Clinical Policy Title: Pharmocogenetic testing for warfarin (Coumadin ) sensitivity

Clinical Policy Title: Pharmocogenetic testing for warfarin (Coumadin ) sensitivity Clinical Policy Number: 02.01.13 Effective Date: September 1, 2013 Initial Review Date: May 15, 2013 Most Recent Review Date: May 17, 2017 Next Review Date: May 2018 Policy contains: Pharmacogenetic testing. CYP2C9 gene. VKORC1 gene. Warfarin. Related policies: CP# 05.01.02 CP# 00.01.03 Prothrombin international normalized ratio-self-testing Genetic testing for cytochrome P450 polymorphisms (CYP2C19) ABOUT THIS POLICY: AmeriHealth Caritas Pennsylvania has developed clinical policies to assist with making coverage determinations. AmeriHealth Caritas Pennsylvania s clinical policies are based on guidelines from established industry sources, such as the Centers for Medicare & Medicaid Services (CMS), state regulatory agencies, the American Medical Association (AMA), medical specialty professional societies, and peer-reviewed professional literature. These clinical policies, along with other sources, such as plan benefits and state and federal laws and regulatory requirements, including any state- or plan-specific definition of medically necessary, and the specific facts of the particular situation are considered by AmeriHealth Caritas Pennsylvania when making coverage determinations. In the event of conflict between this clinical policy and plan benefits and/or state or federal laws and/or regulatory requirements, the plan benefits and/or state and federal laws and/or regulatory requirements shall control. AmeriHealth Caritas Pennsylvania s clinical policies are for informational purposes only and not intended as medical advice or to direct treatment. Physicians and other health care providers are solely responsible for the treatment decisions for their patients. AmeriHealth Caritas Pennsylvania s clinical policies are reflective of evidence-based medicine at the time of review. As medical science evolves, AmeriHealth Caritas Pennsylvania will update its clinical policies as necessary. AmeriHealth Caritas Pennsylvania s clinical policies are not guarantees of payment. Coverage policy AmeriHealth Caritas Pennsylvania considers the use of pharmacogenetic testing for warfarin (Coumadin ) sensitivity to be investigational and, therefore, not medically necessary. Limitations: There are no covered uses for pharmacogenetic testing for warfarin sensitivity. Alternative covered services: Laboratory testing for prothrombin time (PT) and International Normalized Ratio (INR) values. Background 1

Warfarin (trade names: Coumadin ; Jantoven ; Marevan ) is an anticoagulant prescribed to prevent or treat thromboembolic events in high-risk individuals (Hirsh, 2001). The drug has a narrow therapeutic window with variable responses to dosing. Individuals usually receive a starting dose of 2 mg to 5 mg and are frequently monitored until reaching a stable INR value of 2.0 to 3.0. Genetic variants present in the form of alleles have an influence on an individual s response to treatment with warfarin. Pharmacogenomics and pharmacogenetics are bodies of science that involve genetics, and at times the terms are used interchangeably. However, there are differences between the two: Pharmacogenomics pertains to the science of the many genes that determine drug behavior. Pharmocogenetics examines the link between an individual s genetic make-up and their response to a pharmaceutical product. The results of a pharmacogenetic test are used in an attempt to predict an individual s response to a specific pharmaceutical before it is used in a therapy regimen. Warfarin elimination from the body relies almost exclusively upon the metabolic conversion of inactive metabolites by cytochrome P450 (CYP) enzymes in liver (hepatic) cells. The principal cytochrome P450 enzyme that modulates the anticoagulant activity of warfarin is CYP2C9 (Flockhart, 2008). The VKORC1 gene encodes the vitamin K epoxide reductase enzyme, the drug target of warfarin. Different races present different variances with allele types, and different allele types have