Reactivation of human herpes viruses in various forms of cutaneous adverse drug reactions--a comparative study in Taiwan

介紹： 嚴重皮膚藥物不良反應(Severe cutaneous adverse reactions to drugs, SCARs)是一群因為藥物而引發皮膚過敏反應的罕見疾病，發生率雖然不高，但其嚴重度可能造成病患住院、嚴重殘疾甚至死亡。這一群疾病包括了毒性表皮溶解(Toxic epidermal necrolysis, TEN)、史帝芬－強生症候群(Stevens-Johnson syndrome, SJS)、藥物疹合併嗜伊紅血症及全身症狀(Drug reaction with eosinophilia and systemc symptoms, DRESS)、急性廣泛性發疹性膿疱症(Acute generalized exanthematous pustulosis, AGEP)及廣泛性固定型藥物疹(Generalized fixed drug eruption, GFDE)等全身性藥物疹。在臨床上，這些疾病各有其特殊的皮疹表現，病程與預後也都截然不同。 在過去的研究中顯示：人類疱疹病毒(Human herpesvirus, HHV)活化和藥物疹合併嗜伊紅血症及全身症狀有密切的關聯性，然而此概念並未獲得學術界的一致認同，此現象的原因和對藥物不良反應的影響也仍未明。有些學者認為病毒的活化是全身性藥物疹引發免疫反應後的次發現象，而其他學者則認為病毒的活化在全身性藥物疹的反應中扮演了起始誘發的重要角色。另外，在藥物疹合併嗜伊紅血症及全身症狀的病人血液中，許多種人類疱疹病毒的再活化都曾被偵測到過，包括第六型及第七型人類疱疹病毒、艾伯斯坦-巴爾病毒等等。然而受限於以往研究方法的限制，早期的研究僅單純測量抗體的效價變化、後來以周邊血液單核球做聚合酶連鎖反應(Polymerase chain reaction, PCR)測量病毒去氧核醣核酸的量等方法都有其缺陷，而導致敏感度與特異性上的爭議。此外，對於其他嚴重皮膚藥物不良反應表現型，卻未有大規模研究探討人類疱疹病毒活化和這些藥物疹的關聯性。 方法及程序： 本實驗計劃進行前瞻性研究，從2010年9月至2012年2月收集臨床因為全身性皮膚藥物不良反應而住院的病人，在住院當時及藥物疹的病程中，定期收集病人血液來進行以下檢測: 1.定量聚合酶連鎖反應(Quantitative PCR)來定量血清和周邊血液單核球之病毒量 2.酵素聯結免疫分析法(Enzyme-linked immunosorbent assay)偵測血清中抗體改變 3.偵測在藥物疹的病程中，多種人類細胞激素的量值。 本實驗測試三種病毒，包括第六型人類疱疹病毒、艾伯斯坦-巴爾病毒及巨噬細胞病毒，根據定量聚合酶連鎖反應及酵素聯結免疫分析法來綜合判斷病毒在藥物疹病程中是否有活化的現象；也會針對「病毒活化是否和住院使用之類固醇用量有關」此一問題做探討，並觀察病毒活性、病程與預後的關係。 結果： 本實驗共有32個病患參與，男性與女性各16位，平均年齡為47.8歲。其中診斷為藥物疹合併嗜伊紅血症及全身症狀者有14位、毒性表皮溶解症和史帝芬－強生症候群者有10位、斑丘狀毒性紅疹者有6位，另外廣泛性固定型藥物疹及嚴重型多形性紅斑(erythema multiforme majus, EMM)則各有1位。在藥物疹的病程中，總共有19位病患被觀察到有艾伯斯坦-巴爾病毒活化：包括11位藥物疹合併嗜伊紅血症及全身症狀病患(78.6%)；3位毒性表皮溶解症和史帝芬－強生症候群病患(30%)；4位斑丘狀毒性紅疹病患(66.7%)及惟一一位廣泛性固定型藥物疹病患。另一方面，第六型人類疱疹病毒和巨噬細胞病毒的活化卻只在藥物疹合併嗜伊紅血症及全身症狀的病患中發現，分別有5位(35.7%)及4位(28.6%)病患。在四位觀察到巨噬細胞病毒活化的病患中，全部也都有艾伯斯坦-巴爾病毒共同活化的現象，而其中三位則還有第六型人類疱疹病毒的活化。在我們的觀察中，第六型人類疱疹病毒通常是最早活化的，再來則為艾伯斯坦-巴爾病毒或巨噬細胞病毒的連續或同時活化；而後續的艾伯斯坦-巴爾病毒或巨噬細胞病毒活化則常伴隨著臨床症狀的惡化。 根據住院七天內病患所接受的累積全身性皮質類固醇量，我們將所有的病患分成兩組：Prednisolone平均劑量「低於或等於每公斤每天1毫克」以及「高於每公斤每天1毫克」。針對這兩組分別計算艾伯斯坦-巴爾病毒的活化率，在較低劑量組為66.7% 而在較高劑量組則為57.9%，而兩組的差異並未達到統計上的顯著。 在人類細胞激素的檢測中，急性期時介白質第五因子(Interleukin-5, IL-5)在史帝芬－強生症候群病患血漿中的量明顯低於藥物疹合併嗜伊紅血症及全身症狀和斑丘狀毒性紅疹的病患；而介白質-6(IL-6)、介白質-8(IL-8)、介白質-17(IL-17)及腫瘤壞死因子-α(TNF-α)則是在斑丘狀毒性紅疹的病患血中明顯較低於其他兩組。而在藥物疹合併嗜伊紅血症及全身症狀的病患中，根據「有無多種人類疱疹病毒活化」分組比較的結果，發現在病毒活化組有較高的Interferon gamma-induced protein 10 (IP-10, CXCL10)。 結論與討論： 在各種臨床上不同型態的皮膚不良藥物反應中，都能夠觀察到艾伯斯坦-巴爾病毒的活化，表示此病毒的活化並不具特異性；反觀第六型人類疱疹病毒和巨噬細胞病毒的活化則是在藥物疹合併嗜伊紅血症及全身症狀中所特有的。在各式藥物疹中，病毒的活化通常並非在早期就能被觀察到，意謂著病毒活化對藥物疹而言並非起始的導因。在藥物疹合併嗜伊紅血症及全身症狀的病程中，各種人類疱疹病毒的活化都可能伴隨臨床症狀的波動，但只有第六型人類疱疹病毒的活化被證實對預後有負面的影響，而艾伯斯坦-巴爾病毒或巨噬細胞病毒的活化機轉及對預後的影響則未知；反觀其他藥物疹的病患，艾伯斯坦-巴爾病毒的活化則通常不造成臨床症狀的惡化。雖然本實驗無法闡明在藥物疹病人體內疱疹病毒活化的直接原因，但可以澄清的是，病患所使用的全身性皮質類固醇量並非決定的因素；再者，不同型態的藥物疹在病人體內的細胞激素組合會有所差異，而此早期差異及後續的細胞激素交互作用則可能是造成病毒活化的作用因子之一。