【内分泌代谢论坛】亚洲人群的甲状腺眼病状况及其诊治进展
文章来源：中华内分泌代谢杂志2020,36(7):541-562
作者：Jaru-ampornpan Pimkwan 程瑜 焦秦 Douglas Raymond S.
摘要
本文旨在回顾当前对甲状腺眼病(TED)的认识和研究进展，侧重于亚洲人群的相关资料。TED可以显著影响患者的生活质量，并且带来诸如容貌损毁和失明等严重后果。近期研究发现，替妥木单抗(teprotumumab)，一种可抑制胰岛素样生长因子1受体(IGF-1R)的人源单克隆抗体，已成为TED的新兴靶向治疗方案。该药可逆转活动性中重度TED的疾病进程。随机临床试验表明，与安慰剂相比，替妥木单抗可显著改善患者的临床活动性评分、眼球突出度、主观性复视和随访4个月后的生活质量。本文将综述TED的流行病学、病理生理学及其诊治方面的最新进展。
一、前言
甲状腺眼病(TED)，也被称为格雷夫斯眼病、格雷夫斯眼眶病和甲状腺相关眼病，是由自身免疫导致的可威胁视力的眼部疾病。TED与自身免疫性甲状腺疾病密切相关，多数患者在TED病程中同时存在甲状腺功能亢进[1,2,3]。TED通常在诊断Graves病(GD)后的18个月左右发生[4]，但也有二者同时发生、在GD诊断前发生或在GD诊断后很久才发生的情况。TED的临床表现包括眼球突出、限制性斜视、眶周水肿、眼睑退缩、暴露性角膜炎和结膜充血等。视力丧失可见于重度TED，通常继发于压迫性视神经病变或者角膜失代偿[5]。临床上TED的诊断一般基于眼睑退缩、眼球突出或眼眶病变，往往还需同时具有GD的实验室证据或其他临床表现[6]。如果患者无眼睑退缩，则需具备GD的实验室证据和以下临床表现之一：眼球突出、视神经功能障碍、眼外肌受累[6]。TED的发病机制为透明质酸沉积和脂肪的从头合成，造成眶组织扩张、肌肉肥大和多种免疫细胞浸润所致的眼眶炎症。眼眶成纤维细胞被认为是TED发病的主要效应细胞。TED的已知危险因素包括年龄、性别(女性)、遗传因素、甲状腺功能异常、吸烟和妊娠。TED的管理策略取决于病情的严重程度和病程时机，一般包括药物治疗和手术治疗。遗憾的是，TED尚无法治愈。现有的治疗方式，包括高剂量糖皮质激素冲击和眼眶放疗，都仅能减轻疾病活动期的炎症反应，不能纠正疾病的远期后遗症。康复手术包括眶减压术和眼睑整形术，是治疗炎症消退后软组织纤维化所致容貌损害的有效手段。但是，随着对参与TED发生发展的分子通路的认识不断加深，替妥木单抗可望成为治疗TED的新方法。这种新型靶向药物是胰岛素样生长因子1受体(IGF-1R)的抑制剂，能够有效改善活动期、中重度TED患者的突眼和眼眶炎症情况，很有希望在未来替代手术治疗。
由于解剖结构特点、遗传易感性和生活方式不同，亚洲人与其他种族人群的疾病模式不尽相同。就TED而言，尽管亚洲人TED的发病率较高，但多数患者疾病程度较轻。吸烟是TED发病和进展的高危因素，而亚洲男性的吸烟率较高[7,8]，因此，该因素可能会影响亚洲人群的TED模式。亚洲的TED研究多在某些国家/地区(包括中国、韩国、日本、印度、马来西亚和新加坡)开展，覆盖多个种群亚群。本文旨在以亚洲人群为重点，简要概述TED相关研究进展，并介绍TED的最新管理策略。
二、TED的流行病学
目前尚无普通人群中TED的确切发病率和患病率数据。在不同种族人群中，TED的流行病学特征不同。美国白人中，TED患病率为0.25%[9]，发病率为18.9·100 000人－1·年－1(女性16，男性2.9)[1]。在甲状腺功能异常者中，20%～50%有不同程度的TED，患病率明显高于普通人群[9,10]。值得注意的是，亚临床TED，如影像学上表现为眼外肌受累，可见于67%～90%的GD患者[11,12]。尽管TED多与GD有关(90%)，但也可见于甲状腺功能正常(6%)、桥本甲状腺炎(3%)和原发性甲状腺功能减退(1%)者[1,3,9,13]。性别方面，北美和欧洲地区女性患者显著多于男性(患者比例女性∶男性＝5∶1)[1,2]。TED可发生于任何年龄，但年龄分布成双峰曲线，40～50岁和60～70岁的患者比例最高[1]。
亚洲人群中TED患病率的流行病学数据更加有限，相关研究多开展于已经确诊为GD的患者，而非普通人群。Weetman[14]曾认为高加索人和亚洲人GD的患病率相近。然而，已有研究显示亚洲GD患者中的TED患病率与高加索人不同。Tellez等[15]报道亚洲人TED的患病率远远低于高加索人(7.7%对42%)。该研究是基于小样本印度人与来自英国的高加索人的对比；研究中，对高加索人和亚洲人均使用Werner标准(突眼度≥20 mm)作为突眼的诊断标准，但由于2个种族在面部解剖上存在差异，并且亚洲人平均突眼度较低、突眼容易被忽视，因此，该研究中得到的亚洲人TED患病率可能低于真实情况。其他一些研究报道亚洲人TED的患病率为17.3%～74.2%，与高加索人相似(表1)。需要指出：不同研究中采用的TED诊断标准、研究方法学和样本量并不一致，这些因素可能会影响其TED患病率结果。较早的研究一般采用Werner诊断标准，而后期的研究则更多采用改良标准或新标准来诊断TED。此外，一些研究表明吸烟率不同也能造成不同种族的TED患病率差异[12]。
亚洲人群中，TED也是多见于女性，但女性和男性的比例低于高加索人(表1)。
三、TED的病理生理机制和自然史
根据经典的Rundle曲线，TED常以波动的炎性症状为首发临床表现，起病后6个月到2年间症状达到高峰，即活动期[22]。活动期的特点是炎症和眼眶组织被1型辅助性T细胞(Th1)、B淋巴细胞、肥大细胞和巨噬细胞浸润。眶成纤维细胞(OFs)在与以上细胞相互作用后被激活，导致间质性水肿和细胞因子及趋化因子的大量产生，维持炎症级联反应[23,24]。这些激活的OFs亚群可以进一步分化为肌成纤维细胞和脂肪细胞，分别导致了纤维化和眶脂肪组织增生[25,26,27,28,29]。眼外肌透明质酸(一种亲水性氨基多糖)的合成和沉积增加，以及眼眶脂肪增加，都促进了眼眶组织扩张[30,31]。在1～2年后，随着炎症消退，进入疾病的非活动期或稳定期，以组织纤维化为主。当前的药物干预只能针对疾病活动期进行，因此一般用于TED发病后的1～2年间[22]。
TED的病理生理机制尚不明确。有假说提出，该病与机体丧失对促甲状腺激素受体(TSH-R)的免疫耐受，从而将这些受体错误识别为异体有关。TSH-R抗体，或称甲状腺刺激性免疫球蛋白(TSIs)的产生激活了TSH-R相关的炎症级联反应。TSI水平通常与TED的活动性和严重程度有关，但并非所有患者均能检测到该抗体的存在[14,32,33,34,35,36]。因此，对于可疑病例，TSI阳性有助于确诊TED，但阴性不能完全排除TED。
基于眶组织和甲状腺共有某些'甲状腺特异性'蛋白，作为自身抗原引起免疫交叉反应这一假说，一些研究致力于探讨TED的其他发病机制。GD患者中可检出抗IGF-1R的免疫球蛋白[37]，该免疫球蛋白可抑制TSH-R和IGF-1R信号通路[38]。IGF-1R是一种跨膜酪氨酸激酶受体，在发育、代谢和免疫中起至关重要的作用[39]。有研究表明TED患者的OFs和T细胞表面过表达TSH-R和IGF-1R[40,41]。此外，在TED患者的眶成纤维细胞中，TSH-R和IGF-1R可形成一种结构和功能上的信号复合体，可能在TED发病中起关键作用[42]。最近的研究支持针对TSH-R和IGF-1R的免疫识别均会刺激眶成纤维细胞，促进促炎细胞因子和透明质酸的合成[43]，而后者亦是TED的重要标志(图1)。
注：TSH-R抗体(Ab)激活TSH-R，从而引起相关炎症级联反应。自身反应性T细胞向眼眶组织迁移，与眶成纤维细胞(OFs)相互作用，使以上2种细胞均被激活。活化的OFs继而分泌各种炎性细胞因子，维持了炎症的恶性循环。高表达于这些细胞表面的TSH-R，可与TSH-R抗体相互作用并激活细胞。IGF-1R可与OFs表面的TSH-R协同作用，形成一种能与TSH-R抗体相互作用的结构和功能复合体。被激活的OFs可以分化为脂肪细胞或肌成纤维细胞，促进透明质酸的合成。总而言之，炎症级联反应和OFs的分化导致了TED中的眼眶软组织扩张和充血
图1 甲状腺眼病的发病机制
