PredictIO Sequencing Data

# Data for our paper titled "Leveraging Big Data of Immune Checkpoint Blockade Response Identifies Novel Potential Targets".

------------------------------------------------
------------------------------------------------

*Yacine Bareche, Deirdre Kelly, Farnoosh Abbas-Aghababazadeh, Minoru Nakano, Parinaz Nasr Esfahani, Denis Tkachuk, Hassan Mohammad, Robert Samstein, Chung-Han Lee, Luc G. T. Morris, Philippe L. Bedard, Benjamin Haibe-Kains & John Stagg*

## Summary
Several genomics and gene expression signatures have been proposed as predictive biomarkers of clinical response to immune checkpoint blockade (ICB), with questionable reproducibility. We here report a large-scale comparative analysis of candidate biomarkers of ICB responses in a pan-cancer meta-analysis of over 3,500 ICB-treated patients representing 12 different tumor types. A web-application (predictIO.ca) was developed to allow researchers to further interrogate this data compendium. We tested the hypothesis that a de novo pan-cancer gene expression analysis would bring forth critical ICB resistance pathways and novel therapeutic targets. At the genomic level, we confirmed that non-synonymous tumour mutational burden (nsTMB) was significantly associated with ICB responses across tumor types, with the exception of kidney cancer. At the transcriptional level, 21 out of 39 published gene expression signatures were significantly associated with pan-cancer ICB responses. Strikingly, the predictive value of a de novo gene expression signature (referred to as PredictIO_100) composed of the top 100 genes most significantly associated with ICB responses at pan-cancer level was superior to nsTMB and other gene expression biomarkers. Within PredictIO_100, two genes, F2RL1 (encoding protease-activated receptor-2) and RBFOX2 (encoding RNA binding motif protein 9), were concomitantly associated with worse ICB clinical outcomes, T cell dysfunction in ICB-naive patients and resistance to dual PD-1/CTLA-4 blockade in preclinical mouse cancer models. Taken together, our study underlined the relative impact of candidate biomarkers of ICB responses in a large pan-cancer cohort, demonstrated the potential of de novo pan-cancer gene expression signatures and identified F2RL1, previously involved in tumor immune regulation, and RBFOX2, a critical regulator of epithelial-to-mesenchymal transition, as potential therapeutic targets to overcome ICB resistance.

------------------------------------------------
------------------------------------------------

## mouseModel_Zemek
Expression data of the mouse model study from Zemek et al. (https://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GSE117358)

## process_data
Expression and SNV data of the discovery cohort

## validation_cohort
Expression and SNV data of the validation cohort