Identification of Transcriptional Regulators of Immune Evasion Across Cancers: An Alternative Immunotherapeutic Strategy for Cholangiocarcinoma

Abstract

Background: Cancer immune evasion is a multifaceted process that synchronizes pro-tumoral immune infiltration, immunosuppressive inflammation, and inhibitory immune checkpoint expression (IC). Current immunotherapies combat this issue by reinstating immunosurveillance of tumors; however, it benefits a limited patient population. Thus, a more effective immunotherapeutic strategy is warranted to cater to specific patient populations. This investigation introduces a novel immunotherapeutic strategy via inhibition of master regulators of immune evasion (MR-IE).

Methods: Samples of the TCGA Pan-Cancer Atlas transcriptomic data were subset and stratified based on IC and estimated immune cell infiltration. Transcriptomic analysis was conducted to unravel pathways associated with the immune evasion process. Transcription factor enrichment and survival analyses were conducted to identify and rank candidate MR-IEs per cancer type.

Results: Inhibition of the top-ranking MR-IE candidate of cholangiocarcinoma (CCA), MYC, modulated the gene and protein expression of PD-L1. Moreover, pro-tumoral inflammatory markers, IFNA21 and CX3CL1, were downregulated, and anti-tumoral cytokines, IL-18 and IL-16, were upregulated. Lastly, MYC inhibition potentiated fourth-generation anti-folate receptor alpha (FRα) CAR-T cell therapy against CCA cells.

Conclusions: Cumulatively, this study highlights the promise of MR-IE inhibition as a novel potent immunotherapeutic strategy for the treatment of CCA and offers a candidate list of MR-IEs per cancer type for further validation.

Keywords: cancer; cholangiocarcinoma; immune checkpoints; immunotherapy; transcriptomics.

Figure 1

Schematic workflow of integrated in silico analysis to identify MR-IE. Blue boxes represent analyses conducted, and yellow boxes represent details of the analyses. This figure was made with Biorender.com.

Figure 2

Representative analysis of low CD4+ and CD8+ T-cell estimated TCGA samples. (a) Unsupervised hierarchical clustering of TCGA Pan-Cancer samples, a subset with low estimates of CD4+ and CD8+ T-cells against co-inhibitory IC gene expression. (b) Multidimensional Scaling Plot of clusters defined at high (red) and low (blue) IC. (c) Volcano Plot of differentially expressed genes defined at p < 0.001. Red dots denote significantly differentially expressed genes (d) Signaling Pathway Impact Analysis of all three analyses. Red dots represent inactivated pathways, and green dots represent activated pathways. (e) Correlation Analysis of enriched transcription factors against ICs. Red dots represent a positive correlation, whereas blue dots represent a negative correlation. Empty spaces denote insignificant correlation.

Figure 3

Top-ranking candidate MR-IE for each TCGA tumor type based on the composite score. The composite score was calculated using the Hazard Raito (the effect of the identified transcription factor’s effect on patients’ overall survival in each tumor type) multiplied by the statistical enrichment score obtained from the Expression2Kinases application. The composite score for the candidate MR-IE of KICH, KIRC, and KIRP is represented by KIPAN. The composite score for the candidate MR-IE for UCS, UVEM, is represented by UVEM. Candidates not represented in this figure can be found in Supplementary Table S2.

Figure 4

Inhibition or knockdown of master regulator results in the downregulation of PD-L1 expression in CCA cell lines, KKU-213, but not RBE. (a) Western Blot of PD-L1 after MYC inhibitor treatment in KKU-213 cells. (b) Densitometry analysis of PD-L1 bands normalized to GAPDH in KKU-213. (c) The mRNA expression of PD-L1 normalized to 18S expression after MYC inhibitor treatment in KKU-213 cells. (d) The mRNA expression of PD-L1 normalized to 18S after MYC siRNA knockdown in KKU-213 cells. (e) Western Blot of PD-L1 protein after MYC inhibitor treatment in RBE cells. (f) Densitometry analysis of PD-L1 bands normalized to GAPDH in RBE. (g) The mRNA expression of PD-L1 normalized to 18S after MYC inhibitor treatment in RBE cells. (h) The mRNA expression of PD-L1 normalized to 18S after MYC siRNA knockdown in RBE cells. The statistical analyses and data represent an analysis of a minimum of three experiments conducted. Statistically significant values were taken at p-value < 0.05 (*), 0.01 (**), else it is considered ‘non significant’ (ns). The original Western blot figures can be found in Supplementary File S1.

Figure 5

Proteomics analysis reveals the inhibition or knockdown of master regulator results in the modulation of immune-related markers in the cancer cell line proteome in the KKU-213 cell line. (a) Principle Component Analysis of the overall expression of identified proteins of KKU213 cell samples, treated either with the MYC inhibitor or Vehicle control. (b) Volcano Plot of differentially expressed proteins considered at Log2Fold change of two and p-value < 0.05. (c) Heatmap of differentially expressed proteins between KKU-213 treated samples and control. (d) Pathway enrichment of differentially expressed proteins from the WikiPathways Database. (e) Relative expression of immune-related proteins after MYC inhibitor treatment compared to the control.

Figure 6

Inhibition of MYC (MR-IE of CCA) results in the potentiation of immune-mediated cell death through CAR-T cell-based therapy in the KKU-213 cell line. (a) Dose-response curves of CCA cell lines KKU-213 and RBE, treated with varying doses of MYC inhibitor for 24, 48, and 72 h. The X-axis represents log concentrations of the MYC inhibitor (10074-G5) at points 0.1, 1, 10, and 100 µM. (b) Fluorescent images (taken at magnification of 10×; Scale bar represents 100 µm) of KKU-213 tagged with mCherry, treated with MYC inhibitor in varying doses and co-cultured with anti-FR-α CAR-T cells. (c) Cell survival plot compared between target cell alone (TA), and co-cultured with anti-FR-α at an effector to the target cell ratio of 2.5:1. The statistical analyses and data represented are an analysis of a minimum of three experiments conducted. Statistically significant values were taken at p-value < 0.05 (*), 0.001 (***).