Gene-Environment Interaction in Youth Depression: Differential Susceptibility?

To the Editor: I commend Dr. Rocha and colleagues for their study (1), published in the October 2015 issue of the Journal, which closely replicates the 2003 study by Caspi et al. (2) of the connection between serotonin-transporter-linked polymorphisms and depression in the context of childhood maltreatment. I wish to bring attention to one of the replicated findings that Rocha and colleagues note only in passing and whose implications they overlook. The authors’ principal focus is on reporting their replication of the finding in Caspi’s study that individuals with homozygous short alleles, compared with individuals with homozygous long alleles, are at an increased risk of depression in the presence of a history of childhood maltreatment. However, an equally noteworthy finding in Caspi’s study and in Rocha’s replication is that individuals with homozygous short alleles are also at the lowest risk of depression in the absence of childhood maltreatment—even lower than the risk among those with homozygous long alleles. Rocha and colleagues note this in passing on page 984 without emphasizing the heuristic importance of this replicated finding.

As Belsky et al. (3) noted in their comments on Caspi’s study, this interesting finding suggests that, in our focus on genes that confer “vulnerability” in the face of adverse environmental experience, we are at risk of overlooking the way genes may actually confer not “vulnerability” but “plasticity” in response to good and bad environmental experience.

Other investigations (4, 5) have made similar observations of this kind of “differential susceptibility” related to serotonin-transporter-linked polymorphisms in depression, bipolar disorder, and suicide.

A single-minded focus on a model of vulnerability to disease states may lead us to overlook the merits of a plasticity model; that is, of genes conferring greater responsiveness to both adverse and nonadverse environmental experience. The plasticity model may well prove to be a more accurate and useful heuristic model as we work to deepen our understanding of the interactions between genes and environment and how they relate to disease and to resilience.