Genomic and transcriptomic analysis of checkpoint blockade response in advanced non-small cell lung cancer.

Anti-PD-1/PD-L1 agents have transformed the treatment landscape of advanced non-small cell lung cancer (NSCLC). To expand our understanding of the molecular features underlying response to checkpoint inhibitors in NSCLC, we describe here the first joint analysis of the Stand Up To Cancer-Mark Foundation cohort, a resource of whole exome and/or RNA sequencing from 393 patients with NSCLC treated with anti-PD-(L)1 therapy, along with matched clinical response annotation. We identify a number of associations between molecular features and outcome, including (1) favorable (for example, ATM altered) and unfavorable (for example, TERT amplified) genomic subgroups, (2) a prominent association between expression of inducible components of the immunoproteasome and response and (3) a dedifferentiated tumor-intrinsic subtype with enhanced response to checkpoint blockade. Taken together, results from this cohort demonstrate the complexity of biological determinants underlying immunotherapy outcomes and reinforce the discovery potential of integrative analysis within large, well-curated, cancer-specific cohorts.

High-throughput genetic clustering of type 2 diabetes loci reveals heterogeneous mechanistic pathways of metabolic disease.

AIMS/HYPOTHESIS: Type 2 diabetes is highly polygenic and influenced by multiple biological pathways. Rapid expansion in the number of type 2 diabetes loci can be leveraged to identify such pathways. METHODS: We developed a high-throughput pipeline to enable clustering of type 2 diabetes loci based on variant-trait associations. Our pipeline extracted summary statistics from genome-wide association studies (GWAS) for type 2 diabetes and related traits to generate a matrix of 323 variants × 64 trait associations and applied Bayesian non-negative matrix factorisation (bNMF) to identify genetic components of type 2 diabetes. Epigenomic enrichment analysis was performed in 28 cell types and single pancreatic cells. We generated cluster-specific polygenic scores and performed regression analysis in an independent cohort (N=25,419) to assess for clinical relevance. RESULTS: We identified ten clusters of genetic loci, recapturing the five from our prior analysis as well as novel clusters related to beta cell dysfunction, pronounced insulin secretion, and levels of alkaline phosphatase, lipoprotein A and sex hormone-binding globulin. Four clusters related to mechanisms of insulin deficiency, five to insulin resistance and one had an unclear mechanism. The clusters displayed tissue-specific epigenomic enrichment, notably with the two beta cell clusters differentially enriched in functional and stressed pancreatic beta cell states. Additionally, cluster-specific polygenic scores were differentially associated with patient clinical characteristics and outcomes. The pipeline was applied to coronary artery disease and chronic kidney disease, identifying multiple overlapping clusters with type 2 diabetes. CONCLUSIONS/INTERPRETATION: Our approach stratifies type 2 diabetes loci into physiologically interpretable genetic clusters associated with distinct tissues and clinical outcomes. The pipeline allows for efficient updating as additional GWAS become available and can be readily applied to other conditions, facilitating clinical translation of GWAS findings. Software to perform this clustering pipeline is freely available.

Biomarker correlates with response to NY-ESO-1 TCR T cells in patients with synovial sarcoma.

Autologous T cells transduced to express a high affinity T-cell receptor specific to NY-ESO-1 (letetresgene autoleucel, lete-cel) show promise in the treatment of metastatic synovial sarcoma, with 50% overall response rate. The efficacy of lete-cel treatment in 45 synovial sarcoma patients (NCT01343043) has been previously reported, however, biomarkers predictive of response and resistance remain to be better defined. This post-hoc analysis identifies associations of response to lete-cel with lymphodepleting chemotherapy regimen (LDR), product attributes, cell expansion, cytokines, and tumor gene expression. Responders have higher IL-15 levels pre-infusion (p = 0.011) and receive a higher number of transduced effector memory (CD45RA- CCR7-) CD8 + cells per kg (p = 0.039). Post-infusion, responders have increased IFNγ, IL-6, and peak cell expansion (p < 0.01, p < 0.01, and p = 0.016, respectively). Analysis of tumor samples post-treatment illustrates lete-cel infiltration and a decrease in expression of macrophage genes, suggesting remodeling of the tumor microenvironment. Here we report potential predictive and pharmacodynamic markers of lete-cel response that may inform LDR, cell dose, and strategies to enhance anticancer efficacy.

