The heterogeneous sensitivity of pediatric brain tumors to different oncolytic viruses is predicted by unique gene expression profiles

Abstract

Despite decades of research, the prognosis of high-grade pediatric brain tumors (PBTs) remains dismal; however, recent cases of favorable clinical responses were documented in clinical trials using oncolytic viruses (OVs). In the current study, we employed four different species of OVs: adenovirus Delta24-RGD, herpes simplex virus rQNestin34.5v1, reovirus R124, and the non-virulent Newcastle disease virus rNDV-F0-GFP against three entities of PBTs (high-grade gliomas, atypical teratoid/rhabdoid tumors, and ependymomas) to determine their in vitro efficacy. These four OVs were screened on 14 patient-derived PBT cell cultures and the degree of oncolysis was assessed using an ATP-based assay. Subsequently, the observed viral efficacies were correlated to whole transcriptome data and Gene Ontology analysis was performed. Although no significant tumor type-specific OV efficacy was observed, the analysis revealed the intrinsic biological processes that associated with OV efficacy. The predictive power of the identified expression profiles was further validated in vitro by screening additional PBTs. In summary, our results demonstrate OV susceptibility of multiple patient-derived PBT entities and the ability to predict in vitro responses to OVs using unique expression profiles. Such profiles may hold promise for future OV preselection with effective oncolytic potency in a specific tumor, therewith potentially improving OV responses.

Keywords: Gene Ontology; MT: Regular Issue; in vitro; oncolytic viruses; pediatric brain tumors; resistance; response prediction; sensitivity; spheres.

Graphical abstract

Figure 1

Delta24-RGD, rQNestin34.5v1, R124, and rNDV-F0-GFP entry-related genes in HGG, AT/RT, and EPN Every point represents a unique patient-derived cell culture (n = 14) with every sign being a cell culture of specific tumor entity (n = 3). Each boxplot corresponds to the median expression of one gene per tumor entity. The y axis represents the transcription levels of the selected genes on a free scale after log2 transformation of the normalized counts per million (CPM), while the x axis represents the tumor entities in which the cell cultures were grouped.

Figure 2

Cell viability assays after OV infection (A) Patient-derived cell cultures were incubated with various dilutions multiplicities of infection (M.O.I.) of Delta24-RGD, rQNestin34.5v1, R124, and rNDV-F0-GFP and cell viability was measured 5 days post OV infection with CTG assay. The data are shown as the percentage of cells alive after infection with Delta24-RGD, rQNestin34.5v1, R124, or rNDV-F0-GFP at the indicated log10 (M.O.I.s) relative to the non-infected cells, with each point corresponding to the mean ± (SD) values of independent CTG assays per cell line (n = 2 to 4) (see also Table S1). (B) EC₅₀ obtained from (A), the color scaling is independent per OV from green (lowest EC₅₀) to red (highest EC₅₀). (C) Mean EC₅₀ comparisons grouped per tumor entity. Each point in the bar plots corresponds to the EC₅₀ of a specific PBT cell culture. For the comparison of the EC50 per tumor entity the non-parametric Kruskal-Wallis test with Dunn’s multiple comparison correction. ns = not statistically significant.

Figure 3

Correlating genes of OV sensitivity and resistance Volcano plots of Spearman’s correlation coefficient (ρ) values demonstrating significantly correlated genes for sensitivity and resistance for (A) Delta24-RGD, (B) rQNestin34.5v1, (C) R124, and (D) rNDV-F0-GFP, resulting from correlating gene expression and EC₅₀ values (see also Table S2). The x axis represents the Spearman’s correlation coefficient (ρ), and the y axis represents the −log10 (p value). Each dot represents a gene, with black the non-significant (absolute coefficient <0.5 and p value <0.05), green the sensitivity-related genes (coefficient < −0.5 and p value <0.05), while red the resistance-related genes (coefficient >0.5 and p value <0.05). Labeled are the genes related to viral entry using their gene symbol. Volcano plots were created with R (https://www.r-project.org/). Spearman correlation was employed.

Figure 4

Top 10 most enriched biological processes terms of Gene Ontology The bar plots demonstrate the top 10 Gene Ontology biological processes terms based on the numbers of genes enriched after GO enrichment analysis for Delta24-RGD (A), rQNestin34.5v1 (B), R124 (C), and rNDV-F0-GFP (D). The bar plots are color-coded with green being related to sensitivity and red related to resistance. Significant GO terms (p value <0.05) are depicted (see also Table S3). The terms with the lowest p-adjusted value are demonstrated with a darker shade of red or green while terms with higher p value are depicted with lighter shades of red or green, respectively. Bar plots were created with R (https://www.r-project.org/). The p-adjusted value was calculated through the FDR method.

Figure 5

In vitro validation of OV sensitivity signatures Each heatmap represents the expression of genes that significantly correlated with Delta24-RGD (A), rQNestin34.5v1 (C), R124 (E), and rNDV-F0-GFP (G) resistance and sensitivity. Each column represents a PBT culture cluster based on the average expression of the genes. Each line represents the Z score expression levels of all correlated genes for resistance and sensitivity. Cell viability graphs of four confirmation PBT cultures in gray and four newly tested validation PBT cultures in black after 5-day incubation with Delta24-RGD (B), rQNestin34.5v1 (D), R124 (F), and rNDV-F0-GFP (H). Cell viability was measured for the validation of the predictive profiles as described in Figure 2A. Each point represents the mean ± (SD) (n = 3). Heatmaps were plotted in R (https://www.r-project.org/).