Mutational Signatures as Sensors of Environmental Exposures: Role of Smoking in COVID-19 Vulnerabilities

Background: Environmental exposures such as smoking are widely recognized risk factors in the emergence of lung diseases such as lung cancer and acute respiratory distress syndrome (ARDS). However, the strength of environmental exposures is difficult to measure, making it challenging to understand their impacts. On the other hand, some COVID-19 patients develop ARDS in an unfavorable disease progression and smoking has been suggested as a potential risk factor among others. Yet initial studies on COVID-19 cases reported contradictory results on the effects of smoking on the disease. Some suggest that smoking might have a protective effect against it while other studies report an increased risk. A better understanding of how the exposure to smoking and other environmental factors affect biological processes relevant to SARS-CoV-2 infection and unfavorable disease progression is needed. Approach: In this study, we utilize mutational signatures associated with environmental factors as sensors of their exposure level. Many environmental factors including smoking are mutagenic and leave characteristic patterns of mutations, called mutational signatures, in affected genomes. We postulated that analyzing mutational signatures, combined with gene expression, can shed light on the impact of the mutagenic environmental factors to the biological processes. In particular, we utilized mutational signatures from lung adenocarcinoma (LUAD) data set collected in TCGA to investigate the role of environmental factors in COVID-19 vulnerabilities. Integrating mutational signatures with gene expression in normal tissues and using a pathway level analysis, we examined how the exposure to smoking and other mutagenic environmental factors affects the infectivity of the virus and disease progression. Results: By delineating changes associated with smoking in pathway-level gene expression and cell type proportions, our study demonstrates that mutational signatures can be utilized to study the impact of exogenous mutagenic factors on them. Consistent with previous findings, our analysis showed that smoking mutational signature (SBS4) is associated with activation of cytokines mediated singling pathways, leading to inflammatory responses. Smoking related changes in cell composition were also observed, including the correlation of SBS4 with the expansion of goblet cells. On the other hand, increased basal cells and decreased ciliated cells in proportion were associated with the strength of a different mutational signature (SBS5), which is present abundantly but not exclusively in smokers. In addition, we found that smoking increases the expression levels of genes that are upregulated in severe COVID-19 cases. Jointly, these results suggest an unfavorable impact of smoking on the disease progression and also provide novel findings on how smoking impacts biological processes in lung.

Pan-cancer evaluation of gene expression and somatic alteration data for cancer prognosis prediction

Background Over the past decades, approaches for diagnosing and treating cancer have seen significant improvement. However, the variability of patient and tumor characteristics has limited progress on methods for prognosis prediction. The development of high-throughput omics technologies now provides multiple approaches for characterizing tumors. Although a large number of published studies have focused on integration of multi-omics data and use of pathway-level models for cancer prognosis prediction, there still exists a gap of knowledge regarding the prognostic landscape across multi-omics data for multiple cancer types using both gene-level and pathway-level predictors. Methods In this study, we systematically evaluated three often available types of omics data (gene expression, copy number variation and somatic point mutation) covering both DNA-level and RNA-level features. We evaluated the landscape of predictive performance of these three omics modalities for 33 cancer types in the TCGA using a Lasso or Group Lasso-penalized Cox model and either gene or pathway level predictors. Results We constructed the prognostic landscape using three types of omics data for 33 cancer types on both the gene and pathway levels. Based on this landscape, we found that predictive performance is cancer type dependent and we also highlighted the cancer types and omics modalities that support the most accurate prognostic models. In general, models estimated on gene expression data provide the best predictive performance on either gene or pathway level and adding copy number variation or somatic point mutation data to gene expression data does not improve predictive performance, with some exceptional cohorts including low grade glioma and thyroid cancer. In general, pathway-level models have better interpretative performance, higher stability and smaller model size across multiple cancer types and omics data types relative to gene-level models. Conclusions Based on this landscape and comprehensively comparison, models estimated on gene expression data provide the best predictive performance on either gene or pathway level. Pathway-level models have better interpretative performance, higher stability and smaller model size relative to gene-level models.

Characterization of the tumor immune microenvironment of triple-negative breast cancer (TNBC) patients who self-identify as African American (AA) or non-African American (NonAA).

