PRenatal and Obstetric Maternal Exposures and ISlet Autoantibodies in Early Life

Overview

This research study is called 'PRenatal and Obstetric Maternal Exposures and ISlet Autoantibodies in Early Life: The PROMISE Study'. The purpose of this study is to find out more about how exposures during pregnancy, such as having an infection, diet and growth may impact later risk of type 1 diabetes (TID) and islet autoimmunity in the child. We are also interested in finding out more about why having a father or sibling with T1D increases risk of autoimmunity in the child more than having a mother with T1D. We are enrolling women who are pregnant and either have T1D or another first degree relative (father or full sibling) of the baby has T1D. The biological father is also invited to enroll in study, as it is important to understand how the father's health and genetics may contribute to the child's risk of developing T1D. The study procedures for the mother, father and baby are explained below. Mother: Pregnant women will be asked to complete a visit once per trimester (3 visits) during pregnancy and one visit up to 12 weeks after delivery. At each visit, mothers will consent to a blood draw, collection of biological samples and the completion of questionnaires. . Mothers who have T1D will also be asked to download any diabetes device data they have, such as continuous glucose monitor or insulin pump data. Father: The (biological) father will be invited to enroll in a single visit. He will consent to a blood draw and completion of questionnaires. Fathers with T1D will also be asked to download any diabetes device data they have, such as continuous glucose monitor or insulin pump data. Baby: The baby will have blood collected at birth to determine the genetic risk for T1D. Families will consent to the completion of questionnaires about growth, health and diet at 6, 12, 18 and 24 months of age and between 5-7 years of age, and to complete blood testing for islet autoantibodies at 24 months and between 5-7 years of age. For those children with a high genetic risk score, we will also collect blood for autoantibody testing at 6, 12, and 18 months of age.

We will assemble a prenatal cohort where the offspring has a first degree relative (FDR) with T1D (mother, father or full sibling). In order to enroll sufficient participants, we are conducting a multi-center study at the Barbara Davis Center for Diabetes (BDC) in Aurora, CO \[coordinating center\], Mt Sinai Hospital in New York, NY, the Joslin Diabetes Center (JDC) in Boston, MA, and Ohio State University (OSU) in Columbus, OH. These sites were chosen based on a history of successful prior collaborations, access to large established populations of people with T1D (these centers are some of the largest in the United States with respect to pregnancy care for women with T1D, and also have adult and pediatric clinics), site-specific connections to obstetric practices, and a history of successful recruitment into study cohorts. In addition to the four study sites named, we have approached additional potential sites in order to ensure success with recruitment. We have identified an additional potential site at the Mayo Clinic College of Medicine under the direction of Yogish Kudva, and this site is prepared to activate should recruitment in the first year of the grant at the four sites fall short of expectations. Our goal is to assemble a pregnancy cohort of 800 during this initial funding period, and to enroll up to an additional 1,200 in phases 2 and 3 (total N=2,000). D. Description, Risks and Justification of Procedures and Data Collection Tools: Screening and Consent Procedures: Participants will be screened, consented (available through paper and electronic consent process) and will complete questionnaires through an online database with electronic survey capabilities (REDCap) that is HIPAA compliant and can be accessed anywhere with internet access. If needed, REDCap access can be provided in the clinic using a tablet. We have successfully used the REDCap database system for numerous studies involving survey collection and have used online screening and consenting procedures extensively in studies of adults with T1D (including pregnant women) during the Covid-19 pandemic. Study staff will be available to review the consent form with potential participants over the phone, using a HIPAA-compliant video platform such as Zoom, or in person, depending on the circumstances. Screening questions will include the history of T1D in the pregnant woman or father/sibling of the fetus, maternal age, and obstetric history for the pregnant woman, including current pregnancy dates (last menstrual period date, expected date of delivery). Visit and Specimen Collection Schedule: Study visits and specimen