Abstract Accurate biomarker measurement is critical for understanding biopsychosocial and cognitive aging. However, biomarker measurement often relies on fasting samples, whereas cognitive testing is typically conducted non-fasting. The effects of eating on endocrine levels were tested in a well-controlled experiment. Older adults (N=30, mean 67.5 years, 70% women) each attended two, morning laboratory sessions with four, hourly blood draws after >8 hours fasting. A meal was provided after the second or fourth blood draw (session order counter-balanced). Eight hormones were assayed via LC/MS; bioavailable hormones were estimated using SHBG (EIA assays). Bayesian multilevel models (minimally informative priors, Stan MCMC sampler) compared pre-prandial (6 samples from both visits) vs. post-prandial (2 samples) concentrations (covariates: time-of-day, visit order, age, gender, BMI). These models provided decisive evidence for eating increasing cortisol and reducing total testosterone and estrone; weak evidence for increasing corticosterone and DHEAS and decreasing progesterone, free testosterone and estrone. Decisive evidence was observed for cortisone, and total and free estradiol being unaffected by eating. Exploratory analyses of gender moderation provided decisive evidence that effects of eating on total testosterone were most strongly observed among men, although the pattern was similar for women. These findings indicate that cortisol, total testosterone, and total estrone are susceptible to changes due to eating among older adults, suggesting careful consideration related to diet/fasting should be made in studies incorporating these biomarkers. Cortisone and estradiol appear relatively unaffected by eating; further research is necessary for all remaining biomarkers that had weak evidence of change.

Abstract Blood-based neurodegenerative biomarkers hold vast potential to serve as early detectors of neuropathology, including Alzheimer’s disease. However, accurate biomarker assessment is critical and extraneous factors that may affect circulating concentrations need greater understanding. To determine the effects of fasting status on blood-based levels of neurodegenerative biomarkers, 30 older adults (Mage = 67.52; 70% women) fasted >8h prior to two laboratory visits. Each visit included 4 blood draws, spaced 1 hour apart. The visits (counterbalanced) were differentiated by a meal being provided after either the second or fourth blood draw. This design enabled examination of fasting versus post-prandial effects, accounting for circadian variation of biomarkers. Plasma levels of neurodegenerative biomarkers (amyloid β(Aβ) 42/40 ratio, neurofilament light chain (NfL), glial fibrillary acidic protein (GFAP), phosphorylated tau (pTau)181, pTau217) were quantified at each timepoint. Bayesian multilevel models (minimally informative priors, Stan MCMC sampler) compared all pre-prandial (6 timepoints) vs. post-prandial (2 timepoints) concentrations (covariates: time-of-blood-draw, visit order, age, gender, BMI). Results provided strong and very strong evidence for no changes in Aβ42/40 ratio and pTau181, respectively. However, there was very strong evidence that eating decreased GFAP levels, substantial evidence that eating decreased pTau217 levels, and weak evidence that eating decreased NfL levels. These findings indicate that plasma levels of some, but not all, neurodegenerative biomarkers are susceptible to changes due to eating. One implication is that eating should be controlled (e.g., via fasting) in studies incorporating GFAP and pTau217. Further research is needed for NfL, which exhibited less certain (weak) evidence of change.

Abstract BACKGROUND Anti-tumor necrosis factor (anti-TNF) therapies have become first-line therapy for children with inflammatory bowel disease (IBD). Studies describing drug durability and loss of response to anti-TNF therapies are limited in children with IBD. This study evaluates predictors of primary non-response and long-term durability in children with IBD. METHODS This was a single-center retrospective review of patients with IBD aged 4-17 years that initiated anti-TNF therapy (infliximab or adalimumab) from January 1, 2014 - December 31, 2019. Clinical and laboratory data were recorded at time of anti-TNF initiation, at 14 weeks, 12 months, and 36 months. Kaplan-Meier analysis and log-rank tests evaluated long-term durability of anti-TNF therapy. Predictors of primary non-response (discontinuation of anti-TNF within 14 weeks) and long-term durability were assessed using Cox regression analysis and time-dependent receiver operating characteristic curves identified cut points. RESULTS A total of 456 patients were initiated on anti-TNF therapy (183 adalimumab (86% CD and 14% UC/IC) and 273 infliximab (69% CD and 31% UC/IC)). Eighty one patients were lost to follow up or transferred care. Thirty seven (8%) patients were considered primary non-responders. Seven of these patients received adalimumab (3 CD, 4 UC/IC) and 30 received infliximab (7 CD, 23 UC/IC). In CD, baseline ESR >31 mm/h was predictive of primary non-response (p<0.05). In UC/IC, mono-therapy at diagnosis, baseline albumin <4 g/dL, and age at diagnosis <15.6 years were significant predictors of primary non-response. Eighty patients discontinued anti-TNF therapy between 14 weeks and 36 months. Reasons for drug discontinuation included loss of response (56%), drug reaction (24%), low therapeutic level/antibodies (19%), miscellaneous (1%). Amongst patients with CD that completed induction, week 14 lab values of albumin >3.9 g/dL, hemoglobin >11.8 g/dL, CRP <1.2 mg/dL and calprotectin <650 ug/g were associated with improved drug durability at 1 year. Patients with UC/IC had improved drug durability at 1 year if they received infliximab compared to adalimumab. The three year drug durability for both adalimumab and infliximab was slightly above 70%. Amongst patients with CD the three year drug durability was nearly 80% for both therapies. The three year drug durability with UC/IC was 37% for adalimumab and 56% for infliximab. CONCLUSIONS Less than 10% of patients are primary non-responders to anti-TNF therapies; higher baseline ESR is a predictor in patients with CD, and a lower baseline albumin and younger age are predictors for patients with UC/IC. At 1 year, patients with a lesser inflammatory burden after induction appeared to have improved drug durability. Three-year durability is >75% for patients with CD on either therapy, while the durability is less for patients with UC/IC.

