GWAS and meta-analysis identifies 49 genetic variants underlying critical COVID-19.

Critical illness in COVID-19 is an extreme and clinically homogeneous disease phenotype that we have previously shown1 to be highly efficient for discovery of genetic associations2. Despite the advanced stage of illness at presentation, we have shown that host genetics in patients who are critically ill with COVID-19 can identify immunomodulatory therapies with strong beneficial effects in this group3. Here we analyse 24,202 cases of COVID-19 with critical illness comprising a combination of microarray genotype and whole-genome sequencing data from cases of critical illness in the international GenOMICC (11,440 cases) study, combined with other studies recruiting hospitalized patients with a strong focus on severe and critical disease: ISARIC4C (676 cases) and the SCOURGE consortium (5,934 cases). To put these results in the context of existing work, we conduct a meta-analysis of the new GenOMICC genome-wide association study (GWAS) results with previously published data. We find 49 genome-wide significant associations, of which 16 have not been reported previously. To investigate the therapeutic implications of these findings, we infer the structural consequences of protein-coding variants, and combine our GWAS results with gene expression data using a monocyte transcriptome-wide association study (TWAS) model, as well as gene and protein expression using Mendelian randomization. We identify potentially druggable targets in multiple systems, including inflammatory signalling (JAK1), monocyte-macrophage activation and endothelial permeability (PDE4A), immunometabolism (SLC2A5 and AK5), and host factors required for viral entry and replication (TMPRSS2 and RAB2A).

Adaptation of a mutual exclusivity framework to identify driver mutations within oncogenic pathways.

Distinguishing genomic alterations in cancer-associated genes that have functional impact on tumor growth and disease progression from the ones that are passengers and confer no fitness advantage have important clinical implications. Evidence-based methods for nominating drivers are limited by existing knowledge on the oncogenic effects and therapeutic benefits of specific variants from clinical trials or experimental settings. As clinical sequencing becomes a mainstay of patient care, applying computational methods to mine the rapidly growing clinical genomic data holds promise in uncovering functional candidates beyond the existing knowledge base and expanding the patient population that could potentially benefit from genetically targeted therapies. We propose a statistical and computational method (MAGPIE) that builds on a likelihood approach leveraging the mutual exclusivity pattern within an oncogenic pathway for identifying probabilistically both the specific genes within a pathway and the individual mutations within such genes that are truly the drivers. Alterations in a cancer-associated gene are assumed to be a mixture of driver and passenger mutations with the passenger rates modeled in relationship to tumor mutational burden. We use simulations to study the operating characteristics of the method and assess false-positive and false-negative rates in driver nomination. When applied to a large study of primary melanomas, the method accurately identifies the known driver genes within the RTK-RAS pathway and nominates several rare variants as prime candidates for functional validation. A comprehensive evaluation of MAGPIE against existing tools has also been conducted leveraging the Cancer Genome Atlas data.

Identifying somatic fingerprints of cancers defined by germline and environmental risk factors.

Numerous studies over the past generation have identified germline variants that increase specific cancer risks. Simultaneously, a revolution in sequencing technology has permitted high-throughput annotations of somatic genomes characterizing individual tumors. However, examining the relationship between germline variants and somatic alteration patterns is hugely challenged by the large numbers of variants in a typical tumor, the rarity of most individual variants, and the heterogeneity of tumor somatic fingerprints. In this article, we propose statistical methodology that frames the investigation of germline-somatic relationships in an interpretable manner. The method uses meta-features embodying biological contexts of individual somatic alterations to implicitly group rare mutations. Our team has used this technique previously through a multilevel regression model to diagnose with high accuracy tumor site of origin. Herein, we further leverage topic models from computational linguistics to achieve interpretable lower-dimensional embeddings of the meta-features. We demonstrate how the method can identify distinctive somatic profiles linked to specific germline variants or environmental risk factors. We illustrate the method using The Cancer Genome Atlas whole-exome sequencing data to characterize somatic tumor fingerprints in breast cancer patients with germline BRCA1/2 mutations and in head and neck cancer patients exposed to human papillomavirus.

Topical hidden genome: discovering latent cancer mutational topics using a Bayesian multilevel context-learning approach.

