BETHESDA, Md. -- The success of treatment with the antidepressant citalopram (Celexa) depends partly on a common genetic variation, researchers here said.
BETHESDA, Md., Aug. 1 -- The success of treatment with the antidepressant citalopram (Celexa) depends partly on a common genetic variation, researchers here said.
The finding offers hope that researchers are on track to tease out the genetic reasons why some people respond to medication while others do not, according to Francis McMahon, M.D., of the National Institute of Mental Health, and colleagues.
The effect of the new variation -- found in the GRIK4 gene, which codes for the kainic acid-type glutamate receptor KA1 -- is modest, the researchers reported in the Aug. issue of the American Journal of Psychiatry.
Patients carrying two copies of the protective allele have the risk of non-response to citalopram reduced by only 11%, the researchers said, adding the small effect size "probably precludes any immediate clinical relevance."
However, they calculated that carrying two copies of the protective GRIK4 allele and two copies of the previously discovered protective variant -- in the serotonin receptor gene HTR2A -- reduces the risk of non-response by 23%.
"We know that a number of biological mechanisms underlie depression and affect treatment," Dr. McMahon said in a statement. "Findings like this one are building a picture of what they are and how they interact, and can reveal potential molecular targets for faster-acting and more effective medications."
This new association of treatment response with a single-letter genetic variation -- a single nucleotide polymorphism (SNP) -- is the second to come out of the Sequenced Treatment Alternatives to Relieve Depression (STAR*D) trial.
To find the GRIK4 genetic variation, the researchers studied 768 SNPs in 68 genes thought to be potentially involved in depression. With DNA from 1,816 patients, they looked for genetic differences between patients who responded to the medication and those who didn't.
The patients were divided into a "discovery" group of 1,199 and a "replication" group of 617.
The same process was earlier used to find the HTR2A variant but with a smaller cohort.
The researchers' analysis found that the only SNPs significantly associated with response to the drug were rs7997012 in the HTR2A gene and rs1954787 in the GRIK4 gene, at P=0.000013 and P=0.000232, respectively.
The variants were also significantly associated with remission, at P=0.000095 and P=0.00034, respectively, the researchers said.
The authors also reported "exploratory analyses that suggest other genes that may be involved in treatment outcome, but these findings do not hold up to stringent correction for multiple testing, and thus should be considered preliminary. Association was observed with the rs6416623 marker in GRIN2A, a gene coding for an N-methyl-D-aspartate-type glutamate receptor. Association was also detected between treatment response and markers in GRIN2A and GRIK1, but only in female patients."
While the immediate clinical utility of the findings is limited, the study "hints at the eventual benefit promised by a more and more specific understanding of the function of the genes identified in the human genome project," according to Mary Anne Badaracco, M.D., of Harvard Medical School and Beth Israel Deaconess Medical Center.
In an accompanying editorial, Dr. Badaracco noted that currently patients face the "unavoidable curse of a trial-and-error approach" because their doctors don't know which medications will work.
And, since the drugs often take weeks to have an effect, patients can progress significantly during the wait-and-see period, she said.
Personalized medicine, based on a patient's genetic make-up, "could combat discouragement, increase the likelihood of the patient remaining in treatment, and lead to real remission," she said.
Among the authors, John Rush, M.D., reported financial links with Advanced Neuromodulation Systems, Best Practice Project Management, Bristol-Myers Squibb, Cyberonics, Eli Lilly, Forest Pharmaceuticals, Gerson Lehman Group, GlaxoSmith-Kline, Healthcare Technology Systems, Jazz Pharmaceuticals, Merck, Neuronetics, Ono Pharmaceutical, Organon USA, Personality Disorder Research Corp., Pfizer, Wyeth-Ayerst, Guilford Publications, and Healthcare Technology Systems. Stephen Wisniewski, Ph.D., reported financial links with Cyberonics, ImaRx Therapeutics, and Bristol-Myers Squibb. The remaining authors reported no conflicts.