Understanding how a person’s genetic makeup impacts their response to drugs, or pharmacogenomics, is just getting started. As genotyping gets less expensive, more information will be available on patients’ genes, and clinicians can use this to select the best treatments. Still, it’s not immediately obvious how to turn this know-how into mainstream healthcare. The obstacles extend beyond technology alone.
“Genotyping is the core technique that makes pharmacogenomics possible,” says Jorge Fonseca—senior product manager, genetic sciences, Thermo Fisher Scientific. “It enables characterization of genetic variants in individuals.” With genomics technology, a clinician “can analyze a person’s genome for variants that are known to influence the absorption, distribution, metabolism, and excretion [ADME] of prescription drugs in our bodies,” Fonseca explains. “Analyzing the combination of these genetic variants makes it possible to predict how a given person will respond to different medicines.”
As scientists better understand pharmacogenomics, it can be used in more ways. Beyond revealing how one patient will respond to a drug based on genes, pharmacogenomics can also “determine a genetic pattern that might predict a specific drug response in a defined patient cohort,” says Julia Schueler, research director at Charles River Laboratories.
The interaction of a patient and a drug can be seen in two ways: pharmacokinetics (PK), which is how the body processes the drug; and pharmacodynamics (PD), which is how the drug impacts the body. “While PD genes are very drug specific, the genes involved in PK are compound-agnostic, and studies in this area help to understand differences between individuals in different modes of actions and therapeutic areas,” Schueler explains. “Based on the data determined by genotyping in pharmacogenomics, it is possible to estimate adverse drug reactions and anticipate inter-individual differences.”
Multiple methods
Clinicians and research scientists can use various technologies for genotyping for pharmacogenomics. One method is the quantitative polymerase chain reaction (qPCR), which “is ideal for routine testing of samples up to about 120 genetic targets because it enables a low cost per sample and a fast time-to-results,” Fonseca says. “The Applied Biosystems Real-Time PCR Solutions for Pharmacogenomics, featuring TaqMan Assays and QuantStudio instruments, is a good example of this technology.” Thermo Fisher Scientific’s Digital Science offering includes cloud-based data analysis apps for genotyping, which Fonseca says, “enable users to maximize the value of their data, through features such as improved visuals and integrated traces of allelic discrimination plots that allow thorough quality control of SNP assays to accurately reflect the true signals versus background noise.
Microarrays, on the other hand, can test thousands of genetic targets. “This is the best choice when you want the maximum amount of data available, such as in translational research applications or preemptive screening of participants in pharmaceutical clinical trials,” Fonseca notes. The Applied Biosystems PharmacoScan Solution is a good example of this technology. Combining this with Thermo Fisher Scientific’s “Precision Medicine Diversity Array with the Axiom Plus workflow can characterize both SNP variants and copy-number variation with a single assay,” Fonseca says.
Thoughts on throughput
From the genotyping core lab at the University of Florida’s Center for Pharmacogenomics and Precision Medicine, director Taimour Langaee explains that labs can run genotyping at various throughput levels. “If you have automation with robots, you can do more,” he says.
Typically, he uses low-to-medium throughput methods, such as using TaqMan probes with QuantStudio real-time PCR. He also uses chips from Illumina that include 5 million SNPs.
Amongst all of the options for genotyping, from PCR to sequencing, scientists and clinicians must consider many factors. “A beneficial genotyping test must identify most or all of the mutations that have a significant impact on the expression or function of drug-metabolizing enzymes, transporter proteins, and/or drug receptors,” says Schueler. “Selection of the appropriate technology will be based on several factors, including prior knowledge of the mutation/polymorphism, sensitivity/specificity, sample requirements, and cost.”
Much more pharmacogenomics testing surely lies ahead. As an example, Langaee envisions a future, maybe a near one, in which people have their genomic data on a smartphone. “This way, people can be proactive, looking at potential drug interactions,” he says. “More than 100,000 people in the U.S. die every year from drug side effects, and having genotype information can help prevent many cases.”
Image: The sequence of bases in DNA—a person’s genetic makeup—determines the response to drugs, or pharmacogenomics, which can be used to prescribe more effective and safer medications. Image courtesy of National Human Genome Research Institute, National Institutes of Health.
Even with more pharmacogenomic testing ahead, Schueler sees it still relying on microarray-based assays and next generation sequencing (NGS) technologies, which cover the whole exome or even genome. In both cases, more automation will be required to do tomorrow’s levels of testing. Plus, she adds that NGS “is getting more and more attention in recent years as it enables researchers to see the full picture with respect to individual ADME-toxicity gene variants.”
A twig in the forest
Despite the thousands of human genes, just one can make a big difference. An example is CYP2D6, which Marin Jukic, a postdoc in the department of physiology and pharmacology at the Karolinksa Institute, explains, “is the enzyme that metabolizes many psychotropic drugs, and changes in CYP2D6 enzymatic capacity affect drug exposure.”
As an example of the importance of CYP2D6, Jukic points out that about 15% of people have two defective alleles for this gene, which make them slow CYP2D6 metabolizers. As a result, these people “express increased levels of drugs, such as risperidone and aripiprazole,” Jukic says. “Potentially, this puts them under risk of being overdosed and by knowing CYP2D6 genotype before treatment initiation, psychiatrists may prevent this from happening.”
With risperidone, for instance, a slow CYP2D6 metabolizer can be exposed to 50% more of the drug than someone with fully functional CYP2D6 alleles. The slow CYP2D6 metabolizers “are under higher risk of overdose if the dose is higher than 4 milligrams a day, and the drug label defines the maximum daily dose of 6 milligrams a day.”
So, this is a gene worth looking at in people who might be prescribed these kinds of drugs.
Obstacles to overcome
To make use of the increasing levels of pharmacogenomic data, scientists face a few challenges. One that Schueler points out is the “amount of data that must be stored, analyzed, and interpreted.” On top of that, she says, “Technical aspects like reproducibility of the platform, definition of candidate genes and drug response, as well as statistical method development must be addressed.”
The challenges, though, are not all technical. Some are ethical. “The availability of large genomic datasets of individuals in combination with clinical and personal data makes the ethical approval and patient informed consent process very challenging,” Schueler explains. “The public perception of genetic testing differs widely between positive feedback for diagnostics tests, and the willingness of the individual to give unrestricted access to genetic information for research.”
And one thing that cannot be ignored is cost. The more pharmacogenomic testing that will be done, the more it will cost the healthcare system and the drug development industry, as well.
Despite the collection of challenges to pharmacogenomic testing, it provides too many benefits to resist. As Schueler concludes: “In general, genotyping based on the recently available high-throughput platforms has the potential to induce the maturation of individualized healthcare by considering each person’s genomic profile alongside his or her clinical condition to personalize therapeutic interventions.”