Is acute-phase serum amyloid a protein a risk factor for type 2 diabetes

Type 2 diabetes is a metabolic disorder with globally increasing
prevalence. Therefore, the identification of etiological factors is
of ascending relevance for the understanding, treatment, and
prevention of the disease. Levels of the acute-phase serum amyloid
A (A-SAA) protein have been found to be elevated in type 2 diabetic
subjects, but little is known about their causal implication in the
development of type 2 diabetes so far. This doctoral thesis
presents an epidemiological perspective on the association between
circulating levels of A-SAA and risk of type 2 diabetes and
assesses a possible causality in this association using a genetic
approach. Three studies were conducted. In a prospective cohort
study, A-SAA levels were measured in 836 initially non-diabetic,
elderly, Western European subjects without clinically overt
inflammation who participated in a seven-year follow-up
examination. Results of this study provided first evidence that
levels of A-SAA are elevated years before the manifestation of type
2 diabetes independent of other type 2 diabetes risk factors.
However, adjustment for parameters related to glucose metabolism,
particularly levels of 2h-glucose, attenuated the association
suggesting a potential link via post-challenge hyperglycemia in the
association between elevated levels of A-SAA and type 2 diabetes
or, alternatively, a possible reverse causality between levels of
A-SAA and 2h-glucose. In a meta-analysis of genome-wide association
studies (GWAS) on levels of A-SAA conducted in three
population-based studies and one prospective case-cohort study
including a total of 4,212 participants of European descent two
biologically highly plausible genetic susceptibility loci for A-SAA
proteins at chromosome 11p15.5-p13 and chromosome 1p31 were
identified. One of these loci represented a suitable candidate for
a Mendelian Randomization study. In Mendelian Randomization
studies, genetic variants are used as proxies for a biomarker.
These studies benefit from the fact that genotypes are randomly
assorted at meiosis and are largely independent of non-genetic
confounding and disease processes. Thus, they constitute a genetic
approach to assess whether the association between a biomarker and
a disease is causal. The associations between genetic variants of
the candidate locus and type 2 diabetes were extracted from the
results of a meta-analysis of eight GWAS (8,130 cases, 38,987
controls) published by DIAGRAM, a large diabetes and genetic
consortium. In spite of sufficient power, the above mentioned
associations were not significant suggesting that there are genetic
mechanisms that raise plasma levels of A-SAA without translating
into an increase in type 2 diabetes risk. In conclusion, results of
this doctoral thesis indicated that levels of A-SAA are elevated
years before the manifestation of type 2 diabetes but could not
provide evidence that the association is truly causal using a
genetic approach. Rather it seems likely that the association
between levels of A-SAA and risk of type 2 diabetes is
substantially influenced by post-challenge hyperglycemia.
Time-series studies are warranted to elucidate the role of
post-challenge hyperglycemia in this association.