Introduction and Objective: Pancreas size declines in individuals progressing from Stage 1/2 to Stage 3 T1D. Whether teplizumab treatment, which delays progression to Stage 3 T1D, affects pancreas size is not known. Methods: We performed longitudinal MRI using a standardized protocol at two sites (27 MRIs total) in ten individuals with Stage 2 T1D (mean age 16, range 9 - 64, 30% female) treated with teplizumab. Pancreas volume was quantified by a blinded reader and divided by body weight to yield pancreas volume index (PVI). Results: Individuals treated with teplizumab displayed a wide range of baseline PVI at infusion (0.37 ml/kg to 1.68 ml/kg). The mean baseline PVI was 0.85 ml/kg, similar to established PVI in Stage 1 and 2 T1D (PMID: 38151474). PVI transiently increased in individuals treated with teplizumab at ~six months post-infusion. In contrast, PVI was stable in individuals with Stage 1 or Stage 2 T1D not treated with Teplizumab (PMID: 38151474) who do not progress to Stage 3. Conclusion: These data suggest that teplizumab induces a transient increase in pancreas size six months post infusion which returns to baseline one year post infusion. Ongoing studies will determine how initial volume and dynamic changes correspond to treatment outcomes. Disclosure J. Virostko: None. J. Wright: None. J. Williams: None. D. Braun: None. M.A. Hilmes: None. S.W. Greeley: None. A.C. Powers: None. D.J. Moore: Advisory Panel; Current; Sanofi-Aventis U.S. Funding Helmsley Charitable Trust (2201-05374)

Introduction and Objective: Some individuals with T2D have islet autoantibodies (aab). To determine if the presence of autoantibodies represents a form of islet-directed autoimmunity, we integrated and compared serologic, functional, genetic, and histologic data from HPAP-T2D donors with (T2Daab+, n=6) or without (T2Daab-, n=47) aab, and donors with no diabetes (ND, N=96). Methods: Donor clinical and demographic data, islet perifusion results, genetic risk scores (GRS), and multiplex immunofluorescent images (CODEX) are available on the HPAP website, PancDB. Immunofluorescent images were quantified using AI-assisted segmentation and classification algorithms in QuPath software. Statistical significance (p<0.05) was tested by ANOVA. Results: Compared to ND donors, T2D donors had higher BMI and T2D-GRS, lower residual C-peptide, but no difference in relative pancreas weight (RPW). Compared to T2Daab-, T2Daab+ (4 GADA+, 2 IA2+) individuals were similar in age, BMI, disease duration, T1D-GRS, and T2D-GRS. T2Daab+ individuals had reduced islet function, with lower C-peptide levels (1.2±0.3 vs 3.7±0.5 ng/mL, p=0.001), lower glucose-stimulated insulin release from isolated islets (1.2±0.1 vs 1.6±0.2 stimulation index, p=0.07), and higher HbA1c (8.8±0.5 vs 7.4±0.3 %, p=0.03) than T2Daab-. RPW of T2Daab+ individuals was intermediate between T2Daab- and HPAP-T1D donors (0.86±0.08 vs 1.03±0.07 vs 0.68±0.04 g/kg, p=0.001). In pancreatic sections, the T cell/islet cell ratio was higher in T2Daab+ than in ND controls (0.38±0.09 vs 0.11±0.02; p=0.03), but was not different between T2Daab- and controls. Sections from one of six T2Daab+ donors was amyloid positive, compared to 24/41 T2Daab-donors being amyloid positive. Conclusion: Among T2D donors, the presence of islet autoantibodies was associated with histologic evidence of latent autoimmunity, reduced islet function, and decreased pancreas size, suggesting that autoimmune processes play a role in their diabetes. Disclosure M.M. Wimalarathne: None. A. Eskaros: None. S. Spearman: None. F. Feng: None. A.L. Hopkirk: None. K.S. Lee: None. P.B. Jackson: None. J. Haynes: None. R. Jenkins: None. R.A. Brantley: None. A. Pandey: None. S.A. Sharp: None. H. Sun: None. E. Manduchi: None. R. O’Flynn: None. N.M. Doliba: None. J. Cartailler: None. S. Shapira: None. C. Dai: None. P. Wilson: None. D.C. Saunders: None. M. Brissova: None. A.L. Gloyn: Speaker's Bureau; Ended; Novo Nordisk. Other - Spouse is an employee and holds stock options in Roche; Current; Genentech, Inc. A. Naji: None. K. Kaestner: None. A.C. Powers: None. J. Wright: None. Funding National Institutes of Health (DK112217, DK123594, DK123716, DK112232), Veterans Administration (BX005910, BX000666)

