Special Lecture

  1. Special Lecture
  2. Clinical Research Educational Seminar
  3. Symposiums
  4. Lunceon Seminar
  5. Oral Abstracts
  6. Poster Abstracts

SL1

History of pharmacogenetics in relation to clinical psychopharmacology

Leif Bertilsson

Dept of Clinical Pharmacology, Karolinska Institute
Clinical Pharmacology at Karolinska Institutet, Karolinska University Hospital, Huddinge, Stockholm, Sweden
The polymorphisms of debrisoquine and sparteine metabolism were discovered in the 1970ies and found to have the same molecular origin in mutations in the CYP2D6 gene. In Caucasians the most common mutated allele CYP2D6*4 gives rise to the poor metabolisers (PMs). This allele is almost absent in Asians, who instead have CYP2D6*10 with a SNP expressing an enzyme with decreased activity (see review 1). In Africans we also discovered a population specific mutated allele, CYP2D6*17, with decreased activity of the expressed enzyme. We showed (2) that the tricyclic antidepressant nortriptyline is metabolized by CYP2D6, and later that most antidepressants and neuroleptics are metabolized by this enzyme (cf 1). We reported a patient, who was an ultrarapid metabolizer of nortriptyline and had to be treated with doses as high as 600 mg per day to achieve normal plasma concentrations. Later we could show that she had a CYP2D6 gene duplication as the molecular basis for this ultrarapid metabolism (3). The frequency of individuals having a CYP2D6 gene duplication/multiplication is only about 1 % in Sweden, but as high as 29 % among black Ethiopians. We have in one Swedish family found that a father, a daughter and a son have 13 copies of the CYP2D6 with super rapid metabolism of many psychotropic drugs (4). We were able to show that among Swedish patients who did not respond to antidepressants being CYP2D6 substrates, 10 % had a CYP2D6 gene duplication (5). This clearly indicates that ultrarapid metabolism is an important factor for nonresponse to antidepressants.
The different polymorphisms in the CYP2D6, cause a pronounced variation in the enzyme activity between individuals and populations and the dosage of substrate drugs needs to be individualized. Early we showed that personality seems to be different between CYP"D6 phenotypes (6). The late professor Werner Kalow, the father of pharmacogenetics, ascribes his eventful life to his CYP2D6 PM phenotype (7). Recently Zackrisson et al (8) showed that the CYP2D6 gene duplication is related to suicidal behavior. In a commentary to this study on suicide I discuss several possible connections between CYP2D6 and personality/suicide. One is the presence of CYP2D6 in the human brain and production of serotonin from 5-methoxytryptamine (9).
CYP2C19 is also highly polymorphic and 3 and 20 % of Caucasians and Asians are PM of probe drugs such as omeprazole(1). Tertiary amine psychotropic drugs such as amitriptyline and citalopram are N-demethylated by CYP2C19.
Clomipramine clearance is significantly lower in Japanese compared to Swedish patients due to the higher frequency of CYP2C19*2 and *3 in Asians (10).

1. Bertilsson L. Clin Pharmacol Ther 2007;82:606-9.
2. Bertilsson L et al. Life Sci 1980;27:1673-7.
3. Bertilsson L et al. Lancet 1993;341:63.
4. Johansson I et al. PNAS 1993;90:11825-9.
5. Kawanishi C et al. Europ J Clin Pharmacol 2004;59:803-7.
6. Bertilsson L et al. Lancet 1989;1:555.
7. Kalow W. Pharmacol Toxicol 1995;76:221-7.
8. Zackrisson AL et al. Clin Pharmacol Ther 2010, in press.
9. Bertilsson L et al. Clin Pharmacol Ther 2010, in press.
10. Shimoda K et al. J Clin Psychopharmacol 1999;19:393-400.

Page Top

SL-2

Genome wide association data and Neuropsychopharmacology

George R Uhl

Molecular Neurobiology Branch, Intramural Research Program,National Institute on Drug Abuse, USA
Studies of the complex genetics of neuropsychiatric disorders provide a major source of new information about the brain, disease and drugs. Current genome wide association (GWA) studies use 1 million SNP and CNV probe sets to identify markers that differ in allelic frequency between disease and control samples. To reduce type I error, conventional analyses of such data seeks oligogenic influences and require "genome wide significance" in each of multiple independent samples. Such approaches can identify "major gene" or "oligogenic" influences on disease vulnerability and drug actions. Most of these findings have been in "primary" ADME pharmacogenomics and "secondary" pharmacogenomics that describes individual differences in drug "receptors". However, only a small fraction of the genetic influences on brain disorders has been discovered by these searches for large "oligogenic" effects of single variants in single genes, supporting large type II error for these approaches and suggesting that most of the genetic contributions to neuropsychiatric disorders come from a) "polygenic" variants that make individually-small contributions and b) rare variants that are likely to be more prominent in disorders associated with reduced fecundity.
We and others have developed strategies for collecting and analyzing GWA data that have reproducibly identified genes that likely to contain variants that provide "polygenic" influences on disease with substantial allelic heterogeneity. By focusing on identification of genomic areas in which several independent samples provide nominally-significant association from clusters of nearby SNPs, we have identified associations that have been replicated in many independent samples.
Elucidating genetic influences on vulnerability to addiction and on ability to quit smoking provides a good example, along with the corresponding classical genetic studies. These data now provide evidence for:
a) strong genetic and environmental influences on these phenotypes,
b) genetic influences on addiction that are specific to individual substances vs those that appear to influence vulnerability
to dependence on many different addictive substances,
c) a few "oligogenic" effects of individual "primary" and "secondary" pharmacogenomic influences on alcohol and
smoking related phenotypes that provide modest overall contributions to addiction vulnerability,
d) a number of (individually small) effects of variants at a number of (overlapping) gene loci on vulnerability to dependence and success in quitting that, collectively, provide large overall contributions to these phenotypes,
e) specific families of genes that are identified by these studies more than expected by chance,
f) population-specific variants that contribute to addiction vulnerability,
g) substantial allelic heterogeneity (as well as locus heterogeneity),
h) ability to predict ability to quit smoking using a complex genotype score with weighted contributions from 12000 SNPs, in replicated samples.
I will discuss implications of this sort of data and these genetic architectures for clinical neuropsychopharmacology in clinical trials and in practice.

Page Top

  1. Special Lecture
  2. Clinical Research Educational Seminar
  3. Symposiums
  4. Lunceon Seminar
  5. Oral Abstracts
  6. Poster Abstracts