High-Resolution Profiling of Lung Adenocarcinoma Identifies Expression Subtypes with Specific Biomarkers and Clinically Relevant Vulnerabilities.

Lung adenocarcinoma (LUAD) is one of the most common cancer types and has various treatment options. Better biomarkers to predict therapeutic response are needed to guide choice of treatment modality and to improve precision medicine. Here, we used a consensus hierarchical clustering approach on 509 LUAD cases from The Cancer Genome Atlas to identify five robust LUAD expression subtypes. Genomic and proteomic data from patient samples and cell lines was then integrated to help define biomarkers of response to targeted therapies and immunotherapies. This approach defined subtypes with unique proteogenomic and dependency profiles. Subtype 4 (S4)-associated cell lines exhibited specific vulnerability to loss of CDK6 and CDK6-cyclin D3 complex gene (CCND3). Subtype 3 (S3) was characterized by dependency on CDK4, immune-related expression patterns, and altered MET signaling. Experimental validation showed that S3-associated cell lines responded to MET inhibitors, leading to increased expression of programmed death-ligand 1 (PD-L1). In an independent real-world patient dataset, patients with S3 tumors were enriched with responders to immune checkpoint blockade. Genomic features in S3 and S4 were further identified as biomarkers for enabling clinical diagnosis of these subtypes. Overall, our consensus hierarchical clustering approach identified robust tumor expression subtypes, and our subsequent integrative analysis of genomics, proteomics, and CRISPR screening data revealed subtype-specific biology and vulnerabilities. These LUAD expression subtypes and their biomarkers could help identify patients likely to respond to CDK4/6, MET, or PD-L1 inhibitors, potentially improving patient outcome. SIGNIFICANCE: Integrative analysis of multiomic and drug dependency data uncovers robust lung adenocarcinoma expression subtypes with unique therapeutic vulnerabilities and subtype-specific biomarkers of response.

Combined tumor and immune signals from genomes or transcriptomes predict outcomes of checkpoint inhibition in melanoma.

Immune checkpoint blockade (CPB) improves melanoma outcomes, but many patients still do not respond. Tumor mutational burden (TMB) and tumor-infiltrating T cells are associated with response, and integrative models improve survival prediction. However, integrating immune/tumor-intrinsic features using data from a single assay (DNA/RNA) remains underexplored. Here, we analyze whole-exome and bulk RNA sequencing of tumors from new and published cohorts of 189 and 178 patients with melanoma receiving CPB, respectively. Using DNA, we calculate T cell and B cell burdens (TCB/BCB) from rearranged TCR/Ig sequences and find that patients with TMBhigh and TCBhigh or BCBhigh have improved outcomes compared to other patients. By combining pairs of immune- and tumor-expressed genes, we identify three gene pairs associated with response and survival, which validate in independent cohorts. The top model includes lymphocyte-expressed MAP4K1 and tumor-expressed TBX3. Overall, RNA or DNA-based models combining immune and tumor measures improve predictions of melanoma CPB outcomes.

Genetic analysis of dietary intake identifies new loci and functional links with metabolic traits.