564 Background: What tumor biological differences, if any, contribute to the higher incidence and worse prognosis of triple negative breast cancer (TNBC) in African American (AA) compared to NonAA patients are unknown. We hypothesized that differences in the tumor immune microenvironment may contribute to the outcome disparities. The purpose of this study was to characterize and compare the immune microenvironment of TNBC between patients self-identified as NonAA or AA. Methods: Formalin fixed paraffin embedded surgically resected cancer and paired normal tissues collected before any systemic therapy and the corresponding clinical data were collected for NonAA (n = 56) and AA (n = 54) stage I-III TNBC treated at Yale Cancer Center between 2000-2017. The two cohorts were matched for clinical stage, age of diagnosis, and year of diagnosis. We performed somatic and germline whole exome sequencing (WES), bulk RNA sequencing, and immunohistochemistry to assess PD-L1 expression (SP142). Stromal tumor infiltrating lymphocytes (sTILs) were assessed on H&E slides. Mutation load, mutation frequencies, and gene expression differences were compared at gene and pathway level. Immune cell composition was estimated through gene expression deconvolution analyses (TIDE). Results: Tumor mutational burden was similar between the two cohorts. At gene level, few genes had significantly different somatic mutation frequencies, or differential mRNA expression between AA and NonAA samples. Pathway level alterations showed inflammation, immunity (adaptive; innate), antigen presentation, and allograft rejection pathways were more affected by somatic mutations in AA samples. The affected genes differed from cancer to cancer and were not recurrent and therefore were missed at gene level analysis. Gene set enrichment and co-expression analysis also showed higher immune related pathway expression in AA samples. Unsupervised co-expression cluster analysis confirmed coordinated overexpression of genes involved in immunity, inflammation, and cytokine/chemokine signaling in AA patients. Two immunotherapy response predictive signatures, immune inflamed and the IFNG as well as sTILs score and PD-L1 positivity were also higher in AA samples. These findings raise the possibility that immune checkpoint inhibitors might be particularly effective in AA patients. In NonAA samples, the EMT transition, angiogenesis, adipogenesis, myogenesis, fatty acid metabolism, TGFβ signaling, UV-response, and hypoxia pathways were overexpressed. TIDE analysis suggested higher levels of TAM M2, overall TIDE score, and the Immune Exclusion score in NonAA samples. Conclusions: TNBC in AA patients more frequently harbor somatic mutations in genes involved with immune functions and overexpress immune and inflammatory genes compared to NonAA patients.

Set-Based Rare Variant Expression Quantitative Trait Loci in Blood and Brain from Alzheimer Disease Study Participants

Because studies of rare variant effects on gene expression have limited power, we investigated set-based methods to identify rare expression quantitative trait loci (eQTL) related to Alzheimer disease (AD). Gene-level and pathway-level cis rare-eQTL mapping was performed genome-wide using gene expression data derived from blood donated by 713 Alzheimer’s Disease Neuroimaging Initiative participants and from brain tissues donated by 475 Religious Orders Study/Memory and Aging Project participants. The association of gene or pathway expression with a set of all cis potentially regulatory low-frequency and rare variants within 1 Mb of genes was evaluated using SKAT-O. A total of 65 genes expressed in the brain were significant targets for rare expression single nucleotide polymorphisms (eSNPs) among which 17% (11/65) included established AD genes HLA-DRB1 and HLA-DRB5. In the blood, 307 genes were significant targets for rare eSNPs. In the blood and the brain, GNMT, LDHC, RBPMS2, DUS2, and HP were targets for significant eSNPs. Pathway enrichment analysis revealed significant pathways in the brain (n = 9) and blood (n = 16). Pathways for apoptosis signaling, cholecystokinin receptor (CCKR) signaling, and inflammation mediated by chemokine and cytokine signaling were common to both tissues. Significant rare eQTLs in inflammation pathways included five genes in the blood (ALOX5AP, CXCR2, FPR2, GRB2, IFNAR1) that were previously linked to AD. This study identified several significant gene- and pathway-level rare eQTLs, which further confirmed the importance of the immune system and inflammation in AD and highlighted the advantages of using a set-based eQTL approach for evaluating the effect of low-frequency and rare variants on gene expression.