varying warfarin sensitivity. An individual may undergo testing of CYP2C9 or VKORC1 alleles in an attempt to predict the individual s response to warfarin prior to the initiation of drug therapy (Rieder, 2005). Though pharmacogenetic testing is used to gain a better approximation of the optimal initial dosing of warfarin, it does not remove the need for PT/INR testing as the standard diagnostic test for assessing an individual s reaction to warfarin dosing. The U.S. Food and Drug Administration (FDA) has approved several genetic tests for warfarin sensitivity. Searches AmeriHealth Caritas Pennsylvania searched PubMed and the databases of: UK National Health Services Centre for Reviews and Dissemination. Agency for Healthcare Research and Quality s National Guideline Clearinghouse and other evidence-based practice centers. The Centers for Medicare & Medicaid Services (CMS). We conducted searches on April 20, 2017. Search terms were: "Warfarin [MeSH], "Pharmacogenetics [MeSH], and free text terms pharmacogenetic and warfarin. We included: Systematic reviews, which pool results from multiple studies to achieve larger sample sizes and greater precision of effect estimation than in smaller primary studies. Systematic 2

reviews use predetermined transparent methods to minimize bias, effectively treating the review as a scientific endeavor, and are thus rated highest in evidence-grading hierarchies. Guidelines based on systematic reviews. Economic analyses, such as cost-effectiveness, and benefit or utility studies (but not simple cost studies), reporting both costs and outcomes sometimes referred to as efficiency studies which also rank near the top of evidence hierarchies. Findings There is insufficient evidence in the literature to warrant recommending routine warfarin sensitivity pharmacogenetic testing for initial dosing for individuals prior to receiving warfarin therapy. The predictive values for CYP2C9 genotype testing and VKOR SNPs testing are approximately 96 percent to 98 percent and 83 percent to 85 percent, respectively. The inconsistency in the predictive values of these tests has an impact on any determination for standard of care. Clinical variables such as age, race, smoking status, concomitant drugs, and weight account for a great degree of variability in warfarin response (Hirsh, 2001). Algorithms should be developed that incorporate genetic variation (genotype) to warfarin sensitivity and other significant factors to predict the best likely, stable starting warfarin dose to limit high INR values. More studies are needed that show warfarin pharmacogenetic testing actually reduces the risk and incidence of serious bleeding and mortality. More evidence and outcomes from large prospective clinical trials are needed that link genotype to warfarin dosing recommendations before endorsing warfarin sensitivity genotype testing (Flockhart, 2008; Holbrook, 2012; Ruaño, 2010). Policy updates: In 2014, we identified one new cost-effectiveness analysis and two new randomized controlled trials (RCTs) but no new systematic reviews or guidelines for this policy update. The two RCTs provided conflicting results (Kimmel, 2013; Pirmohamed, 2013). One large RCT found no improvement in anticoagulation control during the first four weeks of therapy using genotype-guided dosing of warfarin and no between-group differences in rates of the combined outcome of any INR 4, major bleeding or thromboembolism. However, they noted a significant interaction between dosing strategy and race; among African American patients, the mean percentage of time in the therapeutic range was lower in the genotype-guided group than in the clinically guided group (P = 0.003) (Kimmel, 2013; Clarification of Optimal Anticoagulation through Genetics [COAG] ClinicalTrials.gov number NCT00839657). Conversely, in another study, pharmacogenetic-based dosing was associated with both significantly less time to reach a therapeutic INR and higher percentage of time in the therapeutic INR range than standard dosing during the initiation of warfarin therapy (Pirmohamed, 2013). A cost-effectiveness analysis based on a clinical trial simulation found pharmacogenetic-guided warfarin therapy and apixaban (Eliquis ; Bristol Meyers Squibb) appear to be more cost effective than clinically guided warfarin treatment (Pink, 2014). 3

