Genome-wide association study of Stevens-Johnson Syndrome and Toxic Epidermal Necrolysis in Europe
1 Inserm U946, F-75010, Paris, France
2 Institut Universitaire d'Hématologie, Université Paris Diderot, F-75010, Paris, France
3 Institute of Medical Biometry and Medical Informatics, University Medical Center, D-79095Freiburg, Germany
4 Inserm U448, F-94010, Créteil, France
5 Service Dermatologie, Hôpital Henri-Mondor, Université Paris-Est, F-94010, Créteil, France
6 Department of Dermatology, Azienda Ospedaleria Ospedali Riuniti di Bergamo, Milano University, Bergamo, Italy
7 Dokumentationszentrum schwerer Hautreaktionen (dZh), Department of Dermatology, D-79095, Freiburg, Germany
8 INSERM U781, F-75743, Paris, France
9 Université René Descartes, F-75743, Paris, France
10 Centre Hospitalier Universitaire Necker-Enfants malades, Departments of Genetics and Dermatology, F-75743, Paris, France
Orphanet Journal of Rare Diseases 2011, 6:52 doi:10.1186/1750-1172-6-52Published: 29 July 2011
Stevens-Johnson syndrome (SJS) and Toxic Epidermal Necrolysis (TEN) are rare but extremely severe cutaneous adverse drug reactions in which drug-specific associations with HLA-B alleles were described.
To investigate genetic association at a genome-wide level on a large sample of SJS/TEN patients.
We performed a genome wide association study on a sample of 424 European cases and 1,881 controls selected from a Reference Control Panel.
Six SNPs located in the HLA region showed significant evidence for association (OR range: 1.53-1.74). The haplotype formed by their risk allele was more associated with the disease than any of the single SNPs and was even much stronger in patients exposed to allopurinol (ORallopurinol = 7.77, 95%CI = [4.66; 12.98]). The associated haplotype is in linkage disequilibrium with the HLA-B*5801 allele known to be associated with allopurinol induced SJS/TEN in Asian populations.
The involvement of genetic variants located in the HLA region in SJS/TEN is confirmed in European samples, but no other locus reaches genome-wide statistical significance in this sample that is also the largest one collected so far. If some loci outside HLA play a role in SJS/TEN, their effect is thus likely to be very small.