collection (Table 2) will be scheduled at one of the study sites, at one time point for fathers and at four time points for mothers: first trimester (8-\<14 weeks), second trimester (14-\<28 weeks), third trimester (28+ weeks), and up to 12 weeks postpartum. Medical records will be obtained using medical record release forms for the pregnancy, delivery and post-partum visits, and for the infants' birth records. All study sites have access to electronic health records (EHR) that will allow investigators to obtain medical records for prenatal care and delivery for study participants who deliver within their hospital network, and all sites have protocols in place and experience obtaining records from outside facilities using medical record release forms. At the BDC coordinating center, we have previously collected medical records for pregnancy, delivery and infant birth records and have abstracted these data into REDCap databases. We will follow our established procedures for abstraction of obstetric and newborn records for this study. Maternal and fetal outcomes of women included in this study that are reported to BDC providers will be obtained by manual chart review of EHR charts. Maternal and fetal hospitalizations at any hospital system utilizing EPIC and for whom patients authorize sharing of records can be viewed through EPICS's interoperable sharing tool called Care Everywhere. Manual review of records will be used to obtain specific outcome measures. All records will be collected and abstracted at the Barbara Davis Center, to maintain quality control standards and to streamline training and oversight of staff. Pregnancy Visits: Pregnant women who are eligible for the study will be enrolled in the first trimester (before 14 weeks gestation), if at all possible, but after 8 weeks gestation to reduce the possibility of early pregnancy loss following enrollment. Fathers will be enrolled at the same time and will be asked to complete their study visit (including a blood sample of up to 33 ml or 2 1/4 tablespoons of blood) at the same time as the woman's first pregnancy visit, if at all possible. Women who are already in the second trimester (up to 24 weeks) when approached about the study will be enrolled and will complete second and third trimester visits and the post-partum visit. Routine pregnancy visit data, including weight and blood pressure will be collected in the clinic or from the medical record. Routine pregnancy visit data, including weight and blood pressure, routine laboratory tests (including HbA1c for women with T1D, lipids, etc) will be collected from the clinic or from the medical record. Study visit timing and procedures are shown in Table 3 below. Specimen Collection: Samples will be collected at each study visit at the participating clinic location. External centers will ship all samples to the BDC for storage. Sample collection will be standardized in our Manual of Operations to ensure consistent collection at all sites. Further, sample collection, timing and types have been carefully designed to be similar to the collection completed in existing cohorts, in particular the ENDIA study, to allow for comparison of results across studies. However, timing of trimester-based and post-partum visits are determined by standard of care in our clinics and may not align exactly with the ENDIA study. For example, women typically complete a post-partum visit at 4-6 weeks in our clinical centers, and so we will schedule our post-partum data and sample collection at the same time, to minimize participant burden. In the ENDIA study, mothers underwent a post-partum visit at 3 months. A sample of up to 33 ml or 2 ¼ tablespoons of blood will be obtained by phlebotomy. Plasma will be collected in EDTA tubes, kept on ice and separated by centrifugation. Serum will be collected in serum separator tubes with gel and separated by centrifugation after being allowed to clot. Serum and plasma will be aliquotted into 2ml cryovials and stored at - 80° C. Buffy coats will be stored at each visit for future DNA and methylation studies. Blood RNA samples will be collected using Tempus™ tubes, which allow for stabilization and isolation of total RNA from whole blood for gene expression analysis. These samples can be stored for up to five days at room temperature, and then frozen at 4°C. Optional vaginal swabs will be self-collected by participants using a CultureSwab transport system. Optional stool samples will be self-collected by participants using a commercially available stool collection kit such as the OMNIgene\_GUT. Questionnaire Collection: Participants will be asked to complete questionnaires online using the REDCap database program for each visit, to reduce the time needed to complete the in person appointments for study