Related Experiment Video
Updated: Feb 28, 2026
Predictive Immune Modeling of Solid Tumors
Published on: February 25, 2020
ICIsc: A Deep Learning Framework for Predicting Immune Checkpoint Inhibitor Response by Integrating scRNA-Seq and
Zhenyu Jin1, Di Zhang1, Luonan Chen1,2
1Key Laboratory of Systems Health Science of Zhejiang Province, School of Life Science, Hangzhou Institute for Advanced Study, University of Chinese Academy of Sciences, Chinese Academy of Sciences, Hangzhou 310024, China.
Abstract:
Immune checkpoint inhibitors (ICIs) targeting PD-1/PD-L1 and CTLA-4 are widely used in the treatment of several cancers and have significantly improved survival outcomes in responsive patients. However, a substantial proportion of patients fail to benefit from these therapies, underscoring the urgent need for accurate prediction of ICI response. We propose a deep learning framework, ICIsc, to accurately predict ICI response by integrating single-cell RNA sequencing (scRNA-seq) data with protein large language models. Specifically, patient representations are constructed using transcriptomic profiles and immune-related gene set scores as latent embedding features, while drug representations are derived from amino acid sequences of ICI encoded by the Evolutionary Scale Modeling 2 (ESM2). For bulk data, ICIsc employs a bilinear attention module to fuse patient and drug embeddings for response prediction. For scRNA-seq data, ICIsc infers cell-cell interactions using a single-sample network (SSN) approach and applies GATv2 to model immune microenvironment heterogeneity at the single-cell level. Benchmark evaluations and independent validation demonstrate that ICIsc consistently outperforms baseline models and exhibits robust generalization performance. SHAP-based interpretability analysis further identifies key genes (e.g., GAPDH) associated with immunotherapy response and patient prognosis. Overall, ICIsc provides an accurate and interpretable framework for predicting immunotherapy outcomes and elucidating underlying mechanisms.
Insights
A new deep learning framework, ICIsc, accurately predicts patient response to cancer immunotherapy by integrating single-cell RNA sequencing data with protein language models. This approach enhances treatment prediction and identifies key genes for better patient outcomes.
Area of Science:
- Computational biology
- Immunology
- Oncology
Background:
- Immune checkpoint inhibitors (ICIs) like PD-1/PD-L1 and CTLA-4 have improved cancer survival but many patients do not respond.
- Accurate prediction of ICI response is crucial for effective cancer treatment.
Purpose of the Study:
- To develop a deep learning framework (ICIsc) for predicting patient response to ICIs.
- To integrate single-cell RNA sequencing (scRNA-seq) data with protein large language models for enhanced prediction.
Main Methods:
- ICIsc constructs patient representations from transcriptomic profiles and immune gene set scores.
- Drug representations are derived from ICI amino acid sequences using Evolutionary Scale Modeling 2 (ESM2).
- A bilinear attention module fuses patient and drug embeddings for bulk data; for scRNA-seq, a single-sample network (SSN) and GATv2 model immune microenvironment heterogeneity.
Main Results:
- ICIsc significantly outperforms baseline models in predicting ICI response.
- The framework demonstrates robust generalization performance on benchmark evaluations and independent validation.
- SHAP analysis identified key genes, such as GAPDH, linked to immunotherapy response and prognosis.
Conclusions:
- ICIsc offers an accurate and interpretable computational framework for predicting immunotherapy outcomes.
- The model aids in understanding the mechanisms underlying patient response to cancer immunotherapies.
- This approach has the potential to optimize clinical decision-making for cancer patients undergoing immunotherapy.
