Abstract
This secondary analysis of a randomized clinical trial evaluates the effectiveness of ergocalciferol vs placebo in youths with newly diagnosed type 1 diabetes.
Introduction
Approximately 30% to 50% residual β-cell function may remain at the time of type 1 diabetes (T1D) diagnosis, and this may persist for months or years.1,2 A prolonged partial remission (PR) phase of T1D leads to improved glycemic control and decreased long-term complications.1 We previously reported1 that ergocalciferol significantly decreased circulating tumor necrosis factor (TNF)-α and temporal trends in both hemoglobin A1c (HbA1c) and insulin dose–adjusted A1c (IDAA1c), a marker of PR, compared with placebo. Here, we report the effect size of high-dose ergocalciferol (50 000 IU/wk for 2 months, then biweekly for 10 months) vs placebo on β-cell function, denoted by the ratio of fasting proinsulin to C-peptide (PI:C) and the percent change from baseline in the area under the curve (%∆AUC) of C-peptide.
Methods
We conducted a post hoc secondary analysis of a single-center, double-blind, placebo-controlled, parallel-group randomized clinical trial of ergocalciferol vs placebo in youths (aged 10-21 years) with newly diagnosed T1D. The trial was conducted at the University of Massachusetts Medical Center (UMMC) in Worcester from October 19, 2017, to April 12, 2021 (ClinicalTrials.gov identifier: NCT03046927). The methodology was published previously.1 The UMMC Institutional Review Board approved the study protocol (Supplement 1) with modifications (Supplement 2).3 We obtained written informed consent from adults and parents and assent from youths (aged <18 years). The study followed the CONSORT reporting guideline.
Exclusion and inclusion criteria were reported previously1 and included fasting C-peptide (>0.1 nmol/L [0.3 ng/mL]) or stimulated C-peptide (≥0.2 nmol/L [≥0.6 ng/mL]) and a positive diabetes-associated autoantibody profile. Participants entered a run-in phase of 1 to 2 months, maintained a treat-to-target insulin regimen, and were subsequently randomized to ergocalciferol or placebo. Participants completed mixed-meal tolerance tests to estimate C-peptide, fasting glucose, and proinsulin (normal range, 3.6-22 pmol/L) at 0, 3, 6, 9, and 12 months (Table).
Table. Fasting Proinsulin, Fasting C-Peptide, and Corresponding PI:Cs During the Randomized Clinical Trial.
| Group | Mean (SD) | ||
|---|---|---|---|
| Proinsulin, pmol/L | C-peptide, pmol/L | PI:C | |
| Ergocalciferol, mo | |||
| 0 | 9.56 (16.91) | 265.2 (271.6) | 0.028 (0.015) |
| 3 | 5.13 (5.25) | 195.7 (133.2) | 0.025 (0.011) |
| 6 | 5.25 (6.60) | 152.7 (95.7) | 0.052 (0.099) |
| 9 | 4.68 (4.11) | 140.0 (78.4) | 0.040 (0.053) |
| 12 | 2.75 (2.11) | 114.5 (69.4) | 0.026 (0.013) |
| Placebo, mo | |||
| 0 | 10.08 (19.01) | 225.9 (131.7) | 0.039 (0.071) |
| 3 | 6.31 (5.26) | 215.0 (126.9) | 0.026 (0.016) |
| 6 | 8.13 (10.99) | 238.7 (187.4) | 0.031 (0.017) |
| 9 | 5.14 (5.01) | 146.1 (78.0) | 0.029 (0.012) |
| 12 | 6.82 (9.51) | 166.2 (155.6) | 0.039 (0.022) |
Abbreviation: PI:C, proinsulin to C-peptide ratio.
The sample size and power calculation were reported previously.1 Statistical analysis was based on the intent-to-treat principle. We compared differences in fasting PI:C trends between groups using a repeated-measures generalized linear model with normal distribution, logarithmic link function, and unstructured correlations.
We calculated %∆AUC C-peptide using the trapezoidal method4,5 and used a random intercept model adjusted for sex, age, and race to characterize mean %∆AUC C-peptide from 0 to 12 months. Statistical analyses were performed using SAS, version 9.4 (SAS Institute Inc). Box plots were generated using Prism, version 10 (GraphPad Inc). All tests were 2 sided; P < .05 was considered significant. Data analysis was performed from October 5 to November 15, 2023.
