Association between viral infections and glioma risk: a two-sample bidirectional Mendelian randomization analysis.

BACKGROUND: Glioma is one of the leading types of brain tumor, but few etiologic factors of primary glioma have been identified. Previous observational research has shown an association between viral infection and glioma risk. In this study, we used Mendelian randomization (MR) analysis to explore the direction and magnitude of the causal relationship between viral infection and glioma. METHODS: We conducted a two-sample bidirectional MR analysis using genome-wide association study (GWAS) data. Summary statistics data of glioma were collected from the largest meta-analysis GWAS, involving 12,488 cases and 18,169 controls. Single-nucleotide polymorphisms (SNPs) associated with exposures were used as instrumental variables to estimate the causal relationship between glioma and twelve types of viral infections from corresponding GWAS data. In addition, sensitivity analyses were performed. RESULTS: After correcting for multiple tests and sensitivity analysis, we detected that genetically predicted herpes zoster (caused by Varicella zoster virus (VZV) infection) significantly decreased risk of low-grade glioma (LGG) development (OR = 0.85, 95% CI: 0.76-0.96, P = 0.01, FDR = 0.04). No causal effects of the other eleven viral infections on glioma and reverse causality were detected. CONCLUSIONS: This is one of the first and largest studies in this field. We show robust evidence supporting that genetically predicted herpes zoster caused by VZV infection reduces risk of LGG. The findings of our research advance understanding of the etiology of glioma.

Evaluating hematologic parameters in newly diagnosed and recurrent glioblastoma: Prognostic utility and clinical trial implications of myelosuppression.

Background: Glioblastoma (GBM) patients are treated with radiation therapy, chemotherapy, and corticosteroids, which can cause myelosuppression. To understand the relative prognostic utility of blood-based biomarkers in GBM and its implications for clinical trial design, we examined the incidence, predictors, and prognostic value of lymphopenia, neutrophil-to-lymphocyte ratio (NLR), and platelet count during chemoradiation (CRT) and recurrence. Methods: This cohort study included 764 newly diagnosed glioblastoma patients treated from 2005 to 2019 with blood counts prior to surgery, within 6 weeks of CRT, and at first recurrence available for automatic extraction from the medical record. Logistic regression was used to evaluate exposures and Kaplan-Meier was used to evaluate outcomes. Results: Among the cohort, median age was 60.3 years; 87% had Karnofsky performance status ≥ 70, 37.5% had gross total resection, and 90% received temozolomide (TMZ). During CRT, 37.8% (248/656) of patients developed grade 3 or higher lymphopenia. On multivariable analysis (MVA), high NLR during CRT remained an independent predictor for inferior survival (Adjusted Hazard Ratio [AHR] = 1.57, 95% CI = 1.14-2.15) and shorter progression-free survival (AHR = 1.42, 95% CI = 1.05-1.90). Steroid use was associated with lymphopenia (OR = 2.66,1.20-6.00) and high NLR (OR = 3.54,2.08-6.11). Female sex was associated with lymphopenia (OR = 2.33,1.03-5.33). At first recurrence, 28% of patients exhibited grade 3 or higher lymphopenia. High NLR at recurrence was associated with worse subsequent survival on MVA (AHR = 1.69, 95% CI = 1.25-2.27). Conclusions: High NLR is associated with worse outcomes in newly diagnosed and recurrent glioblastoma. Appropriate eligibility criteria and accounting and reporting of blood-based biomarkers are important in the design and interpretation of newly diagnosed and recurrent glioblastoma trials.

Predicting glioblastoma molecular subtypes and prognosis with a multimodal model integrating convolutional neural network, radiomics, and semantics.

OBJECTIVE: The aim of this study was to build a convolutional neural network (CNN)-based prediction model of glioblastoma (GBM) molecular subtype diagnosis and prognosis with multimodal features. METHODS: In total, 222 GBM patients were included in the training set from Sun Yat-sen University Cancer Center (SYSUCC) and 107 GBM patients were included in the validation set from SYSUCC, Xuanwu Hospital Capital Medical University, and the First Hospital of Jilin University. The multimodal model was trained with MR images (pre- and postcontrast T1-weighted images and T2-weighted images), corresponding MRI impression, and clinical patient information. First, the original images were segmented using the Multimodal Brain Tumor Image Segmentation Benchmark toolkit. Convolutional features were extracted using 3D residual deep neural network (ResNet50) and convolutional 3D (C3D). Radiomic features were extracted using pyradiomics. Report texts were converted to word embedding using word2vec. These three types of features were then integrated to train neural networks. Accuracy, precision, recall, and F1-score were used to evaluate the model performance. RESULTS: The C3D-based model yielded the highest accuracy of 91.11% in the prediction of IDH1 mutation status. Importantly, the addition of semantics improved precision by 11.21% and recall in MGMT promoter methylation status prediction by 14.28%. The areas under the receiver operating characteristic curves of the C3D-based model in the IDH1, ATRX, MGMT, and 1-year prognosis groups were 0.976, 0.953, 0.955, and 0.976, respectively. In external validation, the C3D-based model showed significant improvement in accuracy in the IDH1, ATRX, MGMT, and 1-year prognosis groups, which were 88.30%, 76.67%, 85.71%, and 85.71%, respectively (compared with 3D ResNet50: 83.51%, 66.67%, 82.14%, and 70.79%, respectively). CONCLUSIONS: The authors propose a novel multimodal model integrating C3D, radiomics, and semantics, which had a great performance in predicting IDH1, ATRX, and MGMT molecular subtypes and the 1-year prognosis of GBM.