Inferring the cancer-type specificities of ultra-rare, genome-wide somatic mutations is an open problem. Traditional statistical methods cannot handle such data due to their ultra-high dimensionality and extreme data sparsity. To harness information in rare mutations, we have recently proposed a formal multilevel multilogistic "hidden genome" model. Through its hierarchical layers, the model condenses information in ultra-rare mutations through meta-features embodying mutation contexts to characterize cancer types. Consistent, scalable point estimation of the model can incorporate 10s of millions of variants across thousands of tumors and permit impressive prediction and attribution. However, principled statistical inference is infeasible due to the volume, correlation, and noninterpretability of mutation contexts. In this paper, we propose a novel framework that leverages topic models from computational linguistics to effectuate dimension reduction of mutation contexts producing interpretable, decorrelated meta-feature topics. We propose an efficient MCMC algorithm for implementation that permits rigorous full Bayesian inference at a scale that is orders of magnitude beyond the capability of existing out-of-the-box inferential high-dimensional multi-class regression methods and software. Applying our model to the Pan Cancer Analysis of Whole Genomes dataset reveals interesting biological insights including somatic mutational topics associated with UV exposure in skin cancer, aging in colorectal cancer, and strong influence of epigenome organization in liver cancer. Under cross-validation, our model demonstrates highly competitive predictive performance against blackbox methods of random forest and deep learning.

Subsampling based variable selection for generalized linear models.

A novel variable selection method for low-dimensional generalized linear models is introduced. The new approach called AIC OPTimization via STABility Selection (OPT-STABS) repeatedly subsamples the data, minimizes Akaike's Information Criterion (AIC) over a sequence of nested models for each subsample, and includes in the final model those predictors selected in the minimum AIC model in a large fraction of the subsamples. New methods are also introduced to establish an optimal variable selection cutoff over repeated subsamples. An extensive simulation study examining a variety of proposec variable selection methods shows that, although no single method uniformly outperforms the others in all the scenarios considered, OPT-STABS is consistently among the best-performing methods in most settings while it performs competitively for the rest. This is in contrast to other candidate methods which either have poor performance across the board or exhibit good performance in some settings, but very poor in others. In addition, the asymptotic properties of the OPT-STABS estimator are derived, and its root-n consistency and asymptotic normality are proved. The methods are applied to two datasets involving logistic and Poisson regressions.

Adapting an Undergraduate Summer Internship to a Virtual Format: Implementing a Mentored Cancer Research Experience to Meet Rising Demand for Flexible Learning Environments.

To meet the rising demand for flexible learning in data-driven health research, we adapted an in-person undergraduate research program (Quantitative Sciences Undergraduate Research Experience (QSURE)) to an all-virtual framework in summer 2020 and 2021. We used Web-conferencing and remote computing to implement virtual hands-on research training within a comprehensive cancer center. We designed the program to achieve research and career development goals: students completed faculty-mentored quantitative research projects and received education in the responsible conduct of research and practical skills, such as oral and written presentation. We assessed virtual program efficacy using pre- and post-program quantitative and qualitative student feedback. Eighteen students participated (nine each year); they reported high satisfaction with the virtual format. Compared with baseline, students reported improved perceived competence in quantitative skills and research knowledge post-program; these improvements were comparable to the in-person program. Defined benchmarks and consistent communication (with mentors, program directors, other students) were crucial to students' success; however, students noted challenges in building camaraderie online. With adequate resources, Web-based technology can be leveraged as an effective format for hands-on quantitative research training. Our framework can be tailored to an institution's needs, particularly those for which available resources better align with a virtual research program.

Exome-Wide Pan-Cancer Analysis of Germline Variants in 8,719 Individuals Finds Little Evidence of Rare Variant Associations.

BACKGROUND: Many cancer types show considerable heritability, and extensive research has been done to identify germline susceptibility variants. Linkage studies have discovered many rare high-risk variants, and genome-wide association studies (GWAS) have discovered many common low-risk variants. However, it is believed that a considerable proportion of the heritability of cancer remains unexplained by known susceptibility variants. The "rare variant hypothesis" proposes that much of the missing heritability lies in rare variants that cannot reliably be detected by linkage analysis or GWAS. Until recently, high sequencing costs have precluded extensive surveys of rare variants, but technological advances have now made it possible to analyze rare variants on a much greater scale. OBJECTIVES: In this study, we investigated associations between rare variants and 14 cancer types. METHODS: We ran association tests using whole-exome sequencing data from The Cancer Genome Atlas (TCGA) and validated the findings using data from the Pan-Cancer Analysis of Whole Genomes Consortium (PCAWG). RESULTS: We identified four significant associations in TCGA, only one of which was replicated in PCAWG (BRCA1 and ovarian cancer). CONCLUSIONS: Our results provide little evidence in favor of the rare variant hypothesis. Much larger sample sizes may be needed to detect undiscovered rare cancer variants.

Using the "Hidden" genome to improve classification of cancer types.

