ABSTRACT
Deep learning has potential to automate screening, monitoring and grading of disease in medical images. Pretraining with contrastive learning enables models to extract robust and generalisable features from natural image datasets, facilitating label-efficient downstream image analysis. However, the direct application of conventional contrastive methods to medical datasets introduces two domain-specific issues. Firstly, several image transformations which have been shown to be crucial for effective contrastive learning do not translate from the natural image to the medical image domain. Secondly, the assumption made by conventional methods, that any two images are dissimilar, is systematically misleading in medical datasets depicting the same anatomy and disease. This is exacerbated in longitudinal image datasets that repeatedly image the same patient cohort to monitor their disease progression over time. In this paper we tackle these issues by extending conventional contrastive frameworks with a novel metadata-enhanced strategy. Our approach employs widely available patient metadata to approximate the true set of inter-image contrastive relationships. To this end we employ records for patient identity, eye position (i.e. left or right) and time series information. In experiments using two large longitudinal datasets containing 170,427 retinal optical coherence tomography (OCT) images of 7912 patients with age-related macular degeneration (AMD), we evaluate the utility of using metadata to incorporate the temporal dynamics of disease progression into pretraining. Our metadata-enhanced approach outperforms both standard contrastive methods and a retinal image foundation model in five out of six image-level downstream tasks related to AMD. We find benefits in both a low-data and high-data regime across tasks ranging from AMD stage and type classification to prediction of visual acuity. Due to its modularity, our method can be quickly and cost-effectively tested to establish the potential benefits of including available metadata in contrastive pretraining.
ABSTRACT
PURPOSE: Real world evidence is crucial to understanding the diffusion of new oncologic therapies, monitoring cancer outcomes, and detecting unexpected toxicities. In practice, real world evidence is challenging to collect rapidly and comprehensively, often requiring expensive and time-consuming manual case-finding and annotation of clinical text. In this Review, we summarise recent developments in the use of artificial intelligence to collect and analyze real world evidence in oncology. METHODS: We performed a narrative review of the major current trends and recent literature in artificial intelligence applications in oncology. RESULTS: Artificial intelligence (AI) approaches are increasingly used to efficiently phenotype patients and tumors at large scale. These tools also may provide novel biological insights and improve risk prediction through multimodal integration of radiographic, pathological, and genomic datasets. Custom language processing pipelines and large language models hold great promise for clinical prediction and phenotyping. CONCLUSIONS: Despite rapid advances, continued progress in computation, generalizability, interpretability, and reliability as well as prospective validation are needed to integrate AI approaches into routine clinical care and real-time monitoring of novel therapies.
Subject(s)
Artificial Intelligence , Medical Oncology , Neoplasms , Humans , Medical Oncology/methods , Medical Oncology/trends , Neoplasms/therapyABSTRACT
BACKGROUND AND OBJECTIVE: Polygenic risk scores (PRSs) have been developed to identify men with the highest risk of prostate cancer. Our aim was to compare the performance of 16 PRSs in identifying men at risk of developing prostate cancer and then to evaluate the performance of the top-performing PRSs in differentiating individuals at risk of aggressive prostate cancer. METHODS: For this case-control study we downloaded 16 published PRSs from the Polygenic Score Catalog on May 28, 2021 and applied them to Michigan Genomics Initiative (MGI) patients. Cases were matched to the Michigan Urological Surgery Improvement Collaborative (MUSIC) registry to obtain granular clinical and pathological data. MGI prospectively enrolls patients undergoing surgery at the University of Michigan, and MUSIC is a multi-institutional registry that prospectively tracks demographic, treatment, and clinical variables. The predictive performance of each PRS was evaluated using the area under the covariate-adjusted receiver operating characteristic curve (aAUC), and the association between PRS and disease aggressiveness according to prostate biopsy data was measured using logistic regression. KEY FINDINGS AND LIMITATIONS: We included 18 050 patients in the analysis, of whom 15 310 were control subjects and 2740 were prostate cancer cases. The median age was 66.1 yr (interquartile range 59.9-71.6) for cases and 56.6 yr (interquartile range 42.6-66.7) for control subjects. The PRS performance in predicting the risk of developing prostate cancer according to aAUC ranged from 0.51 (95% confidence interval 0.51-0.53) to 0.67 (95% confidence interval 0.66-0.68). By contrast, there was no association between PRS and disease aggressiveness. CONCLUSIONS AND CLINICAL IMPLICATIONS: Prostate cancer PRSs have modest real-world performance in identifying patients at higher risk of developing prostate cancer; however, they are limited in distinguishing patients with indolent versus aggressive disease. PATIENT SUMMARY: Risk scores using data for multiple genes (called polygenic risk scores) can identify men at higher risk of developing prostate cancer. However, these scores need to be refined to be able to identify men with the highest risk for clinically significant prostate cancer.
