第一部分：IgG4相关性疾病血清学活动临床稳定患者特征及预后的队列研究
中文摘要
背景和目的：IgG4相关性疾病（IgG4-related disease, IgG4-RD）是一种以高血清IgG4水平和多脏器受累为特征的慢性纤维炎症性疾病。在维持治疗期间，患者常处于以下两种状态：1）达到临床缓解但血清IgG4水平仍维持高水平；2）临床缓解且血清IgG4水平正常。我们将上述两种状态首次定义为“血清学活动临床稳定（serologically active clinically quiescent，SACQ）”和“血清学稳定临床稳定（serologically quiescent clinically quiescent，SQCQ）”。本研究旨在定义并描述IgG4-RD中SACQ患者的特征，评估其预后，并与SQCQ患者进行比较。
方法：2012年1月至2022年8月期间，在本中心风湿免疫科连续招募IgG4-RD患者，规律随访并收集患者在基线和随访中的临床数据。SACQ和SQCQ分别定义为处于持续临床缓解状态至少两年，且血清IgG4水平持续高于正常（SACQ）或维持正常（SQCQ）。本研究的结局事件为复发。本研究比较了SACQ和SQCQ组的特征和预后，同时采用流式细胞学检测两组患者在SACQ和SQCQ状态下的B细胞亚群，采用Kaplan-Meier分析比较SACQ和SQCQ患者的结局，以及不同维持治疗方案下SACQ患者的结局。进一步采用倾向性评分匹配（propensity score matching，PSM）平衡混杂因素，以分析不同治疗方案对SACQ组结局的影响。最后，应用单因素和多因素Logistic回归探索SACQ患者复发的风险因素。
结果：本研究纳入268名SACQ患者和108名SQCQ患者。两组患者在诊断时的年龄相似（57.0 [48.0, 63.0]岁 vs. 56.0 [49.0, 61.0]岁，P = 0.402），且均以男性为主（66.4% vs. 58.3%，P = 0.174）。与SQCQ组相比，SACQ组基线IgG4水平更高（13600 [6425, 24450] vs. 3640 [2388, 6995]，P < 0.001），且增殖型的比例更高（83.2% vs. 61.1%，P < 0.001）。SACQ组较SQCQ组的浆母细胞增加，而未转化记忆B细胞和调节性B细胞减少。在平均55个月的随访期内，两组患者都经历了糖皮质激素减量，且复发率相似（16.8% vs. 16.7%，Log-rank P  =  0.659），但SACQ组较SACQ在随访终点时的激素维持剂量更高 [(4.12 ± 3.22) mg/d vs. (2.68 ± 2.92) mg/d，P < 0.001]。在SACQ组中，激素减量的患者与激素剂量维持或增加的患者具有相似的复发率（Log-rank P = 0.895）。Logistic回归发现血清IgG4水平重新升高至≥SACQ状态起点的1.269倍是SACQ患者复发的唯一风险因素（OR 2.56，95% CI：1.80-3.85；AUC = 0.791）。
结论：SACQ患者在谨慎减量激素的情况下，复发率与SQCQ无显著差异。IgG4-RD中SACQ患者的激素减量是可行的。在SACQ状态的长期管理中，监测血清IgG4水平可能对预测复发有一定帮助。

第二部分：IgG4相关性疾病损伤评分建立、初步验证及其在真实世界IgG4-RD患者的应用研究

中文摘要
目的：建立并初步验证IgG4相关性疾病（IgG4-related disease，IgG4-RD）损伤指数（damage index，DI）。用IgG4-RD DI描述长期随诊IgG4-RD患者的器官损伤概况及损伤累积趋势，及探索损伤积累的危险因素。
方法：首先由中国IgG4-RD协作组（the Chinese IgG4-RD Consortium，CIC）的专家制定了评估IgG4-RD损伤的评分条目草案，该损伤评分草案通过改良德尔菲法完善，经专家组共识形成最终版本CIC IgG4-RD DI。随后，构建了代表“疾病活动-损伤增加”、“疾病活动-损伤稳定”、“疾病稳定-损伤增加”、“疾病稳定-损伤稳定”等四种临床情景的40个IgG4-RD模拟病例，每例设定随访时间不少于3年，设置2个损伤评估时间点。邀请来自全国35家医院的48名风湿科医生，使用CIC IgG4-RD DI对40个模拟病例评估损伤。对IgG4-RD DI，采用组内相关系数（intraclass coefficient，ICC）和Kendall-W一致性系数（Kendall-W coefficient of concordance，KW）评估评分者间一致性。通过集成受试者工作特征（summary receiver operative characteristic，SROC）曲线法验证效标效度。进一步，在2012年1月至2022年8月期间，在本中心风湿免疫科招募了真实世界中的IgG4-RD患者，使用CIC IgG4-RD DI进行评分。描述所纳入患者的基线特征、DI评分的分布以及随访期间的增加情况。比较了损伤增加、损伤稳定和无损伤组的临床特征和治疗方案。采用单因素和逐步逻辑回归分析了5年内损伤积累的风险因素。