Objective This multicenter prospective cohort study compared pancreas volume as assessed by MRI, metabolic scores derived from oral glucose tolerance testing (OGTT), and a combination of pancreas volume and metabolic scores for predicting progression to Stage 3 type 1 diabetes (T1D) in individuals with multiple diabetes-related autoantibodies. Research Design and Methods Pancreas MRI were performed in 65 multiple autoantibody-positive participants enrolled in the Type 1 Diabetes TrialNet Pathway to Prevention study. Prediction of progression to Stage 3 T1D was assessed using pancreas volume index, OGTT-derived Index60 and Diabetes Prevention Trial-Type 1 Risk Score (DPTRS) scores, as well as a combination of pancreas volume index and DPTRS. Results Pancreas volume index, Index60, and DPTRS were all significantly different at study entry in 11 individuals who subsequently progressed to Stage 3 T1D compared with 54 participants who did not progress (p < 0.005). Pancreas volume indexdid not correlate with metabolic testing across individual study participants. Pancreas volume index declined longitudinally in the 11 individuals diagnosed with Stage 3 T1D, while Index60 and DPTRS increased. The area under the receiver operating curve for predicting progression to Stage 3 from measurements at study entry was 0.76 for pancreas volume index, 0.79 for Index60, 0.79 for DPTRS, and 0.91 for pancreas volume index and DPTRS. Conclusions These findings suggest that measures of pancreas volume and metabolism reflect distinct components of risk for developing Stage 3 type 1 diabetes and that combining these measures may provide superior prediction than either alone.

Background: Growing evidence suggests that medical students and faculty preceptors have limited familiarity with continuous glucose monitors (CGMs), which may negatively affect how care teams counsel patients and monitor the use of CGMs. Although studies have shown that structured training for care teams may improve knowledge and comfort with various healthcare tools, to our knowledge, no study has attempted to do so for CGMs. Methods: We designed a user experience course for medical students and faculty mentors to address this gap at a student-run free clinic. This course allowed twenty participants to wear a CGM for two weeks and participate in three interactive didactic sessions and group reflections. We evaluated how knowledge and comfort with CGMs among participants changed after the course with a survey and focus group. Results: The cohort showed improvement in self-reported confidence in using the device, teaching patients how to use the device, and interpreting data (p<0.001). The majority of participants demonstrated improvement across all survey domains with higher post-intervention scores than pre-intervention scores. Qualitative analysis of group reflections elucidated three primary themes across participant experiences: the emotional impact of wearing the device, attitude changes with prolonged use, and behavior modification in response to glucose data. Conclusion: This novel educational initiative may improve knowledge about CGMs, ability to counsel patients to use the device, and understanding patients' experiences among medical students and faculty mentors. We plan to expand this educational opportunity to additional clinic volunteers, include patient perspectives, and share the curriculum with other student-run clinics.

Introduction & Objective: Islet amyloid composed of islet amyloid polypeptide (IAPP) is often present in the pancreas in type 2 diabetes. Cell culture and transgenic rodent models have been used to study islet amyloid toxicity, but less is known about the effects of endogenously-produced IAPP on human islet function and health. Methods: To evaluate acute and chronic effects of amyloid formation, we established a pseudoislet platform with adenoviral delivery of constructs expressing human IAPP (huIAPP) or mouse IAPP (moIAPP), which is not amyloidogenic. After 7 days in culture +/- exposure to the fatty acid palmitate, pseudoislets were evaluated for secretory function, islet structure and amyloid formation by ThioS staining. Pseudoislets were also transplanted into the anterior chamber of the eye of immunodeficient mice for assessment of secretory function for up to 3 months. Results: After 7 days in culture, transduced pseudoislets had 6.3 fold greater cellular IAPP content than untransduced pseudoislets. HuIAPP expression resulted in greater amyloid formation (0.49 ± 0.12 vs 0.17 ± 0.07 % islet area), but no changes in insulin or glucagon secretion or cellular toxicity, compared with moIAPP pseudoislets. High resolution microscopy showed ThioS+ fibrils in the ER of huIAPP-expressing beta cells. No extracellular amyloid was seen. Exposure to 0.5mM palmitate for 3 days further increased amyloid deposition and impaired insulin secretion by huIAPP pseudoislets, but not moIAPP. HuIAPP pseudoislets transplanted into immunodeficient mice had normal insulin secretion after 4 and 8 weeks, but impaired secretion after 12 weeks, compared to moIAPP pseudoislets. Conclusion: In vitro overexpression of IAPP in human pseudoislets induces intracellular amyloid formation, and in combination with lipotoxic stress, impairs insulin secretion. In vivo, human IAPP overexpression impairs insulin secretion after three months. This model will be valuable to elucidate mechanisms of islet dysfunction in type 2 diabetes. Disclosure J.J. Wright: None. P.B. Jackson: None. R. Aramandla: None. C. Davis: None. M. Brissova: None. A.C. Powers: None. Funding National Institutes of Health (K08DK133691); Veterans Administration (IK2BX005910)

