The Wall Street Journal Discusses PGx

An article in the Wall Street Journal this week by Health Editor Melinda Beck called “Is Your Medicine Right for Your Metabolism?” discusses the current state of pharmacogenetics and obstacles to implementation of pharmacogenetics (PGx) in routine clinical care. The article and an accompanying video explain that variations in enzymes can drastically alter the metabolism of many commonly prescribed medications and may affect drug response, efficacy, and safety. The article goes on to discuss how some professional medical groups have hesitated to implement PGx testing because of insufficient evidence from large randomized controlled trials, but it also highlights medical research centers such as Vanderbilt University, the University of Pittsburgh, the Mayo Clinic, and St. Jude’s Children’s Research Hospital that are currently implementing PGx testing into care. Information about FDA approved drug labels mentioned in the article as well as detailed information about all of the genes and drugs discussed in the article can be found by searching PharmGKB. In addition, some of the cited gene-drug pair interactions such as CYP2D6 and codeine, clopidogrel and CYP2C19, warfarin and CYP2C9, and statins and SLCO1B1, are also the subjects of Clinical Pharmacogenetic Implementation Consortium (CPIC) dosing guidelines, which are available on the CPIC and PharmGKB websites.  For more information about research institutions currently implementing pharmacogenomics, see the lists on the PharmGKB and CPIC websites.

Integrating PGx throughout the Drug Development Process

Nelson, et al. argue for integrating more pharmacogenomic research into drug screening and the pre-approval process, with benefits for patient outcomes and drug development. Their article  “The genetics of drug efficacy: opportunities and challenges,” published this week in Nature Reviews Genetics reviews currently known associations of genetic variants with drug efficacy. In addition to using other sources, they selected genes from the list of CPIC guidelines. Clinical annotations describing the relationships and evidence for the variants that appear on their list of robust associations (Table 1 of the review), such as for the association of CYP2D6 and codeine efficacy, can be found on the PharmGKB website.

With nearly all pharmacogenetic relationships discovered post-approval, dozens of drugs currently on the market and in development are likely to benefit from clinical application of pharmacogenetics. Many relevant studies simply haven’t been done. Additionally, with genetic testing platforms becoming progressively cheaper, now at $1K for whole genome sequencing, it is possible to interrogate rare variation in addition to common variation, which until now has limited the discoveries of pharmacogenetic relationships to common variants with large effect sizes. 

Even while the majority of drugs may not have clinically useful genetic predictors of efficacy, Nelson, et al. have calculated that somewhere between 2 and 20% of drugs likely do, with important implications for clinical outcomes. We would like to add that beyond their statistic for drug efficacy, genetic variation can have important implications for drug safety, including risk of toxicity and hypersensitivity reactions. By integrating pharmacogenetic research earlier and more comprehensively into the drug development process, the benefit of personalized medicine related to efficacy AND safety can achieved more quickly and more completely, with additional benefits for understanding the biological mechanism of drug activity and targeted development of future drugs.

DTC genetic testing and the role of primary care providers

An original research manuscript and a corresponding editorial published recently in Annals of Internal Medicine analyze interactions of patients with their primary care providers (PCP) regarding their direct to consumer (DTC) genetic testing results. Van der Wouden et al. found that out of 1026 respondents, 277 discussed their results with their PCP.  Of those, 35% were very satisfied while 18% were not at all satisfied with the interaction (the remainder were “somewhat” satisfied). Those who were very satisfied were more likely to report that their provider was willing to discuss their test results (83.7% compared to 7.8% of not satisfied) and use the results in medical care (67.4% vs. 1%).   The study, however, does not report any health care provider perspective on these interactions. 

