Large Consortia Research Articles

Mobilizing data during a crisis: Building rapid evidence pipelines using multi-institutional real world data

The COVID-19 pandemic generated tremendous interest in using real world data (RWD). Many consortia across the public and private sectors formed in 2020 with the goal of rapidly producing high-quality evidence from RWD to guide medical decision-making, public health priorities, and more. Experiences were gathered from five large consortia on rapid multi-institutional evidence generation during the COVID-19 pandemic. Insights have been compiled across five dimensions: consortium composition, governance structure and alignment of priorities, data sharing, data analysis, and evidence dissemination. The purpose of this piece is to offer guidance on building large-scale multi-institutional RWD analysis pipelines for future public health issues.The composition of each consortium was largely influenced by existing collaborations. A central set of priorities for evidence generation guided each consortium, however different approaches to governance emerged. Challenges surrounding limited access to clinical data due to various contributors were overcome in unique ways. While all consortia used different methods to construct and analyze patient cohorts ranging from centralized to federated approaches, all proved effective for generating meaningful real-world evidence. Actionable recommendations for clinical practice and public health agencies were made from translating insights from consortium analyses.Each consortium was successful in rapidly answering questions about COVID-19 diagnosis and treatment despite all taking slightly different approaches to data sharing and analysis. Leveraging RWD, leveraged in a manner that applies scientific rigor and transparency, can complement higher-level evidence and serve as an important adjunct to clinical trials to quickly guide policy and critical care, especially for a pandemic response.

Abstract 1249: cBioPortal for Cancer Genomics

Abstract cBioPortal for Cancer Genomics is an open-source platform for interactive, exploratory analysis of large-scale clinico-genomic data. cBioPortal provides a suite of user-friendly visualizations and analyses, including OncoPrints, mutation lollipop plots, variant interpretation, group comparison, survival analysis, expression correlation analysis, alteration enrichment analysis, cohort and patient-level visualization. The public site (https://www.cbioportal.org) is accessed by >35,000 unique visitors each month and hosts data from >390 studies spanning individual labs and large consortia. All data is available in the cBioPortal Datahub: https://github.com/cBioPortal/datahub; in 2023 we added 35 studies (~16,000 samples). In addition, >86 instances of cBioPortal are installed at academic institutions and companies worldwide. We also host a dedicated instance for AACR Project GENIE, enabling access to the GENIE cohort of >197,000 clinically sequenced samples from 19 institutions (https://genie.cbioportal.org). The GENIE Biopharma Collaborative (BPC) enables the collection of comprehensive clinical annotations including response, outcome, and treatment history, which can all be visualized in cBioPortal. BPC cohorts for non-small cell lung cancer (~2,000 samples) and colorectal cancer (~1,500 samples) are available, with more cancer types to come. This past year, we significantly enhanced existing features. Group comparison now includes a Mutations tab showing a mirrored lollipop plot and outcomes analysis now supports hazard ratios and landmark analysis. Arm-level copy number can now be compared across groups, as can any data in the generic assay format. We also enhanced the study view. Users can now add charts to summarize the copy number of a specific gene across the cohort. The option to add custom data charts now supports numerical data as well as categorical data. Filters in study view can now be manually submitted, which provides a performance improvement when applying multiple filters in a large cohort. Larger-scale performance improvements are currently underway. We continue to improve support for multimodal datasets incorporating derived data elements, including cell type counts and fractions per sample from imaging or single cell data. The MSK-SPECTRUM ovarian cancer study has samples profiled with bulk sequencing, scRNASeq, H&E and mpIF imaging. Through integrations with CELLxGENE (single cell data) and Minerva (imaging), users can explore these data modalities in detail in isolation and query them jointly in cBioPortal. cBioPortal is open source (https://github.com/cBioPortal/). Development is a collaborative effort among groups at Memorial Sloan Kettering Cancer Center, Dana-Farber Cancer Institute, Childrens Hospital of Philadelphia, Princess Margaret Cancer Centre, Caris Life Sciences, Bilkent University and The Hyve. We welcome open source contributions from others in the cancer research community. Citation Format: Tali Mazor, Ino de Bruijn, Rima AlHamad, Calla Chennault, Corey Dubin, Jeremy Easton-Marks, Zhaoyuan Fu, Benjamin Gross, Charles Haynes, David M. Higgins, Jason Hwee, Prasanna K. Jagannathan, Mirella Kalafati, Karthik Kalletla, James Ko, Tim Kuijpers, Sowmiyaa Kumar, Priti Kumari, Ritika Kundra, Bryan Lai, Xiang Li, James Lindsay, Aaron Lisman, Qi-Xuan Lu, Ramyasree Madupuri, Angelica Ochoa, Yusuf Z. Özgül, Oleguer Plantalech, Matthijs N. Pon, Baby A. Satravada, Jessica Singh, Selcuk Onur Sumer, Pim van Nierop, Floris Vleugels, Avery Wang, Manda Wilson, Hongxin Zhang, Gaofei Zhao, Ugur Dogrusoz, Allison Heath, Adam Resnick, Trevor J. Pugh, Chris Sander, Ethan Cerami, Jianjiong Gao, Nikolaus Schultz. cBioPortal for Cancer Genomics [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2024; Part 1 (Regular Abstracts); 2024 Apr 5-10; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2024;84(6_Suppl):Abstract nr 1249.