The endocrine and exocrine compartments of the pancreas are spatially related but functionally distinct. Multiple diseases affect both compartments, including type 1 diabetes (T1D), pancreatitis, cystic fibrosis, and pancreatic cancer. To better understand how the exocrine pancreas changes with age, obesity, and diabetes, we performed systematic analysis of wellpreserved tissue sections from the pancreatic head, body, and tail of organ donors with T1D (n = 20), type 2 diabetes (T2D, n = 25), and donors with no diabetes (ND, n = 74). Among ND donors, we found that acinar-to-ductal metaplasia (ADM), angiopathy, and pancreatic adiposity increased with age, while ADM and adiposity also increased with BMI. Compared to age- and sex-matched ND organs, T1D pancreata had greater acinar atrophy and angiopathy with fewer intralobular adipocytes. T2D pancreata had greater ADM, angiopathy, and total T lymphocytes, but no difference in adipocyte number, compared to ND organs. While total pancreatic fibrosis was increased in both T1D and T2D, the pattern was different with T1D pancreata having greater periductal and perivascular fibrosis, whereas T2D pancreata had greater lobular and parenchymal fibrosis. Thus, the exocrine pancreas undergoes distinct changes as individuals age or develop T1D or T2D.

Increased blood amino acid levels (hyperaminoacidemia) stimulate pancreas expansion by unclear mechanisms. Here, by genetic and pharmacological disruption of glucagon receptor (GCGR) in mice and zebrafish, we found that the ensuing hyperaminoacidemia promotes pancreatic acinar cell proliferation and cell hypertrophy, which can be mitigated by a low protein diet in mice. In addition to mammalian target of rapamycin complex 1 (mTORC1) signaling, acinar cell proliferation required slc38a5, the most highly expressed amino acid transporter gene in both species. Transcriptomics data revealed the activation signature of yes-associated protein (YAP) in acinar cells of mice with hyperaminoacidemia, consistent with the observed increase in YAP-expressing acinar cells. Yap1 activation also occurred in acinar cells in gcgr-/- zebrafish, which was reversed by rapamycin. Knocking down yap1 in gcgr-/- zebrafish decreased mTORC1 activity and acinar cell proliferation and hypertrophy. Thus, the study discovered a previously unrecognized role of the YAP/Taz pathway in hyperaminoacidemia-induced acinar cell hypertrophy and hyperplasia.

Importance Natural language processing tools, such as ChatGPT (generative pretrained transformer, hereafter referred to as chatbot), have the potential to radically enhance the accessibility of medical information for health professionals and patients. Assessing the safety and efficacy of these tools in answering physician-generated questions is critical to determining their suitability in clinical settings, facilitating complex decision-making, and optimizing health care efficiency.Objective To assess the accuracy and comprehensiveness of chatbot-generated responses to physician-developed medical queries, highlighting the reliability and limitations of artificial intelligence-generated medical information.Design, Setting, and Participants Thirty-three physicians across 17 specialties generated 284 medical questions that they subjectively classified as easy, medium, or hard with either binary (yes or no) or descriptive answers. The physicians then graded the chatbot-generated answers to these questions for accuracy (6-point Likert scale with 1 being completely incorrect and 6 being completely correct) and completeness (3-point Likert scale, with 1 being incomplete and 3 being complete plus additional context). Scores were summarized with descriptive statistics and compared using the Mann-Whitney U test or the Kruskal-Wallis test. The study (including data analysis) was conducted from January to May 2023.Main Outcomes and Measures Accuracy, completeness, and consistency over time and between 2 different versions (GPT-3.5 and GPT-4) of chatbot-generated medical responses.Results Across all questions (n = 284) generated by 33 physicians (31 faculty members and 2 recent graduates from residency or fellowship programs) across 17 specialties, the median accuracy score was 5.5 (IQR, 4.0-6.0) (between almost completely and complete correct) with a mean (SD) score of 4.8 (1.6) (between mostly and almost completely correct). The median completeness score was 3.0 (IQR, 2.0-3.0) (complete and comprehensive) with a mean (SD) score of 2.5 (0.7). For questions rated easy, medium, and hard, the median accuracy scores were 6.0 (IQR, 5.0-6.0), 5.5 (IQR, 5.0-6.0), and 5.0 (IQR, 4.0-6.0), respectively (mean [SD] scores were 5.0 [1.5], 4.7 [1.7], and 4.6 [1.6], respectively; P = .05). Accuracy scores for binary and descriptive questions were similar (median score, 6.0 [IQR, 4.0-6.0] vs 5.0 [IQR, 3.4-6.0]; mean [SD] score, 4.9 [1.6] vs 4.7 [1.6]; P = .07). Of 36 questions with scores of 1.0 to 2.0, 34 were requeried or regraded 8 to 17 days later with substantial improvement (median score 2.0 [IQR, 1.0-3.0] vs 4.0 [IQR, 2.0-5.3]; P < .01). A subset of questions, regardless of initial scores (version 3.5), were regenerated and rescored using version 4 with improvement (mean accuracy [SD] score, 5.2 [1.5] vs 5.7 [0.8]; median score, 6.0 [IQR, 5.0-6.0] for original and 6.0 [IQR, 6.0-6.0] for rescored; P = .002).Conclusions and Relevance In this cross-sectional study, chatbot generated largely accurate information to diverse medical queries as judged by academic physician specialists with improvement over time, although it had important limitations. Further research and model development are needed to correct inaccuracies and for validation.