四、TED的临床特点和分类
TED的炎症级联反应导致眼眶脂肪和肌肉的扩张，因受限于眶部骨性结构，故可进一步引起眼眶充血和眶内压增高。眶部脂肪扩张导致突眼、眼睑挛缩和角膜暴露等临床表现。眼肌肥大导致限制性斜视、复视、可能还有视神经周围物理性拥挤，进而出现视敏度下降、色觉异常、视野缺损以及相对性瞳孔传入障碍。这也被称作甲状腺疾病相关的视神经病变(DON)，见于3%～7%的TED患者[5]。临床实践中，一般通过病史和临床表现来诊断TED。必要时需结合甲状腺功能异常的实验室证据(血清TSH、TSI或TSH-R抗体水平)。放射影像学可提供眼外肌肥大的证据并排除突眼的其他病因(如眼眶肿物等)，故也有助于疑难病例的确诊。
亚洲人和高加索人的眼眶解剖结构特点不同，因此在TED的临床表现上也有差别。研究表明亚洲人的平均突眼度低于高加索人[17,18,19,44]。研究测得的亚洲人突眼度正常上限范围在14～19 mm[17,19,44,45,46,47,48]，而高加索人突眼度正常上限为20～21 mm[49]。因此，亚洲人TED的临床诊断标准应不同于高加索人，但目前尚无一致认可的诊断阈值，这可能是由于亚洲不同种族亚群的诊断标准也存在差异[18]。此外，与高加索人相比，亚洲人的眼窝较浅、眶尖较窄，因此更容易出现眶尖结构受压和视神经病变[50]。在亚洲TED患者中，最常见的软组织受累表现为眼球突出和眼睑退缩(图2)[18,44]，眶周水肿和突眼的程度往往低于高加索人，更常出现下眼睑退缩(发生率约20%，有建议将此表现作为亚洲人TED的诊断标准之一)[18,45]。约3.4%～8.5%的亚洲TED患者有获得性眼睑赘皮[47]。这是由于亚洲人眶隔较紧，下眼睑皱襞多，联同TED造成的突眼和板层后部回缩，最终导致板层前部覆盖眼睑边缘、睫毛与角膜接触[50]。高加索人TED最常累及的眼外肌依次为下直肌、内直肌、上直肌和外直肌，而亚洲人更常见单组肌肉受累和垂直肌群受累(上直肌和下直肌)[50]。TED对眼表的影响包括因突眼和眼睑退缩所致的角膜病变和眼表炎[51,52]。干眼症是引起TED眼部不适的最常见原因，其在无TED亚洲人中的患病率高于无TED的高加索人[53,54]，因此，很可能TED相关的眼表疾病在亚洲人中的患病率也高于高加索人。
注：常见临床表现包括突眼和眼睑退缩
图2 亚洲人甲状腺眼病临床表现的高度异质性
对于甲状腺眼病的严重程度和活动性，目前已有一些分类标准。最常用的疾病严重程度分类标准为欧洲Graves眼眶病专家组[European Group on Graves′ Orbitopathy(EUGOGO)]分类标准[55]，即根据主观症状和客观表现将疾病分为轻度GO(60%)、中重度GO(30%)和威胁视力的GO(4%～8.6%)[56](图3)。该标准的优势是可根据其分类结果指导治疗方案。但是，EUGOGO分类也存在一些不足。一方面，多种临床表现混杂于同一类别中，如突眼、眼球运动异常和软组织受累。另一方面，对于严重程度的定义没有考虑疾病给患者带来的感受。因此，TED患者的生活质量问卷调查结果与其EUGOGO严重程度分级的符合度不高，表明客观的临床评估结果与患者自身角度感受到的疾病影响之间存在差异[57]。
图3 欧洲Graves眼病专家组(EUGOGO)对甲状腺眼病的分级
从疾病的活动度出发，TED可分为活动期和非活动期(或稳定期)。常用的疾病活动度评估方法为临床活动度评分(CAS)(表2)。CAS重点关注患者初次就诊时的7种眼部炎症症状，以二分类(是/否)方式评价，有助于判别出对免疫抑制治疗可能有良好反应的患者。CAS评分≥3为疾病活动期。该评分仅是炎症反应的替代指标，CAS评分≥4对糖皮质激素治疗有效的阳性预测值为80%[58]。CAS评分的不足之处在于以二分类方式同等对待各种参数，不能预测疾病的进展。此外，若患者有眼部慢性充血性疾病，其CAS评分即使在稳定期也可能升高；在脂肪病变为主的TED以及部分甲状腺疾病相关视神经病变病例中，CAS评分可能会假性低值，而这种病例在亚洲人中较为常见。尽管日常临床实践中多数医生不常使用CAS评估系统，但由于对TED相关炎症的治疗反应尚无更好的评价指标，CAS在临床试验领域的重要性仍不可替代。
VISA评分兼顾TED的临床活动度和严重程度，在北美更为常用[59,60]。其评估参数包括主观和客观视力评估、炎症、斜视和TED相关的眼部外观改变(表3)。但其不足在于未明确轻、中、重度外观改变的定义，不同医生的界定可能存在差异。并且，VISA评分系统的数据收集较为繁杂，在临床中推广使用有一定困难。
五、TED的危险因素
TED的危险因素分为不可控(内源性)和可控(外源性)两类。不可控因素包括年龄、种族、性别和基因。可控因素包括吸烟、甲状腺疾病状态和放射性碘(RAI)治疗。
对于不可控(内源性)因素的研究表明，高加索人和亚洲人的TED严重程度均与增龄有关[45,61,62]。尽管GD和TED均为女性多见，使得女性性别成为TED的危险因素[21]，但较为严重的TED却多出现于男性中，高加索人和亚洲人中均可见到这一现象[45,61]。
在遗传方面，已有研究发现了一些与GD相关的基因位点，但与TED发病风险的关系尚不明确。这些基因包括人类白细胞抗原(HLA)、细胞毒性T淋巴细胞抗原-4(CTLA-4)、TSH-R基因，以及报道较少的肿瘤坏死因子-α(TNF-α)、细胞间黏附分子1(ICAM-1)和干扰素-γ(IFN-γ)基因[63]。已有研究调查了日本人、韩国人和中国人的GD遗传易感基因。高加索人中最主要的易感性HLA等位基因DRB1*03：01在亚洲人中出现频率较低[50]，提示亚洲人GD和TED的易感性HLA等位基因另有其主。对CTLA-4基因的研究包括不同位点的基因多态性，例如，在CTLA-4基因A49G多态性与GD关联性研究中，不同人群中的研究结果不一致[64]。一项近期发表的meta分析显示，与亚洲人相比，该基因多态性与TED易感性的关系在欧洲人中更为明显[65]。众所周知，TSH-R基因多态性与GD相关，因此已有研究探讨其与TED的联系，多数研究于GD合并TED的患者中开展，但尚未得到最终定论[66,67]。
吸烟是TED发病和进展的重要可控危险因素。我们需要意识到不同种族人群的吸烟率不同，这可能会影响无眼病患者进展为TED的比例和TED的患病率。例如，在Tellez等[15]的报道中，其研究人群中欧洲人和亚洲人的吸烟率分别为61.2%和23%，欧洲人患TED的总风险比亚洲人高6.4倍，这种患病风险的差异既可能源于种族差异，也可能与吸烟率不同有关。在高加索人中，经常吸烟者患TED的风险比非吸烟者高3～4倍，其疾病严重程度和对治疗无反应的概率也更高[7,10,68,69]。有研究表明吸烟与TED有剂量效应关系，每日吸烟数量越大，TED的相对风险越高[70]。亚洲人群中的研究也得出类似结果。在北印度地区，主动吸烟与TED风险增高至3.9倍有关[20]，但吸烟者和非吸烟者的GO严重程度没有差异。有趣的是，由于韩国人和日本人中男性吸烟者多于女性，对其进行的研究显示性别影响了吸烟对TED的作用。韩国研究中，仅在女性中发现了吸烟与TED显著相关[21]，但是，该研究中男性与女性的吸烟率存在巨大差异(男性吸烟率为60%，而女性吸烟者仅有寥寥数例)。日本研究中，发现吸烟仅与女性GD风险增加有关[71]，但未发现吸烟与TED的关联[64]。
另一个重要的可控危险因素是甲状腺功能状态。甲状腺功能控制差与较为严重的TED有关[72]。此外，RAI治疗可导致15%～35%患者发生TED或TED病情恶化，活动期中重度TED患者和治疗后甲减的患者更易发生[73,74]。一项基于韩国人群的研究表明RAI治疗与TED发生有关，进一步支持了上述观点[21]。推荐在RAI治疗时同时口服低剂量泼尼松(0.2～0.5 mg·d－1)以预防TED进展[75]。
六、TED的管理
鉴于TED的异质性，其管理应根据患者症状、疾病病程(活动期或稳定期)和疾病严重程度而采用个体化策略(表4)。很久以来，缺乏有效逆转或预防TED所致眶组织重塑的治疗方案。本综述主要侧重于对TED的通用治疗策略，而非就TED单个症状(如眼睑退缩、突眼或眼表暴露)的特异性治疗展开介绍。
