Proximity extension assay inflammatory profiling cannot distinguish the presence of residual C-peptide in patients with long-standing type 1 diabetes

Abstract

Objective: Many patients with long-standing type 1 diabetes (T1D) have remaining low levels of C-peptide, i.e. and indirect sign of remaining functional beta-cells. This study focused on identifying differences in immunological and inflammatory biomarkers in patients with longstanding T1D and remaining C-peptide.

Research design and methods: Adult patients (n = 120) with long-standing T1D (≥ 10 years) and healthy controls (HC) (n = 50) were recruited at Uppsala University Hospital. Residual beta-cell function was determined with an ultrasensitive C-peptide ELISA under fasting conditions. T1D patients were divided into two groups (C-peptide positive vs. C-peptide negative). Using the OLINK Explore Inflammation proximity extension assay (PEA), 368 circulating immunological and inflammatory biomarkers were analyzed in plasma.

Results: The three groups could not be distinguished by principal component analysis and when correcting for multiple testing we found no differences in circulating biomarkers. However, based on uncorrected p-values there were six biomarkers that were different when comparing all T1D patients with HC and eight markers that were different when comparing C-peptide positive vs. negative T1D patients.

Conclusion: A wide inflammatory assay analysis cannot distinguish patients with longstanding T1D and remaining C-peptide from patients with a complete loss of C-peptide nor from HC.

Keywords: Biomarkers; C-peptide; Proximity extension assay; Type 1 diabetes.

Fig. 1

Principal component analysis (PCA) of the analyzed biomarkers. A PCA plot comparing healthy controls (green) and all patients with longstanding T1D (red), however no clear distinction between the two groups could be identified. B PCA plot comparing healthy controls (green), longstanding T1D patients with remaining C-peptide (blue) and those without remaining C-peptide (red). Again, no distinction between the groups was observed. Each point in the PCA-plots represents one and the position of the point is based on all measured protein values. The percentages displayed show the percentage of explained variance per principal component

Fig. 2

Comparison of individual biomarkers in healthy controls (HC) vs. longstanding type 1 diabetes (T1D). Boxplots of the top six assays in healthy controls vs. patients with longstanding T1D that were found to be significantly different based on uncorrected p-values. The boxplots show the distribution of data (NPX) for the grouping variable. The thick black horizontal line represents the median, the colored area - the interquartile range, the ends of the vertical middle line represent the minimum and the maximum values except for the black dots which are potential assay outliers

Fig. 3

Comparison of individual biomarkers in patients with longstanding type 1 diabetes with- or without-remaining C-peptide. Boxplots of the top eight assays found to be significantly different based on uncorrected p-values in T1D patients with remaining C-peptide (red, C-pepPOS) compared to those without remaining C-peptide (blue, C-pepNEG). The boxplot shows the distribution of data (NPX) for the grouping variable. The thick black horizontal line represents the median, the colored area—the interquartile range. The ends of the vertical middle line represent the minimum and the maximum values except for the black dots which are potential assay outliers