The editorial piece by Burke and Trinidad focuses on problems with DTC testing, comparing them with tests such as whole body scans, stating “They are tests that deliver uncertain information and create patient expectations that may align poorly with evidence; clinical priorities; or, in some cases, the patient's best interests.”  While they cite the numbers for the customers who were not at all satisfied with their PCP interaction in the van der Wouden study, there is no mention of those who were satisfied.  They conclude the lack of satisfaction suggests that the DTC companies create unrealistic expectations of the health care system. They even go so far as to say that DTC companies have been able to “[externalize] the dissatisfaction consumers feel when their experience falls short of marketing promises”, presumably deflecting customer dissatisfaction onto health care providers.  This statement is surprising given the van der Wouden study showed that the customers who discussed their results with their PCP expressed the most satisfaction with the test overall (as compared to those who consulted with another health care provider other than the PCP, or did not consult with a health care provider at all).  Not a single customer in this group reported being completely dissatisfied with their decision to obtain the DTC test.

Both the original article and the editorial raise questions around the customer expectations of DTC testing, and around the responsibilities and capabilities of healthcare providers in understanding and interpreting genetic test results. The conclusions differ markedly.  Regardless, there is a need to address the disconnect in patient/customer expectations and the responses of healthcare providers as genetic testing becomes more prevalent in precision medicine.

FDA-approved drug label for abacavir updated

In an update to the FDA-approved drug label for abacavir (ZIAGEN), the wording within the Boxed Warning section has changed. It now uses stronger language to support screening for HLA-B*57:01 prior to treatment with abacavir. The following is an excerpt from the boxed warning:

"All patients should be screened for the HLA-B*5701 allele prior to initiating therapy with ZIAGEN or reinitiation of therapy with ZIAGEN, unless patients have a previously documented HLA-B*5701 allele assessment"

The previous label used the following wording:

"Prior to initiating therapy with abacavir, screening for the HLA-B*5701 allele is recommended; this approach has been found to decrease the risk of hypersensitivity reaction".

The new stronger wording on the drug label has changed its PGx level ranking on PharmGKB from "Genetic testing recommended" to "Genetic testing required". The label now has the same PGx level ranking as the European Medicines Agency (EMA) label. PharmGKB blogged in 2013 about the vague language used within FDA-approved labels that often stops short of requiring action. With this recent change, the wording on this label now has firm and direct guidance regarding testing.

Patients who carry the HLA-B*57:01 allele are at a significantly increased risk of developing a hypersensitivity reaction to abacavir as compared to those without the allele, with odds ratios well into the quadruple digits in some studies. Symptoms include fever, rash, cough, gastrointestinal symptoms, dyspnea and fatigue. A 2008 study (PREDICT-1) found that screening for this allele eliminated immunologically-confirmed hypersensitivity reactions.

---

See the new drug label annotation on PharmGKB

Read more about drug labels on PharmGKB

Read the PREDICT-1 study:
Mallal S, Phillips E, Carosi G, Molina JM, Workman C, Tomazic J, Jagel-Guedes E, Rugina S, Kozyrev O, Cid JF, Hay P, Nolan D, Hughes S, Hughes A, Ryan S, Fitch N, Thorborn D, Benbow A; PREDICT-1 Study Team. HLA-B*5701 screening for hypersensitivity to abacavir. N Engl J Med. 2008 Feb 7;358(6):568-79. PMID 18256392.

The Precision Medicine Initiative: Consultative PGx and Data Sharing at Stanford University

On February 25th, one year after the announcement of the Precision Medicine Initiative (PMI), the White House hosted the Precision Medicine Summit. A White House press release described how various public and private groups will be taking steps to advance the goals of the PMI. As part of the PMI Stanford will launch a consultative pharmacogenetics practice for physicians to refer patients with unusual drug responses. Dr. Russ Altman, Co-Principal Investigator (Co-PI) of PharmGKB will lead the pharmacogenetics consultative practice. The consultative practice will evaluate genetic variants in patients with unusual drug responses and will use Clinical Pharmacogenetics Implementation Consortium (CPIC) dosing guidelines to focus attention on gene-drug pairs with the best evidence for actionability. In addition, Stanford University will also make the genomic data of 77 individuals of Iranian descent publicly available. The genomes of those 77 individuals were sequenced as part of the Iranian Genome Project, for which Dr. Altman is also the PI. Dr. Roxana Daneshjou, a former student in Dr. Altman’s lab is the lead researcher for the project.  Although Dr. Daneshjou is leading the data analysis for the project, Dr. Altman explained “we wanted to get the data out into the public as soon as possible, so (we) have taken the slightly unusual measure of making the data available before we get our paper out… we want people to use it.” A major goal of the PMI is to “open up data and technology tools to invite citizen participation, unleash new discoveries, and bring together diverse collaborators to share their unique skills.”