对活动期TED，一线治疗方案是保守治疗，即球后注射、口服或静脉用糖皮质激素，以暂时缓解患者的眼部炎症和水肿。研究表明，静脉大剂量糖皮质激素冲击治疗的有效性和安全性均优于口服给药，是目前优先推荐的治疗方案[76]。但是，糖皮质激素仅对50%～80%的患者起效，并且一些研究发现激素治疗在突眼和斜视纠正方面的长期疗效并不理想[42]。激素治疗后TED复发或进展并不少见。另外，糖皮质激素治疗的不良反应也不容忽视。随着激素剂量的增加，糖尿病、精神失常和抑郁等不良事件的发生率也相应增加。研究报道在静注甲基强的松龙的累积剂量超过8 g时，可出现威胁生命的不良事件，如急性肝衰竭和死亡[77,78,79,80]。总而言之，糖皮质激素治疗是一种可有效暂时缓解组织水肿、改善症状和DON病情的治疗措施，但临床医生在用药时需谨慎衡量其风险和益处。
20世纪40年代第一次提出了对TED患者进行放射治疗(RT)，采用的方案为累积剂量20 Gy/眼，于10 d内分10次照射[81]。研究表明这是较为安全的治疗方式，不增加白内障的风险[82]，仅有1例射线诱发肿瘤的病例报道[83]，轻度视网膜病变的发生率仅为1%[82]。在4项对比RT和伪治疗的随机对照试验中[84,85,86,87]，3项研究结果支持RT在改善软组织炎症和运动功能方面有益[84,85,86]。然而，RT不能有效改善突眼和眼睑退缩。Kazim等[88]发现RT可预防DON发生或延迟DON组患者的眶减压手术。目前正在进行的一项多中心随机对照试验(CRISEPTED研究，临床试验注册号NCT02339142)对比了RT联合静脉注射糖皮质激素与单用糖皮质激素治疗早期TED的疗效。总之，RT不是活动性TED的一线治疗，而是静脉注射糖皮质激素的辅助治疗，其有效性有待进一步研究。
处于活动期威胁视力(DON和角膜破裂)的TED患者，应考虑手术治疗。DON最有效和直接的治疗方式是眶内侧壁减压术。解除眶尖后内侧骨性压迫有助于快速改善视力。为最大程度减轻DON患者的眶内壁压迫，应切除眶内下壁骨性连接的后2/3以及部分腭骨[89]。保留眶内下壁骨性连接的前部可减少术后复视的风险[90]。为改善眶后充血，手术可额外进行眶底和(或)眶外侧壁的减压[89]。
七、TED靶向治疗的最新进展
理论上，理想的TED治疗方案应通过改善炎症状态和减少眼部并发症风险，改变疾病的病程；同时，治疗的不良反应和容貌损毁风险应非常小，并且可以改善患者的生活质量、恢复眼功能，避免手术治疗。因此，根据Rundle曲线(图4)，该治疗应起始于疾病的活动期，以阻断疾病进程并避免远期并发症的发生。遗憾的是，现有的治疗措施既不能逆转眶组织的长期改变，也不能改善突眼和斜视，并且其众多的不良反应也不容忽视。随着分子生物学的进展，靶向针对疾病分子机制的免疫调节治疗已经出现。现已开展了多项利妥昔单抗、托珠单抗和替妥木单抗治疗TED的临床试验，部分结果令人惊喜。
注：在活动期，炎症会导致波动性的水肿、红斑、充血和突眼；1～2年后，疾病进入稳定期，此时炎症消退，但眶组织纤维化和眶内容物扩张持续存在(横轴为时间，纵轴为疾病严重程度)
图4 TED的自然史(Rundle曲线)
1．利妥昔单抗(RTX)：
RTX是一种人鼠嵌合性单克隆抗体，可与前体B或成熟B淋巴细胞表面的CD20分子特异性结合。由于RTX可有效清除产生自身抗体的浆细胞的前体——B细胞，其在治疗TED等自身抗体介导的疾病方面显示出美好前景。一些前瞻性和回顾性研究探讨了RTX治疗中重度和威胁视力的TED的效果，但其结论并不一致[91,92,93,94,95,96,97,98,99,100,101]。例如，Salvi等[91]开展的一项小型RCT研究比较了RTX(500或2 000 mg)与常规一线治疗(静脉注射甲泼尼龙，IVMP)对活动性中重度TED的治疗效果。结果显示2种方案均可显著改善CAS评分，但第24周时RTX组对治疗有反应的受试者(100%)多于IVMP组(69%)；IVMP组出现了疾病的复发，但RTX未出现复发。另一项RCT研究比较了利妥昔单抗(每次1 000 mg，间隔2周)与生理盐水的疗效，结果表明利妥昔单抗并不能明显改善CAS评分[92]。但该研究中RTX和对照组在试验开始前均接受了糖皮质激素治疗，因而对照组的疾病改善可能源自先前激素治疗的影响。最近一项关于RTX治疗TED的meta分析表明，RTX治疗中重度TED的疗效优于糖皮质激素和生理盐水，但其对TED的主要严重后果(突眼和复视)没有显著改善[102]。
2．托珠单抗：
白细胞介素6(IL-6)是一种促炎因子，作为支持B细胞功能和部分辅助T细胞发育的因子，其与炎性反应和宿主防御机制密切相关。IL-6在其他自身免疫疾病和TED的进展中起着重要作用。此外，研究发现TED患者的血清IL-6水平升高[103]。托珠单抗(TCZ)是一种重组人单克隆抗体，因其对IL-6受体的拮抗作用，有学者提出该药可通过直接抑制炎症反应而有效治疗TED。2015年12月完成的一项RCT研究(注册号NCT01297699)在西班牙的10个医疗中心开展了试验，以探究托珠单抗治疗活动性TED的有效性和安全性。在该研究中，32例激素治疗无效的活动期TED患者被随机分配为治疗组和安慰剂组，分别于第0、4、8、12周注射托珠单抗和安慰剂，观察周期为40周。第16周时，TCZ组CAS评分降低≥2分的患者比例为93.3%，而安慰剂组相应比例为58.5%(P＝0.04)；治疗组患者突眼度较基线水平下降，但安慰剂组突眼无改善[－1.5(－2.0～0.5)mm对0.0(－0.1～0.5)] mm；P＝0.1]。在副作用方面，研究报道TCZ治疗组的部分患者出现了不良反应，包括第8周时1例转氨酶升高、第32周时另一例患急性肾盂肾炎[104]。TCZ未能改善复视和斜视。总体而言，TCZ有望成为西班牙TED患者的有效治疗药物，但其疗效仍需要更多RCT研究证实。
3．替妥木单抗：
尽管TED的临床表现异质性较大，但其背后的分子机制却有共同之处。随着对自身抗原——TSH-R和IGF-1R在TED发病机制中重要作用的认识(图5)，针对TED的特异性治疗也有了进展。与其他靶向调节免疫系统的药物不同，替妥木单抗的开发实现了从基础向临床的跨越，成为一种特异性调节TED分子通路的药物。替妥木单抗是一种人源单克隆抗体，能以高度亲和性和特异性与胞外的IGF-1R结构域特异性结合[42,105]。该药物与IGF-1R结合后会诱导抗体-受体复合物的内化，从而诱发其降解。由于IGF-1R与TSH-R在结构上和功能上均密切相关，替妥木单抗也能抑制TSH-R信号通路并阻断自身抗体攻击眶成纤维细胞[105,106]。因此，该药物减少了透明质酸的产生和细胞因子的刺激(图6)。
注：TSH-R：促甲状腺激素受体；IGF-1R：胰岛素样生长因子1受体；TSI：甲状腺刺激性免疫球蛋白；GD-IgG：GD免疫球蛋白
图5 TSH-R和IGF-1R协同作用，在TED发病过程中起着关键作用
注：TSH-R：促甲状腺激素受体；IGF-1R：胰岛素样生长因子1受体；TSI：甲状腺刺激性免疫球蛋白；GD-IgG：GD免疫球蛋白
图6 替妥木单抗是一种IGF-1R拮抗剂，可抑制眶成纤维细胞的TSH-R信号通路
一项近期完成的双盲、安慰剂对照Ⅱ期临床试验评估了替妥木单抗的有效性与安全性[106]。该研究纳入了88例研究对象，均为近期发病、处于活动期的中重度TED患者(距TED症状出现小于9个月，CAS≥4)。以1∶1的比例将研究对象随机分配至替妥木单抗治疗组和安慰剂组。治疗方案为每3周1次静脉给药，共治疗24周(共8次给药)。主要终点为第24周时研究侧眼的CAS评分较基线评分下降≥2且突眼减少≥2 mm，同时非研究侧眼病情无加重。分别于第6、12、18和24周进行随访，评估患者的CAS评分和突眼情况。研究结果非常明显，替妥木单抗组患者在治疗第6周(2次治疗)后已经出现CAS评分以及突眼的改善[106]，该组所有患者都出现不同程度的TED改善，受益最大的是疾病较为严重的患者[107]。比如，在24周随访时，治疗组中40%(17/42)的患者突眼度减少≥4 mm，而安慰剂组改善率为0%(0/45)[108]。替妥木单抗带来的突眼度降低至少可持续至用药后1年。迄今为止，没有其他药物可达到与该药相似的突眼缓解效果，仅有手术治疗可与该药的疗效相提并论。性别和吸烟因素可能对替妥木单抗的疗效无影响[107]。另外值得高兴的是，接受替妥木单抗治疗的患者其生活质量和主观性复视都有所改善[106]。
