Carbamazepine-induced SJS/TEN & HLA-B*1502
Carbamazepine is an anticonvulsant and mood-stabilizing drug, used primarily in the treatment of epilepsy and bipolar disorder, but also in the treatment of attention deficit hyperactivity disorder, schizophrenia and trigeminal neuralgia. Carbamazepine acts by stabilizing the inactive state of the voltage-dependent sodium channels of neurons in a use- and voltage-dependent manner.[39,40]
Carbamazepine can cause several serious adverse reactions, including maculopapular exanthema (MPE), SJS/TEN and HSS. SJS and TEN are severe cutaneous adverse reactions, characterized by bullae resulting from epithelial cell apoptosis on both epidermal and mucosal surfaces and have mortality rates of 5 and 30%, respectively. They are considered different parts of the same disease spectrum, subdivided on the level of epidermal detachment; less than 10%, 10–30%, and greater than 30% detachment are classified as SJS, SJS/TEN overlap and TEN in that order. Lesions contain a high level of CD3+ T-cell infiltrate, dominated by the CD8+ T-cell subset, and the MHC-I pathway is implicated in keratinocyte cytotoxicity.[26–29] Carbamazepine–SJS/TEN occurs within 2 months of drug administration, with a median onset of 15 days, and is estimated to be responsible for up to 35% of drug-induced SJS/TEN cases in the Han Chinese population, but only 6% of cases in Caucasians.
Like carbamazepine–SJS/TEN, carbamazepine–MPE also displays delayed onset (median onset 23 days) and is characterized by CD3+ T-cell infiltrate at the dermoepidermal junction and perivascular dermis, the former featuring approximately equal numbers of CD4+ and CD8+ cells, and the later dominated by CD4+ cells. Carbamazepine–HSS is classified as a cutaneous reaction (nonbullous) plus two of fever, lymphadenopathy and hematological abnormalities, and involvement of at least one internal organ such as the liver, lungs kidney or heart, and has a median onset of 33 days.
Of these hypersensitivities, carbamazepine–SJS/TEN is strongly associated with HLA-B*1502 in the Han Chinese population appearing in 100% of affected individuals but only 3% of carbamazepine-tolerant individuals and 8.6% of the general population. This association does not extend to Caucasians where the incidence of HLA-B*1502 is lower,[45,46] but has been observed in a Thai population where an association of HLA-B*1502 with the phenytoin-induced SJS/TEN (but not with MPE) was also observed. Phenytoin is another anticonvulsant used in similar circumstances to carbamazepine and thought to elicit its action via binding of the sodium-dependent voltage channels in a carbamazepine-like fashion. An individual from a Han Chinese population experiencing phenytoin-induced SJS was also found to possess HLA-B*1502 in an unrelated study.
Carbamazepine–MPE and HSS are not associated with HLA-B*1502, although there is preliminary evidence of their association with other parts of the MHC region of chromosome 6, including HLA-A*3101 for carbamazepine–MPE.
Currently available in vitro data for carbamazepine hypersensitivity show that carbamazepine-specific T cells, displaying different effector functions and homing characteristics, apparently persist in hypersensitive patients' blood for many years after resolution of clinical symptoms. Both CD4+ and CD8+ T cells can be reactivated in vitro in response to carbamazepine and its various metabolites in the absence of antigen processing. Although CD8+ cells proliferate less readily in response to carbamazepine they demonstrate a higher carbamazepine-specific cytotoxicity, effectively killing target cells via an MHC-restricted, perforin-dependent pathway.[48–50] Unfortunately the patient cohorts in these studies were not genotyped for the HLA loci, nor their ethnicities commented upon. As a result no comment on the nature of T-cell recruitment in reference to the HLA can be made, and the interaction between the HLA molecule and T-cell receptor (TCR) remains unresolved.
The strength of the association between HLA-B*1502 and carbamazepine–SJS/TEN in both the Han Chinese and Thai population, but not the Caucasian population suggests that, as with AHS, the HLA association is not the only factor involved in disease etiology. This has led to the suggestion that HLA-B*1502 is not mechanistically involved in the reaction, but is in strong LD with a causative polymorphism in Asian populations and that this LD does not extend to European populations, where HLA-B*1502 is of lower prevalence. However, as carbamazepine–SJS/TEN follows a DTH profile of progression and involves CD8+ T cells, it seems a more favorable hypothesis that HLA-B*1502 is mechanistically involved in carbamazepine–SJS/TEN against an Asian genetic background, whilst other factors may be involved in disease progression in European populations. Regardless, it appears that HLA-B*1502 is a strong marker for susceptibility to carbamazepine–SJS/TEN in the Han Chinese population and other Asian populations. Screening for HLA-B*1502 in those of Asian background prior to carbamazepine (and maybe phenytoin) administration may significantly reduce the occurrence of this hypersensitivity; however, it is clear that such a strategy would be of little use in those of Caucasian ancestry.
Personalized Medicine. 2010;7(5):495-516. © 2010
Cite this: Personalized Medicine for HLA-associated Drug-hypersensitivity Reactions - Medscape - Sep 01, 2010.