In this observational case–control study, 107 cutaneous adverse reaction (CAR) cases (CAR+) manifesting up to 12 weeks after the start of treatment with antiseizure medication (ASM) were identified. Control groups consisted of 98 epilepsy patients without a history of CAR (CAR‐) and 3965 healthy individuals in the Brazilian National Registry of Bone Marrow Donors. All participants were HLA typed by high‐resolution Next Generation Sequencing for HLA‐A, B, C, DQB1 and DRB1; HLA‐DPA1, DPB1, DQA1, DRB3, DRB4 and DRB5 were also sequenced in samples from CAR+ and CAR‐ individuals. The relationship between the carrier frequency of each allele, CAR type and ASM for all participants was investigated. The ASMs most frequently associated with CAR were carbamazepine (48% of CAR+ subjects), lamotrigine (23%), phenytoin (18%), phenobarbital (13%) and oxcarbazepine (5%). The main alleles associated with a risk of CAR were HLA‐A*02:05 (OR = 6.28; p = 0.019, carbamazepine or oxcarbazepine); HLA‐DPA1*02:02 (OR = 4.16, p = 0.003, carbamazepine); HLA‐B*53:01 (OR = 47.9, p = 0.014, oxcarbazepine), HLA‐DPA1*03:01/DPB1*105:01 (OR = 25.7, p = 0.005, phenobarbital); HLA‐C*02:10 (OR = 25.7, p = 0.005, phenobarbital) and HLA‐DRB1*04:02 (OR = 17.22, p = 0.007, phenytoin). HLA‐A*03:01 increased the risk for phenytoin‐induced maculopapular exanthema 4.71‐fold (p = 0.009), and HLA‐B*35:02 was associated with a 25.6‐fold increase in the risk of carbamazepine‐induced Stevens‐Johnson syndrome (p = 0.005). None of the 4170 subjects carried the HLA‐B*15:02 allele, and HLA‐A*31:01 was not associated with CAR. Hence, HLA‐A*31:01 and HLA‐B*15:02 were not associated with CAR in this population. Although other HLA class I and II alleles tested were associated with a risk of CAR, none of these associations were strong enough to warrant HLA testing before prescribing ASM.