总体而言，替妥木单抗的药物相关不良反应并不严重且可控。一些患者用药后出现了高血糖，在糖尿病患者中会更加明显，但可以通过药物治疗有效控制[106]。其他替妥木单抗相关的不良反应多数程度较轻，可以自行缓解。总之，替妥木单抗对于活动性中重度TED患者有益，可降低CAS评分、改善突眼和复视，提升患者生活质量的同时仅有轻微的不良反应。2016年8月，替妥木单抗因疗效显著而被美国食品和药物管理局评为'突破性疗法'。
最近一项替妥木单抗的Ⅲ期临床试验(注册号NCT03298867)结果与其Ⅱ期试验结果类似，进一步证实了其有效性和安全性[109]。研究中2组的失访率均较低(<5%)。研究报告了3例不良反应的发生：安慰剂组1例(视野缺损，因需行眶减压术而退出试验)，替妥木单抗组2例(1例气胸，但似与用药无关；1例输液反应，导致治疗中断)。由于其特有的作用机制，不同种族患者的用药剂量可能无须调整。
八、结论
TED是一种异质性自身免疫性疾病，在各种族人群中均可导致严重的视力减退和容貌损毁。遗传和环境因素可能会对疾病的临床表现和严重程度造成一定影响。由于亚洲人与高加索人的眼部解剖结构特点不同，尽管多数亚洲TED患者病情较轻，但一些患者可出现程度较重的DON。根据现有对疾病自然史的了解，在TED活动期有相对较窄的时间窗可通过干预改变病程以减少发病率。几十年来，TED的治疗效果难以令人满意，通过对TED特异性药物的不懈探索，研究者成功研发了靶向调节TSH-R/IGF-1R信号通路的新药物。替妥木单抗是一种针对IGF-1R的抑制性抗体，在活动期中重度TED患者的Ⅱ期和Ⅲ期RCT研究中都表现出显著疗效。该药物可逆转突眼和复视等临床表现，开辟了TED治疗的崭新道路。
1．Introduction
Thyroid eye disease(TED), also known as Graves′ ophthalmopathy, Graves′ orbitopathy, and thyroid-associated ophthalmopathy, results from an autoimmune process that can be vision-threatening. TED is strongly associated with autoimmune systemic thyroid disorders and most patients have hyperthyroidism at some point during the course of the eye disease[1,2,3]. On average, TED occurs approximately 18 months following a diagnosis of Graves′ disease(GD)[4], but can occur simultaneously, prior to, or long after the diagnosis of GD as well. Clinical findings of TED include proptosis, restrictive strabismus, periorbital swelling, eyelid retraction, exposure keratopathy, and conjunctival injection. Vision loss can be found in the severe form of TED, secondary to compressive optic neuropathy or corneal decompensation[5]. The diagnosis of TED is clinically based upon the presence of eyelid retraction, proptosis, or orbital findings and is often made in conjunction with either laboratory evidence or other clinical findings of GD[6]. If eyelid retraction is absent, then laboratory evidence is needed plus one of the following clinical findings—exophthalmos, optic nerve dysfunction, or extraocular muscle involvement[6]. The pathogenesis of TED consists of hyaluronan deposition and de novo adipogenesis, causing orbital tissue expansion, muscle hypertrophy, and orbital inflammation resulting from several types of immune cells′ infiltration. Orbital fibroblasts are believed to be the primary effector cells in TED pathogenesis. Known risk factors for TED are age, female gender, genetics, dysthyroid status, cigarette smoking, and pregnancy. The management of TED depends upon TED severity and timing, but usually involves both medical and surgical treatments. Unfortunately, TED remains incurable. Current treatments, including high-dose systemic steroids and orbital radiation, can decrease inflammation during the active phase, but do not correct the long-term sequalae of the disease. Surgical rehabilitation, including orbital decompression and eyelid reconstruction, remains the gold standard to improve the facial disfigurement caused by soft tissue fibrosis after the inflammation has subsided. However, better understanding the molecular pathways involved in TED pathogenesis led to the development of teprotumumab, a novel targeted therapy that inhibits the insulin-like growth factor-1 receptor(IGF-1R)and has proved to reduce proptosis and orbital inflammation in patients with active, moderate-to-severe TED. This novel disease-modifying therapy has a strong potential to replace surgery for these patients in the future.