Dietary intake is a major contributor to the global obesity epidemic and represents a complex behavioural phenotype that is partially affected by innate biological differences. Here, we present a multivariate genome-wide association analysis of overall variation in dietary intake to account for the correlation between dietary carbohydrate, fat and protein in 282,271 participants of European ancestry from the UK Biobank (n = 191,157) and Cohorts for Heart and Aging Research in Genomic Epidemiology Consortium (n = 91,114), and identify 26 distinct genome-wide significant loci. Dietary intake signals map exclusively to specific brain regions and are enriched for genes expressed in specialized subtypes of GABAergic, dopaminergic and glutamatergic neurons. We identified two main clusters of genetic variants for overall variation in dietary intake that were differently associated with obesity and coronary artery disease. These results enhance the biological understanding of interindividual differences in dietary intake by highlighting neural mechanisms, supporting functional follow-up experiments and possibly providing new avenues for the prevention and treatment of prevalent complex metabolic diseases.

Genetic determinants of daytime napping and effects on cardiometabolic health.

Daytime napping is a common, heritable behavior, but its genetic basis and causal relationship with cardiometabolic health remain unclear. Here, we perform a genome-wide association study of self-reported daytime napping in the UK Biobank (n = 452,633) and identify 123 loci of which 61 replicate in the 23andMe research cohort (n = 541,333). Findings include missense variants in established drug targets for sleep disorders (HCRTR1, HCRTR2), genes with roles in arousal (TRPC6, PNOC), and genes suggesting an obesity-hypersomnolence pathway (PNOC, PATJ). Association signals are concordant with accelerometer-measured daytime inactivity duration and 33 loci colocalize with loci for other sleep phenotypes. Cluster analysis identifies three distinct clusters of nap-promoting mechanisms with heterogeneous associations with cardiometabolic outcomes. Mendelian randomization shows potential causal links between more frequent daytime napping and higher blood pressure and waist circumference.

Subependymal giant cell astrocytomas are characterized by mTORC1 hyperactivation, a very low somatic mutation rate, and a unique gene expression profile.

Subependymal giant-cell astrocytomas (SEGAs) are slow-growing brain tumors that are a hallmark feature seen in 5-10% of patients with Tuberous Sclerosis Complex (TSC). Though histologically benign, they can cause serious neurologic symptoms, leading to death if untreated. SEGAs consistently show biallelic loss of TSC1 or TSC2. Herein, we aimed to define other somatic events beyond TSC1/TSC2 loss and identify potential transcriptional drivers that contribute to SEGA formation. Paired tumor-normal whole-exome sequencing was performed on 21 resected SEGAs from 20 TSC patients. Pathogenic variants in TSC1/TSC2 were identified in 19/21 (90%) SEGAs. Copy neutral loss of heterozygosity (size range: 2.2-46 Mb) was seen in 76% (16/21) of SEGAs (44% chr9q and 56% chr16p). An average of 1.4 other somatic variants (range 0-7) per tumor were identified, unlikely of pathogenic significance. Whole transcriptome RNA-sequencing analyses revealed 190 common differentially expressed genes in SEGA (n = 16, 13 from a prior study) in pairwise comparison to each of: low grade diffuse gliomas (n = 530) and glioblastoma (n = 171) from The Cancer Genome Atlas (TCGA) consortium, ganglioglioma (n = 10), TSC cortical tubers (n = 15), and multiple normal tissues. Among these, homeobox transcription factors (TFs) HMX3, HMX2, VAX1, SIX3; and TFs IRF6 and EOMES were all expressed >12-fold higher in SEGAs (FDR/q-value < 0.05). Immunohistochemistry supported the specificity of IRF6, VAX1, SIX3 for SEGAs in comparison to other tumor entities and normal brain. We conclude that SEGAs have an extremely low somatic mutation rate, suggesting that TSC1/TSC2 loss is sufficient to drive tumor growth. The unique and highly expressed SEGA-specific TFs likely reflect the neuroepithelial cell of origin, and may also contribute to the transcriptional and epigenetic state that enables SEGA growth following two-hit loss of TSC1 or TSC2 and mTORC1 activation.

Genomic Predictors of Good Outcome, Recurrence, or Progression in High-Grade T1 Non-Muscle-Invasive Bladder Cancer.