In 2015, we identified two new meta-analyses with conflicting results. While Tang (2014) found some benefit with pharmacogenetic dosing of warfarin and its analogues, Stergiopoulos (2014) found a pharmacogenetic dosing strategy did not result in increased time of the INR in the therapeutic range, fewer patients with an INR greater than 4, or a reduction in major bleeding or thromboembolic events compared with clinical dosing algorithms. In 2016, we identified two new meta-analyses for this policy (Tang, 2015; Liao, 2015). Patients who receive a loading dose of warfarin often reach a supratherapeutic INR level that can place a patient at risk for bleeding and prolonged hospital stay. While pharmacogenetic warfarin dosing offered no clear advantage over clinical dosing based on patient variables, it was superior to using a fixed initial standard loading dose. Meta-analyses note the variation in study populations, lengths of follow-up, genotypebased and single clinical algorithms, and outcomes that challenge the ability to draw firm conclusions from the evidence base. In light of these results, there remains insufficient evidence in the literature to warrant recommending routine warfarin sensitivity genotype testing for initial dosing for individuals prior to receiving warfarin therapy. In 2017, new evidence from one systematic review and meta-analysis suggests that, as with adults, CYP2C9 polymorphisms are associated with warfarin maintenance dose requirements in pediatric patients (Zhang, 2017). There is significant uncertainty in these findings because of the low frequencies of the CYP2C9 *2/*2, CYP2C9 *2/*3, and CYP2C9 *3/*3 genotypes in the study populations and insufficient ethnic diversity in study populations, number of studies, and sample sizes. Additional research that addresses these limitations is needed to build pharmacogenetic dosing algorithms for pediatric patients and assess associated patient outcomes. The COAG trial investigators examined the frequency and reasons for varying from algorithmic warfarin management in 1,015 trial enrollees who were randomized to either genotype-guided or clinically guided warfarin dosing algorithms (Kasner, 2016; COAG trial ClinicalTrials.gov number NCT00839657). Fifteen percent of participants in the genotype arm and 17 percent in the clinical arm required at least one exception to the protocol-defined warfarin dose, and 90 percent of dose exceptions occurred after the first five days of dosing. Multivariable analysis found only congestive heart failure, and not study arm or genotype, was associated with dose exceptions. While existing algorithms for predicting warfarin dosing can account for approximately half of the interindividual variability in percentage of time in therapeutic INR range (PTTR), the evidence is insufficient to support incorporating genotype to improve INR control or other patient outcomes. No policy changes are warranted. Summary of clinical evidence: Citation Zhang (2017) Content, Methods, Recommendations 4

Citation Cytochrome P450 2C9 gene polymorphism and warfarin maintenance dosage in pediatric patients Kasner (2016) for the COAG trial (ClinicalTrials.gov NCT00839657) Frequency and reasons for varying from algorithmic warfarin management Tang (2015) PG- versus standardversus clinically guided dosing Liao (2015) PG + clinical algorithm versus single clinical algorithm dosing Pink (2014) Content, Methods, Recommendations Systematic review and meta-analysis of eight studies of mixed designs (507 total patients ages <19 years). Overall quality: low. Limited by mostly retrospective designs and low numbers of pediatric patients studied and correspondingly low frequencies of these genotypes. Most studies included Caucasian males. Results from these studies cannot be applied to Asian or African populations. Maintenance warfarin doses in patients with CYP2C9 *1/*2, *1/*3, and variant