Given differences in anatomy, genetic predisposition, and lifestyle, Asians are known to have different disease patterns compared to other ethnic groups. As for TED, the prevalence of the disease is believed to be higher in Asians, but the majority of the affected individuals exhibit a milder form of disease. Smoking, a known risk factor for TED development and worsening, is found at a higher rate among Asian men[7,8], and therefore is likely to affect the disease pattern. Most TED studies in Asians have been conducted among different ethnic subgroups mainly based on geographic locations, such as China, Korea, Japan, India, Malaysia, and Singapore. In this paper, we aim to provide a concise overview of TED, focusing on an Asian population, and an update on disease management.
2．Epidemiology of TED
TED epidemiology varies among different ethnic groups, although there is limited data on the exact incidence and prevalence of TED in the general population. For Caucasians, the estimated prevalence of TED is suggested to be 0.25% among the US population[9], with an incidence(new onset of TED)of 18.9(16 females and 2.9 males)per 100，000 people per year [1]. In those with thyroid dysfunction, the prevalence of TED is significantly higher compared to the general population. Twenty to 50%of patients with thyroid dysfunction have some form of clinical TED[9,10]. Interestingly, subclinical TED, with findings such as extraocular muscle enlargement on imaging, is suggested to be present in 67%-90% of GD individuals [11,12]. Although TED is most often associated with GD(90%), it can occur in those with euthyroidism(6%), Hashimoto′s thyroiditis(3%), and primary hypothyroidism(1%)[1,3,9,13]. In terms of gender, women were found to be affected approximately five times more frequently than men in North America and Europe(female∶male ratio of 5∶1)[1,2]. TED affects all ages, but incidence has a bimodal distribution, with patients in the fourth and sixth decades being the most affected[1]．
As for Asians, epidemiologic data on TED prevalence are even more limited and studies were conducted on patients with a diagnosis of GD, not in the general population. Weetman[14] stated that the prevalence of GD is similar among Caucasians and Asians.However, the reported prevalence of TED among GD patients seems to vary among Asian ethnic groups. Tellez et al [15] reported the TED prevalence in Asians to be much lower compared to Caucasians(7.7%, 42%), although their study was based on a small number of Asian immigrants, mostly from India, that were compared to Caucasians in the United Kingdom. This study utilized the Werner classification of proptosis(an exophthalmometry reading of 20 mm or more)for both Caucasian and Asian subgroups. Given the differences in facial anatomy and the lower average exophthalmometry reading in Asians, proptosis was likely under-detected in the Asian subgroup and this may attribute to the lower prevalence of TED in this study. Other studies have reported the prevalence of TED in Asians ranging from 17.3% to 74.2%, which is more similar to that of Caucasians(Table 1). However, it is important to recognize that the diagnostic criteria of TED, research methodologies, and sample sizes vary among studies and this may impact the observed prevalence of the disease. The older studies tend to use the Werner classification and newer ones may use modified or new criteria for TED diagnosis. Additionally, some studies have suggested that a difference in smoking rates contributes to the ethnic differences in prevalence[12]．
In terms of gender, TED appears to be more prevalent in Asian women than men, but the predominance of females over males in TED is considerably less than that in Caucasians(Table 1)．
3．Pathophysiology and natural history of TED
Based on the classic Rundle′s curve, TED usually begins with fluctuating inflammatory symptoms that peak at 6 months to 2 years from symptom onset, which is considered as an active phase[22]. The active phase is characterized by inflammation and infiltration of orbital tissues by Type 1 T helper(Th1)cells, B-lymphocytes, mast cells, and macrophages. Orbital fibroblasts(OFs)become activated after interacting with these cells, causing interstitial edema and overproduction of cytokines and chemoattractants that perpetuate the inflammatory cascades[23,24]. Subpopulations of these activated OFs can then differentiate into myofibroblasts and adipocytes[25,26,27,28,29], leading to fibrosis and orbital fat expansion, respectively. Upregulation of hyaluronan(a hydrophilic glycosaminoglycan)synthesis and deposition in extraocular muscles and orbital fat also promotes orbital tissue expansion[30,31]. After 1-2 years, inflammation subsides and reaches a plateau of an inactive or stable phase, during which fibrosis occurs. The current pharmacotherapy can only affect the active phase and therefore is usually initiated during the first 1-2 years after the onset of TED[22]．
The underlying pathophysiology of TED is not clearly understood, but it is hypothesized that loss of immune tolerance to the thyroid stimulating hormone receptor(TSH-R)leads to misrecognition of these receptors as non-self. Production of TSH-R antibody, otherwise known as thyroid-stimulating immunoglobulins(TSIs), subsequently causes TSH-R activation of inflammatory cascades. TSI level usually correlates with TED activity and severity, but the antibodies are not always detectable in TED patients[14,32,33,34,35,36]. Therefore, its presence is helpful in aiding a TED diagnosis in ambiguous cases, but its absence does not necessarily rule out the disease.
A search for additional mediators of TED pathogenesis was based on the hypothesis that cross-reactivity against autoantigens is shared by both orbital tissues and the thyroid gland, via shared ' thyroid-specific' proteins. Immunoglobulins against the IGF-1R have been detected in patients with GD[37] and appear to attenuate the TSH-R and IGF-1R signalling pathways[38]. IGF-1R is a transmembrane tyrosine kinase receptor and is known for its role in development, metabolism, and immunity[39]. TSH-R and IGF-1R are overexpressed on the cell surface of OFs and T cells in patients with TED[40,41]. Additionally, they have been shown to form a physical and functional signalling complex in orbital fibroblasts from patients with TED[42], which may be critical in TED pathogenesis. Recent studies supported that immune recognition of both TSH-R and IGF-1R lead to orbital fibroblast stimulation, production of proinflammatory cytokines, and hyaluronan synthesis[43], which are the hallmarks of TED(Figure 1).