It is increasingly common clinically for cancer specimens to be examined using techniques that identify somatic mutations. In principle, these mutational profiles can be used to diagnose the tissue of origin, a critical task for the 3% to 5% of tumors that have an unknown primary site. Diagnosis of primary site is also critical for screening tests that employ circulating DNA. However, most mutations observed in any new tumor are very rarely occurring mutations, and indeed the preponderance of these may never have been observed in any previous recorded tumor. To create a viable diagnostic tool we need to harness the information content in this "hidden genome" of variants for which no direct information is available. To accomplish this we propose a multilevel meta-feature regression to extract the critical information from rare variants in the training data in a way that permits us to also extract diagnostic information from any previously unobserved variants in the new tumor sample. A scalable implementation of the model is obtained by combining a high-dimensional feature screening approach with a group-lasso penalized maximum likelihood approach based on an equivalent mixed-effect representation of the multilevel model. We apply the method to the Cancer Genome Atlas whole-exome sequencing data set including 3702 tumor samples across seven common cancer sites. Results show that our multilevel approach can harness substantial diagnostic information from the hidden genome.

Testing tumors from different anatomic sites for clonal relatedness using somatic mutation data.

A common task for the cancer pathologist is to determine, in a patient suffering from cancer, whether a new tumor in a distinct anatomic site from the primary is an independent occurrence of cancer or a metastasis. As mutational profiling of tumors becomes more widespread in routine clinical practice, this diagnostic task can be greatly enhanced by comparing mutational profiles of the tumors to determine if they are sufficiently similar to conclude that the tumors are clonally related, that is, one is a metastasis of the other. We present here a likelihood ratio test for clonal relatedness in this setting and provide evidence of its validity. The test is unusual in that there are two possible alternative hypotheses, representing the two anatomic sites from which the single clonal cell could have initially emerged. Although evidence for clonal relatedness is largely provided by the presence of exact mutational matches in the two tumors, we show that it is possible to observe data where the test is statistically significant even when no matches are observed. This can occur when the mutational profile of one of the tumors is closely aligned with the anatomic site of the other tumor, suggesting indirectly that the tumor originated in that other site. We exhibit examples of this phenomenon and recommend a strategy for interpreting the results of these tests in practice.

Human genes differ by their UV sensitivity estimated through analysis of UV-induced silent mutations in melanoma.

We hypothesized that human genes differ by their sensitivity to ultraviolet (UV) exposure. We used somatic mutations detected by genome-wide screens in melanoma and reported in the Catalog Of Somatic Mutations In Cancer. As a measure of UV sensitivity, we used the number of silent mutations generated by C>T transitions in pyrimidine dimers of a given transcript divided by the number of potential sites for this type of mutations in the transcript. We found that human genes varied by UV sensitivity by two orders of magnitude. We noted that the melanoma-associated tumor suppressor gene CDKN2A was among the top five most UV-sensitive genes in the human genome. Melanoma driver genes have a higher UV-sensitivity compared with other genes in the human genome. The difference was more prominent for tumor suppressors compared with oncogene. The results of this study suggest that differential sensitivity of human transcripts to UV light may explain melanoma specificity of some driver genes. Practical significance of the study relates to the fact that differences in UV sensitivity among human genes need to be taken into consideration whereas predicting melanoma-associated genes by the number of somatic mutations detected in a given gene.

Testing clonal relatedness of two tumors from the same patient based on their mutational profiles: update of the Clonality R package.

SUMMARY: The Clonality R package is a practical tool to assess the clonal relatedness of two tumors from the same patient. We have previously presented its functionality for testing tumors using loss of heterozygosity data or copy number arrays. Since then somatic mutation data have been more widely available through next generation sequencing and we have developed new methodology for comparing the tumors' mutational profiles. We thus extended the package to include these two new methods for comparing tumors as well as the mutational frequency estimation from external data required for their implementation. The first method is a likelihood ratio test that is readily available on a patient by patient basis. The second method employs a random-effects model to estimate both the population and individual probabilities of clonal relatedness from a group of patients with pairs of tumors. The package is available on Bioconductor. AVAILABILITY AND IMPLEMENTATION: Bioconductor (http://bioconductor.org/packages/release/bioc/html/Clonality.html). SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

Optimized variable selection via repeated data splitting.

