Genetic risk and long-term outcomes associated with Drug-induced Stevens-Johnson syndrome and toxic epidermal necrolysis in survivors

项目摘要 Stevens-Johnson综合征（SJS）和中毒性表皮坏死松解症（TEN）与广泛的 表皮坏死，临床表现包括广泛的水疱和大疱， 粘膜和眼睛。SJS/TEN的病死率高达50%，长期的身心损害， 健康发病率相当高，而且研究不足。在成人中，超过80%的SJS/TEN与药物相关， 有希望的发现与I类特异性主要组织相容性复合体等位基因的变异有关， 如HLA-B * 15：02和HLA-B * 58：01，与卡马西平和别嘌呤醇SJS/TEN相关。 然而，驱动SJS/TEN风险的遗传因素尚不清楚。在别嘌呤醇的情况下，虽然HLA- B * 58：01在东南亚具有接近100%的阴性预测值，仅在欧洲和非洲具有50 - 60%的阴性预测值 发生别嘌呤醇SJS/TEN的患者携带HLA-B * 58：01。美国SJS/TEN的常见原因包括 甲氧苄啶-磺胺甲恶唑、别嘌呤醇和芳香族抗惊厥药，如拉莫三嗪、苯妥英和 卡马西平在美国人群中的流行遗传关联尚未确定， 预防工作和执行工作停滞不前。SJS/TEN的罕见性和缺乏获得大队列的机会， 幸存者的死亡削弱了确定遗传风险因素和长期发病率的能力。我们将使用注册表 由SJS基金会（http：//sjsupport.org/）开发，以开发表型的数据和DNA生物库 判定的SJS/TEN幸存者。我们的可验证假设是，我们将确定遗传风险因素， 与SJS/TEN相关的最常见药物以及长期并发症的性质和风险 与SJS/TEN相关，两者都可能对SJS/TEN产生重大影响 预防和病人的结果。在具体目标1中，我们将建立一个大型SJS/TEN幸存者队列 和相关的DNA生物库参与者将通过SJS基金会网站招募， Facebook页面和注册表，并同意医疗记录审查和口腔DNA收集。 药物诱导的SJS/TEN的独立裁定将确定合格的受试者。具体目标 2我们将定义与SJS/TEN相关的长期精神和身体健康并发症， 相关的风险因素。经验证的问卷将评估跨- 在SJS/TEN后的不同时间点，通过经验证的心理学工具对患者进行分段 痛苦、创伤后应激障碍和与健康有关的生活质量。在具体目标3中，我们将确定 HLA和其他遗传风险因素与药物诱导的SJS/TEN相关高分辨率HLA、ERAP 将对1000名患者进行KIR分型以及扩展的多种族基因分型阵列（MegaEx） 由三名皮肤科医生组成的独立小组证实患有药物诱导的SJS/TEN。 对照将是采用MegaEx分型、插补HLA/KIR/ERAP分型的100，000名BioVu参考人群 以及在年龄、性别、种族和基础疾病上2：1匹配的BioVu药物耐受对照。

PROJECT SUMMARY Stevens-Johnson syndrome (SJS) and toxic epidermal necrolysis (TEN) are associated with widespread epidermal necrosis with the clinical presentation consisting of widespread blisters and bullae and involvement of mucous membranes and the eyes. The mortality of SJS/TEN is up to 50% and long-term physical and mental health morbidity is considerable and understudied. In adults over 80% of SJS/TEN is drug associated, and promising discoveries have associated variation in class I specific major histocompatibility complex alleles such as HLA-B*15:02 and HLA-B*58:01 which are associated with carbamazepine and allopurinol SJS/TEN For most drugs, however, genetic factors driving risk for SJS/TEN are unknown. In the case of allopurinol although HLA- B*58:01 has close to 100% negative predictive value in Southeast Asia only 50-60% of Europeans and Africans who develop allopurinol SJS/TEN carry HLA-B*58:01. Common causes of SJS/TEN in the US include trimethoprim-sulfamethoxazole, allopurinol and aromatic anticonvulsants such as lamotrigine, phenytoin and carbamazepine where the prevalent genetic associations in US populations have yet to be defined which has stalled preventive efforts and implementation. The rarity of SJS/TEN and lack of access to large cohorts of survivors has impaired the ability to define genetic risk factors and long-term morbidity. We will utilize a registry developed by the SJS Foundation (http://sjsupport.org/) to develop a data and DNA biobank of phenotype adjudicated SJS/TEN survivors. Our testable hypothesis is that we will determine genetic risk factors for the most common drugs associated with SJS/TEN and the nature and risk of long-term complications associated with SJS/TEN both of which will have the potential to have significant impact on SJS/TEN prevention and patient outcomes. In Specific Aim 1 we will establish a large cohort of SJS/TEN survivors with an associated DNA biobank. Participants will be recruited through the SJS Foundation website, Facebook page and registry and consented for medical record review and oral DNA collection. Independent adjudication for drug-induced SJS/TEN will determine eligible participants. In Specific Aim 2 we will define long-term mental and physical health complications associated with SJS/TEN and associated risk factors. Validated questionnaires will assess mental and physical health complications cross- sectionally in patients at different time points following SJS/TEN through instruments validated for psychological distress, post-traumatic stress disorder and health-related quality of life. In Specific Aim 3 we will determine HLA and other genetic risk factors associated with drug-induced SJS/TEN. High resolution HLA, ERAP and KIR typing as well as expanded multi-ethnic genotyping array (MegaEx ) will be performed on 1000 patients who have been verified as having drug-induced SJS/TEN by an independent panel of three dermatologists. Controls will be the BioVu reference population of 100,000 with MegaEx typing, imputed HLA/KIR/ERAP typing as well as BioVu drug tolerant controls matched 2:1 on age, sex and race and underlying disease.