participants completing visits at the clinics, and to streamline data collection/entry procedures and reduce data entry errors. Questionnaires will include dietary intake using a validated food frequency questionnaire43, tobacco and alcohol using validated questionnaires44,45, and we will collect screening questionnaires addressing general \[Patient Health Questionnaire-8\]46 and perinatal \[Edinburgh Postnatal Depression Scale\]47 depression. Women who screen positive (a PHQ8 score of 14 or higher, or an Edinburgh Postnatal Depression score of 10 or higher) on either of these questionnaires will be referred to their regular provider for further evaluation. Among mothers and fathers with T1D, assessments of the T1D distress scale (T1-DDS), diabetes self-management and hypoglycemia fear will be assessed using validated instruments. Fathers with T1D will complete the questionnaires at a single visit, whereas mothers with T1D will be asked to complete these questionnaires at each study visit (once per trimester, and post-partum). Diabetes care records, including HbA1cs and other laboratory tests (lipids, urinary albumin to creatinine ratio, serum creatinine), retinopathy exams and examination records will be requested for the year prior to enrollment for determination of complications and glycemic control. Pregnancy history and outcomes: We will collect data on maternal and fetal outcomes of the pregnancy with medical records; we will obtain gestational age at delivery, infant birth weight, delivery complications, and maternal and fetal outcomes by obtaining records for each hospital admission for labor and delivery. As a backup for these hospital records and in case we are unable to access medical records (such as if a woman moves outside of our electronic health record area and we can't obtain records using a medical record release), we will collect the same data using a survey we developed for one of our previous pregnancy studies. It inquires about delivery date, gestational age, infant birth weight and length, living status of the infant, infant sex, preterm labor, type of delivery, 16 specific pregnancy complications, 15 specific labor complications, 5 specific neonatal complications, and admission to the neonatal intensive care unit (NICU). Breastfeeding practices will be collected at the post-partum visit. We will also collect data through a medical chart review for health information prior to pregnancy for women with T1D (blood pressure, kidney function, retinopathy, height and weight). We will verify gestational dating by obtaining records of ultrasounds performed for dating. Glycemic factors among women with T1D: HbA1c and glucose values (CGM and glucose meter downloads), current insulin dosing and insulin delivery method are collected during routine pregnancy visits and will be collected from the medical record. We will also access data downloaded from CGM devices (for those women who use a CGM regularly) during routine pregnancy visits to evaluate indices of glycemic variability such as time spent in pertinent glucose ranges (time in range: 63-140 mg/dL \[pregnancy\] and 70-180 mg/dL \[post-partum\], hypoglycemia: \<63 mg/dL \[pregnancy\] and \<70 mg/dL \[post-partum\], and hyperglycemia: \>140 mg/dL \[pregnancy\] and \>180 mg/dL \[post-partum\]50, coefficient of variation, and SD). Cord blood and stool collection: Collection of cord blood up to 20 ml will be prioritized, since it is an ideal sample for examining the DNA methylation in the infant at birth and before exposure to the outside environment, including breastfeeding or bottle feeding, exposure to microbes and other environmental factors, as a contrast to later samples obtained during early life. Collecting cord blood has been accomplished in our prior studies through agreements with screening hospitals, and through the use of cord blood collection kits sent to the mother for obtaining and mailing back the samples. Given that women will be giving birth at any number of hospitals nationally in this cohort, we will plan to send/give cord blood collection kits to the pregnant mothers with detailed instructions for collection and return of the samples, using our established protocol. In the case that cord blood collection isn't possible, we will include a kit for collecting a heel stick sample of up to 1ml of blood as soon as possible after birth (preferably within a week of delivery). We also plan to give women stool kits for collection of a stool/meconium sample soon after delivery. Screening for High Risk Genotypes among the Offspring: Newborns will be screened for genetic risk for T1D at birth using dried blood spot collections obtained within the first week of life. Samples will be sent to the laboratory of Dr. William Hagopian for determination of the genetic