A pan-cancer clinical platform to predict immunotherapy outcomes and prioritize immuno-oncology combinations in early-phase trials

A computational pipeline for identifying gene targets and signalling pathways in cancer cells to improve lymphocyte infiltration and immune checkpoint therapy efficacy

“…To be robust, each group should have at least 30 samples to be included for further analyses. For the genes that passed this sample number requirement, we considered the following 5 components to further identify genes in cancer cells that could block TILs: 1) Cytotoxic T cells are the basis for ICB to be effective; thus, we considered the expression level of cancer genes that are negatively correlated with tumour-infiltrating activated CD8 + T cell abundance which could be inferred from tumour mRNA-seq data using an activated CD8 + T cell gene signature 32 ; 2) Other infiltrating immune cells such as NK cells and DCs are also important for anti-tumour immunity and T cell infiltration; thus, we also considered the expression level of cancer genes that are negatively correlated with immune response score, which could be inferred from tumour mRNA-seq data using a previously identified gene signature (i.e., VIGex signature) 13 ; 3) Enhanced T cell infiltration is a strong indicator of patient prognosis; thus, we considered cancer genes whose expression correlates to poor survival; 4) Tumour cell immune-related pathways are important components for tumour cells to exert effects within TME; thus, we considered the association between the expression level and CNV of cancer genes and the strength (i.e., count number) of immune-related pathways to infer potential causality; 5) Single tumour cell sequencing data can reflect the signalling pathways in only cancer cells; thus, we also considered the association between the gene expression level and strength of immune-related pathways in these single cells, to confirm the identified genes are linked to cancer cells but not other cell types.…”

“…To conduct the second component, we calculated the ssGSEA score using the normalized gene expression data and the VIGex signature which represent the immune response score. 13 Genes that were in Geneset 1 and whose expression is negatively correlated with the VIGex ssGSEA score (Spearman coefficient ≤ −0.2, FDR-adj. p ≤ 0.01 [Spearman's correlation coefficient]) were designated as Geneset 2.…”

“… 10 T cell infiltration is associated with the expression and secretion of multiple cytokine factors within the TME, 9 , 11 and different signatures for “hot” tumours such as the TIP signature (hot tumour-related) ( CD274 , PDCD1 , CCL5 , CXCL9 , CXCL10 , CXCL11 , CXCR3 , CD3E , CD4 , CD8A , CD8B , CXCR4 ) and the VIGex signature ( CXCL9 , CXCL10 , CXCL11 , IFNG , PRF1 , IL7R , GZMA , GZMB , PDCD1 , CTLA4 , CD274 , and FOXP3 ) were developed for melanoma and pan-cancer, respectively. 12 , 13 Tumour-associated antigens can be presented on the cell surface by MHC class I molecules and be recognised by T cells, enhancing T cell infiltration and homing; however, tumour cells often escape immune surveillance by downregulating MHC-I mediated antigen presentation. 14 , 15 …”

Transcriptome Deconvolution Reveals Absence of Cancer Cell Expression Signature in Immune Checkpoint Blockade Response

“…We evaluated the model performance on six independent test cohorts treated with ICB ( 28–33 ), comprising 232 patients with melanoma, colorectal, lung, and other cancers (Materials and Methods). We calculated the area under the receiver operating curve (AUC) of the model, and benchmarked it against established biomarkers of ICB response, namely CXCL9+TMB ( 16 ), T-cell GEP ( 11 ), VIGex ( 51 ), cytolytic score ( 50 ), TIDE ( 13, 49 ), TMB alone ( 2, 8 ), and PD-L1 expression ( 3 ). The IOSelect classifier achieved a mean AUC of 0.72, significantly outperforming existing biomarkers with mean AUCs ranging from 0.60 (TMB alone) to 0.68 (T-cell GEP; Fig.…”

“…Vigex score is computed as the mean of z-score scaled expression of 12 genes ( CXCL9 , CXCL10 , CXCL11 , IFNG , PRF1 , IL7R , GZMA , GZMB , PDCD1 , CTLA4 , CD274 , FOXP3 ; ref. 51 ).…”