Abstract Inflammation may serve as a biological precursor to the development of mild cognitive impairment (MCI) and dementia. Although interleukin (IL)-6 may be more strongly associated with executive functioning (EF) or processing speed (PS) measures versus other cognitive domains, this has not been assessed simultaneously in a structural equation modeling (SEM) framework among older adults. The present study sought to address this gap. As part of the ongoing Einstein Aging Study, older adults (n=159, 65.4% female, 49.7% non-Hispanic White) without MCI at baseline completed neurocognitive testing and had two blood draws spaced two weeks apart at their baseline wave of data collection. Inflammation was assessed by averaging IL-6 levels obtained from these blood draws. Based on prior factor-analytic studies, a confirmatory factor analysis (CFA) was conducted that included an attention factor (Number Span Forward and Back, Verbal Fluency Test), memory factor (Free Selective Reminding Test, Logical Memory Test, Verbal Fluency Test), and EF/PS factor (Trail Making Test Parts A and B, Digit Symbol Test). A SEM model with pathways from IL-6 predicting cognitive factors was then specified, adjusting for relevant covariates based on fit statistics (age, years of education, sex, race/ethnicity). Both the CFA and SEM models fit the data well (RMSEA< 0.05, CFI>0.95). As expected, IL-6 was negatively associated with the EF/PS factor (β=-0.251, S.E.=0.073, p< 0.01). IL-6 was not associated with either the attention or memory factors (p>0.5). Based on these results, we predict that higher levels of IL-6 may predict nonamnestic MCI but not amnestic MCI.

Objectives: Outpatient inflammatory bowel disease (IBD) care shifted from office visits (OVs) to a model with integrated telemedicine during the 2020 COVID-19 pandemic. We describe the impact of this shift on delivery of pediatric IBD care.Methods: We collected electronic medical record data from office and telemedicine visits for pediatric patients with IBD at a single center from April 2019 to December 2020. We compared visit volume, duration, and test ordering between 2019 and 2020, and between OV and telemedicine, and assessed for differences in telemedicine adoption by sociodemographic factors.Results: Visit volume was maintained between 2019 and 2020. Median overall appointment time was shorter for telemedicine versus OV [46 (interquartile range, IQR 35-72) vs 62 (IQR 51-80) minutes; P < 0.001] with no significant difference in time spent with provider [28 (IQR 21-41) vs OV 30 (IQR 24-39) minutes; P = 0.08]. Accounting for drive time, telemedicine visits were 2.6 times shorter than office visits in 2020 (P < 0.001). In univariate analyses, there was no difference in telemedicine utilization by race or gender. Variables significantly associated with telemedicine were older age, English as primary language, being non-Hispanic, commercial insurance, living in an area of very high opportunity, and having a longer drive time to the office (P < 0.05 for all comparisons). In multivariate analyses, visits among patients with commercial insurance were significantly more likely to be conducted via telemedicine (P = 0.02). Among those with a telemedicine visit, multivariate analyses demonstrated multiracial patients were significantly more likely to have video visits (vs audio-only; P = 0.02), while patients with public insurance, no or missing insurance, and whose primary language was Arabic were significantly less likely to have video visits (P < 0.05 for all comparisons).Conclusions: Integrated telemedicine allowed for continued delivery of pediatric IBD care and significantly decreased appointment time. While telemedicine may improve access for those who live further from the office, concerns remain about the introduction of disparities.