EPIC Integrates Genelex's Pharmacogenomics Decision Support System into Electronic Health Records

On Feb 23^rd, Pharmagenomics (PGx) testing company Genelex announced enhanced connection with  Epic, the most widely used electronic health record (EHR) system in US. With this new connectivity, healthcare systems and providers using Epic EHRs will be able to use Genelex's YouScript Precision Prescribing software to better identify patients who are at greater risk of adverse drug events based on pharmacogenetic information. In a GenomeWebnews article, Genelex CEO Kristine Ashcraft said “It's an indicator of market adoption and market demand for pharmacogenetic information”.

Translating pharmacogenomics knowledge into clinical practice has not been an easy task. Barriers that prevent the widespread use of PGx diagnostics include the lack of incentives for health care systems to conduct tests as well as lack of clear clinical guidelines for translating genetic variations into actionable recommendations. Additionally, computational tools for clinical decision support (CDS) will need to be developed and integrated in the EHR to prompt and guide clinicians on how to use genetic information when prescribing drugs. CPIC was formed in 2009 as a collaborative project between PharmGKB and the Pharmacogenomics ResearchNetwork to address some of the challenges in clinical implementation. Its goal was to create guidelines to aid clinicians on how genetic test results should be translated into specific prescribing actions. CPIC guidelines are simultaneously published and maintained on PharmGKB website, in both plain text as well as computable form for easy integration into clinical decision support tools. Many of the CPIC guidelines have been endorsed by professional societies such as The American Society of Health-System Pharmacists (ASHP). CPIC have also established informatics subgroup to support the adoption of the CPIC guidelines by developing tools to combine clinical information from the EHR with the information from the CPIC guidelines and use them for clinical decision support. Preemptive pharmacogenomic testing programs using EHRs and decision support tools have been deployed at several academic medical centers, eg. St. Jude,  Vanderbilt University and University of Florida.

Read the GenomeWeb article:

EpicIntegrating Genelex PGx Decision Support Into EHR Systems

Read the press release from Genelex:

http://genelex.com/genelex-announces-integration-with-epic/

More information about CPIC and CPIC informatics working group:

https://cpicpgx.org/

https://cpicpgx.org/informatics/

A Study of the Transcriptomic Variation Affecting Pharmacogenes

In an effort to capture differences in the expression of pharmacogenes between individuals and between tissue types and to characterize novel pharmacogene splice variants, researchers conducted a transcriptomic analysis of 389 pharmacologically important genes in liver, kidney (cortex), heart (left ventricle), and adipose tissue from 139 individuals as well as 45 lymphoblastoid cell lines (LCLs). Tissue samples came from various Pharmacogenomics Reseach Network (PGRN) groups. The list of pharmacogenes was compiled from several sources, including PharmGKB. The results were recently published in The Pharmacogenomics Journal.

The study, “Transcriptomic variation of pharmacogenes in multiple human tissues and lymphoblastoid cell lines”, reports that many pharmacogenes were highly expressed in liver or kidney when compared to other tissues, although some were consistently expressed at high or low levels in all tissue types, and that many pharmacogenes showed variable expressed when comparing between individuals and tissue types. The authors suspect that some differences in expression between individuals may reflect environmental and health differences between people. Expression of most pharmacogenes was almost always lower in LCLs when compared to the other tissue types that were examined. Finally, the authors report the discovery of several novel pharmacogene splicing events and splice variant differences between tissue types.

These findings suggest that variable expression of pharmacogenes between tissue types and individuals as well as differences in alternative splicing patterns could contribute to the observed variation in drug dosing, response, and toxicities.