ABSTRACT
OBJECTIVES: To develop recommendations regarding the use of weights to reduce selection bias for commonly performed analyses using electronic health record (EHR)-linked biobank data. MATERIALS AND METHODS: We mapped diagnosis (ICD code) data to standardized phecodes from 3 EHR-linked biobanks with varying recruitment strategies: All of Us (AOU; n = 244 071), Michigan Genomics Initiative (MGI; n = 81 243), and UK Biobank (UKB; n = 401 167). Using 2019 National Health Interview Survey data, we constructed selection weights for AOU and MGI to represent the US adult population more. We used weights previously developed for UKB to represent the UKB-eligible population. We conducted 4 common analyses comparing unweighted and weighted results. RESULTS: For AOU and MGI, estimated phecode prevalences decreased after weighting (weighted-unweighted median phecode prevalence ratio [MPR]: 0.82 and 0.61), while UKB estimates increased (MPR: 1.06). Weighting minimally impacted latent phenome dimensionality estimation. Comparing weighted versus unweighted phenome-wide association study for colorectal cancer, the strongest associations remained unaltered, with considerable overlap in significant hits. Weighting affected the estimated log-odds ratio for sex and colorectal cancer to align more closely with national registry-based estimates. DISCUSSION: Weighting had a limited impact on dimensionality estimation and large-scale hypothesis testing but impacted prevalence and association estimation. When interested in estimating effect size, specific signals from untargeted association analyses should be followed up by weighted analysis. CONCLUSION: EHR-linked biobanks should report recruitment and selection mechanisms and provide selection weights with defined target populations. Researchers should consider their intended estimands, specify source and target populations, and weight EHR-linked biobank analyses accordingly.
Subject(s)
Biological Specimen Banks , Electronic Health Records , Humans , Selection Bias , Female , Male , Adult , Middle Aged , Medical Record Linkage , United States , Aged , United Kingdom , MichiganABSTRACT
Trace elements are important for human health but may exert toxic or adverse effects. Mechanisms of uptake, distribution, metabolism, and excretion are partly under genetic control but have not yet been extensively mapped. Here we report a comprehensive multi-element genome-wide association study of 57 essential and non-essential trace elements. We perform genome-wide association meta-analyses of 14 trace elements in up to 6564 Scandinavian whole blood samples, and genome-wide association studies of 43 trace elements in up to 2819 samples measured only in the Trøndelag Health Study (HUNT). We identify 11 novel genetic loci associated with blood concentrations of arsenic, cadmium, manganese, selenium, and zinc in genome-wide association meta-analyses. In HUNT, several genome-wide significant loci are also indicated for other trace elements. Using two-sample Mendelian randomization, we find several indications of weak to moderate effects on health outcomes, the most precise being a weak harmful effect of increased zinc on prostate cancer. However, independent validation is needed. Our current understanding of trace element-associated genetic variants may help establish consequences of trace elements on human health.