genotypes were 15% to 41% lower than doses in patients with the wild-type allele (CYP2C9 *1/*1) (P-values <0.05). A history of Fontan procedure was associated with a lower warfarin maintenance dose (P <0.05), but target INR range did not influence dosage requirements. Results require confirmation from research with larger numbers of pediatric patients of different ethnicities to provide data to build pharmacogenetic dosing algorithms for pediatric patients that include genotype and clinical factors. Prospective RCT of 1,015 patients at 18 clinical centers in the U.S. randomized to receive pharmacogenetic (PG)-guided or clinically guided warfarin dosing algorithms. The genetic variants in the pharmacogenetic algorithms were CYP2C9 and VKORC1. Frequency of at least one dose exception during the first four weeks of therapy: 15% in pharmacogenetic arm, 17% in clinical arm, P =0.55). Ninety percent of dose exceptions occurred after the first five days of dosing. In multivariable analyses adjusted for age, sex, race, intervention, and clinical center, only congestive heart failure was associated with the odds of any dose exception (odds ratio 2.12; 95% confidence interval, 1.49 to 3.02, P <0.001). Neither study arm nor genotype was associated with dose exceptions. Meta-analysis of eight RCTs (number of subjects not reported). Primary outcome: percentage of time within the therapeutic INR range. Secondary outcomes: INR 4 events, major bleeding events and thromboembolic events. PG-guided dosing significantly increased the effectiveness and safety of Coumarin therapy compared with standard dosing but did not have advantages compared with clinically guided dosing. Meta-analysis of seven RCTs (1,910 total participants). Primary outcome: percentage of time within the therapeutic INR range (TTR). PG-guided group had improved outcomes compared with a fixed initial standard dose, but not with a non-fixed initial dose. No significant difference in rate of adverse events or death. Further experiments needed to confirm these findings. 5

Citation Cost-effectiveness analysis Stergiopoulos (2014) PG-guided initial dosing versus clinical dosing protocols Tang (2014) PG-guided versus clinical or standard initial dosing protocols Content, Methods, Recommendations Clinical trial simulation of S-warfarin used to predict TTR for different dosing algorithms based on data from meta-analyses. Neither dabigatran nor rivaroxaban were cost-effective options. In relation to clinically dosed warfarin, PG-guided warfarin and apixaban had incremental cost-effectiveness ratios of 13,226 and 20,671 per quality adjusted life year gained, respectively. Meta-analysis of nine RCTs (2,812 total patients). Follow-up ranged from four weeks to six months. A PG-guided dosing strategy did not result in a greater percentage of TTR, fewer patients with an INR >4, or a reduction in major bleeding or thromboembolic events compared with clinical dosing algorithms. Systematic review and meta-analysis of 10 RCTs and prospective cohort studies (5,299 total patients), including eight RCTs addressing initial warfarin dosing. PG-based warfarin dosing increased the rate of TTR and reduced the risk for bleeding complications in persons starting warfarin therapy. No difference in number of patients with an INR >4. References Professional society guidelines/other: Flockhart DA, O'Kane D, Williams MS, et al. Pharmacogenetic testing of CYP2C9 and VKORC1 alleles for warfarin. Genet Med. 2008; 10(2): 139 150. Guyatt GH, Akl EA, Crowther M, Gutterman DD, Schuunemann HJ. Executive summary: Antithrombotic Therapy and Prevention of Thrombosis, 9th ed: American College of Chest Physicians Evidence-Based Clinical Practice Guidelines. Chest. 2012; 141(2 Suppl): 7S 47S. Holbrook A, Schulman S, Witt DM, et al. Evidence-based management of anticoagulant therapy: Antithrombotic Therapy and Prevention of Thrombosis, 9th ed: American College of Chest Physicians Evidence-Based Clinical Practice Guidelines. Chest. 2012; 141(2 Suppl): e152s 184S. Keeling D, Baglin T, Tait C, et al. Guidelines on oral anticoagulation with warfarin - fourth edition. Br J Haematol. 2011; 154(3): 311 324. Musunuru K, Roden DM, Boineau R, et al. Cardiovascular pharmacogenomics: current status and future directions-report of a national heart, lung, and blood institute working group. J Am Heart Assoc. 2012; 1(2): e000554. Peer-reviewed references: 6

