Dr. Grossi's Blog

Not So Fast

Dr. Philip Grossi
Monday, 30 August 2010

In 2003 it appeared that Avshalom Caspi had made a massive discovery when he and his colleagues discovered a gene variant which was thought to play a major role in people developing depression in response to life's stresses. This polymorphism related to  serotonin transporter region (5HTTLPR) was hailed as an example of the interaction of a stressful environment with genetic predisposition. The gene comes in two versions: the long (l) allele and the short (s) alleles, and the 5HTTLPR genotypes were ss,sl,ll.  Animal studies in mice demonstrated that they became more fearful in response to loud noises when they had one or two of the s alleles and monkeys with the s allele had reduced serotonin transmission. 

Caspi and Moffitt studied 847 New Zealanders between the ages of 21 to 26.  They counted stressful life events such as romantic breakups, illnesses, bereavements, job problems etc. All of the 26  participants were asked whether they had been depressed in the prior year and 17% had been depressed and 3% reported being suicidal.  In those that had not had major stresses, the probability of depression was the same regardless of 5-HTT allele status.  For those who had stressful events and had an s allele, the probability of depression increased and for those with two s alleles the probability of a depression rose to 43% with four stressful events.  This was twice the rate for individuals with two l alleles and their score on the depression inventory was 50% of what the two s allele group manifested.  The researchers also looked at childhood abuse records and found that only those who had at least one s allele experienced depression by age 18.  In a group with severe maltreatment, the depressive risk was 63% in double s allele participants and the single l risk was 30% regardless of abuse history.  There is additional work showing that there is increased brain activity in people when exposed to fearful stimuli-- a topic for another time.

illustration to not so fast blogIn May 2009 Neil Risch and Kathleen Merikangas published an extensive meta-analysis in the Journal of the American Medical Association  of 14 studies including Caspi's  and found that the combined data did not support the connection between gene, stress, and depression. It is true that many studies were eliminated for methodological reasons - some of which were supportive.  I wonder whether selecting a candidate gene in advance and looking for supporting evidence isn't likely to lead to false positives. I have read a number of letters to the editor by various experts and there are a variety of criticisms and questions raised especially about various methods employed.  I should also note that various physiological, cellular, neural circuit and animal data are omitted from consideration.  Therefore, in my view the evidence is insufficient to draw a certain conclusion though I suspect that the Caspi's findings will eventually be shown to have merit when all type of studies are considered..