Concurrent Mutations Associated With Trastuzumab-resistance Revealed by Single Cell Sequencing

Purpose: HER2-positive breast cancer patients benefit from HER2 targeted therapies, among which the most commonly used is trastuzumab. However, acquired resistance typically happens within one year. The cellular heterogeneity of it is less clear. Methods: Here we generated trastuzumab-resistant cells in two HER2-positive breast cancer cell lines, SK-BR-3 and BT-474. Cells at different time points during the resistance induction were examined by exome sequencing to study changes of genomic alterations over time. Single cell targeted sequencing was also used to identify resistance associated concurrent mutations.Results: We found a rapid increase of copy number variation (CNV) regions and gradual accumulation of single nucleotide variations (SNVs). On the pathway level, nonsynonymous SNVs for SK-BR-3 cells were enriched in the MAPK signaling pathway, while for BT-474 cells were enriched in mTOR and PI3K-Akt signaling pathways. However, all of the three signaling pathways were in the downstream of the HER2 kinase. Putative trastuzumab-resistance associated SNVs included AIFM1 P548L and ERBB2 M833R in SK-BR-3 cells, and OR5M9 D230N, COL9A1 R627T, ITGA7 H911Q and ADAMTS19 V451L in BT-474 cells. Single cell targeted sequencing identified several concurrent mutations. By validation, we found that concurrent mutations (AIFM1 P548L and IL1RAPL2 S546C) led to a decrease of trastuzumab sensitivity. Conclusion: Taken together, our study revealed a common pathway level trastuzumab-resistance mechanism for HER2-positive breast cancer cells. In addition, our identification of concurrent SNVs associated with trastuzumab-resistance may be indicative of potential targets for the treatment of trastuzumab-resistant breast cancer patients.

Multi-Level Attention Graph Neural Network for Clinically Interpretable Pathway-Level Biomarkers Discovery

Precision medicine, regarded as the future of healthcare, is gaining increasing attention these years. As an essential part of precision medicine, clinical omics have been successfully applied in disease diagnosis and prognosis using machine learning techniques. However, existing methods mainly make predictions based on gene-level individual features or their random combinations, none of the previous work has considered the activation of signaling pathways. Therefore, the model interpretability and accuracy are limited, and reasonable signaling pathways are yet to be discovered. In this paper, we propose a novel multi-level attention graph neural network (MLA-GNN), which applies weighted correlation network analysis (WGCNA) to format the omic data of each patient into graph-structured data, and then constructs multi-level graph features, and fuses them through a well-designed multi-level graph feature fully fusion (MGFFF) module to conduct multi-task prediction. Moreover, a novel full-gradient graph saliency mechanism is developed to make the MLA-GNN interpretable. MLA-GNN achieves state-of-the-art performance on transcriptomic data from TCGA-LGG/TCGA-GBM and proteomic data from COVID-19/non-COVID-19 patient sera. More importantly, the proposed model's decision can be interpreted in the signaling pathway level and is consistent with the clinical understanding.

Cancer prognosis prediction using somatic point mutation and copy number variation data: a comparison of gene-level and pathway-based models

Background Genomic profiling of solid human tumors by projects such as The Cancer Genome Atlas (TCGA) has provided important information regarding the somatic alterations that drive cancer progression and patient survival. Although researchers have successfully leveraged TCGA data to build prognostic models, most efforts have focused on specific cancer types and a targeted set of gene-level predictors. Less is known about the prognostic ability of pathway-level variables in a pan-cancer setting. To address these limitations, we systematically evaluated and compared the prognostic ability of somatic point mutation (SPM) and copy number variation (CNV) data, gene-level and pathway-level models for a diverse set of TCGA cancer types and predictive modeling approaches. Results We evaluated gene-level and pathway-level penalized Cox proportional hazards models using SPM and CNV data for 29 different TCGA cohorts. We measured predictive accuracy as the concordance index for predicting survival outcomes. Our comprehensive analysis suggests that the use of pathway-level predictors did not offer superior predictive power relative to gene-level models for all cancer types but had the advantages of robustness and parsimony. We identified a set of cohorts for which somatic alterations could not predict prognosis, and a unique cohort LGG, for which SPM data was more predictive than CNV data and the predictive accuracy is good for all model types. We found that the pathway-level predictors provide superior interpretative value and that there is often a serious collinearity issue for the gene-level models while pathway-level models avoided this issue. Conclusion Our comprehensive analysis suggests that when using somatic alterations data for cancer prognosis prediction, pathway-level models are more interpretable, stable and parsimonious compared to gene-level models. Pathway-level models also avoid the issue of collinearity, which can be serious for gene-level somatic alterations. The prognostic power of somatic alterations is highly variable across different cancer types and we have identified a set of cohorts for which somatic alterations could not predict prognosis. In general, CNV data predicts prognosis better than SPM data with the exception of the LGG cohort.