Clinical Features and Outcomes of Immunoglobulin G4-related Disease

Immunoglobulin G4-related disease (IgG4-RD) is a newly recognized disease, and a few epidemiologic studies about this disorder have been published. This research aimed to describe the clinical, laboratory, and histopathological features and outcomes of IgG4-RD. Ninety-four patients who satised the comprehensive diagnostic criteria on IgG4-RD were included in this study. Fifty-eight (61.7%) were men. The mean age was 54.8 years, and the median follow-up duration was 32.9 months. The clinical feature between single and multiple organ involvement and with or without kidney involvement groups were evaluated based on symptoms and laboratory ndings. The clinical outcome was assessed according to treatment strategies and response. Of 94 patients, 56 (59.6%) had multiple organs involvement. Patients with multiple organ involvement had higher serum IgG and IgG4 levels than those with single organ involvement. Those with IgG4-related kidney disease (IgG4-RKD) had worse renal function, and the incidence of peripheral blood eosinophilia and hypocomplementemia was higher in patients with renal involvement than in those without. Glucocorticoids-based therapy was most commonly used. (79.8%). Thirty-nine (41.5%) achieved complete remission. Eighteen (19.1%) relapsed after response to treatment. IgG4-RD had different clinical features and outcomes. The number of organ involvement and kidney involvement were not associated with clinical outcomes.


Introduction
Immunoglobulin G4-related disease (IgG4-RD) was rst reported as Japanese patients with sclerosing pancreatitis in 2001. 1 Currently, IgG4-RD is considered a systemic disorders based on several reports. It is a unique immune-mediated disease that can occur in each organ. 2,3 Moreover, it is characterized by elevated serum IgG4 concentration, lymphoproliferative in ltration enriched with IgG4-positive plasma cells in affected tissues, and distinct storiform brosis in affected organs. [3][4][5][6] The global incidence of IgG4-RD remains unknown. To date, most studies include case reports, those with a small cohort, and epidemiological research. Most epidemiological studies have been reported in North America, Europe and Japan. [7][8][9][10][11][12][13] The current study aimed to evaluate the clinical features and outcomes of IgG4-RD.

Results
Basic demographic and clinical characteristics of patients with IgG4-RD In total, 94 patients were diagnosed with IgG4-RD, and the number of newly diagnosed patients in our cohort increased annually ( Figure 1). The basic demographic and clinical characteristics of all patients with IgG4-RD are presented in Table 1. The mean standard deviation (SD) age at diagnosis was 54.8 ± 14.2 years, and the median follow-up duration was 32.9 (interquartile range (IQR): 14.9, 55.4) months. In total, 58 (61.7%) of 94 patients were men. Three patients (3.2%) had autoimmune diseases prior to IgG4-RD. One patient had systemic lupus erythematosus, and two patients had Sjögren's syndrome. Five (5.2%) had malignant diseases such as thyroid cancer (n = 1), breast cancer (n = 1), prostate cancer (n = 1), gastric cancer (n = 1) and craniopharyngioma (n = 1). Among them, four had no malignant diseases at the time of IgG4-RD diagnosis. Only one patient with prostate cancer was treated with hormone therapy.
Fifty-six (59.6%) patients with IgG4-RD experienced involvement of two or more organs. Patients presented with various symptoms based on the organs involved. Flank pain (13.5%) was the most common symptom, followed by abdominal pain (12.5%) and eyelid swelling (12.5%). Five (5.2%) patients were incidentally found to have elevated serum creatinine level based on medical examinations. Four (4.2%) presented with fever alone. Further, 10 (10.6%) patients were incidentally diagnosed via imaging studies during health screenings. Imaging studies were performed to diagnose all patients, and computed tomography (CT) scan was the most commonly used method (78.4%).
The distributions of organ involvement are shown in Figure 2. In terms of clinical manifestations, the involved organs were extremely diverse. The lymph nodes (36.1%) were the most commonly involved organs. However, the retroperitoneum was the most frequently affected organ (26.6%) as an isolated organ. The incidence of retroperitoneum involvement was signi cantly higher in men than in women (35.6% vs 13.5%; P = 0.018). The commonly affected organs were the salivary gland (20.2%), lacrimal gland (18.1%), and kidney (13.8%). One patient had bone marrow (BM) involvement. Diffuse 18 Fuorodeoxyglucose ( 18 F-FDG) uptakes in the spine, sternum, and both pelvic bones was con rmed, and the number of plasma cells increased by approximately 5% based on BM biopsy. The involvement of skull, tonsil, breast, pericardium, abdominal wall and bladder were found in each one patient.

