bioRxiv Science⌕ Search
Home/Archive/bioRxiv

bioRxiv · 10.1101/2025.07.31.667894

Epigenetic profile drives accurate survival prediction in breast cancer via a multi-omics machine learning model

Mestrallet, G.·Pierga, A.·Fremont, P.
View original source ↗Export metadata ↓

Abstract

Accurate overall survival (OS) prediction is key for personalized treatment in breast cancer, but mutation burden alone is insufficient. To improve prognostic accuracy, we integrated genomic, transcriptomic, proteomic, epigenetic, and clinical features from 802 breast cancer patients to develop BANDOL (Breast cancer Analysis with Neoplastic Data and Omics Learning), a Random Survival Forest model. BANDOL correctly predicted survival ranking in 73% of patient pairs and outperformed mutation-based models (time-dependent AUC: 0.9-1 vs. 0.4-0.9). Immune activation signatures correlated with a longer OS after therapy, while a shorter OS was linked to TREM2 myeloid cells, B cells, and leptin signaling. The transferability of the model was further explored in three independent TCGA cohorts from distinct cancer types (uterine, ovarian and lower-grade glioma), where moderate predictive performance was maintained despite biological differences between tumors. Epigenetic features were the strongest OS predictors for BANDOL. Current therapies may be combined with strategies to target the methylations of ME3, PPARG, OLIG3 and SLC25A22 genes. This study demonstrates that multi-omics integration via machine learning enhances survival prediction and reveals actionable biomarkers. O_FIG O_LINKSMALLFIG WIDTH=200 HEIGHT=180 SRC="FIGDIR/small/667894v6_ufig1.gif" ALT="Figure 1"> View larger version (61K): org.highwire.dtl.DTLVardef@1f3efc6org.highwire.dtl.DTLVardef@2cff0org.highwire.dtl.DTLVardef@a1c3baorg.highwire.dtl.DTLVardef@392114_HPS_FORMAT_FIGEXP M_FIG C_FIG

Source connections

Explore related subjects

cancer biologyPreprint

Keep this discovery

Explore connections, maps & timelines

BibTeXRIS

Mestrallet, G., Pierga, A., Fremont, P.. 2025-08-02. Epigenetic profile drives accurate survival prediction in breast cancer via a multi-omics machine learning model. https://doi.org/10.1101/2025.07.31.667894

Cite the original work for its findings. Save a collection to share your selection of sources.

KEEP EXPLORING

Related preprints

bioRxiv2026

A low-matrix organoid-T cell co-culture platform for functional evaluation of T cell engagers in patient-derived colorectal cancer organoids

Background Colorectal cancer (CRC) remains a leading cause of cancer-related mortality, and emerging T cell engager (TCE) immunotherapies require predictive preclinical models that capture patient-specific tumour biology and tumour-immune interactions. Although patient-derived organoid (PDO)-immune co-culture systems show promise for evaluating immunotherapy responses, many rely on matrix-embedded cultures that limit scalability and reproducibility. We therefore developed and validated a low-matrix organoid-T cell co-culture platform for functional assessment of TCE activity in patient-derived CRC models. Methods Patient-derived CRC organoids representing diverse molecular and genetic backgrounds were co-cultured with activated allogeneic CD3+ T cells in a suspension low-matrix format. Matrix concentration, medium composition, T cell activation status, and assay duration were optimized. Tumour killing, T cells activation, cytokine secretion, apoptosis, and motility were assessed using flow cytometry, live-cell imaging, cytokine profiling, and immunofluorescence. The platform was evaluated using EGFR- and HER2-targeting bispecific T cell engagers across 21 CRC organoid models and multiple healthy donor-derived T cell populations. Results Optimization identified 1% matrix and a 1:1 organoid: T cell medium that preserved organoid integrity while maintaining T cell viability, activation, and motility. The final workflow enabled reproducible co-culture of 5-day mature organoids with 7-day activated T cells for 72 hours. EGFR-targeting TCEs induced tumour killing, T cell activation, and IFN-{gamma} secretion across donor-organoid combinations. Screening of 21 CRC organoid models revealed substantial inter-patient heterogeneity, with 12 models maintaining [≥]50% baseline viability and supporting functional TCE evaluation. EGFR- and HER2-targeting TCEs produced potent dose-dependent cytotoxicity, with IC50 values ranging from 0.0188 to 54.77 nM across responsive models and up to 75%-85% tumour killing in the most sensitive organoids. These responses were accompanied by increased effector cytokine secretion. Real-time imaging confirmed dynamic T cell engagement and apoptosis-driven organoid destruction following TCE treatment. Conclusions We establish a robust, scalable, human-relevant low-matrix organoid-T cell co-culture platform for functional screening of T cell engagers in patient-derived CRC models. By enabling integrated assessment of tumour killing, immune activation, cytokine responses, and tumour-immune interactions while preserving inter-patient heterogeneity, this system provides a translational framework for immunotherapy development. The platform may support candidate prioritisation, biomarker discovery, patient stratification, and preclinical evaluation of immune-engaging therapeutics.