Subject(s)
Selenium , Trace Elements , Male , Humans , Trace Elements/metabolism , Genome-Wide Association Study , Zinc , Selenium/analysis , ManganeseABSTRACT
Background and Objectives: Amyotrophic lateral sclerosis (ALS) causes profound impairments in neurological function and a cure for this devastating disease remains elusive. Early detection and risk stratification are crucial for timely intervention and improving patient outcomes. This study aimed to identify predisposing genetic, phenotypic, and exposure-related factors for Amyotrophic lateral sclerosis using multi-modal data and assess their joint predictive potential. Methods: Utilizing data from the UK Biobank, we analyzed an unrelated set of 292 ALS cases and 408,831 controls of European descent. Two polygenic risk scores (PRS) are constructed: "GWAS Hits PRS" and "PRS-CS," reflecting oligogenic and polygenic ALS risk profiles, respectively. Time-restricted phenome-wide association studies (PheWAS) were performed to identify pre-existing conditions increasing ALS risk, integrated into phenotypic risk scores (PheRS). A poly-exposure score ("PXS") captures the influence of environmental exposures measured through survey questionnaires. We evaluate the performance of these scores for predicting ALS incidence and stratifying risk, adjusting for baseline demographic covariates. Results: Both PRSs modestly predicted ALS diagnosis, but with increased predictive power when combined (covariate-adjusted receiver operating characteristic [AAUC] = 0.584 [0.525, 0.639]). PheRS incorporated diagnoses 1 year before ALS onset (PheRS1) modestly discriminated cases from controls (AAUC = 0.515 [0.472, 0.564]). The "PXS" did not significantly predict ALS. However, a model incorporating PRSs and PheRS1 improved prediction of ALS (AAUC = 0.604 [0.547, 0.667]), outperforming a model combining all risk scores. This combined risk score identified the top 10% of risk score distribution with a 4-fold higher ALS risk (95% CI: [2.04, 7.73]) versus those in the 40%-60% range. Discussions: By leveraging UK Biobank data, our study uncovers predisposing ALS factors, highlighting the improved effectiveness of multi-factorial prediction models to identify individuals at highest risk for ALS.
ABSTRACT
BACKGROUND: Genetic variants can contribute differently to trait heritability by their functional categories, and recent studies have shown that incorporating functional annotation can improve the predictive performance of polygenic risk scores (PRSs). In addition, when only a small proportion of variants are causal variants, PRS methods that employ a Bayesian framework with shrinkage can account for such sparsity. It is possible that the annotation group level effect is also sparse. However, the number of PRS methods that incorporate both annotation information and shrinkage on effect sizes is limited. We propose a PRS method, PRSbils, which utilizes the functional annotation information with a bilevel continuous shrinkage prior to accommodate the varying genetic architectures both on the variant-specific level and on the functional annotation level. RESULTS: We conducted simulation studies and investigated the predictive performance in settings with different genetic architectures. Results indicated that when there was a relatively large variability of group-wise heritability contribution, the gain in prediction performance from the proposed method was on average 8.0% higher AUC compared to the benchmark method PRS-CS. The proposed method also yielded higher predictive performance compared to PRS-CS in settings with different overlapping patterns of annotation groups and obtained on average 6.4% higher AUC. We applied PRSbils to binary and quantitative traits in three real world data sources (the UK Biobank, the Michigan Genomics Initiative (MGI), and the Korean Genome and Epidemiology Study (KoGES)), and two sources of annotations: ANNOVAR, and pathway information from the Kyoto Encyclopedia of Genes and Genomes (KEGG), and demonstrated that the proposed method holds the potential for improving predictive performance by incorporating functional annotations. CONCLUSIONS: By utilizing a bilevel shrinkage framework, PRSbils enables the incorporation of both overlapping and non-overlapping annotations into PRS construction to improve the performance of genetic risk prediction. The software is available at https://github.com/styvon/PRSbils .
Subject(s)
Genetic Predisposition to Disease , Genome-Wide Association Study , Humans , Bayes Theorem , Genome-Wide Association Study/methods , Multifactorial Inheritance , Software , Risk FactorsABSTRACT
Objective: To explore the role of selection bias adjustment by weighting electronic health record (EHR)-linked biobank data for commonly performed analyses. Materials and methods: We mapped diagnosis (ICD code) data to standardized phecodes from three EHR-linked biobanks with varying recruitment strategies: All of Us (AOU; n=244,071), Michigan Genomics Initiative (MGI; n=81,243), and UK Biobank (UKB; n=401,167). Using 2019 National Health Interview Survey data, we constructed selection weights for AOU and MGI to be more representative of the US adult population. We used weights previously developed for UKB to represent the UKB-eligible population. We conducted four common descriptive and analytic tasks comparing unweighted and weighted results. Results: For AOU and MGI, estimated phecode prevalences decreased after weighting (weighted-unweighted median phecode prevalence ratio [MPR]: 0.82 and 0.61), while UKB's estimates increased (MPR: 1.06). Weighting minimally impacted latent phenome dimensionality estimation. Comparing weighted versus unweighted PheWAS for colorectal cancer, the strongest associations remained unaltered and there was large overlap in significant hits. Weighting affected the estimated log-odds ratio for sex and colorectal cancer to align more closely with national registry-based estimates. Discussion: Weighting had limited impact on dimensionality estimation and large-scale hypothesis testing but impacted prevalence and association estimation more. Results from untargeted association analyses should be followed by weighted analysis when effect size estimation is of interest for specific signals. Conclusion: EHR-linked biobanks should report recruitment and selection mechanisms and provide selection weights with defined target populations. Researchers should consider their intended estimands, specify source and target populations, and weight EHR-linked biobank analyses accordingly.