risk score 2 (GRS2), a panel of 67 single nucleotide polymorphisms (SNPs) that has been shown to discriminate risk for T1D, especially for early-onset T1D (AUC 0.96)1. High-risk human leukocyte antigen (HLA) genotypes are strongly associated with IA and T1D risk, but use of the GRS2 has demonstrated to be nearly twice as efficient as only screening for high-risk HLA genotype. The use of this risk score has been proposed as a more efficient method of newborn screening, allowing for closer monitoring of the children at highest risk for developing IA and T1D, especially early in childhood. All visits and procedures for the offspring are shown in Table 4. Based on data from the TEDDY study, a GRS2 conferring risk equivalent to the 90th percentile in the general population identified 23% of TEDDY participants with an FDR as high risk, compared to 10% of the general population. Given that TEDDY participants with an FDR were selected based on their high risk HLA genotype, we estimate that the percentage identified as high risk based on GRS2 score in this proposed prenatal cohort (where all FDRs will be enrolled regardless of HLA risk) will be slightly lower than the percentage observed in TEDDY. Therefore, we have estimated that \~20% of the FDR offspring in the proposed pregnancy cohort will be identified as high risk and will be followed more closely than those at low-risk. Children with the highest risk on the GRS2 (e.g., risk higher than the equivalent of 90% of general population risk scores) will have blood collected for IA screening and genetic marker storage every six months through 2 years of age and then between 5-6 years. Children with a low risk GRS2 will have blood collected for IA screening at 2 years and between 5-6 years of age. HLA genotyping and IA testing will be done at the Clinical Immunology Lab at the BDC (NIH/NIDDK designated North American Autoantibody/HLA Core) as previously reported40. Other blood samples will be retained for case-control studies of DNA methylation, Virscan and future metabolomic/proteomic markers.. Infant growth and health: Infant growth data will be obtained by self-report from parents through REDCap surveys, and through medical records requests for pediatric records. Immunization records will be obtained. Every 6 months until the age of 2 years, clinical measurements (including collection of up to 20 ml or just over a tablespoon blood), illnesses and infant/child feeding practices will be collected from the high risk group at the clinic visit; this data will be collected by a REDCap questionnaire for the low risk group. Islet Autoantibody Assessments: The main outcome measure of the overall study (all phases) will be persistent positivity for islet autoantibodies (IA) in the child. Serum samples will be obtained for measurement of four islet autoantibodies in high genetic risk offspring at 6, 12, 18, 24 months, and at 5-6 years. The low genetic risk offspring will be tested at 24 months and 5-6 years of age. Samples will be collected, processed and sent from the external sites to the BDC for measurement of IAs. Serum will be stored at -70°C until assayed. The more frequent monitoring for IA among high risk offspring is needed because IA can appear as early as in the first year of life and T1D is more commonly being diagnosed at younger ages. Therefore, we plan to monitor both IA and evolution of dysglycemia (stage 2 T1D), as described below. IA assays will be conducted on coded samples for IAA, GADA, IA-2A and ZnT8A in the Barbara Davis Center Core Autoantibody Laboratory, under the direction of Liping Yu, MD. This laboratory is CLIA certified, and CAP accredited and has been designated as an NIH/NIDDK North American Autoantibody/HLA Core Laboratory. For the past several decades, the Barbara Davis Center Autoantibody laboratory has served a number of major T1D clinical trials, including the TEDDY study, TrialNet, DPT-1, T1DGC, ITN and others. The assays for GAD65 and IA-2 have been harmonized recently in collaboration with the NIH/NIDDK Diabetes Autoantibody Harmonizing Committee using a standard assay protocol with a common antibody unit expressed as DK units/ml (NIDDK units). This made it possible for autoantibody test results to be comparable among all of the ongoing national/international trials between different laboratories worldwide. BDC's Autoantibody/HLA Core Facility is one of three international laboratories helping to accomplish these harmonized assay protocols. All positive samples and 5% of negative samples will undergo blinded duplicate testing. IA will be defined as positivity for at least one of the four islet autoantibodies on ≥2 consecutive visits, or a single positive test followed by diagnosis of T1D. Children with an initial positive autoantibody test will be asked