Biology subjects

Hoang, D.-T.

Publications and source records attributed to Hoang, D.-T..

9 recordsLinked to original sources

bioRxiv2026

Histopathology-inferred spatial transcriptomics characterizes the tumor microenvironment in 1,500 head and neck tumors and predicts clinical outcomes

Head and neck squamous cell carcinoma (HNSC) is a prevalent malignancy associated with poor prognosis despite recent therapeutic advances. We hypothesized that a comprehensive understanding of the spatial heterogeneity and organization of the tumor microenvironment (TME) can substantially improve risk stratification and prediction of treatment response in HNSC. As spatial transcriptomics (ST) remains labor-intensive and costly, we developed HEiST (H&E-Inferred Spatial Transcriptomics), a deep learning framework that predicts spatially resolved gene expression profiles directly from routine hematoxylin and eosin (H&E)-stained histology slides. After rigorous validation across two independent external ST cohorts, we applied HEiST to infer spatial transcriptomes across 1,500 HNSC patient tumors spanning two publicly available datasets and two newly generated cohorts, one treated with concurrent chemoradiotherapy (CCRT) and one with immunotherapy. This large-scale analysis uncovered reproducible spatial clusters characterizing the HNSC TME, defining two distinct prognostic Spatiotypes, Immune-Exhausted and Immune-Activated, with significantly distinct survival outcomes. Critically, spatial cluster composition accurately predicts HPV status and yields treatment response predictors for both CCRT/radiotherapy and immunotherapy that outperform costly gene-expression and direct image-based approaches. Notably, the ST cluster-based predictor of immunotherapy response markedly surpasses the performance of commonly used FDA-approved biomarkers, including CPS, TPS, and their combination. To the best of our knowledge, this represents the first virtual spatial profiling effort and the most comprehensive large-scale spatial TME analysis in HNSC to date. HEiST thus introduces a scalable, low-cost, and spatially grounded biomarker discovery for precision oncology in HNSC.

cancer biology↗

bioRxiv2025

AI-based supervised treatment response prediction from tumor transcriptomics: A large-scale pan-cancer study

Precision oncology aims to guide treatment decisions using biomarkers. While DNA-based panels are increasingly applied, RNA transcriptomics remain underused due to limited datasets and the absence of robust models. We assembled the largest transcriptomic resource for drug response prediction to date, spanning 91 cohorts, 5,675 patients, nine cancer types, and six frontline therapies: anti-PD-1/PD-L1 immune-checkpoint inhibitors, trastuzumab, bevacizumab, BRAF inhibitors, paclitaxel, and FAC/FEC (Fluorouracil-Adriamycin-Cyclophosphamide/Fluorouracil-Epirubicin-Cyclophosphamide) chemotherapy. We developed EXPRESSO (EXpression-Profile-RESponSe-Optimizer), a supervised machine-learning framework that predicts treatment response from pre-treatment transcriptomes by integrating drug targets and context-specific biomarkers. EXPRESSO achieves mean ROC-AUCs of 0.62-0.73 and median odds ratios of 2.4-4.6 across therapies, outperforming 20 published transcriptomic signatures and other machine learning methods. Prospective validation on 22 independent cohorts confirms that performance generalizes beyond cross-validation. The EXPRESSO signature additionally stratifies progression-free survival in immune checkpoint blockade-treated cohorts, demonstrating prognostic value beyond binary response prediction. Robustness analysis reveals that predictive performance plateaued for some therapies with increasing training cohorts but continued to improve for others. These findings suggest inherent limits of supervised brute-force learning for certain treatments, but additional data and deeper mechanistic modeling may further enhance transcriptomics-based predictors. SIGNIFICANCEEXPRESSO forms the next step in studying the feasibility of harnessing bulk transcriptomic data to inform therapeutic decision-making, advancing the role of transcriptomics from exploratory biomarker discovery to actionable predictive modeling.

cancer biology↗

bioRxiv2025

Pathologist-interpretable breast cancer subtyping and stratification from AI-inferred nuclear features

Artificial intelligence (AI) is making notable advances in digital pathology but faces challenges in human interpretability. Here we introduce EXPAND (EXplainable Pathologist Aligned Nuclear Discriminator), the first pathologist-interpretable AI model to predict breast cancer tumor subtypes and patient survival. EXPAND focuses on a core set of 12 nuclear pathologist-interpretable features (NPIFs), composing the Nottingham grading criteria used by the pathologists. It is a fully automated, end-to-end diagnostic workflow, which automatically extracts NPIFs given a patient tumor slide and uses them to predict tumor subtype and survival. EXPANDs performance is comparable to that of existing deep learning non-interpretable black box AI models. It achieves areas under the ROC curve (AUC) values of 0.73, 0.79 and 0.75 for predicting HER2+, HR+ and TNBC tumor subtypes, respectively, matching the performance of proprietary models that rely on substantially larger and more complex interpretable feature sets. The 12 NPIFs demonstrate strong and independent prognostic value for patient survival, underscoring their potential as biologically grounded, interpretable biomarkers for survival stratification in BC. These results lay the basis for building interpretable AI diagnostic models in other cancer indications. The complete end-to-end pipeline is made publicly available via GitHub (https://github.com/ruppinlab/EXPAND) to support community use and reproducibility.