Abstract The temporal stability of basal and stimulated inflammatory cytokines is unclear in older adults, despite often being used to characterize inflammatory status. Further, it is not apparent how the stability of these markers and their multi-dimensional relationships connect to healthy aging. This analysis examines these issues in 227 participants from the Einstein Aging Study [ages: 70-90 (Mean=76.7), 67% women] with blood draws ~2 weeks apart. Temporal stability was assessed by intra-class coefficients (ICCs) obtained from linear mixed models (controlling for days between measurements, age, gender) for a panel of basal and lipopolysaccharide (LPS)-stimulated cytokines (interleukin [IL]-1β, IL-4, IL-6, IL-8, IL-10, tumor necrosis factor [TNF]-α). Sensitivity analyses compared ICCs across genders and individuals reporting high vs. low subjective health. A composite measure of basal cytokines exhibited strong reliability (ICC: 0.854), whereas individual basal cytokines exhibited moderate (>0.5) to very high reliability (>0.9). Conversely, a composite measure of stimulated cytokines demonstrated moderate reliability (ICC: 0.552), with individual stimulated cytokines having only low (< 0.5) to moderate reliability. No appreciable differences were observed by gender, but individuals reporting low (ICC: 0.457) vs. high (ICC: 0.631) subjective health exhibited considerably lower reliability for the stimulated composite measure. In sum, basal cytokines had moderate to high stability, whereas stimulated cytokines exhibited relatively low stability. Given that individuals reporting lower subjective health demonstrated greater variability in stimulated cytokines, higher fluctuations in inflammatory responsivity may reflect suboptimal health in older adults. The inter-relationships of these cytokines, explored via principal component analysis, will also be discussed.

Background: Preterm birth rates are consistently higher in African American (AA) pregnancies compared to White pregnancies in the United States. Neighborhood racial composition, experiences of racial discrimination, and systemic inflammation are factors that have been associated with preterm birth and other adverse pregnancy outcomes that may account for these disparities. Here, we investigated whether perceived neighborhood racial composition and experiences of discrimination were predictive of cytokine levels during pregnancy among AA individuals.Methods: 545 AA individuals completed surveys and had blood samples collected at prenatal clinics in the Midwest at three timepoints (8-18,19-29, and 30-36 weeks gestation) throughout pregnancy. Pro-inflammatory [interferon (IFN)-gamma, interleukin (IL)-6, IL-8, tumor necrosis factor (TNF)-alpha, macrophage migration inhibitory factor (MIF)] and anti-inflammatory cytokines (IL-10) were quantified. Multivariate and multilevel models were used to examine associations of perceived neighborhood racial composition and experiences of racial discrimination with cytokine levels, controlling for relevant covariates.Results: Perceived neighborhood racial composition was significantly associated with MIF at 30-36 weeks gestation in multivariate regression (p < 0.001). Living in neighborhoods with more compared to fewer White people was predictive of higher levels of MIF (b = 0.599, SE = 0.12, p < 0.001). Experiences of discrimination were also associated with higher levels of MIF (beta = 0.141, SE = 0.07, p = 0.036). Neither predictor was associated with other cytokines. Follow-up analyses revealed that neighborhood racial composition was also predictive of higher MIF levels at 8-18 weeks gestation (p = 0.02) and at 19-29 weeks gestation (p = 0.04).Conclusions: Living in neighborhoods with more White individuals and having more lifetime experiences of racial discrimination were positively related to levels of the pro-inflammatory cytokine, MIF, among pregnant AA individuals. MIF's known positive relationships with chronic stress and preterm birth suggest that these elevations in MIF may have negative health consequences. Future studies should explore whether MIF serves as a pathway between neighborhood racial composition or experiences of racial discrimination and preterm birth risk among AA individuals.

Background: Studies assessing adult inflammatory bowel disease (IBD) patient perspectives on biosimilar use revealed that most were unfamiliar with biosimilars and had a negative perception. The objective of this study was to evaluate the perspectives of pediatric patients with IBD and their caregivers regarding biosimilar use and non-medical switches. Methods: A survey was given to a cross section of patients with IBD ages 11-21 years receiving the intravenous anti-tumor necrosis factor originator and caregivers of patients with IBD ages 3-21 years receiving the originator. Recruitment occurred via mail, during clinic visits, and infusions. Fisher exact tests were used to test for statistically significant differences. Results: Response rate amongst caregivers was 49% (n = 98) and among patients was 35% (n = 67). Sixty-four percent of caregivers and 79% of patients had never heard of biosimilars. There was increased discomfort surrounding the use of biosimilars and switching to a biosimilar amongst caregivers who had previously heard of biosimilars compared to caregivers who had not previously heard of biosimilars (P < 0.05). Similar concerns were not seen in patient respondents. The length of time on the originator had no effect on patient or caregiver concerns related to biosimilar efficacy, adverse effects, or switches. Conclusion: The majority of pediatric patients and caregivers had never heard of biosimilars. Caregivers that had heard of biosimilars before the study were more likely to have a negative perception of them. This study highlights the importance of providing thorough and accurate education to pediatric patients and families regarding the safety and efficacy of biosimilars.

