserved transcriptional axes defined by tumor-intrinsic biology including proliferative or metabolic programs and immune-infiltrated states. Consensus clustering resolved nine robust molecular clusters, including an adenocarcinoma-associated subgroup, a neuroendocrine-like adenocarcinoma marked by ASCL1 activation, immune-associated regions, and bifurcation of both small cell and squamous carcinomas into biologically distinct states. Spatially restricted expression of selected clinically relevant transcripts nominated state-specific therapeutic hypotheses requiring future functional and clinical validation. Projection of patient tumors and patient-derived xenografts onto the atlas demonstrated preservation of transcriptional identity and enabled quantitative assessment of model fidelity. This integrated framework organizes lung cancer as a structured continuum of transcriptional states and provides a reference resource for biological interpretation and future translational studies.

Additional datasets, to be included in the next update of LCMAGDB

PCA of reprocessed Mouse430_2 microarray data

Abstract Small cell lung cancer (SCLC) is a highly lethal subtype of lung cancer with a 5-year relative survival rate of less than 10%, even with the addition of immune checkpoint blockade to standard of care therapy. The tumor immune microenvironment of SCLC has been characterized as highly immunosuppressive, and SCLC suppresses the expression of antigen presentation machinery, likely contributing to the lack of effective prolonged immunotherapy responses. Our project investigates the targeting of sialic acid, a sugar molecule overexpressed in many cancers, to increase anti-tumoral immune responses. Studies in other malignancies have shown that cancer cells can hijack this axis as a mechanism of immune evasion through interactions with the SIGLEC family of inhibitory receptors on infiltrating immune cells. We hypothesize that SCLC also utilizes sialic acids for immune masking, and that tumor desialylation would improve anti-tumor responses and could be a potential novel therapeutic approach for SCLC. Using SCLC lines that we derived from genetically engineered mouse models harboring Rb1/Trp53 inactivation, we performed genome wide CRISPR deletion screens and identified Gne as a top gene regulating SCLC sialylation. We then deleted Gne and confirmed robust decreases in sialylation. Under interferon-y stimulated conditions, Gne-deleted cells exhibited elevated MHC-I expression and IFNy signaling pathway, suggesting that loss of sialylation leads to increased antigen presentation and intrinsic immunogenicity. These results are relevant as recent clinical data revealed SCLC patients with low MHC-I expression respond poorer to anti-PDL1 therapy. When the Gne-deleted cells were propagated into mice as flank tumors or as a disseminated metastatic model, we observed delayed tumor kinetics and prolonged survival in syngeneic immunocompetent hosts, but not in immunocompromised recipients, suggesting that the observed effects are immune dependent. Immunophenotyping of syngeneic tumors by flow cytometry revealed increased tumor MHC-I expression and higher infiltration of tumor antigen specific CD8+ T cells upon Gne deletion. Furthermore, co-culturing of Gne-deleted cells expressing ovalbumin with ovalbumin antigen (OT-I) specific CD8+ T cells showed increased tumor cell killing and T cell activation. Lastly, analysis of patient SCLC samples in the IMpower133 clinical trial revealed a survival benefit with chemo-immunotherapy for patients expressing lower transcriptional levels of key sialic acid biosynthesis genes, while this effect is absent in the chemotherapy only group. Altogether, our data suggests that desialylation improves immunogenicity and anti-tumor immunity in SCLC. Ultimately, our work improves the understanding of the mechanisms behind immune responses in SCLC and has potentially uncovered a novel glyco-immunotherapeutic approach to develop new treatments for SCLC in the clinical setting. Citation Format: Alex D. Doan, Kelly Heard, Jackson Fatherree, Mallika Yalangi, Pritha Chanana, Mitchell Kluesner, Daniel S. Hippe, Cody Jenkins, Hannah Kerbyson, Shivani Srivastava, David MacPherson. Targeting sialylation promotes anti-tumor immunity in small cell lung cancer [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2026; Part 1 (Regular Abstracts); 2026 Apr 17-22; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2026;86(7 Suppl):Abstract nr 7016.

