NEW YORK — Simple genetic tests aimed at predicting the risk of drug addiction are still a long way off. But the genomics revolution is slowly changing the way physicians look at their patients and the disorders they treat.
Wade Berrettini, M.D., of the University of Pennsylvania, Philadelphia, said investigators have identified several single nucleotide polymorphisms (SNPs), small but meaningful allelic variants that result in changes to the shape or structure of a specific receptor or enzyme that relate directly to addiction problems.
Among these is a set of SNPs that influence the binding affinity of the mu β-endorphin receptors. Some of the SNPs in this set are proving to have some predictive value for alcohol and nicotine addiction, and for response to addiction treatment.
“The Human Genome Project is really only the beginning,” Dr. Berrettini said at annual conference of the Association for Research in Nervous and Mental Disease. “At this point, it is like knowing the alphabet, without knowing words.”
Dr. Berrettini's lab has been focused on SNPs in the gene that codes for the mu receptor protein, which plays a central role in generating the neurophysiologically rewarding and analgesic effects of morphine.
Endorphins and enkephalins also bind to the mu receptor. Given its place in mediating the brain's response to endogenous as well as exogenous opioids, the mu receptor also plays an important role in the process of addiction to opioids, as well as other substances, such as alcohol and nicotine.
To date, researchers have identified at least 25 SNP variants in the gene coding for the mu receptor. It is important to understand that none of these SNPs constitutes a “gene” for addiction, he said. However, some of them do seem to alter how the mu receptor functions. One such SNP, for example, tends to increase the receptor's binding affinity for β-endorphin.
Dr. Berrettini and his colleagues have been studying mu receptor SNPs in the context of heroin addiction. Though they have not yet identified any single variant that clearly shows an increased prevalence among addicts, compared with nonaddicts, they have found some interesting racial differences.
“In African Americans, we've found some alleles in 10% of the population that we simply have not found in people of European ancestry,” he said at the conference, cosponsored by the New York Academy of Medicine.
This underscores an important guiding principle for genomic research: When looking at the influence of small genetic variations on the risk of a given disease state, it is important to compare ill versus well people of the same racial and ethnic background.
Though Dr. Berrettini's team was unable to identify a specific mu receptor SNP that correlated with heroin addiction, Swedish researchers were able to do so. They found a variant called A118G that does seem to predict risk of heroin dependence. Approximately 18% of Swedish opioid addicts had disease that is caused in part by this SNP.
Some researchers have suggested that the mu receptor may play a role in alcohol dependence since ethanol triggers a release of β-endorphin, the key ligand for the mu receptor.
Available data correlating SNPs in the mu receptor gene and alcoholism are highly variable; there are as yet no studies showing a clear association. However, variants in this gene may predict an alcoholic's response to treatment with naltrexone.
Dr. Berrettini and his group have done a series of studies looking at multiple mu receptor SNPs in alcohol-dependent individuals treated with naltrexone. In the subgroup of patients who had either the A/G or G/G variants of the Asp40 allele, only 10% relapsed after nearly 3 months of posttreatment follow-up. Those with the A/A variant had very high relapse rates, and outcomes were no better for naltrexone than placebo.
What these data suggest is that response to this drug, which is a mu receptor blocker, may be largely determined by genetic variants in a specific receptor. Dr. Berrettini estimated that 25% of the alcoholic population is either homozygous or heterozygous for the G allele, and it predicts better response to naltrexone.
Those who are homozygous for the A allele do poorly on this drug. “We'd like to do a treatment study where we randomize based on mu allele genotype,” Dr. Berrettini said.
Though still in its early stages, this type of research is opening up the possibility of designing treatment protocols based on an individual's genetic predispositions and likelihood of responsiveness to specific medications. In other words, individualized therapy based on pharmacogenomics may soon become the standard of care.