Hemorrhoidal disease and its genetic association with depression, bipolar disorder, anxiety disorders, and schizophrenia: a bidirectional mendelian randomization study

BackgroundHemorrhoids and psychiatric disorders exhibit high prevalence rates and a tendency for relapse in epidemiological studies. Despite this, limited research has explored their correlation, and these studies are often subject to reverse causality and residual confounding. We conducted a Mendelian randomization (MR) analysis to comprehensively investigate the association between several mental illnesses and hemorrhoidal disease.MethodsGenetic associations for four psychiatric disorders and hemorrhoidal disease were obtained from large consortia, the FinnGen study, and the UK Biobank. Genetic variants associated with depression, bipolar disorder, anxiety disorders, schizophrenia, and hemorrhoidal disease at the genome-wide significance level were selected as instrumental variables. Screening for potential confounders in genetic instrumental variables using PhenoScanner V2. Bidirectional MR estimates were employed to assess the effects of four psychiatric disorders on hemorrhoidal disease.ResultsOur analysis revealed a significant association between genetically predicted depression and the risk of hemorrhoidal disease (IVW, OR=1.20,95% CI=1.09 to 1.33, P <0.001). We found no evidence of associations between bipolar disorder, anxiety disorders, schizophrenia, and hemorrhoidal disease. Inverse MR analysis provided evidence for a significant association between genetically predicted hemorrhoidal disease and depression (IVW, OR=1.07,95% CI=1.04 to 1.11, P <0.001).ConclusionsThis study offers MR evidence supporting a bidirectional causal relationship between depression and hemorrhoidal disease.

Mapping tissue microstructure of brain white matter in vivo in health and disease using diffusion MRI

Abstract Diffusion magnetic resonance imaging offers unique in vivo sensitivity to tissue microstructure in brain white matter, which undergoes significant changes during development and is compromised in virtually every neurological disorder. Yet, the challenge is to develop biomarkers that are specific to micrometer-scale cellular features in a human MRI scan of a few minutes. Here, we quantify the sensitivity and specificity of a multicompartment diffusion modeling framework to the density, orientation, and integrity of axons. We demonstrate that using a machine learning-based estimator, our biophysical model captures the morphological changes of axons in early development, acute ischemia, and multiple sclerosis (total N = 821). The methodology of microstructure mapping is widely applicable in clinical settings and in large imaging consortium data to study development, aging, and pathology.

Red and processed meat intake and risk of cardiovascular disease: A two-sample Mendelian randomization study

Previous observational studies have yielded inconsistent findings regarding associations between red/processed meat intake and the risk of cardiovascular disease (CVD). Some studies have suggested positive relationships, while others have demonstrated no significant associations. However, causal effects remain uncertain. This 2023 Mendelianrandomization (MR) study investigated the causal relationship between red and processed meat (porkmeat, mutton meat, beef meat)intake and CVD risk by analyzing summary data from the UK Biobank (exposure), CARDIoGRAMplusC4D (coronary artery disease [CAD]), MEGASTROKE (stroke), Nielsen etal. (atrial fibrillation [AF]), HERMES (heart failure [HF]), and FinnGen (cardiovascular outcomes) public databases. Genome-wide association studies (GWAS) of red meat (pork, beef, and mutton) and processed meat were sourced from the United Kingdom (UK) Biobank. GWAS data on CVD for this study were obtained from the Gene and FinnGen consortia. The primary method employed for the two-sample MR analysis was inverse variance weighting (IVW). Sensitivity analysis was performed to assess the reliability and consistency of the results. Genetically predicted red and processed meat consumption did not demonstrate a causal association with any CVD outcomes when employing the IVW method. For processed meat intake, the odds ratios (ORs) (95% confidence intervals CIs) in large consortia were as follows: 0.88 (0.56-1.39) for CAD, 0.91 (0.65-1.27) for AF, 0.84 (0.58-1.21) for HF, and 1.00 (0.75-1.05) for stroke. In FinnGen, the ORs were as follows: 1.15 (0.83-1.59) for CAD, 1.25 (0.75-2.07) for AF, 1.09 (0.73-1.64) for HF, and 1.27 (0.85-1.91) for stroke. For beef intake, the ORs (95% CIs) in large consortia were as follows: 0.70 (0.28-1.73) for CAD, 0.85 (0.49-1.49) for AF, 0.80 (0.35-1.83) for HF, and 1.29 (0.85-1.95) for stroke. In FinnGen, the ORs were as follows: 2.01 (0.75-5.39) for CAD, 1.83 (0.60-5.56) for AF, 0.80 (0.30-2.13) for HF, and 1.30 (0.62-2.73) for stroke. For pork intake, the ORs (95% CIs) in large consortia were as follows: 1.25 (0.37-4.22) for CAD, 1.26 (0.73-2.15) for AF, 1.71 (0.86-3.39) for HF, and 1.15 (0.63-2.11) for stroke. In FinnGen, the ORs were as follows: 1.12 (0.43-2.88) for CAD, 0.39 (0.08-1.83) for AF, 0.62 (0.20-1.88) for HF, and 0.60 (0.21-1.65) for stroke. For mutton intake, the ORs (95% CIs) in large consortia were as follows: 0.84 (0.48-1.44) for CAD, 0.84 (0.56-1.26) for AF, 1.04 (0.65-1.67) for HF, and 1.06 (0.77-1.45) for stroke. In FinnGen, the ORs were as follows: 1.20 (0.65-2.21) for CAD, 0.92 (0.44-1.92) for AF, 0.74 (0.34-1.58) for HF, and 0.75 (0.45-1.24) for stroke. The results remained robust and consistent in both the meta-analysis and supplementary MR analysis. This MR study demonstrated no significant causal relationships between red/processed meat intake and the risk of the four CVD outcomes examined. Further investigation is warranted to confirm these findings.