Figure 1.Thyroid eye disease pathogenesis. TSH-R antibody(Ab)causes TSH-R activation of the inflammatory cascades. Autoreactive T cells travel to orbital tissues and interact with orbital fibroblasts(OFs), resulting in mutual activation of both cell types. Activated OFs then secrete various inflammatory cytokines, which perpetuate the vicious inflammatory cycle. TSH-R, which is overexpressed on these cell types, can interact with TSH-R antibodies and lead to cellular activation. IGF-1R can synergistically interact with TSH-R on the OF surface to create a physical and functional complex that may interact with TSH-R antibodies. Activated OFs can differentiate into either adipocytes or myofibroblasts and can increase hyaluronan synthesis. Overall, inflammatory cascades and OF differentiation result in orbital soft tissue expansion and congestion in TED.
4．Clinical characteristics and classification of TED
Inflammatory cascades in TED lead to orbital fat and muscle expansion, which are limited by the bony structures of the orbit, resulting in orbital congestion and elevation of intraorbital pressure. The orbital fat expansion causes clinical manifestations of proptosis, eyelid retractions, and corneal exposure. The muscle enlargement results in restrictive strabismus, diplopia, and possibly physical crowding around the optic nerve, leading to decreased visual acuity, dyschromatopsia, visual field defects, and relative afferent pupillary defects. This is defined as dysthyroid optic neuropathy(DON)and occurs in about 3%-7% of those affected by TED[5]. In clinical practice, TED is diagnosed via history and clinical findings. If needed, laboratory evidence of thyroid dysfunction(TSH, TSI, or TSH-R antibody level)may be obtained. Radiographic imaging can also be helpful for equivocal cases, as it can provide evidence of extraocular muscle enlargement and rule out other causes of proptosis(e.g., orbital mass)．
Due to anatomical differences of the orbit, clinical manifestations of TED differ between Asians and Caucasians in multiple aspects. Regarding orbital features, the average exophthalmometry reading has been shown to be lower in Asians than Caucasians[17,18,19,44]. The upper limit of normal exophthalmometry readings in Asians widely varies from 14 to 19 mm in different studies[17,19,44,45,46,47,48], compared to 20 to 21 mm in Caucasians[49]. Therefore, clinical criteria for proptosis in Asians with TED should be different from those of Caucasians, but a clear consensus on the exact threshold value has not been reached. This may be due to differences in values among Asian ethnic groups themselves[18]. Additionally, Asians have shallower orbits and narrower orbital apices, predisposing them to more frequent apical compression and optic neuropathy compared to Caucasians[50]. As for soft tissue involvement, exophthalmos and eyelid retractions are the two most common manifestations of TED in Asians(Figure 2)[18,44]. Asians appear to have a lesser degree of periorbital edema and proptosis compared with Caucasians. Lower eyelid retraction is also more prevalent in Asians with TED(approximately 20%)and has been suggested as a diagnostic criterion for TED in Asians[18,45]. Acquired epiblepharon is also found in approximately 3.4%-8.5%of Asians with TED[47]. This has been attributed to the Asians′ tight orbital septum and lower eyelid crease, which, in conjunction with TED′s proptosis and posterior lamellar retraction, causes the anterior lamella to override the eyelid margin and lash-corneal touch to occur[50]. For extraocular muscles, it is well-known that the most commonly affected one is the inferior rectus, followed by the medial rectus, superior rectus, and lateral rectus in Caucasians. Asians appear to have more single muscle involvement and vertical muscle group(superior and inferior recti)involvement[50]. Regarding the ocular surface, TED can cause exposure keratopathy from proptosis and eyelid retraction, as well as ocular surface inflammation[51,52]. Dry eye, one of the most frequent causes of ocular discomfort in TED, is known to be more prevalent in Asians without TED than in Caucasians without TED[53,54]. Hence, the prevalence of ocular surface disease is likely higher in Asians affected by TED compared to Caucasians.
Figure 2.Heterogeneity of clinical findings in Asians with TED. Common clinical findings include exophthalmos and eyelid retraction.
Several grading schemes have been devised to classify the severity and activity of the disease. For disease severity, the popular classification is the European Group on Graves′ Orbitopathy(EUGOGO)classification[55], which categorizes the disease into mild(60%), moderate-to-severe(30%), and sight-threatening(4%-8.6%)depending on subjective symptoms and objective findings[56](Figure 3). The advantage of this classification is that it further guides management decisions for each category. However, there are a few limitations of the EUGOGO classification, including mixed clinical features within groups. For example, those with proptosis are not distinguished from those with motility or soft tissue problems. Also, the terminology for severity does not take into account the patient′s perspectives about the disease. As a result, several quality of life questionnaires administered to TED patients have only shown moderate correlation with disease severity based on EUGOGO classification, signifying the discrepancy between patient disease perceptions and objective clinical findings[57]．
Figure 3. European Group on Graves′ Orbitopathy(EUGOGO)classification of thyroid eye disease
For disease activity, TED can be classified into active and inactive(or stable)phases. The popular classification for disease activity is the Clinical Activity Score(CAS)(Table 2). It was created based on a binary scale to help identify active patients who will likely respond to immunosuppressive therapy by addressing 7 inflammatory orbital symptoms at the initial visit. A score of 3 or more indicates active disease. Specifically, it was simply used as a surrogate marker for inflammation and those with CAS≥4 have a positive predictive value of 80% to respond to corticosteroid treatment[58]. Disadvantages of the CAS classification are that the binary scale gives equal weight to all parameters and it may not predict progression of the disease. Additionally, CAS may be elevated in chronically congestive disease, even during the inactive phase, and may be falsely low in fat-predominant disease and some cases of dysthyroid optic neuropathy, which is especially common in Asians. Even though CAS is not frequently used by most practitioners in daily clinical practice, it still has a substantial role in the clinical trial arena, as there is no better outcome assessment of clinical response for TED therapy at this time.
VISA classification was conceived to address both clinical activity and severity of TED, and is a popular system used in North America [59,60]. The parameters included in VISA classification address both subjective and objective aspects of vision, inflammation, strabismus, and TED-related appearance changes(Table 3). However, VISA classification does not specify what is considered a mild, moderate, or severe appearance change, which could vary between physicians. Additionally, the complex, extensive data collection required to determine VISA classification have decreased its popularity in routine clinical practice.
5．Risk factors for TED
Risk factors for TED can be categorized as non-modifiable(endogenous)or modifiable(exogenous). Endogenous factors include age, race, gender, and genetics. Exogenous factors include cigarette smoking, dysthyroid status, and radioactive iodine treatment.