High-grade T1 (HGT1) bladder cancer is the highest risk subtype of non-muscle-invasive bladder cancer with unpredictable outcome and poorly understood risk factors. Here, we examined the association of somatic mutation profiles with nonrecurrent disease (GO, good outcome), recurrence (R), or progression (PD) in a cohort of HGT1 patients. Exome sequencing was performed on 62 HGT1 and 15 matched normal tissue samples. Both tumor only (TO) and paired analyses were performed, focusing on 95 genes known to be mutated in bladder cancer. Somatic mutations, copy-number alterations, mutation load, and mutation signatures were studied. Thirty-three GO, 10 R, 18 PD, and 1 unknown outcome patients were analyzed. Tumor mutational burden (TMB) was similar to muscle-invasive disease and was highest in GO, intermediate in PD, and lowest in R patients (P = 0.017). DNA damage response gene mutations were associated with higher TMB (P < 0.0001) and GO (P = 0.003). ERCC2 and BRCA2 mutations were associated with GO. TP53, ATM, ARID1A, AHR, and SMARCB1 mutations were more frequent in PD. Focal copy-number gain in CCNE1 and CDKN2A deletion was enriched in PD or R (P = 0.047; P = 0.06). APOBEC (46%) and COSMIC5 (34%) signatures were most frequent. APOBEC-A and ERCC2 mutant tumors (COSMIC5) were associated with GO (P = 0.047; P = 0.0002). pT1b microstaging was associated with a genomic cluster (P = 0.05) with focal amplifications of E2F3/SOX4, PVRL4, CCNE1, and TP53 mutations. Findings were validated using external public datasets. These findings require confirmation but suggest that management of HGT1 bladder cancer may be improved via molecular characterization to predict outcome. SIGNIFICANCE: Detailed genetic analyses of HGT1 bladder tumors identify features that correlate with outcome, e.g., high mutational burden, ERCC2 mutations, and high APOBEC-A/ERCC2 mutation signatures were associated with good outcome.

Analyses of non-coding somatic drivers in 2,658 cancer whole genomes.

The discovery of drivers of cancer has traditionally focused on protein-coding genes1-4. Here we present analyses of driver point mutations and structural variants in non-coding regions across 2,658 genomes from the Pan-Cancer Analysis of Whole Genomes (PCAWG) Consortium5 of the International Cancer Genome Consortium (ICGC) and The Cancer Genome Atlas (TCGA). For point mutations, we developed a statistically rigorous strategy for combining significance levels from multiple methods of driver discovery that overcomes the limitations of individual methods. For structural variants, we present two methods of driver discovery, and identify regions that are significantly affected by recurrent breakpoints and recurrent somatic juxtapositions. Our analyses confirm previously reported drivers6,7, raise doubts about others and identify novel candidates, including point mutations in the 5' region of TP53, in the 3' untranslated regions of NFKBIZ and TOB1, focal deletions in BRD4 and rearrangements in the loci of AKR1C genes. We show that although point mutations and structural variants that drive cancer are less frequent in non-coding genes and regulatory sequences than in protein-coding genes, additional examples of these drivers will be found as more cancer genomes become available.

The repertoire of mutational signatures in human cancer.

Somatic mutations in cancer genomes are caused by multiple mutational processes, each of which generates a characteristic mutational signature1. Here, as part of the Pan-Cancer Analysis of Whole Genomes (PCAWG) Consortium2 of the International Cancer Genome Consortium (ICGC) and The Cancer Genome Atlas (TCGA), we characterized mutational signatures using 84,729,690 somatic mutations from 4,645 whole-genome and 19,184 exome sequences that encompass most types of cancer. We identified 49 single-base-substitution, 11 doublet-base-substitution, 4 clustered-base-substitution and 17 small insertion-and-deletion signatures. The substantial size of our dataset, compared with previous analyses3-15, enabled the discovery of new signatures, the separation of overlapping signatures and the decomposition of signatures into components that may represent associated-but distinct-DNA damage, repair and/or replication mechanisms. By estimating the contribution of each signature to the mutational catalogues of individual cancer genomes, we revealed associations of signatures to exogenous or endogenous exposures, as well as to defective DNA-maintenance processes. However, many signatures are of unknown cause. This analysis provides a systematic perspective on the repertoire of mutational processes that contribute to the development of human cancer.