Introduction: Severe cutaneous adverse drug reactions (SCARs) are groups of drug hypersensitivity reaction with a heterogeneous clinical presentation. Although the prevalence of SCARs is low, they may result in prolonged hospitalization, significant disability and even death. SCARs include toxic epidermal necrolysis (TEN), Stevens-Johnson Syndrome (SJS), drug reaction with eosinophilia and systemic symptoms (DRESS), acute generalized exanthematous pustulosis (AGEP) and generalized fixed drug eruption (GFDE). All these specific SCARs have different presentations, clinical courses and prognosis. In the past studies, the association of human herpesvirus (HHV) reactivation with DRESS has been reported. Ｈowever, their roles in DRESS are debated and their influences are still elusive. Some authors consider that viral reactivation is a secondary event after an initial immunological reaction directed against the culprit drugs; however, others consider that DRESS is the consequence of viral reactivation from its very beginning. Besides, many types of HHV have been detected to be involved in DRESS. However, previous studies were limited by the use of serologic changes during the disease course or by using PCR method to detect the viral DNAs within PBMC. The sensitivity and specificity of these tests to confirm the roles of HHV were still under debate. Moreover, there is no previous large scale study to investigate the association of HHV reactivation and the development of other SCARs. Materials and Methods: From September 2010 to February 2012, we prospectively included the hospitalized patients being suspected to have cutaneous adverse drug reactions (cADR), including SCARs or generalized maculopapular eruption (MPE). We collected serial blood samples for patients upon admission and during hospitalization course, for: 1. Real time PCR to quantify the viral DNA from serum samples and PBMC. 2. Enzyme-linked immunosorbent assay (ELISA) were carried out to determine the changes of antibody titers in serum samples. 3. Various kinds of cytokine levels were also evaluated during different disease stages. There kinds of HHV were examined, including HHV type 6 (HHV-6), Epstein-Barr virus (EBV) and cytomegalovirus (CMV). According to the results of quantitative-PCR and antibody ELISA, individual viral reactivation rate was evaluated; beside, the correlation between dosage of administrated corticosteroid and viral reactivation is also evaluated. We observe the correlation between viral activities, clinical courses and patient prognosis. Results: Totally, 32 patients were included in this