M41 - DEPRESSION, INFLAMMATION, AND METABOLIC RISK: A GENETICALLY-INFORMED EXPLORATORY STUDY

Abstract

Background Depression is a highly prevalent psychiatric disorder, which has been observed to have greater-than-expected comorbidity with metabolic and inflammatory disorders. Although standard observational study designs encounter limitations when assessing the directionality of the relationships underlying the comorbidities, twin studies offer a means to resolve competing explanations of comorbidity. The purpose of this study is to examine the relationship between depression and biomarkers associated with inflammation and metabolic risk by examining patterns of correlation in monozygotic twin pairs with different depression histories. Methods The sample consisted of 45 monozygotic twin pairs (mean age: 52.8, 69% female, 76% white) assessed twice over a 6-month period. Lifetime history of Major Depressive Disorder (MDD) was assessed using the Diagnostic Interview Schedule. Biological measures included peripheral leukocyte gene expression (data not yet available), hemoglobin A1c (HbA1c, a measure indicative of recent blood sugar levels), and two markers of systemic inflammation, C-reactive Reactive Protein (CRP), and interleukin-6 (IL-6). Intraclass correlation coefficients (ICC) were calculated across members of each twin pair, and within-individual across timepoints. Biomarker levels were also compared between members of twin pairs discordant for MDD, and with twins from pairs in which neither member had a history of MDD. Results Of the 40 pairs with known MDD history, 13 (32.5%) had no history of MDD in either member, 14 (35%) were discordant for MDD, and 13 (32.5%) had a lifetime history of MDD in both twins. HbA1c showed a low ICC between twins (0.363), but a high ICC within individuals over time (0.873). CRP had a low ICC both between twins (0.189) and within individuals over time (0.174). IL-6 had a moderately high ICC both between twins (0.571) and within individuals over time (0.691). In MDD-discordant pairs, the twins with MDD had higher median HbA1c (5.55%) than their non-MDD co-twins (5.45%), while twins from MDD-free pairs were not notably different (5.50%). Change in HbA1c was highest in the twin with MDD in the discordant pairs (0.25%), followed by their non-MDD co-twins (0.20%), and lowest in MDD-free pairs (0.10%). In discordant twin pairs, the twin with a history of MDD had lower median CRP (1810.95 ng/ml) than the twin without MDD (2683.98 ng/ml), while the lowest CRP was found in MDD-free pairs (827.213 ng/ml). For discordant pairs, the twin with MDD had higher median IL-6 (1.53 pg/ml) than their non-MDD co-twins (1.08 pg/ml), and twins from MDD-free pairs had the lowest IL-6 (0.658 pg/ml). Discussion The differing patterns in ICC suggest that HbA1c is greatly influenced by individual-specific environmental factors and moderately influenced by genetic or shared environmental factors. In contrast, circulating levels of CRP are largely influenced by short-term environmental factors. Finally, circulating levels of IL-6 have a notable contribution from genetic or environmental factors shared between twins. Levels of IL-6 and change in HbA1c over time followed a pattern of highest values in twins with MDD, their non-MDD co-twin, followed by twins from MDD-free pairs. This is consistent with the hypothesis that MDD has an influence on these biomarkers, but much of that influence is driven by genetic liability for MDD. These preliminary results demonstrate the value of longitudinal studies using monozygotic twins, although further analysis is needed.