Benchmarking whole exome sequencing in the German network for personalized medicine.

INTRODUCTION: Whole Exome Sequencing (WES) has emerged as an efficient tool in clinical cancer diagnostics to broaden the scope from panel-based diagnostics to screening of all genes and enabling robust determination of complex biomarkers in a single analysis. METHODS: To assess concordance, six formalin-fixed paraffin-embedded (FFPE) tissue specimens and four commercial reference standards were analyzed by WES as matched tumor-normal DNA at 21 NGS centers in Germany, each employing local wet-lab and bioinformatics. Somatic and germline variants, copy-number alterations (CNAs), and complex biomarkers were investigated. Somatic variant calling was performed in 494 diagnostically relevant cancer genes. The raw data were collected and re-analyzed with a central bioinformatic pipeline to separate wet- and dry-lab variability. RESULTS: The mean positive percentage agreement (PPA) of somatic variant calling was 76 % while the positive predictive value (PPV) was 89 % in relation to a consensus list of variants found by at least five centers. Variant filtering was identified as the main cause for divergent variant calls. Adjusting filter criteria and re-analysis increased the PPA to 88 % for all and 97 % for the clinically relevant variants. CNA calls were concordant for 82 % of genomic regions. Homologous recombination deficiency (HRD), tumor mutational burden (TMB), and microsatellite instability (MSI) status were concordant for 94 %, 93 %, and 93 % of calls, respectively. Variability of CNAs and complex biomarkers did not decrease considerably after harmonization of the bioinformatic processing and was hence attributed mainly to wet-lab differences. CONCLUSION: Continuous optimization of bioinformatic workflows and participating in round robin tests are recommended.

Dissecting AI-based mutation prediction in lung adenocarcinoma: A comprehensive real-world study.

INTRODUCTION: Molecular profiling of lung cancer is essential to identify genetic alterations that predict response to targeted therapy. While deep learning shows promise for predicting oncogenic mutations from whole tissue images, existing studies often face challenges such as limited sample sizes, a focus on earlier stage patients, and insufficient analysis of robustness and generalizability. METHODS: This retrospective study evaluates factors influencing mutation prediction accuracy using the large Heidelberg Lung Adenocarcinoma Cohort (HLCC), a cohort of 2356 late-stage FFPE samples. Validation is performed in the publicly available TCGA-LUAD cohort. RESULTS: Models trained on the larger HLCC cohort generalized well to the TCGA dataset for mutations in EGFR (AUC 0.76), STK11 (AUC 0.71) and TP53 (AUC 0.75), in line with the hypothesis that larger cohort sizes improve model robustness. Variation in performance due to pre-processing and modeling choices, such as mutation variant calling, affected EGFR prediction accuracy by up to 7 %. DISCUSSION: Model explanations suggest that acinar and papillary growth patterns are critical for the detection of EGFR mutations, whereas solid growth patterns and large nuclei are indicative of TP53 mutations. These findings highlight the importance of specific morphological features in mutation detection and the potential of deep learning models to improve mutation prediction accuracy. CONCLUSION: Although deep learning models trained on larger cohorts show improved robustness and generalizability in predicting oncogenic mutations, they cannot replace comprehensive molecular profiling. However, they may support patient pre-selection for clinical trials and deepen the insight in genotype-phenotype relationships.

Uncovering convergence and divergence between autism and schizophrenia using genomic tools and patients' neurons.

Autism spectrum disorders (ASDs) are highly heritable and result in abnormal repetitive behaviors and impairment in communication and cognitive skills. Previous studies have focused on the genetic correlation between ASDs and other neuropsychiatric disorders, but an in-depth understanding of the correlation to other disorders is required. We conducted an extensive meta-analysis of common variants identified in ASDs by genome-wide association studies (GWAS) and compared it to the consensus genes and single nucleotide polymorphisms (SNPs) of Schizophrenia (SCZ). We found approximately 75% of the GWAS genes that are associated with ASD are also associated with SCZ. We further investigated the cellular phenotypes of neurons derived from induced pluripotent stem cell (iPSC) models in ASD and SCZ. Our findings revealed that ASD and SCZ neurons initially follow divergent developmental trajectories compared to control neurons. However, despite these early diametrical differences, both ASD and SCZ neurons ultimately display similar deficits in synaptic activity as they mature. This significant genetic overlap between ASD and SCZ, coupled with the convergence towards similar synaptic deficits, highlights the intricate interplay of genetic and developmental factors in shaping the shared underlying mechanisms of these complex neurodevelopmental and neuropsychiatric disorders.