Metrics reloaded: recommendations for image analysis validation.

Increasing evidence shows that flaws in machine learning (ML) algorithm validation are an underestimated global problem. In biomedical image analysis, chosen performance metrics often do not reflect the domain interest, and thus fail to adequately measure scientific progress and hinder translation of ML techniques into practice. To overcome this, we created Metrics Reloaded, a comprehensive framework guiding researchers in the problem-aware selection of metrics. Developed by a large international consortium in a multistage Delphi process, it is based on the novel concept of a problem fingerprint-a structured representation of the given problem that captures all aspects that are relevant for metric selection, from the domain interest to the properties of the target structure(s), dataset and algorithm output. On the basis of the problem fingerprint, users are guided through the process of choosing and applying appropriate validation metrics while being made aware of potential pitfalls. Metrics Reloaded targets image analysis problems that can be interpreted as classification tasks at image, object or pixel level, namely image-level classification, object detection, semantic segmentation and instance segmentation tasks. To improve the user experience, we implemented the framework in the Metrics Reloaded online tool. Following the convergence of ML methodology across application domains, Metrics Reloaded fosters the convergence of validation methodology. Its applicability is demonstrated for various biomedical use cases.

Viral load kinetics and the clinical consequences of cytomegalovirus in kidney transplantation.

Despite advances in clinical management, cytomegalovirus (CMV) infection remains a serious complication and an important cause of morbidity and mortality following kidney transplantation. Here, we explore the importance of viral load kinetics as predictors of risk and potential guides to therapy to reduce transplant failure in a large longitudinal Genome Canada Transplant Consortium (GCTC) kidney transplant cohort. We examined the relationship between CMV infection rates and clinical characteristics, CMV viral load kinetics, and graft and patient outcomes in 2510 sequential kidney transplant recipients in the British Columbia Transplant Program. Transplants were performed between January 1, 2008, and December 31, 2018, were managed according to a standard protocol, and were followed until December 31, 2019, representing over 3.4 million days of care. Longitudinal CMV testing was performed in 2464 patients, of whom 434 (17.6%) developed a first episode of CMV viremia at a median of 120 (range: 9-3906) days post-transplant. Of these patients, 93 (21.4%) had CMV viremia only and 341 (78.6%) had CMV viremia with clinical complications, of whom 21 (4.8%) had resulting hospitalization. A total of 279 (11.3%) patients died and 177 (7.2%) patients lost their graft during the 12 years of follow-up. Patients with CMV infection were at significantly greater risk of graft loss (p=0.0041) and death (p=0.0056) than those without. Peak viral load ranged from 2.9 to 7.0 (median: 3.5) log10 IU/mL, the duration of viremia from 2 to 100 (15) days, and the viral load area under the curve from 9.4 to 579.8 (59.7) log10 IU/mL × days. All three parameters were closely inter-related and were significantly increased in patients with more severe clinical disease or with graft loss (p=0.001). Duration of the first CMV viremic episode greater than 15 days or a peak viral load ≥4.0 log10 IU/mL offered simple predictors of clinical risk with a 3-fold risk of transplant failure. Viral load kinetics are closely related to CMV severity and to graft loss following kidney transplantation and provide a simple index of risk which may be valuable in guiding trials and treatment to prevent transplant failure.

Body Mass Index and Risk of Female Reproductive System Tumors Subtypes: A Meta-Analysis Using Mendelian Randomization.