Background: Ileocecectomy related to stricturing, fistula formation, or medically refractory disease is commonly required in patients with Crohn disease (CD). Limited research exists in endoscopic recurrence (ER) in pediatric inflammatory bowel disease (IBD). In this study, we sought to determine ER rates and the impact of therapy duration before surgery in pediatric patients with CD. Methods: This was a single-center retrospective review of patients with CD between the ages of 2 to 20 years who required ileocecectomy between January 2015 and December 2019 at Nationwide Children's Hospital. Follow-up endoscopies, laboratory values, medications, and sPCDAI scores were recorded at 6, 12, 24, and 36 months post-resection wherever available. Modified Rutgeert scores (mRS) were independently assigned to post-resection colonoscopy images by 3 trained investigators. Post-resection outcomes were compared between patients on CD therapy >30 days before resection (late surgery) to those started on CD therapy <30 days before resection (early surgery). Results: A total of 48 patients underwent ileocecectomy, with a mean age at time of resection of 17 years (+/-2.3). In total, 88% of patients had a post-resection endoscopy and 57% had an endoscopy within 12 months of resection. Twenty-nine percentage had ER with a mRS >= i2. There was no statistical difference in endoscopic and clinical outcomes after resection between the early and late surgery groups. Conclusions: Post-resection endoscopic recurrence after ileocecectomy was found in 29% of our center's pediatric CD population based on mRS. Post-resection outcomes were not affected by therapy duration before resection.

Background: Cinical outcomes of pediatric Crohn's disease (CD) patients presenting in the anti-TNF era with internally penetrating (B3) disease behavior are not well-described.Aims: To describe clinical and radiographic characteristics, initial and subsequent therapy, and need for surgery in children presenting with B3 CD in the last 6 years.Methods: We performed a retrospective review of children ages 4-18 years presenting with B3 CD at 5 major pediatric IBD centers from 1/1/13 to 4/30/19.Demographics, clinical data, laboratory and imaging findings, medical therapy, and surgical information was abstracted from electronic records.Follow-up to 6, 12, and 60 months was obtained when available.Descriptive statistics reported as means±SD or medians(IQR) along with student t-test or Wilcoxon Rank Sum test for comparisons of means and medians across surgical and non-surgical groups.Results: 63 patients are included in this study, median age 16.4(14.8-17.2) years, 56% female, mean PCDAI 42.8±12.5 at presentation.65% (41/63) had symptoms ‡1 month at diagnosis, 24% (15/63) ‡1 week to <1 month, and 11% (7/63) for <1 week.Follow-up was <1 year in 5% (3/63), ‡1 year in 83% (52/63), and ‡5 years in 13% (8/63).36 (57%) had an abdominal/pelvic abscess, 19 (30%) with phlegmon, and 29 (46%) with fistula on imaging.Overall, 67% (42/63) ultimately underwent ileocectomy.Out of the 36 patients that presented with an abscess, 28 (78%) had an abscess ‡2cm with 22 (79%) undergoing surgery and the remaining 8 had an abscess <2cm with 6 (75%) undergoing surgery.Percutaneous drainage was performed in 18 (29%) with 16 (89%) still requiring surgery.The Figure shows the likelihood of remaining surgery free during the year following diagnosis.Of those going to surgery, 33 (83%) went within 6 months of diagnosis, 2 (5%) from 6 months to 1 year, and 5 (12%) after 1 year.97% of patients were initiated on anti-TNF therapy after diagnosis; 70% received infliximab and 29% adalimumab.Median initiation was 17 days (2-47 days) from diagnosis for those that did not undergo surgery and 51 days (33-77 days) for initiation after surgery.Antibiotics were documented in 49 (78%) of patients with 25 (51%) receiving less than 4 weeks, 19 (39%) from 4 to 8 weeks, and 5 (10%) longer than 8 weeks.Initial standard of care lab studies for the surgical and non-surgical groups are shown in the Table .Only total WBC was statistically different: surgery group median WBC 12.4(9.3-16.7)Thou/uL vs. non-surgery group 10.1(6.9-12.7)Thou/uL (p= 0.034).Conclusions: Despite intravenous antibiotics and percutaneous drainage in many, over 75% of children presenting with B3 CD have resectional therapy within the first year.Few pediatric patients presenting with B3 CD are treated with anti-TNF agents prior to surgery.Newer management therapies are needed for these difficult patients.