Preemptive Pharmacogenomic Testing Can Be Successfully Implemented, Integrated Into Care

Pharmacogenomic (PGx) testing is at the epicenter of personalized medicine, with the promise that accessible genomic information will lead to more informed prescribing practices—enabling prescription of the proper drug at the correct dose, the first try. A special issue of the American Journal of Medical Genetics (AJMG) Part C (Seminars in Medical Genetics) was recently dedicated to implementation of genomic medicine. DTET reviewed two of the special issue’s published case reports detailing development of preemptive PGx programs, the implementation of testing and reporting of identified variants, and the acceptance of the program by clinicians. The lessons learned by the University of Chicago (UC; Illinois) and St. Jude Children’s Research Hospital (Memphis, Tenn.) can provide valuable insight to other laboratories and institutions contemplating establishing such programs. In both of these cases, PGx decision support was successfully implemented due to the availability of data reported from high-quality genotyping arrays. University of Chicago The 1,200 Patients Project at UC was considered successful because “patient interest was robust, physician adoption of information was high, and results were routinely utilized,” reports Peter H. O’Donnell, M.D., principal investigator of the project in AJMG. The 1,200 Patients Project offers free, broad, preemptive PGx testing to outpatients seen at UC. Enrollment is continuing at a pace of roughly 30 patients per month. Patients are eligible if, at enrollment, they are regularly using one to six prescription medications. Genotyping was performed using a MASS-ARRAY/matrix-assisted laser desorption/ionization time-of-flight mass spectrometry method (Sequenom) at Knight Diagnostic Laboratories (Oregon Health & Science University). Both a custom-designed and commercially available PGx panel (Sequenom) were used. “Establishment of a validated, custom-designed genotyping panel to generate accurate genotype calls in a CLIA environment was not trivial,” O’Donnell explained. “Validation required repeat testing of reference samples, analysis, and refinement of the panel over a span of approximately six months, and some genotype calls for validated assays nevertheless remain inconsistently reportable, although the overall frequency of missing calls is low.” O’Donnell cited CYP2D6 genotypes as particularly plagued with “technical hurdles” and they were not initially reported. PGx results are made available to enrolled physician-patient pairs through the Genomic Prescribing System (GPS) portal, which also provides real-time prescription guidance. Results were delivered in the form of “traffic light” signals, with green (favorable), yellow (caution), and red (high-risk). Each signal had a corresponding “clinical summary” that was viewable with a mouse click and provided a clinical translation of the PGx result, designed to be read in 30 seconds or less. Clinic providers were given daily reminders of which patients with scheduled appointments were enrolled and had been genotyped. In AJMG, the UC researchers reported 812 patients had participated (90 percent of those approached) along with six physicians (all remained enrolled). Of the 608 patients who had been successfully genotyped, there were 268 clinic encounters at which results were accessible via the GPS. At 86 percent of these encounters, physicians accessed the GPS, receiving 367 result signals for medications patients were currently taking (57 percent green, 41 percent yellow, and 1.4 percent red). Clinical summary click frequencies varied by alert severity, with 100 percent of red clicked, 72 percent of yellow, and 20 percent of green. Workflow analysis suggests, the authors say, that drug comparison information is being “under-utilized so far during new prescription decisions by our early-adopter physicians.” The authors say GPS is being checked ahead of the visit, rather than while in the room with the patient. For the vast majority of visits (85 percent), clinical PGx information was available for at least one drug the patient was taking, “suggesting relevance of the delivered information,” the authors write. “The lifetime impact on drug prescribing is potentially immense: ≈25 percent of [tested] patients have a genotype which would confer a red light signal for at least one of the drugs for which we report results, and over 50 percent of patients would have a level yellow light alert for at least one drug with results in our system.” The authors do acknowledge that the value of preemptive testing will ultimately be determined through the evaluation of outcomes measures. In the future, “other high-throughput genotyping options” will be evaluated in order to expand the service, control marginal costs, and cut turnaround times. St. Jude Children’s Research Hospital Researchers from St. Jude similarly report that preemptive clinical pharmacogenetics has proven “feasible, clinically useful, and scalable.” The institution’s translational research experience, which initially focused on implementation of pharmacogenetics