Introduction: A strong association was previously established between body mass index (BMI) and female reproductive system tumors; however, the causal relationship is unclear. We conducted a Mendelian randomization (MR) study to further explore this association. Methods: Genetic information for BMI was retrieved from a published genome-wide association study involving 339,224 participants. Genetic associations with five common female reproductive system tumors were obtained from the FinnGen, UK Biobank studies, and other large consortia. Results: Genetic predisposition towards BMI exhibits a significant association with multiple tumors of the female reproductive system. Specifically, for every 1-unit increase in BMI log-transformed odds ratio (OR). The OR fluctuations overall for patients with breast cancer ranged from 0.661 to 0.996 (95% confidence interval [CI],0.544-1.000, P < 0.05). When stratified by estrogen receptor (ER) status, the OR for patients with ER (+) breast cancer ranged from 0.782 to 0.844 (95% CI, 0.616-0.994, P < 0.05) and that for those with ER (-) breast cancer ranged from 0.663 to 0.789 (95% CI, 0.498-0.991, P < 0.05). Additionally, ORs were as follows for cancer types: 1.577-1.908 (95% CI, 1.049-2.371, P < 0.05) for endometrial carcinoma; 1.216-1.303 (95% CI, 1.021-1.591, P < 0.05) for high-grade serous ovarian cancer; 1.217 (95% CI, 1.034-1.432, P < 0.05) for low-grade malignant serous ovarian cancer; and 1.502 (95% CI, 1.112-2.029, P < 0.05) for endometrioid ovarian carcinoma. Furthermore, our findings indicated that genetic predisposition towards BMI did not exhibit a causal association with uterine fibroids, cervical precancerous lesions, or cervical cancer itself. Conclusion: A genetic association was established between a high BMI and high risk of developing multiple tumors of the female reproductive system and their associated subtypes. This underscores the significance of taking measures to prevent reproductive system tumors in women who have a high BMI.

Mendelian randomization study reveals the relationship between dietary factors and respiratory diseases

The existence of causal relationship between dietary factors and respiratory diseases is uncertain. We comprehensively investigated the association between dietary factors and respiratory diseases by using Mendelian randomization (MR). Genetic variants linked to dietary factors were selected as instrumental variables with genome-wide significance. These instrumental variables were obtained from large GWAS databases. These databases include Biobank, the FinnGen study, and other large consortia. We used multivariate MR analyses to control the effects of smoking and education. Median analysis was conducted to evaluate whether body mass index (BMI) played a role in dietary factors in respiratory diseases. Dried fruit intake was found to be associated with a decreased risk of chronic obstructive pulmonary disease (COPD) (OR: 0.211; 95% CI 0.117–0.378; P < 0.001) and asthma (OR: 0.539; 95% CI 0.357–0.815; P = 0.003). Conversely, pork intake was associated with an increased risk of international pharmaceutical federation (IPF) (OR: 1.051*102, 95% CI 4.354–2.56*103, P = 0.004). However, no significant associations were observed between the 20 dietary factors and obstructive sleep apnea (OSA). In addition, multivariate MR analyses showed that the above results were unchanged in smoking and nonsmoking populations, while the effect of dried fruit intake on asthma was significantly attenuated after corrective education. The results of the mediator variable analysis indicated that BMI could serve as a mediator of the above results. This study found that dried fruits slowed the progression of COPD and asthma, while pork promoted IPF. However, no effect of dietary factors on OSA was found. Meanwhile, we showed that the above results were unchanged in smoking and non-smoking populations. In contrast, education could influence the role of diet on asthma, and BMI could be used as a mediating variable to influence the above results.

Prediction Models for Glaucoma in a Multicenter Electronic Health Records Consortium: The Sight Outcomes Research Collaborative

ObjectiveAdvances in artificial intelligence have enabled the development of predictive models for glaucoma. However, most work is single-center and uncertainty exists regarding the generalizability of such models. The purpose of this study was to build and evaluate machine learning (ML) approaches to predict glaucoma progression requiring surgery using data from a large multicenter consortium of electronic health records (EHR). DesignCohort study. Participants36,548 patients with glaucoma, as identified by ICD codes from six academic eye centers participating in the Sight OUtcomes Research Collaborative (SOURCE). MethodsWe developed machine learning models to predict whether glaucoma patients would progress to glaucoma surgery in the coming year (identified by CPT codes) using the following modeling approaches: 1) penalized logistic regression (lasso, ridge, and elastic net); 2) tree-based models (random forest, gradient boosted machines, XGBoost), and 3) deep learning models. Model input features included demographics, diagnosis codes, medications, and clinical information (intraocular pressure, visual acuity, refractive status, and central corneal thickness) available from structured EHR data. One site was reserved as an “external site” test set (N=1550); of the patients from the remaining sites, 10% each were randomly selected to be in development and test sets, with the remaining 27999 reserved for model training. Main Outcome and MeasuresEvaluation metrics included area under the receiver operating characteristic curve (AUROC) on the test set and the external site. Results6019 (16.5%) of 36,548 patients underwent glaucoma surgery. Overall, the AUROC ranged from 0.735-0.771 on the random test set and from 0.706-0.754 on the external test site, with the XGBoost and random forest model performing best, respectively. There was greatest performance decrease from the random test set to the external test site for the penalized regression models. ConclusionsML models developed using structured EHR data can reasonably predict whether glaucoma patients will need surgery, with reasonable generalizability to an external site. Additional research is needed to investigate the impact of protected class characteristics such as race or gender on model performance and fairness.

Role of Polyunsaturated Fat in Modifying Cardiovascular Risk Associated With Family History of Cardiovascular Disease: Pooled De Novo Results From 15 Observational Studies.