OBJECTIVE:To determine the mechanism of reduced pancreas size in type 1 diabetes and the significance of islet-derived insulin in pancreatic growth.RESEARCH DESIGN AND METHODS:Using a validated and standardized MRI protocol, we measured pancreas volume and shape in a family with an autosomal-dominant insulin gene mutation that results in insulin deficiency similar in severity to that of type 1 diabetes but without autoimmunity. DNA sequencing confirmed the mutation in all four affected individuals and none of the four control family members. Insulin secretory capacity was determined by measuring postprandial urinary C-peptide.RESULTS:Family members with this form of monogenic diabetes had a markedly smaller pancreas and a severely impaired postprandial C-peptide level than family members without diabetes.CONCLUSIONS:These results suggest that severe insulin deficiency, rather than islet-directed autoimmunity, leads to reduced pancreas size in type 1 diabetes and that insulin is a major trophic factor for the exocrine pancreas.

Background: Natural language processing models such as ChatGPT can generate text-based content and are poised to become a major information source in medicine and beyond. The accuracy and completeness of ChatGPT for medical queries is not known. Methods: Thirty-three physicians across 17 specialties generated 284 medical questions that they subjectively classified as easy, medium, or hard with either binary (yes/no) or descriptive answers. The physicians then graded ChatGPT-generated answers to these questions for accuracy (6-point Likert scale; range 1 – completely incorrect to 6 – completely correct) and completeness (3-point Likert scale; range 1 – incomplete to 3 - complete plus additional context). Scores were summarized with descriptive statistics and compared using Mann-Whitney U or Kruskal-Wallis testing. Results: Across all questions (n=284), median accuracy score was 5.5 (between almost completely and completely correct) with mean score of 4.8 (between mostly and almost completely correct). Median completeness score was 3 (complete and comprehensive) with mean score of 2.5. For questions rated easy, medium, and hard, median accuracy scores were 6, 5.5, and 5 (mean 5.0, 4.7, and 4.6; p=0.05). Accuracy scores for binary and descriptive questions were similar (median 6 vs. 5; mean 4.9 vs. 4.7; p=0.07). Of 36 questions with scores of 1-2, 34 were re-queried/re-graded 8-17 days later with substantial improvement (median 2 vs. 4; p<0.01). Conclusions: ChatGPT generated largely accurate information to diverse medical queries as judged by academic physician specialists although with important limitations. Further research and model development are needed to correct inaccuracies and for validation.

INTRODUCTION:In hospitalized patients, continuous glucose monitoring (CGM) may improve glycemic control, prevent hypoglycemic events, and reduce staff workload compared with point-of-care (POC) capillary glucose monitoring. METHODS:To evaluate CGM accuracy and safety of use in the inpatient setting, two versions of CGM sensors were placed on 43 and 34 adult patients with diabetes admitted to non-intensive care unit (ICU) medical wards, respectively. CGM accuracy relative to POC and safety of use were measured by calculating mean absolute relative difference (MARD) and by Clarke Error Grid (CEG) analysis. RESULTS:CGM version 2 had improved accuracy compared with CGM version 1 with MARD 17.7 compared with 21.4%. CGM accuracy did not change with POC value or with time of sensor wear. On CEG, 98.8% of paired values fell within acceptable zones A and B. CONCLUSION:Despite reduced accuracy compared with the outpatient setting, both versions of CGMs had acceptable safety profiles in the inpatient setting.