Large-scale public data reuse to model immunotherapy response and resistance

Area of Science:

• Oncology
• Immunology
• Bioinformatics

Background:

• Evaluating immune checkpoint blockade (ICB) response and immune evasion is complex, despite numerous trials with omics data.
• Existing data integration methods struggle to comprehensively assess ICB efficacy and resistance mechanisms.

Purpose of the Study:

• To develop a unified web platform, TIDE, integrating large-scale omics data and biomarkers from ICB trials and other cancer datasets.
• To facilitate robust evaluation of ICB response, immune evasion, and identify predictive biomarkers.

Main Methods:

• Integrated omics data from over 33,000 samples across 188 tumor cohorts from public databases.
• Incorporated data from 12 ICB clinical studies (998 tumors) and eight CRISPR screens.
• Developed three interactive analysis modules on the TIDE web platform (http://tide.dfci.harvard.edu).

Main Results:

• Demonstrated the utility of public data reuse for hypothesis generation in cancer immunotherapy.
• Showcased biomarker optimization for predicting ICB response.
• Enabled effective patient stratification based on integrated omics and clinical data.

Conclusions:

• The TIDE platform provides a valuable resource for researchers studying cancer immunology and immunotherapy.
• Integrated omics data analysis on TIDE enhances understanding of ICB mechanisms and improves treatment strategies.
• Facilitates data-driven discovery for advancing personalized cancer medicine.