结果：CIC IgG4-RD DI是一个累积评分量表，包含14个器官系统项目，共39个条目。IgG4-RD DI能够区分疾病活动与损伤累积。所有评分者在各个评分时间点的评分者间一致性总体良好。两次评分时间点的ICC分别为0.69和0.70，KW分别为0.74和0.73。在4个临床情境的亚组分析中，所有亚组的ICC和KW均分别超过0.55和0.61。效标效度检验中IgG4-RD DI的表现良好，SROC曲线示敏感性为0.86（95% CI 0.82-0.88），特异性为0.79（95% CI：0.76-0.82），曲线下面积（area under the curve，AUC）为0.88（95% CI：0.85-0.91）。从前瞻性队列中招募了306名真实世界中的IgG4-RD患者。DI评分在随访期间逐渐累积。最重要的器官损伤项是“胰腺”、“肝脏/胆道”和“其他损伤”。 “其他损伤”中最重要的损伤条目是糖皮质激素相关损伤和新发恶性肿瘤。使用Logistic回归分析得出，随访5年时损伤积累的风险因素是基线胆道受累（OR 2.51，95%CI：1.35-4.67，P = 0.004）。     
结论：CIC IgG4-RD DI是一种量化疾病损伤的有用手段，具有良好的可操作性和可信度。CIC IgG4-RD DI在真实患者中的应用揭示了IgG4-RD患者在长期随访过程中疾病损伤的积累。胰腺胆道、糖皮质激素相关毒性和新发恶性肿瘤是IgG4-RD损伤的重要方面。基线胆道受累是随访5年时损伤积累的风险因素。预计该损伤指数将成为IgG4-RD患者治疗性临床试验和预后研究的有用工具。

The clinical characteristics and long-term outcome of serologically active clinically quiescent IgG4-related disease
Abstract
Background and objectives. IgG4-related disease (IgG4-RD) is a chronic fibroinflammatory disease characterized by elevated serum IgG4 level and multiple organ involvement. During remission maintenance treatment, patients are often under one of the following two status: 1) achieving clinical remission but with persistently high IgG4 levels; 2) achieving clinical remission and with normal serum IgG4. We first define the above-mentioned two statuses as active clinically quiescent (SACQ) and serologically quiescent clinically quiescent (SQCQ). This study aimed to determine and describe the characteristics of SACQ status in patients with IgG4-RD and evaluate its prognosis, in comparison with SQCQ patients.
Methods. Patients diagnosed with IgG4-RD were consecutively enrolled at the Rheumatology and Clinical Immunology clinic of this center from January 2012 to August 2022 and regularly followed up. The clinical data were collected at baseline and every subsequent follow-up visit. SACQ and SQCQ were defined as persistent clinical remission with elevated but steady (for SACQ)/normal (for SQCQ) serum IgG4 for at least two years, respectively. The outcome was defined as relapse. Characteristics and prognoses between the SACQ and SQCQ groups were compared. Flow cytometry was used to investigate the B cell subsets during SACQ or SQCQ status in both groups. Kaplan-Meier analysis was used to compare the outcomes between SACQ and SQCQ patients, and between SACQ patients under different maintenance treatment regimens. Furthermore, propensity score matching (PSM) was used to balance confounding factors when analyzing the association of treatment regimens with outcomes in SACQ patients. Finally, univariate and multivariate Logistic regression were used to investigate risk factors for relapse in SACQ patients.