Model selection in high-dimensional settings has received substantial attention in recent years, however, similar advancements in the low-dimensional setting have been lacking. In this article, we introduce a new variable selection procedure for low to moderate scale regressions (n>p). This method repeatedly splits the data into two sets, one for estimation and one for validation, to obtain an empirically optimized threshold which is then used to screen for variables to include in the final model. In an extensive simulation study, we show that the proposed variable selection technique enjoys superior performance compared with candidate methods (backward elimination via repeated data splitting, univariate screening at 0.05 level, adaptive LASSO, SCAD), being amongst those with the lowest inclusion of noisy predictors while having the highest power to detect the correct model and being unaffected by correlations among the predictors. We illustrate the methods by applying them to a cohort of patients undergoing hepatectomy at our institution.

An EM algorithm to improve the estimation of the probability of clonal relatedness of pairs of tumors in cancer patients.

BACKGROUND: We previously introduced a random-effects model to analyze a set of patients, each of which has two distinct tumors. The goal is to estimate the proportion of patients for which one of the tumors is a metastasis of the other, i.e. where the tumors are clonally related. Matches of mutations within a tumor pair provide the evidence for clonal relatedness. In this article, using simulations, we compare two estimation approaches that we considered for our model: use of a constrained quasi-Newton algorithm to maximize the likelihood conditional on the random effect, and an Expectation-Maximization algorithm where we further condition the random-effect distribution on the data. RESULTS: In some specific settings, especially with sparse information, the estimation of the parameter of interest is at the boundary a non-negligible number of times using the first approach, while the EM algorithm gives more satisfactory estimates. This is of considerable importance for our application, since an estimate of either 0 or 1 for the proportion of cases that are clonal leads to individual probabilities being 0 or 1 in settings where the evidence is clearly not sufficient for such definitive probability estimates. CONCLUSIONS: The EM algorithm is a preferable approach for our clonality random-effect model. It is now the method implemented in our R package Clonality, making available an easy and fast way to estimate this model on a range of applications.

Contralateral breast cancers: Independent cancers or metastases?

A cancer in the contralateral breast in a woman with a previous or synchronous breast cancer is typically considered to be an independent primary tumor. Emerging evidence suggests that in a small subset of these cases the second tumor represents a metastasis. We sought to investigate the issue using massively parallel sequencing targeting 254 genes recurrently mutated in breast cancer. We examined the tumor archives at Memorial Sloan Kettering Cancer Center for the period 1995-2006 to identify cases of contralateral breast cancer where surgery for both tumors was performed at the Center. We report results from 49 patients successfully analyzed by a targeted massively parallel sequencing assay. Somatic mutations and copy number alterations were defined by state-of-the-art algorithms. Clonal relatedness was evaluated by statistical tests specifically designed for this purpose. We found evidence that the tumors in contralateral breasts were clonally related in three cases (6%) on the basis of matching mutations at codons where somatic mutations are rare. Clinical data and the presence of similar patterns of gene copy number alterations were consistent with metastasis for all three cases. In three additional cases, there was a solitary matching mutation at a common PIK3CA locus. The results suggest that a subset of contralateral breast cancers represent metastases rather than independent primary tumors. Massively parallel sequencing analysis can provide important evidence to clarify the diagnosis. However, given the inter-tumor mutational heterogeneity in breast cancer, sufficiently large gene panels need to be employed to define clonality convincingly in all cases.

Estimating the probability of clonal relatedness of pairs of tumors in cancer patients.

Next generation sequencing panels are being used increasingly in cancer research to study tumor evolution. A specific statistical challenge is to compare the mutational profiles in different tumors from a patient to determine the strength of evidence that the tumors are clonally related, that is, derived from a single, founder clonal cell. The presence of identical mutations in each tumor provides evidence of clonal relatedness, although the strength of evidence from a match is related to how commonly the mutation is seen in the tumor type under investigation. This evidence must be weighed against the evidence in favor of independent tumors from non-matching mutations. In this article, we frame this challenge in the context of diagnosis using a novel random effects model. In this way, by analyzing a set of tumor pairs, we can estimate the proportion of cases that are clonally related in the sample as well as the individual diagnostic probabilities for each case. The method is illustrated using data from a study to determine the clonal relationship of lobular carcinoma in situ with subsequent invasive breast cancers, where each tumor in the pair was subjected to whole exome sequencing. The statistical properties of the method are evaluated using simulations, demonstrating that the key model parameters are estimated with only modest bias in small samples in most configurations.

Examining the common aetiology of serous ovarian cancers and basal-like breast cancers using double primaries.