It is unknown whether dietary intake of polyunsaturated fatty acids (PUFA) modifies the cardiovascular disease (CVD) risk associated with a family history of CVD. We assessed interactions between biomarkers of low PUFA intake and a family history in relation to long-term CVD risk in a large consortium. Blood and tissue PUFA data from 40 885 CVD-free adults were assessed. PUFA levels ≤25th percentile were considered to reflect low intake of linoleic, alpha-linolenic, and eicosapentaenoic/docosahexaenoic acids (EPA/DHA). Family history was defined as having ≥1 first-degree relative who experienced a CVD event. Relative risks with 95% CI of CVD were estimated using Cox regression and meta-analyzed. Interactions were assessed by analyzing product terms and calculating relative excess risk due to interaction. After multivariable adjustments, a significant interaction between low EPA/DHA and family history was observed (product term pooled RR, 1.09 [95% CI, 1.02-1.16]; P=0.01). The pooled relative risk of CVD associated with the combined exposure to low EPA/DHA, and family history was 1.41 (95% CI, 1.30-1.54), whereas it was 1.25 (95% CI, 1.16-1.33) for family history alone and 1.06 (95% CI, 0.98-1.14) for EPA/DHA alone, compared with those with neither exposure. The relative excess risk due to interaction results indicated no interactions. A significant interaction between biomarkers of low EPA/DHA intake, but not the other PUFA, and a family history was observed. This novel finding might suggest a need to emphasize the benefit of consuming oily fish for individuals with a family history of CVD.

NIA Genetics of Alzheimer’s Disease Data Storage Site (NIAGADS): 2023 Update

AbstractBackgroundNIAGADS is a national genomics data repository that facilitates access of genotypic and sequencing data to qualified investigators for the study of the genetics of Alzheimer’s disease (AD) and related neurological diseases. Collaborations with large consortia and centers such as the Alzheimer’s Disease Genetics Consortium (ADGC), Cohorts for Heart and Aging Research in Genomic Epidemiology (CHARGE) Consortium, the Alzheimer’s Disease Sequencing Project (ADSP), and the Genome Center for Alzheimer’s Disease (GCAD) allow NIAGADS to lead the effort in managing large AD datasets that can be easily accessed and fully utilized by the research community.MethodNIAGADS is supported by National Institute on Aging (NIA) under a cooperative agreement. All data derived from NIA funded AD genetics studies are expected to be deposited in NIAGADS or another NIA approved site. NIAGADS manages a Data Sharing Service (DSS) that facilitates the deposition and sharing of genomic data and association results with approved users in the neurodegenerative research community. In addition, researchers are able to freely use the NIAGADS Alzheimer’s Genomics Database (www.niagads.org/genomics/) to search annotation resources that link published AD studies to AD‐relevant sequence features and genome‐wide annotations.ResultAs of January 2023, NIAGADS houses 93 datasets comprised of &gt;149,000 samples including GWAS, sequencing, gene expression, annotations, deep phenotypes, and summary statistics. Qualified investigators can retrieve ADSP sequencing data with ease and flexibility through the NIAGADS DSS. To date, the ADSP and other contributing studies have completed whole exome sequencing (WES) of 20,503 samples and whole genome sequencing (WGS) of 36,361 samples. Raw WES and WGS files, quality controlled VCF files, and phenotype data files are available via qualified access. The next round of sequencing currently underway will generate around 30,000 additional genomes to be released in late 2023.ConclusionNIAGADS is a rich resource for AD researchers, with the goal of facilitating advances in Alzheimer’s genetics research. By housing datasets from many projects and institutions, NIAGADS enables AD researchers to meet their research goals more efficiently. Datasets, guidelines, and features are available on our website at https://www.niagads.org.

Novel scores relevant to antimicrobial resistance and hospital-acquired infections developed with data from a multi-hospital consortium in the Parisian region of France

Indicators for comparing and understanding differences in antimicrobial resistance (AMR) and healthcare-associated infections (HAIs) for benchmarking are essential to identify priorities for hospitals. This study measured the incidence of hospital-acquired or resistant Gram-negative bacilli bloodstream infections (GNB-BSIs) in a large public healthcare consortium in the Parisian region of France. Within each hospital, there was a strong positive correlation between the incidence of GNB-BSIs due to resistant GNB and the incidence of hospital-acquired GNB-BSIs. Two scores measuring AMR and HAI rates by combining different GNB-BSI incidence rates were developed as indicators. These scores were highly variable within the hospital consortium. On multi-variate analysis, AMR and HAI scores were significantly associated with the proportion of surgical beds, staff absenteeism and the consumption of alcohol-based hand rub, with the latter two characteristics being amenable to interventions. Carbapenem use was also linked to AMR, but this may be because carbapenems are the preferred drug for treating resistant infections. These results shed light on the incidence of HAIs and AMR in the study hospitals, and suggest possibilities for targeted interventions at healthcare facility level.

Mobilizing Patient and Public Involvement in the Development of Real-World Digital Technology Solutions: Tutorial.