Tangent normalization for somatic copy-number inference in cancer genome analysis.

MOTIVATION: Somatic copy-number alterations (SCNAs) play an important role in cancer development. Systematic noise in sequencing and array data present a significant challenge to the inference of SCNAs for cancer genome analyses. As part of The Cancer Genome Atlas, the Broad Institute Genome Characterization Center developed the Tangent normalization method to generate copy-number profiles using data from single-nucleotide polymorphism (SNP) arrays and whole-exome sequencing (WES) technologies for over 10 000 pairs of tumors and matched normal samples. Here, we describe the Tangent method, which uses a unique linear combination of normal samples as a reference for each tumor sample, to subtract systematic errors that vary across samples. We also describe a modification of Tangent, called Pseudo-Tangent, which enables denoising through comparisons between tumor profiles when few normal samples are available. RESULTS: Tangent normalization substantially increases signal-to-noise ratios (SNRs) compared to conventional normalization methods in both SNP array and WES analyses. Tangent and Pseudo-Tangent normalizations improve the SNR by reducing noise with minimal effect on signal and exceed the contribution of other steps in the analysis such as choice of segmentation algorithm. Tangent and Pseudo-Tangent are broadly applicable and enable more accurate inference of SCNAs from DNA sequencing and array data. AVAILABILITY AND IMPLEMENTATION: Tangent is available at https://github.com/broadinstitute/tangent and as a Docker image (https://hub.docker.com/r/broadinstitute/tangent). Tangent is also the normalization method for the copy-number pipeline in Genome Analysis Toolkit 4 (GATK4). SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

Structural variants shape driver combinations and outcomes in pediatric high-grade glioma.

We analyzed the contributions of structural variants (SVs) to gliomagenesis across 179 pediatric high-grade gliomas (pHGGs). The most recurrent SVs targeted MYC isoforms and receptor tyrosine kinases (RTKs), including an SV amplifying a MYC enhancer in 12% of diffuse midline gliomas (DMG), indicating an underappreciated role for MYC in pHGG. SV signature analysis revealed that tumors with simple signatures were TP53 wild type (TP53WT) but showed alterations in TP53 pathway members PPM1D and MDM4. Complex signatures were associated with direct aberrations in TP53, CDKN2A and RB1 early in tumor evolution and with later-occurring extrachromosomal amplicons. All pHGGs exhibited at least one simple-SV signature, but complex-SV signatures were primarily restricted to subsets of H3.3K27M DMGs and hemispheric pHGGs. Importantly, DMGs with complex-SV signatures were associated with shorter overall survival independent of histone mutation and TP53 status. These data provide insight into the impact of SVs on gliomagenesis and the mechanisms that shape them.

Sequence and vector shapes vaccine induced antibody effector functions in HIV vaccine trials.

Despite the advent of long-acting anti-retroviral therapy able to control and prevent infection, a preventative vaccine remains a global priority for the elimination of HIV. The moderately protective RV144 vaccine trial suggested functional IgG1 and IgG3 antibodies were a potential correlate of protection, but the RV144-inspired HVTN702 validation trial failed to demonstrate efficacy despite inducing targeted levels of IgG1/IgG3. Alterations in inserts, and antigens, adjuvant, and regimen also resulted in vaccine induced target quantitative levels of the immune correlates, but drove qualitative changes to the humoral immune response, pointing to the urgent need to define the influence of vaccine strategies on shaping antibody quality, not just quantity. Thus, defining how distinct prime/boost approaches tune long-lived functional antibodies represents an important goal in vaccine development. Here, we compared vaccine responses in Phase I and II studies in humans utilizing various combinations of DNA/vector, vector/vector and DNA/protein HIV vaccines. We found that adenoviral vector immunization, compared to pox-viral vectors, resulted in the most potent IgG1 and IgG3 responses, linked to highly functional antibody activity, including assisting NK cell related functions. Minimal differences were observed in the durability of the functional humoral immune response across vaccine regimens, except for antibody dependent phagocytic function, which persisted for longer periods in the DNA/rAd5 and rAd35/rAd5 regimen, likely driven by higher IgG1 levels. Collectively, these findings suggest adenoviral vectors drive superior antibody quality and durability that could inform future clinical vaccine studies. Trial registration: ClinicalTrials.gov NCT00801697, NCT00961883, NCT02207920, NCT00125970, NCT02852005).