bioinformatics↗

bioRxiv2025

Pan-cancer prediction of tumor immune activation and response to immune checkpoint blockade from tumor transcriptomics and histopathology

Accurately predicting which patients will respond to immune checkpoint blockade (ICB) remains a major challenge. Here, we present TIME_ACT, an unsupervised 66-gene transcriptomic signature of tumor immune activation derived from TCGA (The Cancer Genome Atlas) melanoma data. First, we demonstrate that TIME_ACT scores accurately identify tumors with activated immune microenvironments across different cancer types. Further, analysis of spatial features reveals that tumor microenvironment regions with dense lymphocyte infiltration near tumor cells have high TIME_ACT scores, successfully marking localized immune activation. Second, across 25 transcriptomic ICB cohorts encompassing nine cancer types, TIME_ACT achieves a mean AUC of 0.76 and a mean odds ratio of 5.77, significantly outperforming 30 established transcriptomic signatures and prediction methods for ICB response, including a recently developed foundation model for immunotherapy response prediction. Third, we show that TIME_ACT scores can be accurately inferred from routine tumor histopathology slides and that slide-inferred TIME_ACT scores predict ICB response across nine new independent patient cohorts spanning eight cancer types, achieving a mean AUC of 0.72 and a mean odds ratio of 4.99. These findings establish TIME_ACT as a robust, pan-cancer biomarker that enables accurate, low-cost, and clinically scalable prediction of ICB response from routine histopathology.

cancer biology↗

bioRxiv2025

Path2Omics: Enhanced transcriptomic and methylation prediction accuracy from tumor histopathology

Precision oncology is becoming increasingly integral to clinical practice, demonstrating notable improvements in treatment outcomes. While molecular data provide comprehensive insights, obtaining such data remains costly and time-consuming. To address this challenge, we developed Path2Omics, a deep learning model that predicts gene expression and methylation from histopathology for 23 cancer types. Path2Omics was trained on 20,497 slides (9,456 formalin-fixed and paraffin-embedded (FFPE) and 11,041 fresh frozen (FF)) from 8,007 patients across 23 The Cancer Genome Atlas cohorts. When tested on FFPE slides, the most readily available format in clinical pathology practice, the integrated model outperformed its individual FF and FFPE components, robustly predicting nearly 5,000 genes on average, approximately five times more than our recently published DeepPT model. Externally evaluated on seven independent cohorts, Path2Omics robustly predicted the expression of approximately 4,400 genes, yielding a 30% increase over the FFPE model alone. Finally, we demonstrate that the inferred gene expression is nearly as effective as the actual values in predicting patient survival and treatment response. These results lay the basis for using Path2Omics to advance precision oncology from histopathology slides in a speedy and cost-effective manner. Statement of significancePath2Omics is a deep learning model that accurately predicts gene expression and methylation from histopathology slides across 23 cancer types. Unlike existing approaches that rely solely on FFPE slides for training, Path2Omics leverages both FFPE and FF slides by constructing two separate models and integrating them. Downstream analyses show that the inferred values from Path2Omics are nearly as effective as actual values in predicting patient survival and treatment response.

bioinformatics↗

bioRxiv2024

Path2Space: An AI Approach for Cancer Biomarker Discovery Via Histopathology Inferred Spatial Transcriptomics

Spatial transcriptomics (ST) is transforming our understanding of tumor heterogeneity by enabling high-resolution, location-specific mapping of gene expression across tumors and their microenvironment. However, the associated high cost of the assay has limited cohort size and hence large-scale biomarker discovery. Here we present Path2Space, a deep learning approach that predicts spatial gene expression directly from histopathology slides. Trained on substantial breast cancer ST data, it robustly predicts the spatial expression of over 4,300 genes in independent validations, markedly outperforming existing ST predictors. Path2Space additionally accurately infers cell-type abundances in the tumor microenvironment (TME) based on the inferred ST data. Applied to more than a thousand breast tumor histopathology slides from the TCGA, Path2Space characterizes their TME on an unprecedented scale and identifies three new spatially-grounded breast cancer subgroups with distinct survival rates. Path2Space-inferred TME landscapes enable more accurate predictions of patients response to chemotherapy and trastuzumab directly from H&E slides than those obtained by existing established sequencing-based biomarkers. Path2Space thus offers a transformative, fast and cost-effective approach to robustly delineate the TME directly from their histopathology slides, facilitating the development of spatially-grounded biomarkers to advance precision oncology.