ABSTRACT
Importance: Posttraumatic stress disorder (PTSD) has been reported to be a risk factor for several physical and somatic symptoms. However, the genetics of PTSD and its potential association with medical outcomes remain unclear. Objective: To examine disease categories and laboratory tests from electronic health records (EHRs) that are associated with PTSD polygenic scores. Design, Setting, and Participants: This genetic association study was conducted from July 15, 2021, to January 24, 2023, using EHR data from participants across 4 biobanks. The polygenic scores of PTSD symptom severity (PGS-PTSD) were tested with all available phecodes in Vanderbilt University Medical Center's biobank (BioVU), Mass General Brigham (MGB), Michigan Genomics Initiative (MGI), and UK Biobank (UKBB). The significant medical outcomes were tested for overrepresented disease categories and subsequently tested for genetic correlation and 2-sample mendelian randomization (MR) to determine genetically informed associations. Multivariable MR was conducted to assess whether PTSD associations with health outcomes were independent of the genetic effect of body mass index and tobacco smoking. Exposures: Polygenic score of PTSD symptom severity. Main Outcomes and Measures: A total of 1680 phecodes (ie, International Classification of Diseases, Ninth Revision- and Tenth Revision-based phenotypic definitions of health outcomes) across 4 biobanks and 490 laboratory tests across 2 biobanks (BioVU and MGB). Results: In this study including a total of 496â¯317 individuals (mean [SD] age, 56.8 [8.0] years; 263â¯048 female [53%]) across the 4 EHR sites, meta-analyzing associations of PGS-PTSD with 1680 phecodes from 496â¯317 individuals showed significant associations to be overrepresented from mental health disorders (fold enrichment = 3.15; P = 5.81 × 10-6), circulatory system (fold enrichment = 3.32; P = 6.39 × 10-12), digestive (fold enrichment = 2.42; P = 2.16 × 10-7), and respiratory outcomes (fold enrichment = 2.51; P = 8.28 × 10-5). The laboratory measures scan with PGS-PTSD in BioVU and MGB biobanks revealed top associations in metabolic and immune domains. MR identified genetic liability to PTSD symptom severity as an associated risk factor for 12 health outcomes, including alcoholism (ß = 0.023; P = 1.49 × 10-4), tachycardia (ß = 0.045; P = 8.30 × 10-5), cardiac dysrhythmias (ß = 0.016, P = 3.09 × 10-5), and acute pancreatitis (ß = 0.049, P = 4.48 × 10-4). Several of these associations were robust to genetic effects of body mass index and smoking. We observed a bidirectional association between PTSD symptoms and nonspecific chest pain and C-reactive protein. Conclusions and Relevance: Results of this study suggest the broad health repercussions associated with the genetic liability to PTSD across 4 biobanks. The circulatory and respiratory systems association was observed to be overrepresented in all 4 biobanks.
Subject(s)
Cardiovascular Diseases , Pancreatitis , Stress Disorders, Post-Traumatic , Humans , Female , Middle Aged , Stress Disorders, Post-Traumatic/genetics , Stress Disorders, Post-Traumatic/psychology , Acute Disease , Risk Factors , Genome-Wide Association StudyABSTRACT
Genome-wide association studies have contributed extensively to the discovery of disease-associated common variants. However, the genetic contribution to complex traits is still largely difficult to interpret. We report a genome-wide association study of 2394 cases and 2393 controls for age-related macular degeneration (AMD) via whole-genome sequencing, with 46.9 million genetic variants. Our study reveals significant single-variant association signals at four loci and independent gene-based signals in CFH, C2, C3, and NRTN. Using data from the Exome Aggregation Consortium (ExAC) for a gene-based test, we demonstrate an enrichment of predicted rare loss-of-function variants in CFH, CFI, and an as-yet unreported gene in AMD, ORMDL2. Our method of using a large variant list without individual-level genotypes as an external reference provides a flexible and convenient approach to leverage the publicly available variant datasets to augment the search for rare variant associations, which can explain additional disease risk in AMD.