Regarding endogenous factors, studies have shown that more advanced age correlates with the severity of TED in both Caucasians and Asians [45,61,62]. Although there is female predominance in both GD and TED, with female gender being a risk factors for TED[21]. Men with TED appear to have more severe disease compared to women in both Caucasians and Asians[45,61]．
As for genetics factors, several gene loci have been identified for their association with GD. However, their association with additional risk of developing TED remains unclear. These include human leucocyte antigen(HLA), cytotoxic T-lymphocyte antigen-4(CTLA-4), TSH-R gene, and, less commonly, tumor necrosis factor-α(TNF-α), intercellular adhesion molecule-1(ICAM-1), and interferon-γ(IFN-γ)[63]. There have been a few studies about susceptibility alleles for GD conducted in Japanese, Koreans, and Chinese populations. For the HLA gene, studies have shown that the most prominent susceptibility HLA allele in Caucasians, DRB1*03：01, has a much lower frequency in Asians[50]. Given this finding, it is likely that different HLA alleles are responsible for the GD and TED predisposition observed in Asians. For the CTLA-4 gene, several polymorphisms at different regions have been studied. For example, studies on the association between A49G polymorphism of the CTLA-4 gene and GD yielded conflicting results among different populations[64]. A recent meta-analysis associates this polymorphism with susceptibility of TED more in Europeans than Asians[65]. Polymorphisms in the TSH-R gene, which is well-known for its likely association with GD, have also been studied, mostly in patients with GD, and the results regarding its association with TED remain inconclusive[66,67]．
Cigarette smoking has been established as a significant modifiable risk factor for the new development of TED as well as worsening of existing TED. It is important to note that smoking rates vary among different ethnic groups and may affect the risk of developing TED and the prevalence of the disease. For example, Tellez et al[15] reported smoking rates of 61.2%in Europeans and 23%in Asians in their study population and the overall risk for developing TED was 6.4 times higher in Europeans than in Asians. This difference in risk may be attributed to ethnic differences, but also to smoking rates. In Caucasians, active smokers with GD carry about 3-4 times the risk of developing TED with a higher degree of disease severity and higher rates of treatment unresponsiveness compared to nonsmokers[7,10,68,69]. A dose-effect has also been found, with increasing relative risk for TED as the number of cigarettes per day increases[70]. Similarly, several studies in Asians have shown an association between smoking and TED. In North Indians, active smoking status was found to be associated with a 3.9-fold risk of TED[20]. However, the severity of GO did not differ between smokers and non-smokers. Interestingly, gender appears to influence the effect of smoking among Koreans and Japanese populations, as more men than women are smokers. In Koreans, the association between smoking and TED was only significant in women[21]. It is notable that smoking rates between men and women in this study population were drastically different(60%in men versus single digits in women). Similarly, smoking was associated with the risk of GD in Japanese women, but not in men[71]. However, an effect of smoking on development of TED has not been shown in the Japanese population[64]．
Another important modifiable risk factor is thyroid status. Poorly controlled thyroid function is associated with a more severe form of TED[72]. Additionally, radioactive iodine(RAI)treatment can lead to the development or progression of TED in 15%-35%of patients, especially in those with active, moderate-to-severe TED and post-treatment hypothyroidism[73,74]. A study in a Korean population supported this by showing an association between RAI therapy and development of TED[21]. Co-administration of low-dose oral prednisone(0.2-0.5 mg·kg－1·day－1)is recommended during RAI treatment to prophylactically prevent the progression of TED[75].
6．Management of TED
Due to the heterogeneity of TED, its management varies depending on each individual′s symptoms, duration of the disease(active versus inactive phase), and disease severity(Table 4). No treatment has been shown to effectively reverse or prevent underlying orbital tissue remodeling caused by TED until recently. In this review, we will focus our discussion on global treatments of TED rather than specific treatments for individual consequences of TED, such as eyelid retraction, lagophthalmos, or ocular surface exposure.
For active disease, a palliative treatment with either intraorbital, oral, or intravenous(IV)corticosteroids is a recommended first-line therapy to provide temporary relief in ocular inflammation and edema. Administration of pulse IV high-dose glucocorticoid is preferable to oral administration due to its better efficacy and safety profiles[76]. Disappointingly, corticosteroids are only effective in 50%-80%of patients and several studies have failed to show their long-term efficacy in a disease-modifying capability, measured as an improvement in proptosis or strabismus[42]. Relapses or flare-ups of TED following corticosteroid treatment are common. Furthermore, the side effect profiles of corticosteroids should not be taken lightly. The rate of adverse events progressively increases as the dose increases and include diabetes mellitus, psychosis, and depression. More severe, life-threatening adverse events have been reported, including acute liver failure and death[77,78,79,80], especially with cumulative doses of more than 8 g of IV methylprednisolone. In summary, corticosteroids are relatively effective as a temporizing measure for tissue edema, symptomatic relief, and DON, but the risk-benefit ratio needs to be seriously considered before use.
Radiotherapy(RT)was first described for TED in the 1940s. It is administered at a cumulative dose of 20 Gy per eye divided into 10 fractions over 10 days[81]. The safety profile is relatively acceptable—no increased risk of cataract[82], one reported case of a radiation-induced tumor[83], and a 1%incidence of mild retinopathy[82]. Retrospective studies have shown approximately 60%efficacy, comparable to oral corticosteroids, but RT takes longer to reach the maximal clinical response. Three[84,85,86] out of four[87] randomized controlled trials supported RT′s benefit in improving soft tissue inflammation and motility when comparing to sham therapy. However, for exophthalmos or eyelid retraction, RT has no benefit. Kazim et al[88] showed RT′s efficacy in preventing DON or delaying orbital decompression among the DON group. An ongoing multi-center randomized controlled trial(CRISEPTED, NCT02339142)is underway to investigate the efficacy of combined RT and IV corticosteroids in comparison to IV corticosteroids alone for early progressive TED. In summary, RT is usually not a first-line therapy in active TED, but rather an adjunctive treatment to IV glucocorticoid therapy, and its efficacy remains controversial.
For active disease, surgical intervention is reserved for sight-threatening cases—those with DON and corneal breakdown. The most effective and direct treatment for DON is a medial orbital wall decompression, commonly performed via a transcaruncular approach. Bony decompression near the posteromedial orbital apex allows vision to improve relatively rapidly. The posterior two-thirds of the inferomedial orbital strut, as well as part of the palatine bone, should be removed to allow maximal medial wall decompression in DON cases[89]. Sparing the anterior inferomedial orbital strut reduces the risk of postoperative diplopia[90]. Additional decompression of the orbital floor and/or lateral wall can be performed to relieve posterior orbital congestion[89]．
7．Recent advances in targeted therapy for TED
In theory, the ideal TED treatment should alter the disease course by improving inflammatory signs and decreasing risk of visual complications. This treatment would also have minimal side effects, minimize facial disfigurement, restore patient′s quality of life, restore ocular functions, and eliminate the need for surgical interventions. Therefore, the therapy should be initiated during the active phase of the disease, according to the Rundle′s curve(Figure 4), to interrupt the disease processes and alter long-term complications of the disease. Unfortunately, current treatments do not reverse the underlying long-term alteration of orbital tissues nor do they improve proptosis or strabismus. Furthermore, they often have substantial side effects. Recent advances in molecular biology have led to the development of immunomodulatory therapies that target the underlying molecular mechanisms of disease pathogenesis. Rituximab, tocilizumab, and teprotumumab have been tested in randomized clinical trials for their efficacy in TED treatment, with some showing promising results.
Figure 4. The natural history of TED follows Rundle′s curve. Time is on the horizontal axis and disease activity is on the vertical axis. During the active phase, inflammation results in fluctuating edema, erythema, injection, and proptosis. After 1-2 years, the disease enters the inactive or stable phase, in which inflammation subsides, but orbital tissue fibrosis occurs and orbital content expansion remains.