A model combining clinical and genomic factors to predict response to PD-1/PD-L1 blockade in advanced urothelial carcinoma.

BACKGROUND: In metastatic urothelial carcinoma (mUC), predictive biomarkers that correlate with response to immune checkpoint inhibitors (ICIs) are lacking. Here, we interrogated genomic and clinical features associated with response to ICIs in mUC. METHODS: Sixty two mUC patients treated with ICI who had targeted tumour sequencing were studied. We examined associations between candidate biomarkers and clinical benefit (CB, any objective reduction in tumour size) versus no clinical benefit (NCB, no change or objective increase in tumour size). Both univariable and multivariable analyses for associations were conducted. A comparator cohort of 39 mUC patients treated with taxanes was analysed by using the same methodology. RESULTS: Nine clinical and seven genomic factors correlated with clinical outcomes in univariable analysis in the ICI cohort. Among the 16 factors, neutrophil-to-lymphocyte ratio (NLR) ≥5 (OR = 0.12, 95% CI, 0.01-1.15), visceral metastasis (OR = 0.05, 95% CI, 0.01-0.43) and single-nucleotide variant (SNV) count < 10 (OR = 0.04, 95% CI, 0.006-0.27) were identified as independent predictors of NCB to ICI in multivariable analysis (c-statistic = 0.90). None of the 16 variables were associated with clinical benefit in the taxane cohort. CONCLUSIONS: This three-factor model includes genomic (SNV count >9) and clinical (NLR <5, lack of visceral metastasis) variables predictive for benefit to ICI but not taxane therapy for mUC. External validation of these hypothesis-generating results is warranted to enable use in routine clinical care.

Genomic analyses of flow-sorted Hodgkin Reed-Sternberg cells reveal complementary mechanisms of immune evasion.

Classical Hodgkin lymphoma (cHL) is composed of rare malignant Hodgkin Reed-Sternberg (HRS) cells within an extensive, but ineffective, inflammatory/immune cell infiltrate. HRS cells exhibit near-universal somatic copy gains of chromosome 9p/9p24.1, which increase expression of the programmed cell death protein 1 (PD-1) ligands. To define genetic mechanisms of response and resistance to PD-1 blockade and identify complementary treatment targets, we performed whole-exome sequencing of flow cytometry-sorted HRS cells from 23 excisional biopsies of newly diagnosed cHLs, including 8 Epstein-Barr virus-positive (EBV+) tumors. We identified significantly mutated cancer candidate genes (CCGs) as well as somatic copy number alterations and structural variations and characterized their contribution to disease-defining immune evasion mechanisms and nuclear factor κB (NF-κB), JAK/STAT, and PI3K signaling pathways. EBV- cHLs had a higher prevalence of genetic alterations in the NF-κB and major histocompatibility complex class I antigen presentation pathways. In this young cHL cohort (median age, 26 years), we identified a predominant mutational signature of spontaneous deamination of cytosine- phosphate-guanines ("Aging"), in addition to apolipoprotein B mRNA editing catalytic polypeptide-like, activation-induced cytidine deaminase, and microsatellite instability (MSI)-associated hypermutation. In particular, the mutational burden in EBV- cHLs was among the highest reported, similar to that of carcinogen-induced tumors. Together, the overall high mutational burden, MSI-associated hypermutation, and newly identified genetic alterations represent additional potential bases for the efficacy of PD-1 blockade in cHL. Of note, recurrent cHL alterations, including B2M, TNFAIP3, STAT6, GNA13, and XPO1 mutations and 2p/2p15, 6p21.32, 6q23.3, and 9p/9p24.1 copy number alterations, were also identified in >20% of primary mediastinal B-cell lymphomas, highlighting shared pathogenetic mechanisms in these diseases.