Subject(s)
Genome-Wide Association Study , Macular Degeneration , Humans , Genome-Wide Association Study/methods , Macular Degeneration/genetics , Genotype , Genetic Testing , Whole Genome Sequencing , Polymorphism, Single Nucleotide/genetics , Genetic Predisposition to Disease , Complement Factor H/geneticsABSTRACT
There is a need to identify accurately prognostic factors that determine the progression of intermediate to late-stage age-related macular degeneration (AMD). Currently, clinicians cannot provide individualised prognoses of disease progression. Moreover, enriching clinical trials with rapid progressors may facilitate delivery of shorter intervention trials aimed at delaying or preventing progression to late AMD. Thus, we performed a systematic review to outline and assess the accuracy of reporting prognostic factors for the progression of intermediate to late AMD. A meta-analysis was originally planned. Synonyms of AMD and disease progression were used to search Medline and EMBASE for articles investigating AMD progression published between 1991 and 2021. Initial search results included 3229 articles. Predetermined eligibility criteria were employed to systematically screen papers by two reviewers working independently and in duplicate. Quality appraisal and data extraction were performed by a team of reviewers. Only 6 studies met the eligibility criteria. Based on these articles, exploratory prognostic factors for progression of intermediate to late AMD included phenotypic features (e.g. location and size of drusen), age, smoking status, ocular and systemic co-morbidities, race, and genotype. Overall, study heterogeneity precluded reporting by forest plots and meta-analysis. The most commonly reported prognostic factors were baseline drusen volume/size, which was associated with progression to neovascular AMD, and outer retinal thinning linked to progression to geographic atrophy. In conclusion, poor methodological quality of included studies warrants cautious interpretation of our findings. Rigorous studies are warranted to provide robust evidence in the future.
Subject(s)
Retinal Drusen , Wet Macular Degeneration , Humans , Prognosis , Angiogenesis Inhibitors , Disease Progression , Visual Acuity , Vascular Endothelial Growth Factor AABSTRACT
BACKGROUND: Post-Acute Sequelae of COVID-19 (PASC) have emerged as a global public health and healthcare challenge. This study aimed to uncover predictive factors for PASC from multi-modal data to develop a predictive model for PASC diagnoses. METHODS: We analyzed electronic health records from 92,301 COVID-19 patients, covering medical phenotypes, medications, and lab results. We used a Super Learner-based prediction approach to identify predictive factors. We integrated the model outputs into individual and composite risk scores and evaluated their predictive performance. RESULTS: Our analysis identified several factors predictive of diagnoses of PASC, including being overweight/obese and the use of HMG CoA reductase inhibitors prior to COVID-19 infection, and respiratory system symptoms during COVID-19 infection. We developed a composite risk score with a moderate discriminatory ability for PASC (covariate-adjusted AUC (95% confidence interval): 0.66 (0.63, 0.69)) by combining the risk scores based on phenotype and medication records. The combined risk score could identify 10% of individuals with a 2.2-fold increased risk for PASC. CONCLUSIONS: We identified several factors predictive of diagnoses of PASC and integrated the information into a composite risk score for PASC prediction, which could contribute to the identification of individuals at higher risk for PASC and inform preventive efforts.