as standard of care for its patients, with single-gene tests (TPMT and CYP2D6), was “substantially” expanded in 2011 through the introduction of the PG4KDS research protocol, which called for preemptively genotyping patients for multiple genes. “We elected to implement array-based clinical PGx in the context of a clinical trial,” write the authors, led by James Hoffman, Pharm.D., the medication outcomes and safety officer at St. Jude. “It is important to highlight that a clinical research approach is not essential for implementation of a hospital-wide pre-emptive PGx program. Other large scale implementations of pre-emptive PGx have been successful in the context of routine clinical care.” The Pharmacogenetic Oversight Committee provides oversight for the PG4KDS study and determines which gene test results should be placed in the EHR, what constitutes priority (high-risk) diplotypes, which drugs should be linked to genetic test results, and preferred methods of notification. In AJMG, the St. Jude researchers report that genotyping occurred for 230 genes, including 1,936 loci relevant to pharmacogenomics. Testing was performed by the Medical College of Wisconsin in a CLIA-certified laboratory using Affymetrix’s Drug Metabolizing Enzymes and Transporters Plus array, supplemented with a CYP2D6 copy number analysis using a quantitative polymerase chain reaction test. Test results for four genes (TPMT, CYP2D6, SLCO1B1, and CYP2C19) coupled to 12 high-risk drugs have been incorporated into electronic health records (EHRs) for clinical implementation with 55 clinical decision support rules. EHR results are tied to an interpretive consult and interruptive alerts at the time of prescription or dispensing. “Our prioritization of new gene/drug pair implementation has relied heavily on the availability of Clinical Pharmacogenetics Implementation Consortium (CPIC) guidelines,” write the authors. “Gene/drug pairs are prioritized for migration to the EHR based on a variety of criteria: inclusion in guidelines by CPIC or other professional organizations, [Food and Drug Administration] labeling recommendations, evidence of reimbursement for genetic testing for that drug’s use, the availability of a stand-alone CLIA-approved test for the individual gene, and the publication of clinical trials linking drug effects to functional pharmacogenetic loci.” Through August 2013, the researchers report in AJMG that 1,559 patients have been enrolled, with genotype test results available for 1,016 patients. More than three-quarters (78 percent) had at least one high-risk (i.e., actionable) genotype result placed in their EHR. Average turnaround time has ranged from 20 days to 137 days, which the authors say is OK given the research protocol and with testing done for preemptive purposes, although they estimate that it could be optimized to a “practical” time of 14 days to 21 days. “Key elements necessary for our successful implementation have included strong institutional support, a knowledgeable clinical laboratory, a process to manage any incidental findings, a strategy to educate clinicians and patients, a process to return results, and extensive use of informatics, especially clinical decision support,” the authors conclude. “We plan to implement at least eight new gene test results into the EHR over the next three years (e.g., DPYD, UGT1A1, G6PD), along with additional drugs, based partly on the output from CPIC over the next several years.” Takeaway: Preemptive PGx genotyping programs have been successfully implemented. A key element of success is the incorporation of PGx-related clinical decision support in the EHRs to ensure the results can inform clinical decisionmaking.

2014 CPIC PGx Dosing Guidelines
The Clinical Pharmacogenetics Implementation Consortium (CPIC) and the Pharmacogenomics Knowledge Base have combined genomic and clinical information to establish clinical practice guidelines enabling clinicians to understand how to use PGx test results to optimize drug therapy. The group says that a key assumption behind their guidelines is that clinical high-throughput and preemptive genotyping will become more widespread. The following dosing guidelines have been issued or updated by CPIC in 2014.
Drug	Variant	Recommendation
Abacavir	HLA-B	For HLA-B*57:01-+, abacavir is not recommended
PEG-interferon-alpha-containing regimens	IFNL3	Patients with the favorable response genotype (rs12979860 CC) have increased likelihood of response (higher sustained virologic response rate) to hepatitis C virus treatments with PEG-IFN alpha
Fluoropyrimidines	DPYD	Alternative drug for patients who are homozygous for DPYD nonfunctional variants; 50 percent reduction in starting dose for heterozygous patients (intermediate activity).
Codeine	CYP2D6	Alternate analgesics are recommended for CYP2D6 ultrarapid and poor metabolizers
Ivacaftor	CFTR	Treatment only recommended in cystic fibrosis patients who are either homozygous or heterozygous
Rasburicase	G6PD	Contraindicated in G6PD deficient patients
Simvastatin	SLCO1B1	In patients with the C allele at SLCO1B1 rs4149056, there are modest increases in myopathy risk even at half doses (40 mg)