Current collection of LCAMGDB datasets

Circulating tumor DNA (ctDNA) profiling from liquid biopsies is increasingly adopted as a minimally invasive solution for clinical cancer diagnostic applications. Current methods for inferring gene expression from ctDNA require specialized assays or ultra-deep, targeted sequencing, which preclude transcriptome-wide profiling at single-gene resolution. Herein we jointly introduce Triton, a tool for comprehensive fragmentomic and nucleosome profiling of cell-free DNA (cfDNA), and Proteus, a multi-modal deep learning framework for predicting single gene expression, using standard depth (~30-120x) whole genome sequencing of cfDNA. By synthesizing fragmentation and inferred nucleosome positioning patterns in the promoter and gene body from Triton, Proteus reproduced expression profiles using pure ctDNA from patient-derived xenografts (PDX) with an accuracy similar to RNA-Seq technical replicates. Applying Proteus to cfDNA from four patient cohorts with matched tumor RNA-Seq, we show that the model accurately predicted the expression of specific prognostic and phenotype markers and therapeutic targets. As an analog to RNA-Seq, we further confirmed the immediate applicability of Proteus to existing tools through accurate prediction of gene pathway enrichment scores. Our results demonstrate the potential clinical utility of Triton and Proteus as non-invasive tools for precision oncology applications such as cancer monitoring and therapeutic guidance.

Decreased availability of the amino acid aspartate constrains cell function across diverse biological contexts, but the temporal interplay between aspartate abundance, downstream metabolic changes and functional effects remains poorly understood. Here we show that succinate dehydrogenase (SDH) inhibition suppresses pyrimidine synthesis via dual effects of cellular aspartate depletion and succinate accumulation. Using an aspartate biosensor and live-cell imaging, we monitor aspartate levels and cell proliferation across several models of aspartate limitation. While complex I inhibition or knockout of aspartate biosynthetic enzymes lead to a strict decrease in aspartate levels and impair proliferation, SDH inhibition produces a unique aspartate rebound, yet fails to restore proliferation. Mechanistically, we find that SDH loss impairs pyrimidine biosynthesis via succinate accumulation, which competitively inhibits aspartate utilization by mammalian aspartate transcarbamylase (ATCase), a key step in pyrimidine biosynthesis. This metabolic interaction occurs in multiple models of SDH deficiency, causing pyrimidine insufficiency, replication stress and sensitivity to ATR kinase inhibition. Taken together, these findings define an unexpected role for succinate in modulating cellular nucleotide homeostasis and demonstrate how cascading metabolic interactions can unfold to impact cell function.

Ascl1 expression across reprocessed RNA-seq datasets

Dotplot of Ezh2 expression across various samples profiled by the Mouse430_2 microarray platform

User interface of the GEMMs table in the LCAMGDB web application

Abstract Small-cell lung cancer (SCLC) is a devastating neuroendocrine carcinoma in critical need of new therapeutic approaches. While a small subset of patients responds well to standard of care chemo-immunotherapy, durable responses are rare. Recent studies have revealed substantial transcriptional and functional heterogeneity, including the identification of multiple low-neuroendocrine SCLC subtypes, which we and others have linked to increased inflammation and response to immunotherapy. A major mediator of neuroendocrine state in SCLC is the transcription factor REST, which is a repressor of neuronal/neuroendocrine genes. While REST is typically silenced in SCLC, it is active in inflamed, low-neuroendocrine SCLC. However, links between REST and immune phenotypes in SCLC have been understudied. To explore the interplay between REST/neuroendocrine status, immune infiltration and response to clinically relevant therapies, we first generated an autochthonous mouse model of SCLC in the Rb/p53-null (RP) background with conditional REST expression (RP-REST). Transcriptionally, REST expression promoted a low-neuroendocrine SCLC phenotype with decreased expression of ASCL1 target genes compared to RP controls. In contrast, RP-REST tumors displayed a striking enrichment of antigen presentation and inflammatory response gene signatures, which correlated with increased infiltration of CD8+ T cells and F4/80+ macrophages, confirming that REST expression generates inflamed, low-neuroendocrine SCLC. We then overexpressed REST in a panel of murine SCLC cell lines derived from the RP mouse model and confirmed REST-driven repression of a neuroendocrine gene signature. Moreover, REST overexpression potentiated responses to interferon-γ resulting in increased MHC-I antigen presentation and phospho-STAT1. Using murine cell lines to generate syngeneic allografts in immunocompetent mice, we found that REST overexpression sensitized tumors to PD1 checkpoint blockade in vivo. Multiparameter flow cytometry reveals that these responses were associated with increased infiltration of activated CD8+ T cells, in addition to an expansion of M1 macrophages. Further, scRNAseq of the tumor compartment identifies a subset of cells in REST-expressing tumors with exceptionally high levels of inflammatory signaling, including secreted cytokines that may serve to recruit the abundant immune cells found in these tumors. Tumor heterogeneity is a critical variable in determining patient outcomes clinically. Here, we show that REST drives a low-neuroendocrine, inflamed SCLC phenotype in an autochthonous mouse model, which is corroborated by isogenic models showing reduced neuroendocrine markers, increased antigen presentation and sensitivity to immunotherapy. Citation Format: Jackson P. Fatherree, Kelly Heard, Alex Doan, Joseph Hiatt, Daniel S. Hippe, Feinan Wu, Shivani Srivastava, David MacPherson, . REST represses neuroendocrine transcriptional programs and enables anti-tumor immunity in SCLC [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2026; Part 1 (Regular Abstracts); 2026 Apr 17-22; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2026;86(7 Suppl):Abstract nr 1333.

