Journal of Psychosocial Nursing and Mental Health Services


Future Prospects for Pharmacogenetics in the Quest for Personalized Medicine

Robert H. Howland, MD


Understanding pharmacogenetic differences in drug response and tolerability has been an important area of research in personalized medicine, but the clinical utility of pharmacogenetics testing has not been established. Identification of genetic polymorphisms due to single nucleotide polymorphisms is the most common approach, but this does not take into account the potential relevance of copy number variants, noncoding RNA gene regulation, gene-gene and gene-interactions, and epigenetic modifications, which increase the complexity of pharmacogenomics research.


Understanding pharmacogenetic differences in drug response and tolerability has been an important area of research in personalized medicine, but the clinical utility of pharmacogenetics testing has not been established. Identification of genetic polymorphisms due to single nucleotide polymorphisms is the most common approach, but this does not take into account the potential relevance of copy number variants, noncoding RNA gene regulation, gene-gene and gene-interactions, and epigenetic modifications, which increase the complexity of pharmacogenomics research.

Pharmacogenomics information is central to the concept of personalized medicine. With a few exceptions, most research has not established the clinical utility of pharmacogenetics testing. Why the usefulness of current testing is limited may be due to epistasis, epigenetics, and other genetic factors that are potentially important but have been relatively unexplored in pharmacogenetics research. In this month’s article, I will describe some of these genetic factors.

Human Genome Basics

The human genome consists of 23 pairs of chromosomes, each containing a single DNA molecule (Court, 2007). Structurally, the DNA molecule is a “double helix” consisting of two bound strands of polynucleotides, much like a spiraling staircase in which the right and left railings are connected by each step. The building blocks of DNA are the four nucleotide bases adenine, guanine, thymine, and cytosine. Each polynucleotide strand is a sequential string of these bases, and the two strands of the DNA molecule are bound to each other base-pair by base-pair. The arrangement of these DNA base-pairs along the chromosome is called the DNA sequence. The basic functional unit of the genome is the gene, which is a shorter segment of DNA within a chromosome. Most genes encode for protein products. The DNA sequence of a gene serves as a template for transcription of messenger RNA (mRNA), and mRNA serves as a template for ribosomes to translate the DNA sequence into a protein product. The entire genome has approximately 3 billion DNA base-pairs, but only 20,000 to 30,000 protein-encoding genes.

The two DNA sequences of each chromosome pair within a person are highly similar, but not identical. Hence, every person has two copies of each gene that are not necessarily identical. The DNA sequences of the same genes in different individuals are also highly similar but not identical. The majority of observed DNA sequence variations (genetic polymorphisms) are due to single nucleotide polymorphisms (SNPs), which are single base-pair substitution mutations. The location of SNPs in relation to a particular gene determines whether the normal function of the gene is affected. Because each individual has two gene copies, SNPs can occur on one copy (heterozygous) or both (homozygous). The detection of SNPs that are associated with altered gene function that can significantly influence drug effects (e.g., metabolism, clinical response, side effects) is the classic basis for pharmacogenetics investigations and the commercial development of pharmacogenetics tests. The net effect of a SNP on gene function might depend on one or both gene copies being affected, so pharmacogenetic testing may identify homozygous or heterozygous genetic polymorphisms.

Copy Number Variation

Copy number variants (CNVs), a type of genomic variation that is distinct from SNPs, are rearrangements of large sections of DNA resulting in partial or entire gene deletions, insertions, or duplications. CNVs are being actively investigated for a possible role in causing disease, including psychiatric disorders (Joober & Boksa, 2009). As a source of genetic variation, CNVs are much more important than SNPs that are used in genome-wide association studies. Two unrelated people can differ from each other in CNVs in thousands of chromosomal locations. Differences in CNV genomic content have been seen in monozygotic (identical) twins and in different tissues within an individual person.

CNVs have not been extensively investigated in pharmacogenomics studies, but one example is with the hepatic drug metabolizing enzyme (DME) cytochrome P450-2D6 (CYP2D6) (Ouahchi, Lindeman, & Lee, 2006). Although most studies have focused on identifying CYP2D6 gene SNPs, one study found that 5% of participants had a CYP2D6 gene deletion CNV (resulting in poor metabolism) and 7% had a gene duplication CNV (resulting in ultra-rapid metabolism). Other undiscovered CNVs could be relevant to drug treatment response, and further studies are needed.