Although the value of patient and public involvement and engagement (PPIE) activities in the development of new interventions and tools is well known, little guidance exists on how to perform these activities in a meaningful way. This is particularly true within large research consortia that target multiple objectives, include multiple patient groups, and work across many countries. Without clear guidance, there is a risk that PPIE may not capture patient opinions and needs correctly, thereby reducing the usefulness and effectiveness of new tools. Mobilise-D is an example of a large research consortium that aims to develop new digital outcome measures for real-world walking in 4 patient cohorts. Mobility is an important indicator of physical health. As such, there is potential clinical value in being able to accurately measure a person's mobility in their daily life environment to help researchers and clinicians better track changes and patterns in a person's daily life and activities. To achieve this, there is a need to create new ways of measuring walking. Recent advancements in digital technology help researchers meet this need. However, before any new measure can be used, researchers, health care professionals, and regulators need to know that the digital method is accurate and both accepted by and produces meaningful outcomes for patients and clinicians. Therefore, this paper outlines how PPIE structures were developed in the Mobilise-D consortium, providing details about the steps taken to implement PPIE, the experiences PPIE contributors had within this process, the lessons learned from the experiences, and recommendations for others who may want to do similar work in the future. The work outlined in this paper provided the Mobilise-D consortium with a foundation from which future PPIE tasks can be created and managed with clearly defined collaboration between researchers and patient representatives across Europe. This paper provides guidance on the work required to set up PPIE structures within a large consortium to promote and support the creation of meaningful and efficient PPIE related to the development of digital mobility outcomes.

New evidence for the effect of type 2 diabetes and glycemic traits on testosterone levels: a two-sample Mendelian randomization study.

Type 2 diabetes mellitus (T2DM) is an endocrine-related disease with an increasing incidence worldwide. Male sexual dysfunction is common in diabetic patients. Therefore, we designed a Mendelian randomization (MR) study to investigate the association of type 2 diabetes and 3 glycemic traits with testosterone levels. Uncorrelated single nucleotide polymorphisms (SNPs) associated with T2DM (N = 228), fasting insulin (N = 38), fasting glucose (N = 71), and HbA1c (N = 75) at the genome-wide significance were selected as instrument variables. Genetic associations with testosterone levels (total testosterone, TT, bioavailable testosterone, BT, and sex hormone-binding globulin, SHBG) were obtained from the UK Biobank studies and other large consortia. Two-sample MR analysis was used to minimize the bias caused by confounding factors and response causality. Multivariable MR analysis was performed using Body mass index (BMI), Triglycerides (TG), LDL cholesterol (LDL), and adiponectin to adjust for the effects of potential confounders. Type 2 diabetes mellitus was associated with the decrease of total testosterone (β: -0.021,95%CI: -0.032, -0.010, p<0.001) and sex hormone binding globulin (β: -0.048,95%CI: -0.065, -0.031, p<0.001). In males, total testosterone (β: 0.058, 95% CI: 0.088, 0.028, p < 0.001) decreased. In females, it was associated with an increase in bioavailable testosterone (β: 0.077,95%CI: 0.058,0.096, p<0.001). Each unit (pmol/L) increase in fasting insulin was associated with 0.283nmol/L decrease in sex hormone-binding globulin (95%CI: -0.464, -0.102, p=0.002) and 0.260nmol/L increase in bioavailable testosterone (95%CI: -0.464, -0.102, p= 0.002). In males, sex hormone binding globulin decreased by 0.507nmol/L (95%CI: -0.960, -0.054, p= 0.028) and bioavailable testosterone increased by 0.216nmol/L (95%CI: 0.087,0.344, p= 0.001). In females, sex hormone binding globulin decreased by 0.714 nmol/L (95%CI: -1.093, -0.335, p<0.001) and bioavailable testosterone increased by 0.467nmol/L (95%CI: 0.286,0.648, p<0.001). Each unit (%) increase in HbA1c was associated with 0.060nmol/L decrease in sex hormone-binding globulin (95%CI: -0.113, -0.007, p= 0.026). In males, total testosterone decreased by 0.171nmol/L (95%CI: -0.288, -0.053, p=0.005) and sex hormone binding globulin decreased by 0.206nmol/L (95%CI: -0.340, -0.072, p=0.003). Total testosterone increased by 0.122nmol/L (95%CI: 0.012,0.233, p=0.029) and bioavailable testosterone increased by 0.163nmol/L (95%CI: 0.042,0.285, p=0.008) in females. Using MR Analysis, we found independent effects of type 2 diabetes, fasting insulin, and HbA1c on total testosterone and sex hormone-binding globulin after maximum exclusion of the effects of obesity, BMI, TG, LDL and Adiponectin.

Factors Associated With Cardiac Radiation Dose Reduction After Hypofractionated Radiation Therapy for Localized, Left-Sided Breast Cancer in a Large Statewide Quality Consortium