study, 16 male and 16 female, and the average age is 47.8-year-old. 14 patients were DRESS, 10 were SJS/TEN, 6 were MPE, and the other two were GFDE and erythema multiforme majus (EMM) respectively. EBV reactivation was detected in 19 patients, including 11 DRESS (78.6%), 3 SJS/TEN (30%), 4 MPE (66.7%) and 1 GFDE patient. Reactivations of HHV-6 and CMV were observed in 5 (35.7%) and 4 (28.6%) DRESS patients respectively. All of those four patients with CMV reactivation also had EBV reactivations and three of them had HHV-6 reactivation. Mostly, HHV-6 reactivated first, and then followed by sequential or simultaneous CMV and EBV reactivations, as well as fluctuating clinical courses. According to the accumulative dose of systemic corticosteroid within 7 days after hospitalization, we divided all patients into two groups: average dose less or equal to 1mg prednisolone/kg/day and average dose more than 1mg prednisolone/kg/day. For the lower and higher dosage groups, EBV reactivation rate were 66.7% and 57.9%, respectively. The p-value of t-test for these two groups was 0.6 and the difference between these two groups was not significant. For the cytokine study, interleukin-5 (IL-5) was significantly lower in SJS/TEN patients during the acute stage. Besides, for MPE patients, IL-6, IL-8, IL-17 and tumor necrosis factor-alpha (TNF-α) were significantly lower than DRESS and SJS/TEN patients. For DRESS patients, they could be devided into two groups according to whether multiple human herpesvirus actived, and in the viral reaction group, interferon gamma-induced protein 10 (IP-10, CXCL10) was significantly higher than another group. Conclusion and discussion: EBV reactivation, which was detected in various forms of cADRs, is not specific to DRESS, while HHV-6 and CMV reactivations seem to be more specific to DRESS. In this observational study, most of viral reactivation was not detected in the early stage of cADRs which may indicate that viral reactivation was not the initiator of these drug eruptions. For DRESS patients, reactivation of various HHV may correlate with fluctuation clinical courses, but only HHV-6 was documented to have negative impacts on patients’ prognosis. On the other hand, EBV reactivation did not associate with clinical deterioration in SJS/TEN or MPE patients. Although our study could not answer what is the direct cause of viral reactivation in SCAR patients, the dose of administrated systemic corticosteroid was not a major determinant. Different cytokine patterns during acute stages of various cADRs and following cytokine cascades may be one of mediators for vrial reactivation pathomechanisms