Frequent Monitoring of C-Peptide Levels in Newly Diagnosed Type 1 Subjects Using Dried Blood Spots Collected at Home

• Published in: The journal of clinical endocrinology and metabolism Vol. 103; no. 9; pp. 3350 - 3358
• Main Authors: Willemsen, Ruben H, Burling, Keith, Barker, Peter, Ackland, Fran, Dias, Renuka P, Edge, Julie, Smith, Anne, Todd, John, Lopez, Boryana, Mander, Adrian P, Guy, Catherine, Dunger, David B
• Format: Journal Article
• Language: English
• Published: United States Copyright Oxford University Press 01.09.2018; Oxford University Press; Endocrine Society
• Subjects: Adolescent; Area Under Curve; Beta cells; Blood Glucose - analysis; Blood Glucose Self-Monitoring - statistics & numerical data; C-Peptide - blood; Child; Clinical s; Correlation of Data; Diabetes; Diabetes mellitus (insulin dependent); Diabetes Mellitus, Type 1 - blood; Diagnosis; Dried Blood Spot Testing - methods; Dried Blood Spot Testing - statistics & numerical data; Fasting; Fasting - blood; Feasibility Studies; Female; Glucose; Humans; Laboratory testing; Male; Peptides; Time Factors
• ISSN: 0021-972X; 1945-7197; 1945-7197
• DOI: 10.1210/jc.2018-00500

To evaluate an approach to measure β-cell function by frequent testing of C-peptide concentrations in dried blood spots (DBSs). Thirty-two children, aged 7 to 17 years, with a recent diagnosis of type 1 diabetes. Mixed-meal tolerance test (MMTT) within 6 and again at 12 months after diagnosis, with paired venous and DBS C-peptide sampling at 0 and 90 minutes. Weekly DBS C-peptide before and after standardized breakfasts collected at home. DBS and plasma C-peptide levels (n = 115) correlated strongly (r = 0·91; P < 0.001). The Bland-Altman plot indicated good agreement. The median number of home-collected DBS cards per participant was 24 over a median of 6.9 months. Repeated DBS C-peptide levels varied considerably within and between subjects. Adjustment for corresponding home glucose measurements reduced the variance, permitting accurate description of changes over time. The correlation of the C-peptide slope over time (assessed by repeated home DBS) vs area under the curve during the two MMTTs was r = 0.73 (P < 0.001). Mixed models showed that a 1-month increase in diabetes duration was associated with 17-pmol/L decline in fasting DBS C-peptide, whereas increases of 1 mmol/L in glucose, 1 year older age at diagnosis, and 100 pmol/L higher baseline plasma C-peptide were associated with 18, 17, and 61 pmol/L higher fasting DBS C-peptide levels, respectively. In addition, glucose responsiveness decreased with longer diabetes duration. Our approach permitted frequent assessment of C-peptide, making it feasible to monitor β-cell function at home. Evaluation of changes in the slope of C-peptide through this method may permit short-term evaluation of promising interventions.