7.1 Rituximab
Rituximab(RTX)is a human/murine chimeric monoclonal antibody to CD20, the protein expressed on the surface of pre-B and mature B-lymphocytes. Because RTX can effectively deplete B cells, which are precursors of autoantibody-producing plasma cells, it has promising potential to treat autoantibody-mediated diseases such as TED. Several prospective and retrospective studies have examined the efficacy of RTX in treating moderate-to-severe and sight-threatening TED, and the findings are conflicting [91,92,93,94,95,96,97,98,99,100,101]. For example, one small randomized controlled study by Salvi et al compared the efficacy of RTX(500 or 2000 mg)with the common first-line therapy, IV methylprednisolone(IVMP), in patients with active moderate-to-severe TED [91]. Both treatment groups showed significant improvement in CAS, but the number of responders were higher in the RTX group(100%)than the IVMP group(69%)at 24 weeks. Reactivation of disease was also found in the IVMP group, but not in the RTX group. Another randomized clinical trial showed that rituximab(1000 mg each, 2 weeks apart)offered no additional benefit in CAS improvement over normal saline[92]. However, both the RTX and placebo groups in the latter study had received prior corticosteroid treatment, and thus the improvement experienced in the control group may have been influenced by a persistent benefit from prior corticosteroid treatment. A recent meta-analysis of RTX treatment in TED suggests that RTX is superior to glucocorticoids or saline for patients with moderate-to-severe TED. However, RTX did not provide significant improvement in proptosis or diplopia, the major serious consequences of TED[102].
7.2 Tocilizumab
Interleukin 6(IL-6), a proinflammatory cytokine, is widely implicated in the inflammatory response and host defense mechanism, as a key B cell support factor and part of T helper cell development. IL-6 has been shown to play roles in other autoimmune diseases and in the TED disease process. Additionally, IL-6 serum levels obtained from TED patients were found to be elevated [103]. Tocilizumab, a recombinant humanized monoclonal antibody, which acts as an antagonist against the IL-6 receptor, has been proposed as a possible TED treatment by directly reducing the inflammatory response. A randomized clinical trial(NCT01297699)was completed in December 2015 to determine the efficacy and safety profile of tocilizumab in active TED in 10 Spanish centers. The study randomized 32 active TED patients who were deemed steroid-resistant to receive either infusions of tocilizumab or placebo on weeks 0, 4, 8, and 12 with a 40-week observation period[104]. Significant CAS improvement of at least 2 points was found in 93.3%of the tocilizumab group versus 58.8%of the placebo group(P＝0.04)at week 16. The change in exophthalmos at week 16 from baseline appeared to occur in the treated group, but not in the placebo group[－1.5(－2.0 to 0.5) mm vs 0.0(－1.0 to 0.5) mm; P＝0.01] Regarding adverse events, the study reported moderately increased transaminase level in one patient at week 8 and an acute pyelonephritis in another patient at week 32 in the tocilizumab group [104]. No improvement in diplopia or strabismus was reported. Overall, tocilizumab appears to be promising as a potential disease-modifying treatment in a Spanish population, but further randomized clinical trials are needed to confirm its efficacy.
7.3 Teprotumumab
Although TED exhibits heterogeneity in clinical manifestations, there exists a common underlying molecular mechanism that initiates the disease processes. The search for 'a TED-specific therapy' involves recent understanding of the roles of the identified autoantigens, TSH-R and IGF-1R, in the disease pathogenesis(Figure 5). Unlike other drugs that target the immune system, teprotumumab was developed as a bench-to-bedside drug and specifically targets the molecular pathways of TED. Teprotumumab is a human monoclonal antibody that binds to the extracellular domain of the IGF-1R with high affinity and specificity[42,105]. Binding of the drug to the IGF-1R induces internalization of the antibody-receptor complex, which leads to its degradation. Because IGF-1R and TSH-R are physically and functionally linked, teprotumumab thus inhibits TSH-R signaling pathways and blocks autoantibodies from attacking orbital fibroblasts[105,106]. Consequently, it reduces hyaluronan production and cytokine stimulation(Figure 6)．
Figure 5. TSH-R and IGF-1R interact synergistically and are key players in TED pathogenesis. TSI：thyroid-stimulating immunoglobulin; GD-IgG：Graves′ disease immunoglobulin
Figure 6. Teprotumumab is an IGF-1R antagonist and, therefore, inhibits TSH-R signaling pathways in orbital fibroblasts
A recent double-masked, placebo-controlled, phase 2 clinical trial was completed to evaluate the efficacy and safety profile of teprotumumab [106]. The study included 88 subjects with recent-onset, active, moderate-to-severe TED(less than 9 months since TED symptom onset, CAS of 4 or more). Subjects were randomized to receive teprotumumab or placebo in a 1∶1 ratio. The regimen consisted of IV study drug infusions every 3 weeks over a 24-week period(total of 8 infusions). The primary outcome was the percentage of patients who had a reduction in CAS of ≥2 points and a decrease in proptosis of≥2 mm in the study eye without deterioration in the fellow eye. Patients were examined for CAS and proptosis at 6, 12, 18, and 24 weeks. The study revealed relatively striking results, showcasing teprotumumab′s efficacy in reducing both CAS and proptosis as early as week 6(after 2 infusions)[106]. It is remarkable that every patient treated with teprotumumab demonstrated some degree of improvement. Specifically, the biggest improvement was found among patients with more severe disease[107]. For instance, a proptosis reduction of at least 4-mm was found in 40%(17/42)of patients who received teprotumumab, but in 0%(0/45)of patients who received placebo at week 24[107]. Proptosis reduction persisted at least 1-year after the last infusion of teprotumumab. So far, no other medical treatment has achieved a similar magnitude of proptosis reduction, which has only been observed after surgical intervention. Gender and smoking status did not appear to influence the degree of response among patients treated with teprotumumab[107]. Importantly, the quality of life and subjective diplopia among these subjects were also shown to improve in the teprotumumab treated group[106]．
Generally, drug-related adverse events were non-serious and manageable. Hyperglycemia occurred in a few patients treated with teprotumumab[106], which was of a worse degree in diabetic patients, but was well-controlled with medication adjustment. Other adverse events associated with teprotumumab were mostly mild and resolved without treatment. In summary, teprotumumab benefited those with active moderate-to-severe TED, as evidenced by a clinical improvement in CAS, proptosis, diplopia, and quality of life, while having minimal side effects. In August 2016, teprotumumab received ' breakthrough therapy' designation from the United States Food and Drug Administration for its disease-modifying properties.
A recent phase 3 trial(NCT03298867)of similar design has further confirmed teprotumumab′s efficacy and safety profiles with similar findings to the phase 2 study[108]. The dropout rate was low(<5%)in both arms. Three adverse events were reported; one in the placebo group(visual field defect, requiring orbital decompression and study discontinuation), and 2 in the teprotumumab group(1 pneumothorax deemed ' unrelated' to the drug; 1 infusion reaction, which lead to treatment discontinuation). Given its mechanism of action, there is no reason why teprotumumab will be dosed differently among different ethnic populations.
8．Conclusion
TED is a heterogenous autoimmune disease that can cause debilitating vision impairment and facial disfigurement, regardless of ethnicity. Genetic and environmental factors partially influence disease manifestations and severity. Although TED in the majority of Asians is of mild severity, some suffer a more severe form of DON compared to Caucasians because of differences in orbital anatomy. With our current knowledge of the natural history of the disease, there is a relatively limited time window to modify the disease course during the active phase, and thus reduce morbidity. After decades of suboptimal treatments and the search for a TED-specific drug, studies indicate that researchers have developed a novel therapy against TED that targets the TSH-R/IGF-1R signaling pathways. Teprotumumab, an inhibiting antibody against IGF-1R, has been shown in phase 2 and 3 randomized clinical trials to effectively alter disease course in patients with active, moderate-to-severe TED. This drug strikingly reversed clinical outcomes of the disease, including proptosis and diplopia, providing new hope for an effective TED treatment.
参考文献 （略）
扫码下载杂志官方App