French B, Joo J, Geller NL, et al. Statistical design of personalized medicine interventions: the Clarification of Optimal Anticoagulation through Genetics (COAG) trial. Trials. 2010; 11: 108. Heneghan C, Ward A, Perera R et al. Self-monitoring of oral anticoagulation: systematic review and meta-analysis of individual patient data. Lancet. 2012; 379(813): 322 334. Hirsh J, Dalen J, Anderson DR, et al. Oral anticoagulants: mechanism of action, clinical effectiveness, and optimal therapeutic range. Chest. 2001; 119(1 Suppl): 8S 21S. Kasner SE, Wang L, French B, et al. Warfarin Dosing Algorithms and the Need for Human Intervention. Am J Med. 2016; 129(4): 431-437. Kimmel SE, French B, Kasner SE, et al. A pharmacogenetic versus a clinical algorithm for warfarin dosing. N Engl J Med. 2013; 369(24): 2283 2293. Liao Z, Feng S, Ling P, Zhang G. Meta-analysis of randomized controlled trials reveals an improved clinical outcome of using genotype plus clinical algorithm for warfarin dosing. J Thromb Thrombolysis. 2015; 39(2): 228 234. Limdi NA, Wadelius M, Cavallari L, et al. Warfarin pharmacogenetics: a single VKORC1 polymorphism is predictive of dose across 3 racial groups. Blood. 2010; 115(18): 3827 3834. Moreau C, Bajolle F, Siguret V, et al. Vitamin K antagonists in children with heart disease: height and VKORC1 genotype are the main determinants of the warfarin dose requirement. Blood. 2012; 119(3): 861 867. Pink J, Pirmohamed M, Lane S, Hughes DA. Cost-effectiveness of pharmacogenetics-guided warfarin therapy vs. alternative anticoagulation in atrial fibrillation. Clin Pharmacol Ther. 2014 Feb; 95(2): 199 207. Rieder MJ, Reiner AP, Gage BF et al. Effect of VKORC1 haplotypes on transcriptional regulation and warfarin dose. N Engl J Med. 2005; 352: 2285 2293. Stergiopoulos K, Brown DL. Genotype-guided vs clinical dosing of warfarin and its analogues: metaanalysis of randomized clinical trials. JAMA Intern Med. 2014; 174(8): 1330 1338. Tang Q, Zou H, Guo C, Liu Z. Outcomes of pharmacogenetics-guided dosing of warfarin: a systematic review and meta-analysis. Int J Cardiol. 2014; 175(3): 587 591. Tang T, Liu J, Zuo K, et al. Genotype-Guided Dosing of Coumarin Anticoagulants: A Meta-analysis of Randomized Controlled Trials. J Cardiovasc Pharmacol Ther. 2015; 20(4): 387 394. 7

The International Warfarin Pharmacogenetics Consortium. Estimation of the Warfarin Dose with Clinical and Pharmacogenetic Data. N Engl J Med 2009; 360: 753 64. Zhang J, Tian L, Huang J, et al. Cytochrome P450 2C9 gene polymorphism and warfarin maintenance dosage in pediatric patients: A systematic review and meta-analysis. Cardiovasc Ther. 2017; 35(1): 26-32. CMS National Coverage Determinations (NCDs): 90.1 Pharmacogenomic Testing for Warfarin Response. CMS website. http://www.cms.gov/medicarecoverage-database/details/ncddetails.aspx?ncdid=333&ncdver=1&ncaid=224&ver=13&ncaname=pharmacogenomic+testing+for+w arfarin+response&bc=beaaaaaacaaaaa%3d%3d&. Accessed April 21, 2017. Local Coverage Determinations (LCDs): No LCDs identified as of the writing of this policy. Commonly submitted codes Below are the most commonly submitted codes for the service(s)/item(s) subject to this policy. This is not an exhaustive list of codes. Providers are expected to consult the appropriate coding manuals and bill accordingly. CPT Code Description Comments 81227 CYP2C9 (cytochrome P450, family 2, subfamily C, polypeptide 9) (e.g., drug metabolism), gene analysis, common variants (e.g., *2, *3, *5, *6). 81355 VKORC1 (vitamin K epoxide reductase complex, subunit 1) (e.g., warfarin metabolism), gene analysis, common variants (e.g., -1639/3673). ICD-10 Code Description Comments No covered diagnoses. HCPCS Level II Code G9143 Description Warfarin responsiveness testing by genetic technique using any method, any number of specimen(s). Comments 8