Results. The study included 268 SACQ and 108 SQCQ patients with IgG4-RD. Both groups of patients were of similar age (57.0 [48.0, 63.0] vs. 56.0 [49.0, 61.0], P = 0.402) at diagnosis and male dominance (66.4% vs. 58.3%, P = 0.174). Compared to the SQCQ group, the SACQ group had higher baseline levels of IgG4 (13600 [6425, 24450] vs. 3640 [2388, 6995], P < 0.001) and a greater prevalence of proliferative subtype (83.2% vs. 61.1%, P <0.001). Plasmablasts were increased, while unswitched memory B cells and regulatory B cells were decreased in SACQ patients compared with SQCQ patients. During an average follow-up of 55 months, both groups experienced glucocorticoid (GC) tapering and had similar relapse rates (16.8% vs. 16.7%, Log-rank P = 0.659), but the SACQ group had a higher GC maintenance dose compared with the SQCQ group [(4.12 ± 3.22) mg/d vs. (2.68 ± 2.92) mg/d，P < 0.001] at the end of follow-up. In the SACQ group, patients who underwent GC tapering had a comparable relapse rate with those who maitained or increased GC dosage (Log-rank P = 0.895). Serum IgG4 level re-elevation to ≥ 1.269 times the start of SACQ status was the sole risk factor for  relapse in SACQ patients (OR 2.56, 95% CI: 1.80-3.85; AUC = 0.791) identified by Logistic regression.
Conclusions.  SACQ patients achieved a relapse rate similar to SQCQ with cautious GC tapering. It is feasible for patients of SACQ status to taper GC. Monitoring serum IgG4 levels might be useful for predicting relapses in the long-term management of SACQ status in IgG4-RD.

The development and initial validation of IgG4-related disease damage index and its Application on Real-world patients with IgG4-RD
Abstract
Objective. To develop and conduct an initial validation of the damage index for IgG4-related disease (IgG4-RD DI). To characterize organ damage and its accumulation trend in IgG4-RD patients during long-term follow-up using the IgG4-RD DI and to explore risk factors for damage accrual.
Methods. First, a draft of index items for assessing organ damages in IgG4-RD patients was generated by experts from the Chinese IgG4-RD Consortium (CIC). The preliminary DI was refined using the modified Delphi method, and a final version was generated by expert consensus. Forty IgG4-RD simulated cases were composed to represent four types of clinical scenarios: Disease Active-Damage Increase, Disease Active-Damage Stable, Disease Stable-Damage Increase, and Disease Stable-Damage Stable. Each case was assigned with over three years of follow-up and two time points for damage assessment. Forty-eight rheumatologists from 35 hospitals nationwide were invited to evaluate damage in the 40 simulated cases using the CIC IgG4-RD DI. The intraclass correlation coefficient (ICC) and the Kendall-W coefficient of concordance (KW) were used to assess the inter-rater reliability. The criterion validity was tested by using the summary receiver operative characteristic (SROC) curve. Next, real-world patients with IgG4-RD were enrolled at the Rheumatology and Clinical Immunology clinic of this center from January 2012 to August 2022, assessed and scored using the CIC IgG4-RD DI. The baseline characteristics of the included patients were described. The distribution of DI score and the increased value between visits were described. The clinical characteristics, treatment regimens were compared between damage increase, damage stable, and no damage groups. Univariate and stepwise Logistic regression was used to investigate the risk factors for damgage accrual at 5 years.
Results. CIC IgG4-RD DI is a cumulative index consisting of 14 domains of organ systems, and a total of 39 items. The IgG4-RD DI can distinguish between disease activity and damage accumumulation. Overall consistency in scores at each assessment by all raters was satisfactory. The ICC at the two assessment time points was 0.69 and 0.70, and the KW was 0.74 and 0.73, respectively. In subgroup analysis of four clinical scenarios, ICC and KW were all over 0.55 and 0.61, respectively. The IgG4-RD DI showed showed a favorable criterion validity, with a sensitivity of 0.86 [95% CI 0.82-0.88], a specificity of 0.79 [95% CI 0.76-0.82], and an area under the curve (AUC) of 0.88 [95% CI 0.85-0.91] indicated by the SROC curve. Three hundred and six real-world patients with IgG4-RD were recruited from a prospective cohort. The DI score accumulated during follow-up. The important damage domians were the “Pancreas”, “Liver/Biliary tree”, and “Other”. The most important damage items within the “Other” domain were glucocorticoid-related damage and malignancy onset. The risk factor for damage accrual at 5 years was baseline bile duct involvement (OR 2.51, 95%CI: 1.35-4.67, P = 0.004).
Conclusions. The CIC IgG4-RD DI is a useful approach to quantify disease damage, with good operability and credibility. The application of CIC IgG4-RD DI on real-world patients with IgG4-RD revealed damage accrual during long-term follow-up. The importance of pancreas/bile duct, glucocorticoid toxicity, and malignancy onset as IgG4-RD damage is underscored. Baseline bile duct involvement was the risk factor of damage accrual at 5 years. The damage index is expected to become a useful tool for therapeutic clinical trials and prognostic research in patients with IgG4-RD.