ABSTRACT
OBJECTIVE: To overcome the limitations associated with the collection and curation of COVID-19 outcome data in biobanks, this study proposes the use of polygenic risk scores (PRS) as reliable proxies of COVID-19 severity across three large biobanks: the Michigan Genomics Initiative (MGI), UK Biobank (UKB), and NIH All of Us. The goal is to identify associations between pre-existing conditions and COVID-19 severity. METHODS: Drawing on a sample of more than 500,000 individuals from the three biobanks, we conducted a phenome-wide association study (PheWAS) to identify associations between a PRS for COVID-19 severity, derived from a genome-wide association study on COVID-19 hospitalization, and clinical pre-existing, pre-pandemic phenotypes. We performed cohort-specific PRS PheWAS and a subsequent fixed-effects meta-analysis. RESULTS: The current study uncovered 23 pre-existing conditions significantly associated with the COVID-19 severity PRS in cohort-specific analyses, of which 21 were observed in the UKB cohort and two in the MGI cohort. The meta-analysis yielded 27 significant phenotypes predominantly related to obesity, metabolic disorders, and cardiovascular conditions. After adjusting for body mass index, several clinical phenotypes, such as hypercholesterolemia and gastrointestinal disorders, remained associated with an increased risk of hospitalization following COVID-19 infection. CONCLUSION: By employing PRS as a proxy for COVID-19 severity, we corroborated known risk factors and identified novel associations between pre-existing clinical phenotypes and COVID-19 severity. Our study highlights the potential value of using PRS when actual outcome data may be limited or inadequate for robust analyses.
Subject(s)
COVID-19 , Population Health , Humans , Genome-Wide Association Study , Genetic Risk Score , COVID-19/genetics , Biological Specimen Banks , Preexisting Condition Coverage , Risk Factors , Genetic Predisposition to DiseaseABSTRACT
We investigated the design and analysis of observational booster vaccine effectiveness (VE) studies by performing a scoping review of booster VE literature with a focus on study design and analytic choices. We then applied 20 different approaches, including those found in the literature, to a single dataset from Michigan Medicine. We identified 80 studies in our review, including over 150 million observations in total. We found that while protection against infection is variable and dependent on several factors including the study population and time period, both monovalent boosters and particularly the bivalent booster offer strong protection against severe COVID-19. In addition, VE analyses with a severe disease outcome (hospitalization, intensive care unit admission, or death) appear to be more robust to design and analytic choices than an infection endpoint. In terms of design choices, we found that test-negative designs and their variants may offer advantages in statistical efficiency compared to cohort designs.
Subject(s)
COVID-19 , Humans , COVID-19/epidemiology , COVID-19/prevention & control , Hospitalization , Intensive Care Units , Michigan/epidemiology , Observational Studies as TopicABSTRACT
Purpose: To investigate and compare novel volumetric microperimetry (MP)-derived metrics in intermediate age-related macular degeneration (iAMD), as current MP metrics show high variability and low sensitivity. Methods: This is a cross-sectional analysis of microperimetry baseline data from the multicenter, prospective PINNACLE study (ClinicalTrials.gov NCT04269304). The Visual Field Modeling and Analysis (VFMA) software and an open-source implementation (OSI) were applied to calculate MP-derived hill-of-vison (HOV) surface plots and the total volume (VTOT) beneath the plots. Bland-Altman plots were used for methodologic comparison, and the association of retinal sensitivity metrics with explanatory variables was tested with mixed-effects models. Results: In total, 247 eyes of 189 participants (75 ± 7.3 years) were included in the analysis. The VTOT output of VFMA and OSI exhibited a significant difference (P < 0.0001). VFMA yielded slightly higher coefficients of determination than OSI and mean sensitivity (MS) in univariable and multivariable modeling, for example, in association with low-luminance visual acuity (LLVA) (marginal R2/conditional R2: VFMA 0.171/0.771, OSI 0.162/0.765, MS 0.133/0.755). In the multivariable analysis, LLVA was the only demonstrable predictor of VFMA VTOT (t-value, P-value: -7.5, <0.001) and MS (-6.5, <0.001). Conclusions: The HOV-derived metric of VTOT exhibits favorable characteristics compared to MS in evaluating retinal sensitivity. The output of VFMA and OSI is not exactly interchangeable in this cross-sectional analysis. Longitudinal analysis is necessary to assess their performance in ability-to-detect change. Translational Relevance: This study explores new volumetric MP endpoints for future application in therapeutic trials in iAMD and reports specific characteristics of the available HOV software applications.
