Lilly, Daiichi Explore Pharmacogenomic Marker For Prasugrel

Circulation study shows prasugrel metabolism unaffected by marker for Plavix non-response. Data won’t surprise FDA, but what will be the regulatory response?

A May 4 Circulation article describing prasugrel's efficacy in patients who carry a genetic marker indicative of non-response to Plavix may be the opening volley in one of the first big marketing battles of the personalized medicine era.

The article reports on an analysis of a subset of patient's from Daiichi Sankyo/Eli Lilly & Co.' s massive TRITON study, which pitted prasugrel against Bristol-Myers Squibb/Sanofi Aventis' clopidogrel in 13,600 acute coronary syndrome patients undergoing percutaneous coronary intervention. The overall study showed superior efficacy for prasugrel in reducing ischemic events, albeit with a higher risk of bleeding (¹ (Also see "Lilly’s Prasugrel Superior To Bristol’s Plavix In Reducing CV Events In Pivotal TRITON Trial" - Pink Sheet, 4 Nov, 2007.)).

The subgroup analysis published in Circulation focused on the efficacy of prasugrel in the subset of 1,466 patients who agreed to provide samples for genetic screening, and compared the response in patients with markers for reduced function in the cytochrome P450 metabolic pathway. One genotype in particular--a reduced function of the CYP 2C19 allele--has been associated with non-response to Plavix, presumably because the enzyme is critical in converting clopidogrel into its active metabolite (² (Also see "Thrombosis Events Tied To Gene Mutation In Stent Patients On Clopidogrel" - Pink Sheet, 1 Jan, 2009.)).

The study, by TIMI group investigator Jessica Mega, MD, et al., found no impact on efficacy of prasugrel from any of the CYP alleles screened, and in particular that "CYP2C19 variants do not significantly affect pharmacological or clinical outcomes in patients treated with prasugrel."

"These genetic observations explain some of the differences in the pharmacological and clinical response to treatment with prasugrel compared with clopidogrel and potentially could be used to help tailor pharmacotherapy in the future," the authors conclude.

A Double-Edged Sword for Prasugrel

On its face, the finding is a double-edged sword for prasugrel. In the traditional primary care blockbuster marketing model, it offers a clear point of differentiation that could encourage prescribers to choose prasugrel first based on a superior pharmacologic profile.

However, in a world of greater emphasis on personalized medicine, the message of an overall better profile could be turned on its head. The marker for Plavix non-response is present in about 30% of the Caucasian population, and about double that in Asians. (In the substudy, 405 of the 1,466 patients carried the non-response marker). That could provide a basis for payors to limit use of prasugrel to the subset of patients who will not respond to Plavix, especially if the non-responder issue is believed to explain the overall difference in efficacy seen in the TRITON study.

The most significant aspect of the Circulation article, however, may not be its conclusions, but rather the fact that Lilly and Daiichi sponsored the research in the first place. Lilly has publicly committed to a vision of personalized medicine, and the Circulation study is a tangible sign of that commitment (³ (Also see "Lilly CEO Taurel Passes Baton To Lechleiter Effective April 1" - Pink Sheet, 18 Dec, 2007.)).

At a practical level, there is currently no commercial test for the CYP2C19 allele, so there will be no immediate impact on medical practice. "This type of genotyping is typically performed in a research capacity and therefore in most cases is not currently being used to alter patient care," says lead author Mega.

Bristol agrees on that point. "These data are not conclusive for providing clinical recommendations," company spokesperson Laura Hortas says.

FDA Aware of Potential Marker

The publication certainly will not surprise FDA, which is continuing to review the application for prasugrel--now seven months past its user fee goal date of September 2008 (⁴ (Also see "Last Gasp Or New Life? 2009’s Novel Products Have Blockbuster Potential" - Pink Sheet, 23 Mar, 2009.), p. 35).

