Young Soo Song

Interaction between ENPP1 and homologous recombination deficiency defines distinct pan-cancer signatures: A retrospective observational study

Ectonucleotide pyrophosphatase/phosphodiesterase 1 (ENPP1), 1st identified in breast cancer and subsequently in multiple other cancer types, is an innate immune checkpoint regulator that recently emerged as a promising biomarker and therapeutic target. Homologous recombination deficiency (HRD) has gained clinical relevance with therapeutic vulnerability, particularly in breast and ovarian cancers. Despite the increasing significance of ENPP1 and HRD in cancer biology and treatment, their potential relationships have not yet been comprehensively investigated. We analyzed the relationship between ENPP1 expression and HRD score across the Cancer Genome Atlas pan-cancer and individual tumor types using the Pearson and Spearman correlations. To account for heterogeneity, pan-cancer samples were clustered using linear regression into 3 groups based on Bayesian Information Criterion. Differential expression, functional enrichment, and survival analyses were performed for these clusters at both the pan-cancer and representative tumor type levels. Although the pan-cancer relationship between ENPP1 expression and HRD score was heterogeneous, significant correlations were observed in 11 tumor types. Linear regression-based clustering resolved this heterogeneity into 3 functionally and clinically distinct groups: Cluster 1 was characterized by proliferation programs; Cluster 2 by extracellular matrix remodeling, differentiation, and immune response; and Cluster 3, by metabolic reprogramming. Clinically, Cluster 3 was associated with better survival than Clusters 1 and 2 in a pan-cancer analysis ( P  < .0001). At the individual tumor type level, these global cluster features were further modified in tissue-specific contexts, reflecting local microenvironment adaptation. Significant survival differences were observed in patients with adrenocortical carcinoma, chromophobe renal cell carcinoma, low grade glioma, and mesothelioma, further underscoring the tissue-specific modification of global cluster features. Our comprehensive pan-cancer analysis revealed the intrinsic heterogeneity of ENPP1 expression and HRD score, which may arise from complex and dynamic interactions with diverse cancer hallmarks, including proliferation, extracellular matrix remodeling, immune response, and metabolic reprogramming, and can be generalized into 3 clusters with distinct molecular and clinical characteristics. At the individual tumor type level, these global cluster features were further modified to adapt to a tissue-specific microenvironment, manifesting distinct tissue-specific patterns. Collectively, these findings provide a foundation for refining biomarker-driven precision medicine strategies for diverse tumor types.

Cancer-specific functional profiling in microsatellite-unstable (MSI) colon and endometrial cancers using combined differentially expressed genes and biclustering analysis

Microsatellite-unstable (MSI) cancers have distinct genetic and clinical features from microsatellite-stable cancers, but the molecular functional differences between MSI cancers originating from different tissues or organs have not been well studied because the application of usual differentially expressed gene (DEG) analysis is error-prone, producing too many noncancer-specific normally functioning genes. To maximize therapeutic efficacy, biomarkers reflecting cancer-specific differences between MSI cancers of different tissue origins should be identified. To identify functional differences between MSI colon and endometrial cancers, we combined DEG analysis and biclustering instead of DEG analysis alone and refined functionally relevant biclusters reflecting genuine functional differences between the 2 tumors. Specifically, using The Cancer Genome Atlas and genome-tissue expression as data sources, gene ontology (GO) enrichment tests were performed after routinely identifying DEGs between the 2 tumors with the exclusion of DEGs identified in their normal counterparts. Cancer-specific biclusters and associated enriched GO terms were obtained by biclustering with enrichment tests for the preferences for cancer type (either colon or endometrium) and GO enrichment tests for each cancer-specific bicluster, respectively. A novel childness score was developed to select functionally relevant biclusters among cancer-specific biclusters based on the extent to which the enriched GO terms of the biclusters tended to be child terms of the enriched GO terms in DEGs. The selected biclusters were tested using survival analysis to validate their clinical significance. We performed multiple sequential analyses to produce functionally relevant biclusters from the RNA sequencing data of MSI colon and endometrial cancer samples and their normal counterparts. We identified 3066 cancer-specific DEGs. Biclustering analysis revealed 153 biclusters and 41 cancer-specific biclusters were selected using Fisher exact test. A mean childness score over 0.6 was applied as the threshold and yielded 8 functionally relevant biclusters from cancer-specific biclusters. Functional differences appear to include gland cavitation and the TGF-β receptor, G protein, and cytokine pathways. In the survival analysis, 6 of the 8 functionally relevant biclusters were statistically significant. By attenuating noise and applying a synergistic contribution of DEG results, we refined candidate biomarkers to complement tissue-specific features of MSI tumors.

Refining of cancer-specific genes in microsatellite-unstable colon and endometrial cancers using modified partial least square discriminant analysis

Despite similarities in microsatellite instability (MSI) between colon and endometrial cancer, there are many clinically important organ-specific features. The molecular differences between these 2 MSI cancers are underexplored because the usual differentially expressed gene analysis yields too many noncancer-specific normally expressed genes. We aimed to identify cancer-specific genes in MSI colorectal adenocarcinoma (CRC) and MSI endometrial carcinoma (ECs) using a modified partial least squares discriminant analysis. We obtained a list of cancer-specific genes in MSI CRC and EC by taking the intersection of the genes obtained from tumor samples and normal samples. Specifically, we obtained publically available 1319 RNA sequencing data consisting of MSI CRCs, MSI ECs, normal colon including the rectum, and normal endometrium from The Cancer Genome Atlas and genome-tissue expression sites. To reduce gene-centric dimensions, we retained only 3924 genes from the original data by performing the usual differentially expressed gene screening for tumor samples using DESeq2. The usual partial least squares discriminant analysis was performed for tumor samples, producing 625 genes, whereas for normal samples, projection vectors with zero covariance were sampled, their weights were square-summed, and genes with sufficiently high values were selected. Gene ontology (GO) term enrichment, protein–protein interaction, and survival analyses were performed for functional and clinical validation. We identified 30 cancer-specific normal-invariant genes, including Zic family members (ZIC1, ZIC4, and ZIC5), DPPA2, PRSS56, ELF5, and FGF18, most of which were cancer-associated genes. Although no statistically significant GO terms were identified in the GO term enrichment analysis, cell differentiation was observed as potentially significant. In the protein–protein interaction analysis, 17 of the 30 genes had at least one connection, and when first-degree neighbors were added to the network, many cancer-related pathways, including MAPK, Ras, and PI3K-Akt, were enriched. In the survival analysis, 16 genes showed statistically significant differences between the lower and higher expression groups (3 in CRCs and 15 ECs). We developed a novel approach for selecting cancer-specific normal-invariant genes from relevant gene expression data. Although we believe that tissue-specific reactivation of embryonic genes might explain the cancer-specific differences of MSI CRC and EC, further studies are needed for validation.