Pharmacoepidemiology and Pharmacogenetics of a Statin Adverse Event

BACKGROUND:

Cerivastatin (Baycol), an HMG-CoA reductase inhibitor (statin), was approved and marketed in
early 1998 for the treatment of dyslipidemias. Soon afterwards, suspected adverse drug
reaction (SADR) reports in cerivastatin users often cited rhabdomyolysis, an uncommon
condition in which the breakdown of skeletal muscle cells causes pain, weakness and, in-some
cases, renal failure or death. By the time that cerivastatin was withdrawn from the market
in 2001, the FDA had received 1,899 SADRs for rhabdomyolysis associated with cerivastatin
compared to 1,440 for all other statins combined. In an analysis conducted by FDA
scientists, the relative reporting rate (RRR) of fatal rhabdomyolysis was 40 times higher
among users of cerivastatin than among users of other statins. For the outcome of all fatal
and non-fatal rhabdomyolysis, the RRR was 54 times higher. About half of the rhabdomyolysis
cases occurred in subjects who had taken both cerivastatin and gemfibrozil. Subsequently,
pharmacokinetic studies demonstrated that gemfibrozil inhibits both major metabolic pathways
for cerivastatin-not only the Cytochrome P-450 (CYP) 2C8-mediated oxidation, but also the
glucuronidation by uridine diphosphate glucuronysyltransferases (UGT). The investigators
hypothesize that rhabdomyolysis cases who were taking cerivastatin but not gemfibrozil had
one or both of two types of risk factors: (1) medications that are known inhibitors of these
enzymes; or (2) functional genetic variants in one or more of the CYP or UGT enzymes.

Coronary heart disease (CHD) is the leading cause of morbidity and mortality in the U.S. and
statin agents are commonly employed for the treatment of hyperlipidemia, one of the
principal risk factors for CHD. Given the hundreds of thousands to millions of Americans
presently on statin therapy, and the potentially serious (though relatively rare)
complication of rhabdomyolysis, this is a significant topic to study.

DESIGN NARRATIVE:

The investigators hypothesize that rhabdomyolysis cases who were taking cerivastatin but not
gemfibrozil had one or both of two types of risk factors: (1) medications that are known
inhibitors of these enzymes; or (2) functional genetic variants in one or more of the CYP or
UGT enzymes. This project is a case-control study of factors that increased the risk of
rhabdomyolysis in cerivastatin users. Cases will be cerivastatin users who had
rhabdomyolysis, and statin users from two on-going epidemiologic studies will serve as two
complementary control groups. "For cases, identified with the assistance of attorneys, a
review of medical records and telephone interview will verify the diagnosis and provide
information on medication use. Subjects will also return a mouthwash buccal sample for DNA
extraction. All subjects will be genotyped for known functional variants in CYP2C8, UGT1A1
and UGT1A3. In the hypothesis-testing part of this case-control study, medications known to
be inhibitors of these enzymes and functional variants in their genes will be evaluated as
risk factors for rhabdomyolysis in cerivastatin users. Additionally, pharmacoepidemiologic
analyses of the FDA AERS data will be conducted to identify other potential new drug-drug
interactions. Power to detect expected effect sizes is 80% or greater. In the discovery
phase of the project, the DNA of rhabdomyolysis cases who did not take gemfibrozil will be
sequenced to identify new single nucleotide polymorphisms (SNPs) in CYP2C8, UGT1 Al and UGT1
A3. The functional significance of new pharmacoepidemiologic associations will be evaluated
in in vitro pharmacokinetic inhibition studies. Additionally, newly discovered SNPs in
CYP2C8 will be expressed in E. coli and evaluated for their functional significance.