NEW YORK (GenomeWeb) – A group of researchers led by Samuel Johnson of Kaiser Permanente Colorado is hoping to garner the attention of payors and health systems by showing how CYP2C19 genotyping can help them differentiate which acute coronary syndrome patients would fare best on low-cost generic clopidogrel, and which would be better off on higher priced branded agents, Daiichi Sankyo/Eli Lilly's Effient (prasugrel) and AstraZeneca's Brilinta (ticagrelor).
Studies have shown that patients carrying certain loss-of-function CYP2C19 alleles may not respond well to clopidogrel — the most widely used generic antiplatelet — and that genotyping might help doctors determine which patients should receive more expensive, newer treatments in a cost-effective manner.
For example, a study published in the Annals of Internal Medicine last year by Kazi et al. found that a PGx strategy where testing wasn't more than $358 was cost-effective in guiding antiplatelet therapy for acute coronary syndrome patients following a stent procedure.
Although these cost-effectiveness analyses make compelling arguments in favor of implementing PGx testing in specific scenarios, they haven't presented this information in a way that highlights the benefit to payors or large health systems, according to Johnson. "In recent history, one of the big perceived barriers to pharmacogenetic testing has been the cost/reimbursement challenge," Johnson told GenomeWeb. "We really wanted to present an angle that spoke to payors from that perspective rather than a more academic perspective."
In a paper published this month in the Journal of Managed Care & Specialty Pharmacy, Johnson and colleagues modeled the financial impact of CYP2C19 genotyping in a cohort of 1,000 acute coronary syndrome patients who had a stent procedure. In this scenario, genotyping all patients in order to determine whether to give them clopidogrel, Effient, or Brilinta saved 20 lives and close to $445,000.
According to published consumer reports, generic clopidogrel can cost as low as $1 a day, while a month's supply of the two other drugs can amount to several hundred dollars depending on the patient's insurance plan. In their study, Johnson's group used 2012 wholesale drug prices available to large managed care organizations and national Medicare rates to estimate the cost of adverse events like non-fatal myocardial infarction, stroke, bleeds, and cardiovascular death.
"This analysis was intentionally modeled from the payor perspective to practically demonstrate the magnitude of the financial impact from CYP2C19 genotype-guided antiplatelet therapy," Johnson and colleagues wrote in the paper.
Despite studies showing that variations in CYP2C19 impact people's ability to process clopigorgrel, the majority of doctors in the US still don't order PGx testing to inform anti-platelet prescribing. Their reluctance is due to the fact that treatment guidelines don't yet back routine testing and published studies involving heterogeneous cardiac populations have shed doubt on the association between CYP2C19 genotypes and cardiovascular events. Furthermore, the unavailability of rapid, point-of-care testing in the US, reimbursement barriers, and the challenges of integrating genomic data in electronic medical records have made it even harder for healthcare systems to justify implementing such testing.
Ask testing shops selling CYP2C19 genotyping assays and they'll say that drugmakers haven't helped promote PGx testing not for a lack of evidence but because they don't want to cut the market share of their drugs. Although the makers of the branded version of clopidogrel (Plavix), Bristol-Myers Squibb and Sanofi-Aventis, did update the label of the drug in 2010 with a black box warning that patients with certain CYPC19 gene variations have a diminished ability to process the drug, at the time, "there wasn't much incentive for [them] to advance the notion that it would be helpful to genotype," Johnson said.
The US Food and Drug Administration added the black box warning regarding CYP2C19 testing to Plavix's label during a critical time for its sponsors — two years before the drug lost patent protection. For several years, Plavix had been the second best-selling drug in the world, and in 2010, the drug brought in $9 billion in worldwide sales. But it was also starting to face competition from newcomer Effient and soon Brilinta would enter the market.
With the drug's patent life running short, the makers of Plavix never advanced a companion diagnostic to identify poor responders for the drug. Diagnostics shops like Spartan Bioscience and Autogenomics developed CYP2C19 tests and got FDA clearance for them as tools to guide treatment strategy. But without the financial support of the drugmakers, it was prohibitively expensive for a diagnostic firm to do the necessary studies and develop a companion test specifically for Plavix, Spartan CEO Paul Lem previously told GenomeWeb.
Even without the market incentives, the state of Hawaii has sued BMS and Sanofi, asserting that the companies had an obligation to better educate the public about the diminished response in CYP2C19 poor metabolizers. In the lawsuit, the plaintiffs allege that the drugmakers knew since 1998 that Plavix didn't work as well for approximately 30 percent of the state's population due to genetic traits or the influence of other drugs.
"If clopidogrel had been approved as a new drug in the last two to three years … Plavix would have been approved with a recommendation for genotyping," Johnson said, reflecting on conversations he's had with regulatory experts. "It's interesting what a difference a decade makes."
Impacting market share
Since it lost patent protection in 2012, generic clopidogrel remains the most widely used antiplatelet drug, which according to some in the field perhaps heightens the argument for genotyping since a significant portion of patients prescribed the drug likely won't derive treatment benefit.
