I saw a woman many years ago in the hospital with fever, lymphadenopathy, and liver function abnormalities associated with phenytoin. Her case has stuck with me, probably because her EEG/video monitoring demonstrated nonepileptic seizures and not epilepsy. She suffered from a serious complication of phenytoin, a drug she didn’t even need!
Of all the antiepileptic drugs (AEDs), phenytoin is associated with the highest rate of rashes (5.9%), followed by lamotrigine (4.8%) and carbamazepine (3.7%). Although usually mild, drug rashes can be life-threatening if they take the form of Stevens-Johnson syndrome (SJS), toxic epidermal necrolysis (TEN), or drug reactions with eosinophilia and systemic symptoms (DRESS).
A recent study revealed that phenytoin-related severe cutaneous adverse reactions are associated with CYP2C9*3 missense variants on chromosome 10 in an Asian population. This case-controlled study included 105 cases of phenytoin-related severe cutaneous adverse reactions (61 SJS or TEN, 44 DRESS); 78 cases of maculopapular exanthema; 130 phenytoin-tolerant control individuals; and 3655 population control individuals from Japan, Malaysia, and Taiwan. A genome-wide association study (GWAS) included 60 cases with severe phenytoin-related reactions and 412 Taiwanese controls. The results were validated in Taiwanese, Japanese, and Malaysian populations.
GWAS identified a cluster of 16 single nucleotide polymorphisms in CYP2C genes on chromosome 10 (10q23.33). Of these, the missense variant rs1057910 (CYP2C9*3) had the highest association (odds ratio, 12) with phenytoin-related severe cutaneous adverse reactions.
Pharmacogenomics and Phenytoin
CYP2C9 is one of the most important hepatic metabolizing enzymes in humans; its substrates include phenytoin, S-warfarin, tolbutamide, and numerous nonsteroidal anti-inflammatory drugs. In the genetics study in an Asian population, the CYP2C9*3 isoform was associated with slower phenytoin metabolism, resulting in higher phenytoin serum levels. Delayed phenytoin clearance coupled with heightened immunogenicity may have increased the risk for cutaneous adverse reactions. The CYP2C9*3 variant is much more common in Asian than white populations.
AEDs are an important cause of severe cutaneous adverse reactions in Asian patients. The risk for carbamazepine-related SJS/TEN is approximately 10 times that in white patients. Since 2007, the US Food and Drug Administration has recommended screening for the HLA-B*1502 allele in Asian patients before they are prescribed carbamazepine.
Significant cross-reactivity for rash between phenytoin and carbamazepine also exists. In one retrospective study, 10 of 17 patients (59%) who had a rash from phenytoin had a carbamazepine rash, whereas 10 of 25 patients (40%) with a carbamazepine rash developed a phenytoin rash. In another study, cross-reactivity between carbamazepine and phenytoin was 45% for antihypersensitivity syndrome (defined as two or more of the following symptoms: rash, fever, lymphadenopathy, hepatitis, or hematologic abnormalities). Both phenytoin and carbamazepine are metabolized to reactive aromatic epoxides, which may contribute to their cross-reactivity. CYP2C9 does not appear to play a role in this cross-reactivity because it does not metabolize carbamazepine.
The finding that CYP2C9*3 predisposes to severe cutaneous adverse reactions from phenytoin will allow genetic testing and avoidance of phenytoin-related morbidity and mortality. The work of Chung and colleagues is testimony that the promise of pharmacogenomics to provide personalized therapy that enhances drug safety is slowly but surely materializing. Testing of HLA-B*1502 is already available to prevent carbamazepine-induced SJS-TEN. An explanation for the high cross-reactivity between phenytoin and carbamazepine requires further exploration.
Future studies will no doubt expand our knowledge of genetic susceptibility for AED-related adverse reactions. Moreover, clinicians may one day rely on genetic testing to select AEDs that are not only safer for a particular patient but more likely to be efficacious.
2 of 2Previous Page