Subject(s)
Benchmarking , Macular Degeneration , Humans , Cross-Sectional Studies , Prospective Studies , Visual Field Tests , Macular Degeneration/diagnosis , Retina/diagnostic imagingABSTRACT
The Epidemiologic Questionnaire (EPI-Q) was established to collect broad, uniform, self-reported health data to supplement electronic health record (EHR) and genotype information from participants in the University of Michigan (UM) Precision Health cohorts. Recruitment of EPI-Q participants, who were already enrolled in 1 of 3 ongoing UM Precision Health cohorts-the Michigan Genomics Initiative, Mental Health Biobank, and Metabolism, Endocrinology, and Diabetes cohorts-began in March 2020. Of 54,043 retrospective invitations, 5,577 individuals enrolled, representing a 10.3% response rate. Of these, 3,502 (63.7%) were female, and the average age was 56.1 years (standard deviation, 15.4). The baseline survey comprises 11 modules on topics including personal and family health history, lifestyle, and cancer screening and history. Additionally, 11 optional modules cover topics including financial toxicity, occupational exposure, and life meaning. The questions are based on standardized and validated instruments used in other cohorts, and we share resources to expedite development of similar surveys. Data are collected via the MyDataHelps platform, which enables current and future participants to share non-Michigan Medicine EHR data. Recruitment is ongoing. Cohort data are available to those with institutional review board approval; for details, contact the Data Office for Clinical and Translational Research (DataOffice@umich.edu).
Subject(s)
Electronic Health Records , Mobile Applications , Humans , Female , Middle Aged , Male , Retrospective Studies , Genotype , Surveys and Questionnaires , Health SurveysABSTRACT
Background: Observational vaccine effectiveness (VE) studies based on real-world data are a crucial supplement to initial randomized clinical trials of Coronavirus Disease 2019 (COVID-19) vaccines. However, there exists substantial heterogeneity in study designs and statistical methods for estimating VE. The impact of such heterogeneity on VE estimates is not clear. Methods: We conducted a two-step literature review of booster VE: a literature search for first or second monovalent boosters on January 1, 2023, and a rapid search for bivalent boosters on March 28, 2023. For each study identified, study design, methods, and VE estimates for infection, hospitalization, and/or death were extracted and summarized via forest plots. We then applied methods identified in the literature to a single dataset from Michigan Medicine (MM), providing a comparison of the impact of different statistical methodologies on the same dataset. Results: We identified 53 studies estimating VE of the first booster, 16 for the second booster. Of these studies, 2 were case-control, 17 were test-negative, and 50 were cohort studies. Together, they included nearly 130 million people worldwide. VE for all outcomes was very high (around 90%) in earlier studies (i.e., in 2021), but became attenuated and more heterogeneous over time (around 40%-50% for infection, 60%-90% for hospitalization, and 50%-90% for death). VE compared to the previous dose was lower for the second booster (10-30% for infection, 30-60% against hospitalization, and 50-90% against death). We also identified 11 bivalent booster studies including over 20 million people. Early studies of the bivalent booster showed increased effectiveness compared to the monovalent booster (VE around 50-80% for hospitalization and death).Our primary analysis with MM data using a cohort design included 186,495 individuals overall (including 153,811 boosted and 32,684 with only a primary series vaccination), and a secondary test-negative design included 65,992 individuals tested for SARS-CoV-2. When different statistical designs and methods were applied to MM data, VE estimates for hospitalization and death were robust to analytic choices, with test-negative designs leading to narrower confidence intervals. Adjusting either for the propensity of getting boosted or directly adjusting for covariates reduced the heterogeneity across VE estimates for the infection outcome. Conclusion: While the advantage of the second monovalent booster is not obvious from the literature review, the first monovalent booster and the bivalent booster appear to offer strong protection against severe COVID-19. Based on both the literature view and data analysis, VE analyses with a severe disease outcome (hospitalization, ICU admission, or death) appear to be more robust to design and analytic choices than an infection endpoint. Test-negative designs can extend to severe disease outcomes and may offer advantages in statistical efficiency when used properly.
