The following is the summary of “Leveraging big data of immune checkpoint blockade response identifies novel potential targets” published in the December 2022 issue of Oncology by Bareche, et al.

The introduction of immune checkpoint blockade (ICB) has revolutionized the treatment of several types of cancer. Despite the fact that ICB improves survival rates in the long run for a few different types of cancer, a sizable portion of patients who get treatment for these diseases do not experience any improvement in their condition. The molecular characteristics and processes that regulate response to ICB have been revealed by recent clinical profiling investigations. However, no clinically relevant studies were conducted on any of the observed molecular characteristics.

The purpose of this literature review was to identify studies that were relevant to ICB, including clinical datasets of patients treated with ICB [anti-programmed cell death protein 1 (PD-1)/programmed death-ligand 1 (PD-L1), anti-cytotoxic T-lymphocyte antigen 4 (CTLA-4) or the combination] and available sequencing data. The mutational burden of tumors (TMB) and 37 gene expression signatures (GE) were calculated with reference to the primary literature. Both individual studies and meta-analysis looked into the relationship between biomarkers and ICB response (IR) and survival (progression-free survival/overall survival). Researchers used an open-source online program ( to compare and meta-analyze genomic and transcriptome biomarkers of IRs in over 3,600 patients with 12 different tumor types. Across tumor types, IRs may be predicted by the TMB and 21/37 gene signatures. 

Next, researchers used our pan-cancer analysis to construct a de novo GE signature (PredictIO) and proved its higher predictive power compared to existing biomarkers. We used PredictIO’s T-cell dysfunction score for each gene and its ability to forecast dual PD-1/CTLA-4 inhibition in mice to find new targets. T-cell dysfunction in ICB-naive patients and resistance to simultaneous PD-1/CTLA-4 inhibition in preclinical models were both shown to be related with the presence of 2 genes, F2RL1 (encoding protease-activated receptor-2) and RBFOX2 (encoding RNA-binding motif protein 9). The findings from this study demonstrate the promise of meta-analyses of this magnitude for locating novel biomarkers and potential therapeutic targets for cancer immunotherapy.