Limiting cardiac radiation dose is important for minimizing long-term cardiac toxicity in patients with left-sided early-stage breast cancer. Prospectively collected dosimetric data were analyzed for patients undergoing moderately hypofractionated radiation therapy to the left breast within the Michigan Radiation Oncology Quality Consortium from 2016 to 2022. The mean heart dose (MHD) goal was progressively tightened from ≤2 Gy in 2016 to MHD ≤ 1.2 Gy in 2018. In 2021, a planning target volume (PTV) coverage goal was added, and the goal MHD was reduced to ≤1 Gy. Multivariate logistic regression models were developed to assess for covariates associated with meeting the MHD goals in 2016 to 2020 and the combined MHD/PTV coverage goal in 2021 to 2022. In total, 4165 patients were analyzed with a median age of 64 years. Overall average cardiac metric compliance was 91.7%. Utilization of motion management increased from 41.8% in 2016 to 2020 to 46.5% in 2021 to 2022. Similarly, use of prone positioning increased from 12.2% to 22.2% in these periods. On multivariate analysis in the 2016 to 2020 cohort, treatment with motion management (odds ratio [OR], 5.20; 95% CI, 3.59-7.54; P < .0001) or prone positioning (OR, 3.21; 95% CI, 1.85-5.57; P < .0001) was associated with meeting the MHD goal, while receipt of boost (OR, 0.25; 95% CI, 0.17-0.39; P < .0001) and omission of hormone therapy (OR, 0.65; 95% CI, 0.49-0.88; P=.0047) were associated with not meeting the MHD goal. From 2021 to 2022, treatment with motion management (OR, 1.89; 95% CI, 1.12-3.21; P=.018) or prone positioning (OR, 3.71; 95% CI, 1.73-7.95; P=.0008) was associated with meeting the combined MHD/PTV goal, while larger breast volume (≥1440 cc; OR, 0.34; 95% CI, 0.13-0.91; P=.031) was associated with not meeting the combined goal. In our statewide consortium, high rates of compliance with aggressive targets for limiting cardiac dose were achievable without sacrificing target coverage.

Factors Associated with Cardiac Radiation Dose Reduction Following Hypofractionated Radiation Therapy for Localized, Left-Sided Breast Cancer in a Large Statewide Quality Consortium

Limiting radiation dose to the heart is important for minimizing the risk of long-term cardiac toxicity in patients with left-sided early-stage breast cancer. Prospectively collected dosimetric data were analyzed for patients undergoing hypofractionated radiation therapy to the left breast for localized node-negative breast cancer within the Michigan Radiation Oncology Quality Consortium (MROQC) from 2016-2022. Goals for limiting cardiac dose were adjusted over time. From 2016-2020, the cardiac quality metric focused on total mean heart dose (MHD) from the composite whole breast and boost plans, tightening from a goal of MHD ≤2 Gy to MHD ≤1.2 Gy by 2020. In 2021-2022, the cardiac metric transitioned to a combined goal of MHD ≤1.0 Gy from the whole breast plan and ≥95% lumpectomy cavity planning target volume (PTV) receiving 95% of the prescription dose. Separate multivariate logistic regression models were developed to assess for covariates associated with meeting the MHD goal in 2016-2020 and combined MHD/PTV coverage goal in 2021-2022. In total, 4,165 patients were analyzed with a median age of 64 years. Most patients (86%) had either Tis or T1 disease, and 66% received hormone therapy. Baseline demographic and disease characteristics did not change substantially between treatment periods. Use of breath-hold or motion gating increased from 42% in 2016-2020 to 46% in 2021-2022. Similarly, use of prone positioning increased from 12% to 20%. From 2016-2020, 90.9% of plans achieved the MHD goal, compared to 93.6% of plans achieving the composite MHD/PTV goal from 2021-2022. On multivariate analysis in the 2016-2020 cohort, treatment with motion management (OR 5.20, 95% CI [3.59-7.54], p<0.0001) or prone positioning (OR 3.21, 95% CI [1.85-5.57], p < 0.0001) were associated with meeting the MHD goal, while receipt of boost (OR 0.25, 95% CI [0.17-0.39], p<0.0001) and omission of hormone therapy (OR 0.65, 95% CI [0.49-0.88], p = 0.0047), were associated with not meeting the MHD goal. During the era including composite heart dose and PTV coverage goals (2021-2022), treatment with motion management (OR 1.89, 95% CI [1.12-3.21], p = 0.018) or prone positioning (OR 3.71, 95% CI [1.73-7.95], p = 0.0008) were associated with meeting the combined goal, while larger breast volume (≥1440 cc, OR 0.34, 95% CI [0.13 - 0.91], p = 0.031) and treatment at an academic center (OR 0.36, 95% CI [0.22-0.67], p = 0.0009) were associated with not meeting the combined goal. In our statewide consortium, rates of compliance with aggressive targets for limiting cardiac dose remain high, despite tightening of these goals to include lower mean heart doses and inclusion of a concurrent PTV coverage goal. Treatment using motion management or prone positioning is associated with achieving the cardiac dose goals.

Assessing Patterns of Practice in Early-Stage Lung Cancer Radiation Therapy: Findings from a Large Statewide Consortium Study on Hypofractionation