For the record, Lilly says it "remains engaged with the FDA on the final stages of the NDA review process for prasugrel, and we hope to be able to put this new medication into the hands of physicians for their patients in the United States soon. Daiichi Sankyo, Inc. and Eli Lilly and Company remain optimistic; however, we cannot speculate on the outcome of the FDA's review of the prasugrel New Drug Application or when the FDA will issue an Action Letter."

FDA is well aware of the Plavix non-response data, which has been the subject of several independent publications, and it also served as the background for a new safety warning regarding use of Plavix in combination with proton pump inhibitors (⁵ (Also see "BMS/Sanofi-Aventis To Study Plavix Efficacy Concerns" - Pink Sheet, 26 Jan, 2009.)).

Nor is the idea of using the marker as a way to target prasugrel therapy new for FDA. Several agency officials, including Center for Drug Evaluation & Research Director Janet Woodcock, participated in an Institute of Medicine workshop on pharmacogenomics and drug safety in March, which included a presentation on the Plavix marker and a discussion of the potential utility of genetic screening to select appropriate patients for prasugrel.

[Editor's Note: An in-depth look at the potential for prasugrel to blaze a trail in personalized medicine will appear in the May issue of ⁶ The RPM Report.]

"I think it is well known that one reason prasugrel was developed was that it is not subject to some of the metabolizing enzyme issues that affect clopidogrel," Woodcock said after the meeting. "We are well aware of this" issue.

Further Studies

Lilly is not speculating on what if any reference to the pharmacogenomic data will be in prasugrel labeling.

One key question from FDA's perspective is likely to be how representative the patient sample in the sub-study is of the overall population.

During the IoM meeting, FDA Office of Biostatistics Director Robert O'Neil highlighted the problems with relying on a "convenience sample" as the basis for regulatory decisions. O'Neill focused specifically on the use of KRAS as a potential marker of response for EGFR inhibitor cancer therapies, but made clear that in general FDA will have questions about samples derived from a portion of an overall study population, where there is at least a potential for bias because subjects have to agree to provide a sample for screening.

The data reported in Circulation were part of a pre-specified analysis, which should help address that concern, but the sample includes only one-tenth of the total study population.

Lilly is mum about plans for prospectively designed studies to test the hypothesis that prasugrel works in patients who do not respond to Plavix. "For proprietary reasons, we do not discuss any future clinical trial plans unless they have already disclosed," Lilly says.

Lilly notes that the only study it has publicly discussed is a 10,000 patient trial comparing prasugrel and clopidogrel in acute coronary syndrome (TRILOGY). The ClinicalTrials.gov listing for that study notes that genotyping will be one aspect of that trial.

In addition,ClinicalTrials.gov includes a ⁷ listing for a comparative trial in Asian patients (who are much more likely to carry the Plavix non-response marker).

Impact on Plavix Labeling?

There may also be a regulatory response in the labeling of Plavix. "We continue to work with FDA and other regulatory agencies to ensure that our label includes the most up to date scientific and clinical information on Plavix as it is essential that both physicians and patients receive appropriate guidance on the use of this drug," Bristol's Hortas said.

Not surprisingly, Bristol is clear about one thing it will not be doing in response to the data: the company and Sanofi have no plans to develop a commercial test.

Beyond that, we "are working with the FDA, and conducting studies to obtain additional information that will allow us to understand and characterize the factors that may influence this complex issue," Hortas says. "Sanofi-Aventis/Bristol Myers Squibb have been reviewing this complex issue specifically with regard to metabolism of Plavix, including a review of our extensive database (more than 100,000 patients), evaluating new data conducted by others available through the public domain, and proposing new studies."

FDA, of course, does not have to wait for Bristol if it wants to push the screening test. "We can require info in the labels even if it is not submitted by the sponsors," Woodcock notes. "A lot of the pharmacogenomic data has entered labels through FDA instigation" in other cases.

- Michael McCaughan ([email protected])