Identification of data redundancy by hierarchical clustering

Sample table, providing sample characteristics

Depositor communications, describing input from data depositors

Small cell lung cancer (SCLC) responds exceptionally well to cytotoxic chemotherapy. However, relapse with the emergence of chemoresistant disease is rapid and accompanied by poor treatment outcomes. To understand the genetic basis of chemoresistance in SCLC, we apply in vivo CRISPR deletion screening to patient-derived xenograft (PDX) models. Top screen hits include genes encoding components of the transcriptional co-activator SAGA (Spt-Ada-Gcn5 acetyltransferase) complex. We demonstrate that deletion of the SAGA deubiquitylase USP22 confers cisplatin-etoposide resistance in two chemosensitive PDX models, and that restoring expression in a PDX model harboring homozygous truncating mutation of USP22 re-sensitizes tumors to chemotherapy. USP22 loss increases gene body histone H2AK119 monoubiquitylation at key regulators of neuronal differentiation and suppresses neural and neuroendocrine gene expression including targets of ASCL1. Chemoresistance following USP22 loss reflects attenuated DNA damage-driven phosphorylation events and apoptosis, in conjunction with increased expression of glycolysis and hypoxia-related genes. Glycolysis program upregulation may reflect a targetable vulnerability, as inhibition of GLUT1 re-sensitizes USP22-null tumors to chemotherapy.

Custom sample filtering and selection

Top 100 frequent genetic mutations in realworld lung cancer patients

Lung cancer encompasses multiple histological entities with substantial molecular heterogeneity that remain incompletely resolved at population scale. Here, we constructed a unified reference landscape of lung cancer by analyzing raw RNA sequencing data from 1,558 tumors spanning adenocarcinoma (n=753), squamous cell carcinoma (n=540), small cell lung cancer (n=150), and unclassified non-small cell lung cancer (n=80). Following batch correction, samples were embedded using PaCMAP to generate a continuous molecular atlas annotated with clinical and biological metadata. Rather than segregating strictly by histology, tumors organized along conserved transcriptional axes defined by tumor-intrinsic proliferative or metabolic programs and immune-infiltrated states. Consensus clustering resolved nine robust molecular clusters, including a female non-smoker-enriched adenocarcinoma subgroup, a neuroendocrine-like adenocarcinoma marked by ASCL1 activation, immune-associated regions, and bifurcation of both small cell and squamous carcinomas into biologically distinct states. Spatially-restricted expression of clinically actionable targets revealed state-specific vulnerabilities. Projection of patient tumors and patient-derived xenografts onto the atlas demonstrated preservation of transcriptional identity and enabled quantitative assessment of model fidelity. This unified framework redefines lung cancer as a structured continuum of transcriptional states with translational relevance.

Dataset table, summarizing input datasets for LCMAGDB