There are many different acceptable radiation dose and fractionation regimens for the treatment of early-stage non-small cell lung cancer (NSCLC), including hypofractionation (HypoRT). There are limited data supporting when to use HypoRT. We investigated which factors led physicians to choose HypoRT rather than stereotactic body radiation therapy (SBRT) or conventional fractionation (CRT) for early-stage NSCLC patients in a statewide consortium. We examined patients with T1-3N0M0 NSCLC treated at multiple institutions in a statewide consortium from January 2012-July 2022. We classified treatments as SBRT if 5 fractions or fewer, HypoRT if 6-20 fractions, and CRT if 1.8-2 Gy per fraction (Gy/Fx) for 25 or greater fractions. We excluded patients between 20 and 25 fractions as most appeared to be CRT that did not finish treatment (5% of total). We then performed a classification tree using age, race, gender, smoking status, T stage, PTV within 2cm of esophagus, PTV within 2cm of heart, and concurrent chemotherapy as covariates with a three-variable output (SBRT, HypoRT, and CRT). We excluded ECOG as it was not significant on initial analysis and was missing for 100 patients. We also reran the classification tree without CRT as an output to better discriminate between SBRT and HypoRT. A total of 418 patients were included in the analysis. 184 patients had T1, 123 with T2, and 111 with T3 tumors. In total, 228 patients underwent SBRT (median 50 Gy), 51 patients HypoRT (median 60 Gy), and 139 CRT (median 63 Gy). Covariates significant for discriminating between all three treatment regimens included T2, T3 vs T1, PTV within 2cm of the esophagus, and T3 vs T2. 94% of T1 patients were treated with SBRT. Among T2 and T3 patients, those within 2cm of the esophagus were significantly more likely to be treated with CRT or HypoRT (80% vs 15%). Patients with T3 tumor not within 2cm of the esophagus, were more likely to be treated with CRT or HypoRT than the T2 patients (85% vs 36%). Patients treated with CRT were also more likely to receive chemotherapy, particularly for T3 tumors (80% received concurrent chemotherapy). Excluding CRT as an output variable, proximity to the heart became significant in addition to the other previously described covariates. Notably, the branch with the highest likelihood of HypoRT were patients with T2/T3 tumors within 2 cm of both the esophagus and heart (94% HypoRT vs 6% SBRT). Patients with tumors not close to central structures but with T3 rather than T2 tumors were more likely to be treated with HypoRT as well (62% vs 38%). Based on this large prospective real-world data of early-stage NSCLC, larger tumors and those located near central structures are more likely to be treated with HypoRT. The patient's age, performance status, race, and smoking status were not significant in this analysis. Additional analysis on outcomes and toxicity related to treatments is underway.

Constraints on simulated past Arctic amplification and lapse rate feedback from observations

Abstract. The Arctic has warmed more rapidly than the global mean during the past few decades. The lapse rate feedback (LRF) has been identified as being a large contributor to the Arctic amplification (AA) of climate change. This particular feedback arises from the vertically non-uniform warming of the troposphere, which in the Arctic emerges as strong near-surface and muted free-tropospheric warming. Stable stratification and meridional energy transport are two characteristic processes that are evoked as causes for this vertical warming structure. Our aim is to constrain these governing processes by making use of detailed observations in combination with the large climate model ensemble of the sixth Coupled Model Intercomparison Project (CMIP6). We build on the result that CMIP6 models show a large spread in AA and Arctic LRF, which are positively correlated for the historical period of 1951–2014. Thus, we present process-oriented constraints by linking characteristics of the current climate to historical climate simulations. In particular, we compare a large consortium of present-day observations to co-located model data from subsets that show a weak and strong simulated AA and Arctic LRF in the past. Our analyses suggest that the vertical temperature structure of the Arctic boundary layer is more realistically depicted in climate models with weak (w) AA and Arctic LRF (CMIP6/w) in the past. In particular, CMIP6/w models show stronger inversions in the present climate for boreal autumn and winter and over sea ice, which is more consistent with the observations. These results are based on observations from the year-long Multidisciplinary Drifting Observatory for the Study of Arctic Climate (MOSAiC) expedition in the central Arctic, long-term measurements at the Utqiaġvik site in Alaska, USA, and dropsonde temperature profiling from aircraft campaigns in the Fram Strait. In addition, the atmospheric energy transport from lower latitudes that can further mediate the warming structure in the free troposphere is more realistically represented by CMIP6/w models. In particular, CMIP6/w models systemically simulate a weaker Arctic atmospheric energy transport convergence in the present climate for boreal autumn and winter, which is more consistent with fifth generation reanalysis of the European Centre for Medium-Range Weather Forecasts (ERA5). We further show a positive relationship between the magnitude of the present-day transport convergence and the strength of past AA. With respect to the Arctic LRF, we find links between the changes in transport pathways that drive vertical warming structures and local differences in the LRF. This highlights the mediating influence of advection on the Arctic LRF and motivates deeper studies to explicitly link spatial patterns of Arctic feedbacks to changes in the large-scale circulation.

Age-Related Macular Degeneration in an Ageing Society - Current Epidemiological Research.

Epidemiological studies on age-related macular degeneration (AMD) provide crucial data on the frequency of early and late forms as well as associated risk factors. The increasing number of population-based cross-sectional and longitudinal cohort studies in Germany and Europe with published data is making prevalence and incidence estimators for AMD more robust, although they show mostly method-related fluctuations. This review article brings together the latest published epidemiological measures for AMD from Germany and Central as well as Western Europe. Based on this data and population figures for Germany and Europe, prevalence is projected, and future trends are forecasted. The epidemiological evidence for AMD-associated risk factors is also improving, especially through meta-analyses within large consortia with correspondingly high case numbers. This review article summarizes the latest findings and resulting recommendations for prevention approaches. Additionally, it discusses treatment options and future challenges.