Epidemiology and risk factors of severe cutaneous drug reactions and evaluation through available databases

Severe cutaneous adverse drug reactions (SCARs), comprising mainly Stevens-Johnson syndrome (SJS) and toxic epidermal necrolysis (TEN), drug reaction with eosinophilia and systemic symptoms, acute generalized exanthematous pustulosis, generalized bullous fixed drug eruption (GBFDE), are rare adverse reactions to drugs, hampered by a significant morbidity and mortality. Due to their unpredictable nature, they completely disrupt the benefit-risk balance of drug prescription. Nonetheless, a partial predictability is emerging. Firstly, all drugs do not carry the same risk for SCARs. Drugs carrying the higher risk are allopurinol, carbamazepine, lamotrigine, iodinated contrast media, and antibiotics, particularly sulfonamides and aminopenicillins. In addition, main risk factors comprise genetic polymorphisms of the HLA system and increased drug dosage. A targeted HLA genetic screening together with drug avoidance in case of high risk allowed to reduce the incidence of SCARs associated with carbamazepine and allopurinol. Similarly, a progressive titration of lamotrigine dose allowed to reduce the incidence of SCARs associated with lamotrigine. Pharmacovigilance databases allow the quick identification of culprit drugs, together with the implementation of whistle-blowing policies. Nonetheless, they present with quantitative and qualitative limitations and their own particular biases. Pharmacoepidemiologic databases allow the computation of SCAR incidences and confirm previously identified associations between drugs and SCARs. They may also allow help to identify the main clinical risk factors, e.g., chronic renal disease and Asian ethnicity for allopurinol. Nonetheless, they are limited by measure bias and lack of clinical and biological data. Ultimately, clinical and pharmacogenomic databases allow the identification of more precise risk factors, comprising precise genetic polymorphisms, drug dosage and impaired metabolism. The combination of all allows to establish a personalized risk stratification across SCARs subtypes and drug prescriptions. Ultimately, this allows the establishment of prescribing policies and the decrease of SCARs incidence.