Differences in clinical features between patients with single and those with multiple organ involvement
The differences of clinical features between patients with single and those with multiple organ involvement are presented in Table 2, and the distribution of organ involvement is shown in Figure 2. The mean age at diagnosis, male-to-female ratio, and prevalence of underlying diseases were similar between patients with single and those with multiple organ involvement. Eosinophilia and hypocomplementemia were more commonly observed in patients with multiple than in those with single organ involvement.
However, the results did not signi cantly differ. The serum IgG and IgG4 levels were signi cantly higher in patients with multiple organ involvement than in those with single organ involvement (P = 0.037 and P = 0.020, respectively). The retroperitoneum was the most commonly affected organ in patients with single and multiple organ involvement. The salivary gland was more frequently affected in patients with multiple than in those with single organ involvement (P = 0.014).
The proportion of patients with elevated ESR and CRP levels and those with autoimmune Ab did not differ between the single and multiple organ involvement groups.
Differences in clinical features based on the presence of IgG4-RKD The differences in clinical features according to renal involvement are presented in Table 3. The patients with IgG4-related kidney disease (IgG4-RKD) were older than those without (P = 0.040). A high number of patients had hypertension prior to IgG4-RD diagnosis (P = 0.018). All patients with IgG4-RKD underwent biopsies of affected organs, and the characteristic pathologic ndings were con rmed in all patients. Eleven patients with IgG4-RKD underwent kidney biopsies, and two biopsies of other affected organs, such as the lacrimal gland and bile duct. In terms of kidney biopsy ndings, lymphoplasmacytic in ltration was observed in all tissues, and only one patient presented with obliterative phlebitis. The nal diagnosis was tubulointerstitial nephritis (TIN). At the time of diagnosis, the renal function of patients with IgG4-RKD was worse than of those without. After treatment, eight (61.5%) patients with IgG4-RKD experienced improvement in Chronic Kidney Disease-Epidemiology collaboration eGFR (CKD-EPI-eGFR).
Peripheral blood eosinophilia was more common in patients with IgG4-RKD (P = 0.011). The mean complement C3 and C4 levels were signi cantly lower in patients with IgG4-RKD than in those without (P = 0.026 and P = 0.031, respectively). Hypocomplementemia was more common in patients with IgG4-RKD (P = 0.007). The serum IgG and IgG4 levels did not signi cantly differ between patients with and without IgG4-RKD.

Treatment strategies and response
The treatment strategies and responses are shown in Figure 3 and 4. The most common treatment was glucocorticoids (GCs) monotherapy (n = 50 [53.2%]). GCs were administered in combination with other immunosuppressive agents or surgical resection. Hence, 75 (79.8%) patients received GCs-based therapy. Some patients did not respond to GCs monotherapy. Among them, 16 (17.0%) patients received GCs combined with other medications such as azathioprine, methotrexate, cyclosporine, and mycophenolate mofetil. Nine (9.6%) patients who did not receive treatment were followed-up. Of them, three patients achieved complete remission. The median treatment duration was 14.7 (IQR: 4.2, 43.8) months. The median initial GCs (prednisolone) dose was 45 mg/day, and the median maintenance dose was 5 mg/day. The proportion of patients based on treatment strategies was similar between patients who presented with single and those with multiple organ involvement and those with or without IgG4-RKD.
Three patients were lost to follow-up after diagnosis. