Personalized tumor combination therapy optimization using the single-cell transcriptome.

Area of Science:

• Computational Biology
• Genomics
• Cancer Research

Background:

• Personalized cancer treatment relies on precise characterization of tumors and immune microenvironments.
• Current methods like in vitro assays or bulk transcriptomes overlook tumor heterogeneity and in vivo immune microenvironments.
• Single-cell transcriptomes are crucial for overcoming these limitations in personalized cancer therapy.

Purpose of the Study:

• To present comboSC, a computational proof-of-concept for optimizing personalized cancer combination therapy.
• To stratify patient samples by evaluating their immune microenvironment using single-cell RNA sequencing.
• To identify synergistic drug combinations and immunotherapy pairings for personalized clinical use.

Main Methods:

• comboSC utilizes single-cell RNA sequencing data to quantitatively assess the immune microenvironment.
• It integrates in vitro cellular response data to identify synergistic drug combinations.
• Bipartition graph optimization is employed to prioritize drug combinations for clinical use.

Main Results:

• comboSC was applied to 119 single-cell transcriptome datasets from 15 cancer types.
• Predicted drug combinations were validated using literature, clinical trial data, cell line perturbations, and in vivo samples.
• The study demonstrates the feasibility of comboSC for predicting effective cancer drug combinations.

Conclusions:

• comboSC is a feasible computational prototype for predicting personalized cancer drug combinations.
• It accelerates personalized tumor treatment by reducing screening time and saving clinical time.
• A web server and source code are available for clinical and research users.