Roujeau Jean-Claude
UniversitéParis XII, Department of Dermatology and Inserm U 659, Hôpital Henri Mondor, Créteil, France.
Allergol Int. 2006 Mar;55(1):27-33. doi: 10.2332/allergolint.55.27.
Clinicians had suspected for years that drug eruptions were probably mediated by immune mechanisms because their timing suggested sensitization and specific immunologic memory rather than direct toxicity. An immune response to medications was also demonstrated by positive skin tests and by several types of in vitro tests that evidenced immediate or delayed hypersensitivity. In the last decade several teams of researchers obtained in vitro drug-specific human T-cell clones, in a variety of clinical types of drug eruptions. These clones were produced from blood or skin mononuclear cells of patients with a history of drug reaction by stimulation in vitro with drug. They were either of CD4 or CD8 phenotypes. Drug specific clones were stimulated by the parent drug much more often than by reactive metabolites. That challenged the classical "hapten hypothesis" that the immune response was initiated by reactive metabolites combined to self proteins. The medication usually stimulated specific T-cells after non-covalent binding to major histocompatibility (MHC) molecules on antigen presenting cells. In toxic epidermal necrolysis, T-lymphocytes present at the site of lesions, exhibited a drug specific cytotoxicity against autologous target cells, or allogeneic cells that shared the same HLA than autologous cells. This MHC class I restriction and mediation of death by perforin/granzyme release, is the classical behavior of cytotoxic T lymphocytes, like those operating in the reject of a transplanted organ. MHC restriction could explain the key role of HLA genes as predisposing factors to severe drug reactions. A strong association between HLA and hypersensitivity to abacavir, SJS or TEN to carbamazepine or allopurinol has been recently demonstrated. Resemblance to graft rejection points to the possible therapeution value of immuno suppressive agents. Most drug eruptions appear to result from T-cell mediated delayed hypersensitivity. The secondary activation of different cascades of cytokines, may contribute to the heterogeneity of clinical presentations.
多年来,临床医生一直怀疑药物疹可能是由免疫机制介导的,因为其发作时间提示存在致敏和特异性免疫记忆,而非直接毒性。皮肤试验阳性以及几种体外试验证明了对药物的免疫反应,这些试验证实了速发型或迟发型超敏反应。在过去十年中,几个研究团队在多种临床类型的药物疹中获得了体外药物特异性人T细胞克隆。这些克隆是通过用药物体外刺激,从有药物反应史患者的血液或皮肤单核细胞中产生的。它们要么是CD4表型,要么是CD8表型。药物特异性克隆被母体药物刺激的频率比被反应性代谢产物刺激的频率高得多。这对经典的“半抗原假说”提出了挑战,该假说认为免疫反应是由与自身蛋白质结合的反应性代谢产物引发的。药物通常在与抗原呈递细胞上的主要组织相容性(MHC)分子非共价结合后刺激特异性T细胞。在中毒性表皮坏死松解症中,病变部位的T淋巴细胞对自体靶细胞或与自体细胞具有相同HLA的同种异体细胞表现出药物特异性细胞毒性。这种MHC I类限制以及通过穿孔素/颗粒酶释放介导的细胞死亡,是细胞毒性T淋巴细胞的经典行为,就像在移植器官排斥反应中起作用的那些细胞一样。MHC限制可以解释HLA基因作为严重药物反应易感因素的关键作用。最近已证明HLA与对阿巴卡韦过敏、对卡马西平或别嘌醇发生史蒂文斯-约翰逊综合征或中毒性表皮坏死松解症之间存在密切关联。与移植排斥反应的相似性表明免疫抑制剂可能具有治疗价值。大多数药物疹似乎是由T细胞介导的迟发型超敏反应引起的。不同细胞因子级联的二次激活可能导致临床表现的异质性。