to return for a confirmation test as soon as possible after the initial test. Monitoring for dysglycemia: Families will be educated on the signs and symptoms of T1D and will be instructed to contact study staff to report any concerns or questions about potential signs and/or symptoms in their children. Families of children who develop IA will be provided access to the ASK study team, assuming funding for this study is still active, for enrollment into the ASK follow-up study for education, monitoring and support including educational materials to explain risk factors associated with islet autoantibody status, repeat testing of autoantibodies and HbA1c every three to six months to monitor for progression, provided with a free glucometer, strips, and training. A continuous glucose monitor may be provided free of charge every 3- 6 months. In addition, the ASK the Experts national monitoring program, also funded by Helmsley Charitable Trust, will be available to those families who have local providers or pediatric endocrinologists that can work in tandem with the ASK the Experts team at the Barbara Davis Center for education, monitoring and support. If the ASK/ASK the Experts studies are no longer funded, the same education and support will be provided through this study protocols, including measurement of HbA1c every three to six months and free glucometer, glucose strips and training. Viral exposure assessments: Exposure to viruses will be assessed longitudinally in mothers and offspring using the Virscan assay. For the initial three year study period analysis, serum samples from visits during each trimester of pregnancy and post-partum as well as those collected at 6 months of life among selected high risk infants (as described in the data analysis plan) will be tested on the Virscan pan-viral serological surveillance tool. This platform is based on phage immunoprecipitation sequencing (PhIP-seq) technology that uses a bacteriophase library containing proteome-wide peptides from human-pathogenic viruses. The Virscan library contains peptides representing most known pathogenic, human viruses (\~400 species and strains) as well as other antigens listed in the National Institute of Allergy and Infectious Diseases Immune Epitope Database (www.iedb.org). The goal of the longer-term study once all phases are complete and a sufficient number of offspring have developed IA is to perform a case-control study of viral exposure in mothers and offspring among those who have developed IA versus age and HLA-matched control offspring who have not developed IA. To perform the serological screening, serum samples are mixed with the library allowing antibodies to bind with pathogen specific epitopes displayed on the phage surface. The bacteriophage-antibody complexes are then immunoprecipitated and the phage DNA region encoding the artificially expressed peptide antigens to which an antibody was bound are sequenced by NGS to reveal the repertoire of anti-viral antibodies in a given serum sample. Data analysis to detect antibodies against a wide-range of pathogens relies on statistical assessment (p values) of relative enrichment of pathogen-specific peptides in serum specimens. The resulting data are therefore qualitative in nature and are not directly comparable to antibody titers that can be obtained from conventional, singleplex serological methods. Epigenetic and gene expression studies: DNA methylation will be measured in maternal peripheral whole blood and gene expression (RNA) measured in whole blood RNA samples collected during pregnancy and post-partum, and in offspring whole blood and RNA samples collected in early life prior to the development of IA. Measurements of DNA methylation and gene expression will be compared in the initial three year study period in cord blood from offspring of mothers with T1D (n=86) compared to offspring who have a father with T1D but a mother without diabetes (N=86) in a case-control design matched for maternal age and HLA risk. The goal of the longer-term analysis once all phases are complete is to examine epigenetic changes and gene expression in a case-control study design, selecting IA cases from among infants who are persistently IA positive, and among controls matched on age and HLA risk. Methylation quantification will be performed using the Infinium HumanMethylation450K Beadchip ("450 K") and the Infinium MethylationEPIC Beadchip ("EPIC") as previously described41. The University of Colorado Anschutz Medical Campus houses a state-of-the-art facility and has launched an RNA Bioscience Initiative (https://medschool.cuanschutz.edu/rbi) with the goal to provide a 'fluid pipeline from basic to clinical RNA research at the Anschutz Medical Campus" through combining RNA testing and informatics resources.