The Insulin-Like Growth Factor Axis: A New Player in Gestational Diabetes Mellitus?

Gestational diabetes mellitus (GDM), defined as glucose intolerance of varying severity with first onset and recognition in pregnancy, is a diagnosis that holds relevance for the diabetes epidemic across two generations—that of the mother and her child (1,2). Indeed, despite typically regaining normal glucose tolerance in the immediate postpartum, women with previous GDM have a very high risk of future progression to type 2 diabetes (T2D) (3). Similarly, their offspring have an elevated risk of accruing metabolic abnormalities in childhood that may be partly attributable to the intrauterine environment of the GDM pregnancy (4). As such, enhanced understanding of the pathophysiology of GDM could yield strategies for early identification of at-risk mothers and ideally mitigation of their metabolic risk, to the benefit of both mother and child (2).

Current understanding of the pathophysiology of GDM holds that affected women have a defect in pancreatic β-cell function that first manifests clinically as an inability to fully compensate for the marked insulin resistance of the latter half of pregnancy, resulting in characteristic hyperglycemia in late second or third trimester (1,2). Importantly, although this clinical presentation arises in response to the physiologic stress test posed by pregnancy, affected women have chronic β-cell dysfunction and insulin resistance that is readily apparent in the years thereafter and that contributes to their elevated lifetime risk of T2D (5,6). Moreover, it is now recognized that metabolic dysfunction actually long precedes the development of GDM, leading to the recent emergence of a host of markers that may enable the early identification of at-risk women in first trimester and even prior to pregnancy (7–11) (Fig. 1).

Against this backdrop, in this issue of Diabetes, Zhu et al. (12) report a detailed longitudinal characterization across pregnancy of circulating components of the insulin-like growth factor (IGF) axis in women with and without GDM. The IGF axis is a signal transduction complex consisting of 1) the growth factors IGF-I and IGF-II, 2) a series of IGF binding proteins (IGFBPs) that may regulate their bioavailable fraction, and 3) membrane receptors through which they act (13). Although traditionally linked to the regulation of cellular growth and differentiation, a growing body of evidence has recently implicated components of the IGF axis as potential factors in glucose homeostasis (14,15). For example, in a nested case-control analysis from the Nurses’ Health Study, baseline IGFBP-2 and IGFBP-1 levels were inversely associated with incident diabetes over median 9 years follow-up, whereas IGFBP-3 (which binds >90% of circulating IGF-I) showed a positive association (16). Total circulating IGF-I has yielded conflicting relationships with diabetic risk in a series of studies (16–19), possibly reflecting the necessity of focusing on its bioactive fraction (reflected by unbound free IGF-I or the total IGF-I/IGFBP-3 molar ratio). Taken together, this evolving literature is currently suggestive of a complex physiology between the IGF axis and glucose homeostasis that remains to be fully elucidated at this time.

In this context, the impact of the IGF axis in GDM is of interest, given the pathophysiologic and clinical relationship between GDM and T2D. Zhu et al. (12) thus report a timely case-control study of total IGF-I, IGFBP-2, IGFBP-3, and IGF-I/IGFBP-3 ratio measured at multiple time points in pregnancy in 107 women who developed GDM and 214 non-GDM control subjects. They showed that plasma IGF-I, IGFBP-3, and IGF-I/IGFBP-3 ratio increased across pregnancy, whereas circulating levels of IGFBP-2 decreased. Of note, IGF-I and IGFBP-3 concentrations at 10–14 weeks' gestation were positively associated with the risk of developing GDM later in the pregnancy. An even more robust association was shown for IGFBP-2, higher levels of which at 10–14 weeks predicted a significantly lower risk of subsequent GDM. Collectively, these findings suggest that components of the IGF axis, particularly IGFBP-2, may be relevant to the development of GDM.

An invaluable feature of this study is the serial measurement of multiple elements of the IGF axis across all three trimesters, providing insight into their temporal patterns over time before and after the development of GDM. In addition, the clinical characterization of the study population has allowed for matching and/or adjustment of relevant covariates, including diabetes risk factors. However, these data also raise important new questions.

A critical unresolved question is whether or not the IGF axis contributes to the pathophysiology of GDM. As women with GDM have chronic insulin resistance, they exhibit many associated features (e.g., dyslipidemia, adipokine dysregulation, inflammation) that differ from their peers but that are not necessarily related to the β-cell dysfunction that underlies the development of GDM (2,5). Indeed, although IGF-I, IGFBP-2, and IGF-I/IGFBP-3 at 10–14 weeks' gestation correlated with HOMA of insulin resistance at 15–26 weeks, their relationships with β-cell function are not known. In this regard, it is notable that the only correlation between these IGF axis components and glucose values on the oral glucose tolerance test on which GDM was diagnosed was the inverse association between IGFBP-2 and fasting glucose. From a pathophysiologic perspective, this is mildly disconcerting insofar as postchallenge glycemia is characteristic of GDM (i.e., hence the need for the oral glucose tolerance test for its diagnosis). It is currently unclear whether decreased IGFBP-2 is contributing to the pathophysiology of GDM or whether it is another reflection of the chronic insulin resistance of this patient population.

Irrespective of its potential role in the etiology of GDM, IGFBP-2 could still serve as a predictor for the early identification of at-risk women. Notably, IGFBP-2 at 10–14 weeks' gestation yields a modest incremental improvement in the prediction of GDM beyond conventional factors (12). Given current focus on pregravid predictors of GDM that might enable risk modification prior to conception, future studies should evaluate the predictive capacity of IGFBP-2 measurement prior to pregnancy, ideally with comprehensive adjustment for the factors shown in Fig. 1. The implications of the IGF axis for postpartum progression to T2D following GDM require similar investigation (20). Thus, though its etiologic and clinical role remains to be fully established, it is currently clear that the IGF axis (and particularly IGFBP-2) warrants further study as an emerging player in the natural history of both GDM and T2D.