Junk DNA

When the human genome was first sequenced, the number of genes encoding for proteins (approximately 20,000 to 30,000) was significantly fewer than the 100,000 genes that were expected, suggesting that much of the DNA in our genome was useless “junk.” The Encyclopedia of DNA Elements (ENCODE) project was initiated to study the genome in greater detail (Ecker et al., 2012). A significant finding from ENCODE is that more than 80% of the genome has a functional purpose. In addition to the presence of “traditional” genes (that encode for protein products), much of the remaining DNA sequences seem to have regulatory effects on these genes. Ordinarily, DNA codes for mRNA that is transcribed and then translated into a protein product, but ENCODE found approximately 18,000 RNA molecules (referred to as noncoding RNA) that were not translated into protein products. Noncoding RNA can regulate gene expression and other aspects of cellular function. The role of noncoding RNA and other functional aspects of “junk” DNA sequences are likely to be extremely important in pharmacogenomics research (Tang, Hu, Muallem, & Gulley, 2011).


A candidate gene approach to genetic association studies (looking for genetic correlates of disease) or pharmacogenetic studies (looking for genetic correlates of drug treatment response) is to select a candidate gene that is based on known biological or functional relevance. For example, the serotonin (5HT) transporter (5HTT) regulates the reuptake of 5HT into neurons and is the principal site of action of selective serotonin reuptake inhibitor (SSRI) antidepressant drugs. Multiple genetic polymorphisms of the 5HTT gene have been identified. The 5HTT is one of the most extensively studied candidate genes in genetic studies of depression and other psychiatric disorders and in pharmacogenetics research, but studies have not consistently found or confirmed an association. Because of the potential relevance of many other genes, however, gene-gene interactions (epistasis) are likely to play an even more important role in understanding the genetic basis of disease as well as drug treatment effects (Motsinger, Ritchie, & Reif, 2007). To assess multiple gene-gene interactions in pharmacogenomic studies necessarily involves a higher level of complexity, and few studies have been conducted to date (Lane, Tsai, & Lin, 2012).

The term epistasis has also sometimes been used to refer to gene-environment interactions. One example of the importance of gene-environment interactions is the hepatic DME CYP2C19, which has at least 28 different genetic polymorphisms. A number of studies have demonstrated that environmental effects (rather than inherited genetic effects), such as pregnancy, old age, cancer, and congestive heart failure, can lead to acquired alterations in CYP2C19 activity (Helsby & Burns, 2012). A poor metabolizer status is seen in these patients, even though they have apparently functional CY-P2C19 genes. Another example of gene-environment interactions are studies investigating the relationship among stressful life events, 5HTT genetic polymorphisms, and response to SSRI antidepressant drugs (Keers & Uher, 2012). One particular polymorphism is characterized as a “short” or “long” form of the 5HTT gene. The short form is associated with decreased expression (decreased amount) of the 5HTT protein. Two studies reviewed by Keers and Uher (2012) found that depressed patients carrying the short 5HTT polymorphism have a poor response to fluoxetine (Prozac®) or escitalopram (Lexapro®) but only in those patients with a history of stressful life events. A third study in Keers and Uher’s review, however, did not replicate these findings.


Epigenetics refers to the regulation of DNA sequences that does not involve alteration of their actual base composition (Narayan & Dragunow, 2010). Epigenetic modifications are potentially reversible. The predominant epigenetic mechanisms are DNA methylation and histone acetylation. The addition of a methyl chemical group to a DNA sequence typically operates to suppress the expression of the associated gene. Histones are proteins that package DNA strands into nucleosomes (like thread wound around a spool). The addition of an acetyl chemical group to histones typically operates to “unwind” the DNA and allow gene transcription. The enzyme histone deacetylase (HDAC) removes acetyl groups and suppresses gene transcription. Some antidepressant drugs, antipsychotic drugs, and the anticonvulsant drug valproate (valproic acid or divalproex sodium [Depacon®, Depakene®, Stavzor®, Depakote ER®]), affect epigenetic regulatory mechanisms (Boks et al., 2012). HDAC inhibitor drugs have antidepressant-like and antipsychotic-like effects in animal models (Grayson, Kundakovic, & Sharma, 2010). Some animal and human studies have found that chronic treatment with antipsychotic drugs can induce epigenetic changes (Kurita et al., 2012). In animals, HDAC inhibitor drugs can prevent these epigenetic changes and can favorably alter the behavioral effects of the antipsychotic drugs. Valproate is a nonspecific HDAC inhibitor that has been used clinically to augment the treatment effects of antipsychotic drugs in schizophrenia (Kurita et al., 2012). Epigenetic changes can be induced by environmental effects, such as stress (McGowan & Szyf, 2010). Hence, gene-environment interactions may be partly explained by epigenetic mechanisms.

The potential role of epigenetic modifications in pharmacogenomics, including adverse drug effects (Csoka & Szyf, 2009), is immense, but it is even more complex than assessing gene-gene interactions. Epigenetic modifications are variable and depend on cell type and hormonal and environmental conditions. Within a single person, different tissues (including different brain regions and even individual neurons) could have distinct patterns of epigenetic modifications. Because such epigenetic markers would be distinctly present in the tissue site, blood or saliva samples containing DNA (which are typically used for genetic tests of SNPs) would not be likely to carry the same epigenetic modifications. Being able to effectively identify and locate epigenetic modifications of interest and to use this information in pharmacogenomics research might depend on identifying suitable biomarkers linked to epigenetic modifications that could perhaps be visualized using brain imaging.