BACKGROUND: The somatic molecular profiles of basal-like breast cancers and high-grade serous ovarian cancers share many similarities, leading to the hypothesis that they have similar aetiologies, in which case they should occur together in the same patient more often than expected. METHODS: We identified 545 women with double independent primary cancers of the breast and ovary reported to the California Cancer Registry from 1999 to 2013 and examined the coincidence of subtype combinations. RESULTS: For most subtype combinations the observed frequencies were similar to their expected frequencies, but in 103 observed cases vs 43.8 expected (O/E=2.35; 95% CI 1.90-2.81) a triple-negative breast tumour (typically basal-like) was matched with a serous ovarian tumour (typically high-grade). CONCLUSIONS: The results provide compelling evidence that basal-like breast cancer and high-grade serous ovarian cancer share a much more similar aetiology than breast and ovarian cancers more broadly. Further research is needed to clarify the influence of germ-line BRCA1 mutations and other risk factors on these results.

Proceedings of the third international molecular pathological epidemiology (MPE) meeting.

Molecular pathological epidemiology (MPE) is a transdisciplinary and relatively new scientific discipline that integrates theory, methods, and resources from epidemiology, pathology, biostatistics, bioinformatics, and computational biology. The underlying objective of MPE research is to better understand the etiology and progression of complex and heterogeneous human diseases with the goal of informing prevention and treatment efforts in population health and clinical medicine. Although MPE research has been commonly applied to investigating breast, lung, and colorectal cancers, its methodology can be used to study most diseases. Recent successes in MPE studies include: (1) the development of new statistical methods to address etiologic heterogeneity; (2) the enhancement of causal inference; (3) the identification of previously unknown exposure-subtype disease associations; and (4) better understanding of the role of lifestyle/behavioral factors on modifying prognosis according to disease subtype. Central challenges to MPE include the relative lack of transdisciplinary experts, educational programs, and forums to discuss issues related to the advancement of the field. To address these challenges, highlight recent successes in the field, and identify new opportunities, a series of MPE meetings have been held at the Dana-Farber Cancer Institute in Boston, MA. Herein, we share the proceedings of the Third International MPE Meeting, held in May 2016 and attended by 150 scientists from 17 countries. Special topics included integration of MPE with immunology and health disparity research. This meeting series will continue to provide an impetus to foster further transdisciplinary integration of divergent scientific fields.

A comparison of statistical methods for the study of etiologic heterogeneity.

Cancer epidemiologic research has traditionally been guided by the premise that certain diseases share an underlying etiology, or cause. However, with the rise of molecular and genomic profiling, attention has increasingly focused on identifying subtypes of disease. As subtypes are identified, it is natural to ask the question of whether they share a common etiology or in fact arise from distinct sets of risk factors. In this context, epidemiologic questions of interest include (1) whether a risk factor of interest has the same effect across all subtypes of disease and (2) whether risk factor effects differ across levels of each individual tumor marker of which the subtypes are comprised. A number of statistical models have been proposed to address these questions. In an effort to determine the similarities and differences among the proposed methods, and to identify any advantages or disadvantages, we use a simplified data example to elucidate the interpretation of model parameters and available hypothesis tests, and we perform a simulation study to assess bias in effect size, type I error, and power. The results show that when the number of tumor markers is small enough that the cross-classification of markers can be evaluated in the traditional polytomous logistic regression framework, then the statistical properties are at least as good as the more complex modeling approaches that have been proposed. The potential advantage of more complex methods is in the ability to accommodate multiple tumor markers in a model of reduced parametric dimension.

An efficient basket trial design.

The landscape for early phase cancer clinical trials is changing dramatically because of the advent of targeted therapy. Increasingly, new drugs are designed to work against a target such as the presence of a specific tumor mutation. Because typically only a small proportion of cancer patients will possess the mutational target, but the mutation is present in many different cancers, a new class of basket trials is emerging, whereby the drug is tested simultaneously in different baskets, that is, subgroups of different tumor types. Investigators desire not only to test whether the drug works but also to determine which types of tumors are sensitive to the drug. A natural strategy is to conduct parallel trials, with the drug 's effectiveness being tested separately, using for example, the popular Simon two-stage design independently in each basket. The work presented is motivated by the premise that the efficiency of this strategy can be improved by assessing the homogeneity of the baskets ' response rates at an interim analysis and aggregating the baskets in the second stage if the results suggest the drug might be effective in all or most baskets. Via simulations, we assess the relative efficiencies of the two strategies. Because the operating characteristics depend on how many tumor types are sensitive to the drug, there is no uniformly efficient strategy. However, our investigation demonstrates that substantial efficiencies are possible if the drug works in most or all baskets, at the cost of modest losses of power if the drug works in only a single basket. Copyright © 2017 John Wiley & Sons, Ltd.