Abstract Programmed cell death (PCD) encompasses regulated processes whose dysregulation shapes cancer fitness, yet computational studies use known PCD genes for prognosis rather than discovering regulators. We developed xNNPCD, an interpretable neural-network framework that links CRISPR–Cas9 perturbation signatures from CMap to gene dependency profiles from DepMap. The model constrains hidden neurons to five PCD pathways and iteratively refines a prior gene–PCD pathway mask matrix derived from GO, KEGG, and Reactome using pathway–neuron ablation. This converts binary gene–pathway relationships into continuous-valued associations and improves dependency prediction over random forests, fully connected multi-layer perceptrons, and its own non-iterative variant. The learned mask matrix recovers annotated death regulators and nominates candidate regulators, including RPL23A, HSPA5, SNRPA1, SLC6A2, and ASAH1; combined with dependency scores, it separates pathway coupling from regulatory direction. Transferring the refined mask matrix and learned weights to compound-induced perturbation data enables in silico drug screening, identifying BRD-K19103580 and decitabine as therapeutic agents for apoptosis and ferroptosis, respectively. The pathway-resolved drug profiles can facilitate the design of combination therapies targeting complementary PCD pathways to overcome single-pathway resistance. Overall, xNNPCD offers a pan-cancer, interpretable approach that generalizes across perturbation modalities for mapping the regulatory landscape and elucidating the molecular processes of PCD in cancer.
Abstract High-grade serous ovarian carcinoma (HGSOC) recurs in most patients despite initial responses to chemotherapy and poly(ADP-ribose) polymerase inhibition. A leading explanation for the high rates of recurrence is the presence of minimal residual disease (MRD) comprised of drug-tolerant persister (DTP) cells that survive therapy through reversible epigenetic reprogramming. DTPs have various adaptations that enable them to tolerate therapy and eventually develop into resistant clones, including quiescence, metabolic and hypoxic stress responses, epithelial-mesenchymal plasticity, and immune evasion. Here, we discuss evidence from ovarian cancer of epigenetic mechanisms that coordinate DTP phenotypes, including DNA methylation, histone post-translational modifications, enhancer rewiring, and transcription factor gene-regulatory networks. Based on therapeutic vulnerabilities identified from increased resolution of DTPs, we highlight epigenetic therapies that are currently in clinical trials aimed at depleting MRD to delay or prevent relapse in HGSOC. Finally, we outline emerging technologies that have the potential to detect and monitor persister states in patients, including single-cell transcriptomic and epigenomic profiling, long-read sequencing, and spatial mapping of tumor biopsies.
Abstract Deficiencies in DNA repair pathways create context-specific vulnerabilities that may be exploited therapeutically. Recent studies have established EXO1 as a druggable nuclease, but the genetic contexts that confer sensitivity to EXO1 inhibition remain incompletely defined. Here, we assessed whether pharmacological inhibition of EXO1 selectively targets repair-deficient cancer cells across multiple DNA repair backgrounds. Using potent and selective EXO1 inhibitors, we demonstrate impaired viability in cells deficient for Fanconi Anaemia pathway components (FANCG and APITD1/MHF1), BRCA1-A complex factor BRCC36, and the spliceosome factor ZRSR2. EXO1 inhibition in FANCG or ZRSR2 KO cells induced accumulation of micronuclei, γH2AX, and RPA32 phosphorylation at serine 33, consistent with enhanced replication stress leading to cell death, and supporting a role for EXO1 in processing replication-associated DNA structures and maintaining genome stability under conditions of replication stress. Combining EXO1 inhibitor with PARP inhibitors exacerbated cytotoxicity, suggesting partially nonoverlapping repair functions and therapeutic potential for EXO1/PARP inhibitor combination strategies. Collectively, these findings establish that pharmacological inhibition of EXO1 recapitulates the sensitivity to genetic EXO1 loss in certain DNA repair deficient backgrounds, supporting its development as a targeted therapeutic strategy in molecularly stratified cancers.
Abstract N 6-methyladenosine (m6A) regulates nearly every aspect of messenger RNA (mRNA) processing and function, impacting downstream gene expression programs. Changes in m6A have been implicated in many different types of cancer, and inhibition of m6A installation is emerging as a cancer therapeutic strategy. However, chemoresistance remains a significant clinical challenge in the treatment of glioblastoma (GBM). We established GBM cell culture models of acquired temozolomide (TMZ) resistance, analyzed the role of m6A in controlling resistance-associated pathways, and assessed the effects of METTL3 inhibition. We show that m6A stabilizes key genes and pathways promoting TMZ resistance, and that METTL3 inhibition can reverse this and restore TMZ sensitivity. These findings highlight that TMZ resistance can occur independent of a glioma stem cell population and that changes in m6A need not be driven by changes in METTL3 expression. Collectively, our results suggest that genes associated with TMZ resistance are stabilized by m6A methylation even as the majority of the transcriptome remains subject to m6A-mediated mRNA decay. Moreover, these data highlight METTL3 inhibition as a promising therapeutic approach to overcoming TMZ resistance in GBM.
Abstract Current breast cancer biomarkers rely predominantly on protein-coding transcriptomes, leaving the regulatory information encoded by enhancer RNAs (eRNAs) largely unexploited. Here, we profiled eRNA expression across 1073 The Cancer Genome Atlas Breast Cancer (TCGA-BRCA) samples integrated with The Cancer eRNA Atlas annotations, using GTEx healthy breast tissue (n = 179) as a curated normal reference. For prognostic stratification, least absolute shrinkage and selection operator (LASSO)-Cox regression identified a 10-eRNA signature that stratified patients into high- and low-risk groups in both training (P <.0001) and independent testing cohorts (P = 0.0022), with a 3-year time-dependent area under the curve (AUC) of 0.743. Multivariate Cox analysis confirmed the signature as an independent predictor [hazard ratio = 3.176, 95% confidence interval (CI): 2.075–4.864, P <.001]. Multi-omics network integration linked these prognostic eRNAs to regulatory hubs centered on the ESR1/FOXA1/GATA3/AR axis. For early-stage detection, we developed a 19-eRNA panel via bootstrap LASSO feature selection and benchmarked ten classifiers; a Multilayer Perceptron achieved the best generalization (AUC = 0.959, 95% CI: 0.925–0.986) on an independent cohort (GSE225846). Both modules were deployed in eRNACare, a publicly accessible web application for individualized survival and tumor probability scoring. These findings establish eRNAs as a complementary noncoding layer to conventional messenger RNA classifiers, with clinical potential for breast cancer risk stratification, therapeutic guidance, and early detection.