The study of epistasis, epigenetics, CNVs, regulatory DNA, and other complex areas of genomics research is growing rapidly, and these are likely to be extremely relevant to pharmacogenomic studies of drug treatments. Nurses should be familiar with these topics and how they demonstrate the limitations of current pharmacogenetic testing methods.


  • Boks, M.P., de Jong, N.M., Kas, M.J., Vinkers, C.H., Fernandes, C., Kahn, R.S. & Ophoff, RA. (2012). Current status and future prospects for epigenetic psychopharmacology. Epigenetics, 7, 20–28. doi:10.4161/epi.7.1.18688 [CrossRef]
  • Court, M.H. (2007). A pharmacogenomics primer. Journal of Clinical Pharmacology, 47, 1087–1103. doi:10.1177/0091270007303768 [CrossRef]
  • Csoka, A.B. & Szyf, M. (2009). Epigenetic side-effects of common pharmaceuticals: A potential new field in medicine and pharmacology. Medical Hypotheses, 73, 770–780. doi:10.1016/j.mehy.2008.10.039 [CrossRef]
  • Ecker, J.R., Bickmore, W.A., Barroso, I., Pritchard, J.K., Gilad, Y. & Segal, E. (2012). Genomics: ENCODE explained. Nature, 489(7414), 52–55. doi:10.1038/489052a [CrossRef]
  • Grayson, D.R., Kundakovic, M. & Sharma, R.P. (2010). Is there a future for histone deacetylase inhibitors in the pharmacotherapy of psychiatric disorders?Molecular Pharmacology, 77, 126–135. doi:10.1124/mol.109.061333 [CrossRef]
  • Helsby, N.A. & Burns, K.E. (2012). Molecular mechanisms of genetic variation and transcriptional regulation of CYP2C19 (Article 206). Frontiers in Genetics, 3. doi:10.3389/fgene.2012.00206 [CrossRef]
  • Joober, R. & Boksa, P. (2009). A new wave in the genetics of psychiatric disorders: The copy number variant tsunami. Journal of Psychiatry & Neurosciences, 34, 55–59.
  • Keers, R. & Uher, R. (2012). Gene-environment interaction in major depression and antidepressant treatment response. Current Psychiatry Reports, 14, 129–137. doi:10.1007/s11920-011-0251-x [CrossRef]
  • Kurita, M., Holloway, T., García-Bea, A., Kozlenkov, A., Friedman, A.K., Moreno, J.L. & González-Maseo, J. (2012). HDAC2 regulates atypical antipsychotic responses through the modulation of mGlu2 promoter activity. Nature Neuroscience, 15, 1245–1254. doi:10.1038/nn.3181 [CrossRef]
  • Lane, H.Y., Tsai, G.E. & Lin, E. (2012). Assessing gene-gene interactions in pharmacogenomics. Molecular Diagnosis and Therapy, 16, 15–27. doi:10.1007/BF03256426 [CrossRef]
  • McGowan, P.O. & Szyf, M. (2010). The epigenetics of social adversity in early life: Implications for mental health outcomes. Neurobiology of Disease, 39, 66–72. doi:10.1016/j.nbd.2009.12.026 [CrossRef]
  • Motsinger, A.A., Ritchie, M.D. & Reif, D.M. (2007). Novel methods for detecting epistasis in pharmacogenomics studies. Pharmacogenomics, 8, 1229–1241. doi:10.2217/14622416.8.9.1229 [CrossRef]
  • Narayan, P. & Dragunow, M. (2010). Pharmacology of epigenetics in brain disorders. British Journal of Pharmacology, 159, 285–303. doi:10.1111/j.1476-5381.2009.00526.x [CrossRef]
  • Ouahchi, K., Lindeman, N. & Lee, C. (2006). Copy number variants and pharmacogenomics. Future Medicine, 7, 25–29. doi:
  • Tang, W., Hu, Z., Muallem, H. & Gulley, M.L. (2011). Clinical implementation of RNA signatures for pharmacogenomics decision-making. Pharmacogenomics and Personalized Medicine, 4, 95–107. doi:

Dr. Howland is Associate Professor of Psychiatry, University of Pittsburgh School of Medicine, Western Psychiatric Institute and Clinic, Pittsburgh, Pennsylvania.

The author has disclosed no potential conflicts of interest, financial or otherwise.

Address correspondence to Robert H. Howland, MD, Associate Professor of Psychiatry, University of Pittsburgh School of Medicine, Western Psychiatric Institute and Clinic, 3811 O’Hara Street, Pittsburgh, PA 15213; e-mail:

Posted Online: December 01, 2012


Sign up to receive

Journal E-contents