KRAS and TP53 co-mutation predicts benefit of immune checkpoint blockade in lung adenocarcinoma.

BACKGROUND: Predictive biomarkers in use for immunotherapy in advanced non-small cell lung cancer are of limited sensitivity and specificity. We analysed the potential of activating KRAS and pathogenic TP53 mutations to provide additional predictive information. METHODS: The study cohort included 713 consecutive immunotherapy patients with advanced lung adenocarcinomas, negative for actionable genetic alterations. Additionally, two previously published immunotherapy and two surgical patient cohorts were analyzed. Therapy benefit was stratified by KRAS and TP53 mutations. Molecular characteristics underlying KRASmut/TP53mut tumours were revealed by the analysis of TCGA data. RESULTS: An interaction between KRAS and TP53 mutations was observed in univariate and multivariate analyses of overall survival (Hazard ratio [HR] = 0.56, p = 0.0044 and HR = 0.53, p = 0.0021) resulting in a stronger benefit for KRASmut/TP53mut tumours (HR = 0.71, CI 0.55-0.92). This observation was confirmed in immunotherapy cohorts but not observed in surgical cohorts. Tumour mutational burden, proliferation, and PD-L1 mRNA were significantly higher in TP53-mutated tumours, regardless of KRAS status. Genome-wide expression analysis revealed 64 genes, including CX3CL1 (fractalkine), as specific transcriptomic characteristic of KRASmut/TP53mut tumours. CONCLUSIONS: KRAS/TP53 co-mutation predicts ICI benefit in univariate and multivariate survival analyses and is associated with unique molecular tumour features. Mutation testing of the two genes can be easily implemented using small NGS panels.

Homogenous TP53mut-associated tumor biology across mutation and cancer types revealed by transcriptome analysis.

TP53 is the most frequently mutated gene in human cancer. While no TP53-targeting drugs have been approved in the USA or Europe so far, preclinical and clinical studies are underway to investigate targeting of specific or all TP53 mutations, for example, by restoration of the functionality of mutated TP53 (TP53mut) or protecting wildtype TP53 (TP53wt) from negative regulation. We performed a comprehensive mRNA expression analysis in 24 cancer types of TCGA to extract (i) a consensus expression signature shared across TP53 mutation types and cancer types, (ii) differential gene expression patterns between tumors harboring different TP53 mutation types such as loss of function, gain of function or dominant-negative mutations, and (iii) cancer-type-specific patterns of gene expression and immune infiltration. Analysis of mutational hotspots revealed both similarities across cancer types and cancer type-specific hotspots. Underlying ubiquitous and cancer type-specific mutational processes with the associated mutational signatures contributed to explaining this observation. Virtually no genes were differentially expressed between tumors harboring different TP53 mutation types, while hundreds of genes were over- and underexpressed in TP53mut compared to TP53wt tumors. A consensus list included 178 genes that were overexpressed and 32 genes that were underexpressed in the TP53mut tumors of at least 16 of the investigated 24 cancer types. In an association analysis of immune infiltration with TP53 mutations in 32 cancer subtypes, decreased immune infiltration was observed in six subtypes, increased infiltration in two subtypes, a mixed pattern of decreased and increased immune cell populations in four subtypes, while immune infiltration was not associated with TP53 status in 20 subtypes. The analysis of a large cohort of human tumors complements results from experimental studies and supports the view that TP53 mutations should be further evaluated as predictive markers for immunotherapy and targeted therapies.

Ioncopy: an R Shiny app to call copy number alterations in targeted NGS data.

BACKGROUND: Somatic copy number alterations (CNAs) contribute to the clinically targetable aberrations in the tumor genome. For both routine diagnostics and biomarkers research, CNA analysis in a single assay together with somatic mutations is highly desirable. RESULTS: Ioncopy is a validated method and easy-to-use software for CNA calling from targeted NGS data. Copy number and significance of CNA are estimated for each gene in each sample. Copy number gains and losses are called after multiple testing corrections controlling FWER or FDR. CONCLUSIONS: Ioncopy facilitates calling of CNAs in a cohort of tumors tissues with or without using normal (germline) DNA controls.