cancer biology↗
bioRxiv2026

MYC-Hyperactivated Osteosarcoma Models Exhibit Resistance to Cabozantinib plus TIGIT Blockade

Background: Relapsed and refractory osteosarcoma (OS) remains a major therapeutic challenge, with fewer than 20% of patients surviving beyond 3 years. Increasing evidence indicates that MYC amplification/overexpression is associated with inferior survival. Small molecule inhibitors and immunotherapies have limited single-agent efficacy in pediatric solid tumors. Using syngeneic cell lines derived from p53-driven and MYC-hyperactivated genetically engineered mouse models (GEMMs) of OS, we tested cabozantinib, a multi-tyrosine kinase inhibitor with immunomodulatory properties, with TIGIT immune checkpoint blockade and investigated mechanisms underlying therapeutic response and resistance. Methods: In vitro cabozantinib sensitivity was established in GEMM-derived cell lines. Mice bearing tibial tumors were randomized to vehicle control, cabozantinib, anti-TIGIT antibody, or combination therapy, and tumor growth and survival assessed after a 3-week treatment period. Temporal RNA sequencing was performed at early (8-15 days) and late (18-24 days) time points to characterize transcriptomic changes associated with efficacy. Results: MYC-hyperactivated cell lines were more resistant to cabozantinib in vitro than p53-driven lines (mean IC50 5.51 vs 0.65 mciroM, p=0.0016). In p53-driven orthotopic models, combination therapy significantly decreased tumor growth and improved survival compared to solvent and cabozantinib alone, while in MYC-hyperactivated models cabozantinib-containing regimens delayed tumor progression relative to control or anti-TIGIT monotherapy, however the addition of anti-TIGIT did not significantly improve survival over cabozantinib alone. Temporal transcriptomics revealed upregulated anti-tumor immune-response pathways and decreased M2 macrophages only with combination treatment in the p53-driven model. In contrast, combination-treated MYC-hyperactivated models demonstrated increased TNF signaling and elevated Cxcl5 and Ccr2 expression, indicative of increased myeloid cell recruitment, and upregulation of extracellular matrix (ECM) remodeling pathways suggest a therapy-induced stress adapted state that propagates treatment resistance over time. Conclusion: New therapies are needed for patients with relapse or refractory OS. By targeting tumor-intrinsic resistance mechanisms and modulating the tumor microenvironment using cabozantinib and anti-TIGIT therapy, improved tumor control and survival was achieved in p53-driven orthotopic OS models. MYC-hyperactivated models were able to overcome therapeutic pressure and employ myeloid recruitment and ECM remodeling programs to achieve treatment resistance. Targeting of these programs should be considered in future studies investigating therapeutic strategies in relapsed and refractory OS.

cancer biology↗
bioRxiv2026

Mitochondrial priming in human germ cell tumors is dependent on MCL1 and BCL2L1

Germ cell tumors (GCTs) are highly sensitized to cell death in response to DNA damaging agents, a property that underlies the success of current chemotherapeutic regimens. To address the molecular basis for this, known as apoptotic priming, we evaluated how different BCL2 family members modulate the heightened sensitivity of GCTs to therapy. Our analysis of human GCTs finds consistently high expression of the pro-survival factors MCL1 and BCL2L1 (BCLX) in a cohort of primary tumors and in their embryonic precursor cells, frequently accompanied by copy number gains of these loci and reciprocal losses of their pro-apoptotic interaction partners and inhibitors, PMAIP1 (NOXA) and BAD. We find that co-inhibition of MCL1 and BCLX using selective BH3 mimetics results in a potent synthetic lethality in multiple GCT embryonal carcinoma cell lines. When these cell lines were cultured with the DNA damaging agents cisplatin or etoposide, inhibition of MCL1 or BCLX potentiated their apoptotic effect in undifferentiated embryonal carcinoma cell lines, but not in retinoic acid-differentiated cells. The inhibition of MCL1 also heightened cisplatin sensitivity in p53-deficient or -mutant cell lines, which is associated with resistance to therapy. Employing an in ovo human xenograft model, we validate that the combination of cisplatin and MCL1 inhibition enhanced the therapeutic response by eliminating tumor cells. Our findings identify MCL1 and BCLX as critical factors to maintain GCT viability and as putative therapeutic targets to further augment GCT responsiveness to DNA damaging agents.

cancer biology↗