Drawing on data from a large administrative database, Johnson's group estimated that clopidogrel has 93 percent of the market, while Effient has 5 percent and Brilinta has 2 percent. However, in their financial model, when genotyping was given to 50 percent of the cohort (reflecting a scenario where some patients or doctors decide not to test), clopidogrel had 83 percent of the market, while Effient and Brilinta's share grew to 12.1 percent and 4.9 percent, respectively. When all patients were genotyped, clopidogrel had 73 percent of the market, and Effient and Brillinta had 19.3 percent and 7.7 percent, respectively.
Switching from genotyping none of the patients to half the cohort would save around $222,000, while switching from no genotyping to testing all patients would save approximately $445,000. According to the model, the lower overall costs per patient and lower spending for healthcare systems are due the non-fatal heart-related adverse events, bleeding, cardiovascular deaths that genotyping helps avoid.
In Johnson's model one of the unanticipated findings was that a genotyping strategy appears to positively impact the market share of newer anti-platelet agents. "Looking at the data at face value, it does illustrate that maybe there is a benefit to the pharmaceutical industry supporting this because they might realize a higher market share," he said.
The model took into account intermediate and poor metabolizers, which accounts for 27 percent of the population. However, Johnson pointed out that based on the evidence, most clinicians would likely focus on the poor metabolizers, which would cut down the number of patients that would receive branded agents after gentoyping.
If clopidogrel had been approved as a new drug in the last two to three years … Plavix would have been approved with a recommendation for genotyping.
Moreover, the three antiplatelets aren't interchangeable. For example, Effient's label restricts its use in people with a history of stroke, but clopidogrel isn't restricted in this population. Brilinta is the only drug of the three with effects that aren't irreversible. Moreover, the newer agents have higher risk for serious bleeding. "Those also factor into the market share" of these drugs, Johnson said.
As such, the makers of Brilinta and Effient, focused on growing sales of their drugs, may not see the genotyping argument as favorable to their finances. According to AstraZeneca's 2014 annual report, Brilinta was one of its major growth drivers, bringing in $476 million globally, marking a 70 percent increase from the previous year. Last year, worldwide sales for Effient were $522 million according to Lilly's annual report, marking 3 percent growth.
In launching Effient and Brilinta, these drugmakers have largely pushed their products as more effective at preventing cardiac events than the older drug, without its PGx testing headaches. In the TRITON-TIMI 38 trial that led to Effient's approval, for example, acute coronary syndrome patients treated with the drug had a lower risk of dying from cardiovascular causes and of experiencing nonfatal strokes or myocardial infarctions, compared to patients receiving clopidogrel. Similarly, in PLATO, acute coronary syndrome patients had lower cardiovascular deaths and non-fatal myocardial infarctions on Brilinta compared to those on clopidogrel.
Although these same studies also showed that Effient and Brilinta caused more life-threatening or fatal bleeding events than clopidogrel, the labels of these drugs point out that unlike clopidogrel, Effient and Brilinta aren't affected by CYP2C19 genotypes. Around 75 percent of the general patient population has CYP2C19 genotypes that make them able to benefit from clopidogrel. Based on this, drugmarkers have highlighted head-to-head comparisons of newer agents and clopidogrel and claimed superiority, Johnson said. "So, why would you genotype and use clopidogrel, when you could just prescribe one of the new agents and forget about it?"
But Johnson points out that these trials empirically treated everybody with Brilinta or Effient, and didn't look more closely at whether treating poor metabolizers of clopidogrel with one of the newer drugs had clinical and economic benefits. Subsequent analysis has shown that such a strategy would impact outcomes.
With its study, Johnson's group wants to relay to payors and health systems that genotyping "is a rational approach to consider based on evidence" and "you don't have to empirically prescribe one of the newer agents." This argument is further strengthened, Johnson said, since multiplex testing makes it very cheap to gauge CYP2C19 genotypes.
Colorado test case
Kaiser Permanente Colorado for a year and a half has been performing CYP2C19 genotyping for acute coronary syndrome patients who have undergone a stent procedure. The Colorado system also performs PGx testing for tricyclic antidepressants and thiopurine drugs.
For the time being, Kaiser Colorado has decided to only implement specific gene-drug pairs and not to implement a broad PGx panel. Since not all the markers in such a panel have proven clinical value, these results would need to be stored in a database separate from the one used for medical decision making. "Just setting all that up and the logistics and cost of having that dual registry system, we're still navigating how we want to do that," he said.
According to Johnson, experts are working on streamlining reporting results within patients' EMRs and decision support tools for doctors. They will track how testing impacts outcomes and if it reduces physician office visits, emergency room visits, or hospitalizations due to adverse events.
"Our intent is to validate this model," Johnson said, adding that Kaiser is thinking about how to implement such a program in other regions, such as a Hawaii, where a higher percentage of the population has loss-of-function CYP2C19 alleles and could derive even greater benefit from personalized antiplatelet treatment. "So, we're definitely the test case here."