ABSTRACT
Purpose: To study the individual course of retinal changes caused by healthy aging using deep learning. Design: Retrospective analysis of a large data set of retinal OCT images. Participants: A total of 85 709 adults between the age of 40 and 75 years of whom OCT images were acquired in the scope of the UK Biobank population study. Methods: We created a counterfactual generative adversarial network (GAN), a type of neural network that learns from cross-sectional, retrospective data. It then synthesizes high-resolution counterfactual OCT images and longitudinal time series. These counterfactuals allow visualization and analysis of hypothetical scenarios in which certain characteristics of the imaged subject, such as age or sex, are altered, whereas other attributes, crucially the subject's identity and image acquisition settings, remain fixed. Main Outcome Measures: Using our counterfactual GAN, we investigated subject-specific changes in the retinal layer structure as a function of age and sex. In particular, we measured changes in the retinal nerve fiber layer (RNFL), combined ganglion cell layer plus inner plexiform layer (GCIPL), inner nuclear layer to the inner boundary of the retinal pigment epithelium (INL-RPE), and retinal pigment epithelium (RPE). Results: Our counterfactual GAN is able to smoothly visualize the individual course of retinal aging. Across all counterfactual images, the RNFL, GCIPL, INL-RPE, and RPE changed by -0.1 µm ± 0.1 µm, -0.5 µm ± 0.2 µm, -0.2 µm ± 0.1 µm, and 0.1 µm ± 0.1 µm, respectively, per decade of age. These results agree well with previous studies based on the same cohort from the UK Biobank population study. Beyond population-wide average measures, our counterfactual GAN allows us to explore whether the retinal layers of a given eye will increase in thickness, decrease in thickness, or stagnate as a subject ages. Conclusion: This study demonstrates how counterfactual GANs can aid research into retinal aging by generating high-resolution, high-fidelity OCT images, and longitudinal time series. Ultimately, we envision that they will enable clinical experts to derive and explore hypotheses for potential imaging biomarkers for healthy and pathologic aging that can be refined and tested in prospective clinical trials. Financial Disclosures: Proprietary or commercial disclosure may be found after the references.
ABSTRACT
Background: Genetic variants can contribute differently to trait heritability by their functional categories, and recent studies have shown that incorporating functional annotation can improve the predictive performance of polygenic risk scores (PRSs). In addition, when only a small proportion of variants are causal variants, PRS methods that employ a Bayesian framework with shrinkage can account for such sparsity. It is possible that the annotation group level effect is also sparse. However, the number of PRS methods that incorporate both annotation information and shrinkage on effect sizes is limited. We propose a PRS method, PRSbils, which utilizes the functional annotation information with a bilevel continuous shrinkage prior to accommodate the varying genetic architectures both on the variant-specific level and on the functional annotation level. Results: We conducted simulation studies and investigated the predictive performance in settings with different genetic architectures. Results indicated that when there was a relatively large variability of group-wise heritability contribution, the gain in prediction performance from the proposed method was on average 8.0% higher AUC compared to the benchmark method PRS-CS. The proposed method also yielded higher predictive performance compared to PRS-CS in settings with different overlapping patterns of annotation groups and obtained on average 6.4% higher AUC. We applied PRSbils to binary and quantitative traits in three real world data sources (the UK Biobank, the Michigan Genomics Initiative (MGI), and the Korean Genome and Epidemiology Study (KoGES)), and two sources of annotations: ANNOVAR, and pathway information from the Kyoto Encyclopedia of Genes and Genomes (KEGG), and demonstrated that the proposed method holds the potential for improving predictive performance by incorporating functional annotations. Conclusions: By utilizing a bilevel shrinkage framework, PRSbils enables the incorporation of both overlapping and non-overlapping annotations into PRS construction to improve the performance of genetic risk prediction. The software is available at https://github.com/styvon/PRSbils.
ABSTRACT
Biobanks of linked clinical patient histories and biological samples are an efficient strategy to generate large cohorts for modern genetics research. Biobank recruitment varies by factors such as geographic catchment and sampling strategy, which affect biobank demographics and research utility. Here, we describe the Michigan Genomics Initiative (MGI), a single-health-system biobank currently consisting of >91,000 participants recruited primarily during surgical encounters at Michigan Medicine. The surgical enrollment results in a biobank enriched for many diseases and ideally suited for a disease genetics cohort. Compared with the much larger population-based UK Biobank, MGI has higher prevalence for nearly all diagnosis-code-based phenotypes and larger absolute case counts for many phenotypes. Genome-wide association study (GWAS) results replicate known findings, thereby validating the genetic and clinical data. Our results illustrate that opportunistic biobank sampling within single health systems provides a unique and complementary resource for exploring the genetics of complex diseases.