Genomic analyses of PMBL reveal new drivers and mechanisms of sensitivity to PD-1 blockade.

Primary mediastinal large B-cell lymphomas (PMBLs) are aggressive tumors that typically present as large mediastinal masses in young women. PMBLs share clinical, transcriptional, and molecular features with classical Hodgkin lymphoma (cHL), including constitutive activation of nuclear factor κB (NF-κB), JAK/STAT signaling, and programmed cell death protein 1 (PD-1)-mediated immune evasion. The demonstrated efficacy of PD-1 blockade in relapsed/refractory PMBLs led to recent approval by the US Food and Drug Administration and underscored the importance of characterizing targetable genetic vulnerabilities in this disease. Here, we report a comprehensive analysis of recurrent genetic alterations -somatic mutations, somatic copy number alterations, and structural variants-in a cohort of 37 newly diagnosed PMBLs. We identified a median of 9 genetic drivers per PMBL, including known and newly identified components of the JAK/STAT and NF-κB signaling pathways and frequent B2M alterations that limit major histocompatibility complex class I expression, as in cHL. PMBL also exhibited frequent, newly identified driver mutations in ZNF217 and an additional epigenetic modifier, EZH2. The majority of these alterations were clonal, which supports their role as early drivers. In PMBL, we identified several previously uncharacterized molecular features that may increase sensitivity to PD-1 blockade, including high tumor mutational burden, microsatellite instability, and an apolipoprotein B mRNA editing catalytic polypeptide-like (APOBEC) mutational signature. The shared genetic features between PMBL and cHL provide a framework for analyzing the mechanism of action of PD-1 blockade in these related lymphoid malignancies.

Scaling computational genomics to millions of individuals with GPUs.

Current genomics methods are designed to handle tens to thousands of samples but will need to scale to millions to match the pace of data and hypothesis generation in biomedical science. Here, we show that high efficiency at low cost can be achieved by leveraging general-purpose libraries for computing using graphics processing units (GPUs), such as PyTorch and TensorFlow. We demonstrate > 200-fold decreases in runtime and ~ 5-10-fold reductions in cost relative to CPUs. We anticipate that the accessibility of these libraries will lead to a widespread adoption of GPUs in computational genomics.

Passenger Hotspot Mutations in Cancer.

Current statistical models for assessing hotspot significance do not properly account for variation in site-specific mutability, thereby yielding many false-positives. We thus (i) detail a Log-normal-Poisson (LNP) background model that accounts for this variability in a manner consistent with models of mutagenesis; (ii) use it to show that passenger hotspots arise from all common mutational processes; and (iii) apply it to a ∼10,000-patient cohort to nominate driver hotspots with far fewer false-positives compared with conventional methods. Overall, we show that many cancer hotspot mutations recurring at the same genomic site across multiple tumors are actually passenger events, recurring at inherently mutable genomic sites under no positive selection.

RNA sequence analysis reveals macroscopic somatic clonal expansion across normal tissues.

How somatic mutations accumulate in normal cells is poorly understood. A comprehensive analysis of RNA sequencing data from ~6700 samples across 29 normal tissues revealed multiple somatic variants, demonstrating that macroscopic clones can be found in many normal tissues. We found that sun-exposed skin, esophagus, and lung have a higher mutation burden than other tested tissues, which suggests that environmental factors can promote somatic mosaicism. Mutation burden was associated with both age and tissue-specific cell proliferation rate, highlighting that mutations accumulate over both time and number of cell divisions. Finally, normal tissues were found to harbor mutations in known cancer genes and hotspots. This study provides a broad view of macroscopic clonal expansion in human tissues, thus serving as a foundation for associating clonal expansion with environmental factors, aging, and risk of disease.