cancer biology↗

bioRxiv2022

Protein Structure Prediction with Expectation Reflection

Sequence covariation in multiple sequence alignments of homologous proteins has been used extensively to obtain insights into protein structure. However, global statistical inference is required in order to ascertain direct relationships between amino acid positions in these sequences that are not simply secondary correlations induced by interactions with a third residue. Methods for statistical inference of such covariation have been developed to exploit the growing availability of sequence data. These hints about the folded protein structure provide critical a priori information for more detailed 3D predictions by neural networks. We present a novel method for protein structure inference using an iterative parameter-free model estimator which uses the formalism of statistical physics. With no tunable learning rate, our method scales to large system sizes while providing improved performance in the regime of small sample sizes. We apply this method to 40974 PDB structures and compare its performance to that of other methods. Our method outperforms existing methods for 76% of analysed proteins.

molecular biology↗

bioRxiv2022

Synthetic lethality-based prediction of cancer treatment response from histopathology images

Advances in artificial intelligence have paved the way for leveraging hematoxylin and eosin (H&E)-stained tumor slides for precision oncology. We present ENLIGHT-DeepPT, an approach for predicting response to multiple targeted and immunotherapies from H&E-slides. In difference from existing approaches that aim to predict treatment response directly from the slides, ENLIGHT-DeepPT is an indirect two-step approach consisting of (1) DeepPT, a new deep-learning framework that predicts genome-wide tumor mRNA expression from slides, and (2) ENLIGHT, which predicts response based on the DeepPT inferred expression values. DeepPT successfully predicts transcriptomics in all 16 TCGA cohorts tested and generalizes well to two independent datasets. Importantly, ENLIGHT-DeepPT successfully predicts true responders in five independent patients cohorts involving four different treatments spanning six cancer types with an overall odds ratio of 2.44, increasing the baseline response rate by 43.47% among predicted responders, without the need for any treatment data for training. Furthermore, its prediction accuracy on these datasets is comparable to a supervised approach predicting the response directly from the images, trained and tested on the same cohort in cross validation. Its future application could provide clinicians with rapid treatment recommendations to an array of different therapies and importantly, may contribute to advancing precision oncology in developing countries. Statement of SignificanceENLIGHT-DeepPT is the first approach shown to successfully predict response to multiple targeted and immune cancer therapies from H&E slides. In distinction from all previous H&E slides prediction approaches, it does not require supervised training on a specific cohort for each drug/indication treatment but is trained to predict expression on the TCGA cohort and then can predict response to an array of treatments without any further training. ENLIGHT-DeepPT can provide rapid treatment recommendations to oncologists and help advance precision oncology in underserved regions and low-income countries.

cancer biology↗

bioRxiv2022

Clinically oriented prediction of patient response to targeted and immunotherapies from the tumor transcriptome

BackgroundPrecision oncology is gradually advancing into mainstream clinical practice, demonstrating significant survival benefits. However, eligibility and response rates remain limited in many cases, calling for better predictive biomarkers. MethodsWe present ENLIGHT, a transcriptomics-based computational approach that identifies clinically relevant genetic interactions and uses them to predict a patients response to a variety of therapies in multiple cancer types, without training on previous treatment response data. We study ENLIGHT in two translationally oriented scenarios: Personalized Oncology (PO), aimed at prioritizing treatments for a single patient, and Clinical Trial Design (CTD), selecting the most likely responders in a patient cohort. FindingsEvaluating ENLIGHTs performance on 21 blinded clinical trial datasets in the PO setting, we show that it can effectively predict a patients treatment response across multiple therapies and cancer types. Its prediction accuracy is better than previously published transcriptomics-based signatures and is comparable to that of supervised predictors developed for specific indications and drugs. In combination with the IFN-{gamma}signature, ENLIGHT achieves an odds ratio larger than 4 in predicting response to immune checkpoint therapy. In the CTD scenario, ENLIGHT can potentially enhance clinical trial success for immunotherapies and other monoclonal antibodies by excluding non-responders, while overall achieving more than 90% of the response rate attainable under an optimal exclusion strategy. ConclusionENLIGHT demonstrably enhances the ability to predict therapeutic response across multiple cancer types from the bulk tumor transcriptome. FundingThis research was supported in part by the Intramural Research Program, NIH and by the Israeli Innovation Authority.

bioinformatics↗