Results
Of 48 youths with T1D eligible for the 12-month trial, 36 (24 males [66.7%], 12 females [33.3%]) were randomized to ergocalciferol or placebo (eFigure in Supplement 3). Their mean (SD) age was 13.5 (2.8) years; 2 were Asian (5.6%), 2 were Black (5.6%), 27 (75.0%) were White, and 5 (13.8%) did not report their race. Ergocalciferol significantly decreased fasting PI:C vs placebo (mean [SE], −0.0009 [0.0008] vs 0.0011 [0.0003]; P = .01; Figure, A and B) for the monthly overall difference in trends. The mean (SD) decrease in %∆AUC C-peptide was similar for both groups in the first 3 months (−10.9 [6.3] vs −8.2 [7.0]; P = .99) but subsequently decreased more slowly with ergocalciferol vs placebo (−28.4 [6.2]; P < .001 vs −41.5 [5.9]; P < .001), with a significant reduction in monthly overall temporal trends (mean [SE], −2.8% [0.7] vs −4.7 % [0.6]; P = .03; Figure, C).
Figure. Fasting Proinsulin, Fasting C-Peptide, and Corresponding Proinsulin to C-Peptide Ratios (PI:Cs).
A and B, Observed (A) and model-predicted (B) PI:Cs. Trends were generated from a repeated-measures generalized linear model of fasting PI:Cs. The total number of repeated-measures observations was 149 from 36 participants (18 per group). Three observed values were greater than 0.2; these were considered extreme outliers and were removed (in A). The remaining observations ranged from 0.005 to 0.087. The error distribution was normal, the link function was logarithmic, the repeated-measures correlation was unstructured, and the difference in trends between the 2 groups was significant (P = .01). C, Overall analysis of the trends showed that ergocalciferol significantly slowed the decline in percentage AUC C-peptide from baseline compared with placebo (P = .03).
Discussion
In this study, ergocalciferol significantly decreased PI:C and slowed the decrease in %∆AUC C-peptide among youths with T1D. We previously showed that ergocalciferol significantly reduced temporal trends in HbA1c, IDAA1c, and TNF-α.1 Although this randomized clinical trial was limited by its single-center setting, the results suggest a protective action of ergocalciferol on β cells and possible mechanisms of action to prolong PR. Ergocalciferol’s ∆ effect size for β-cell protection (15%) is comparable to that of imatinib,6 verapamil,4 and other agents (15%-19.4%). Thus, vitamin D may be combined with other treatments (eg, teplizumab and baricitinib) to prolong PR.
Trial Protocol
Trial Protocol. Summary of IRB-Approved Amendments/Modifications to the Investigational Study Protocol With the Associated Dates
eFigure. CONSORT Flow Diagram
Data Sharing Statement
References
- 1.Nwosu BU, Parajuli S, Jasmin G, et al. Ergocalciferol in new-onset type 1 diabetes: a randomized controlled trial. J Endocr Soc. 2021;6(1):bvab179. doi: 10.1210/jendso/bvab179 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Yu MG, Keenan HA, Shah HS, et al. Residual β cell function and monogenic variants in long-duration type 1 diabetes patients. J Clin Invest. 2019;129(8):3252-3263. doi: 10.1172/JCI127397 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Nwosu BU, Parajuli S, Jasmin G, et al. Investigational study protocol. 2021. Accessed January 27, 2024. https://repository.escholarship.umassmed.edu/handle/20.500.14038/28276
- 4.Forlenza GP, McVean J, Beck RW, et al. ; CLVer Study Group . Effect of verapamil on pancreatic beta cell function in newly diagnosed pediatric type 1 diabetes: a randomized clinical trial. JAMA. 2023;329(12):990-999. doi: 10.1001/jama.2023.2064 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Allison DB, Paultre F, Maggio C, Mezzitis N, Pi-Sunyer FX. The use of areas under curves in diabetes research. Diabetes Care. 1995;18(2):245-250. doi: 10.2337/diacare.18.2.245 [DOI] [PubMed] [Google Scholar]
- 6.Gitelman SE, Bundy BN, Ferrannini E, et al. ; Gleevec Trial Study Group . Imatinib therapy for patients with recent-onset type 1 diabetes: a multicentre, randomised, double-blind, placebo-controlled, phase 2 trial. Lancet Diabetes Endocrinol. 2021;9(8):502-514. doi: 10.1016/S2213-8587(21)00139-X [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Trial Protocol
Trial Protocol. Summary of IRB-Approved Amendments/Modifications to the Investigational Study Protocol With the Associated Dates
eFigure. CONSORT Flow Diagram
Data Sharing Statement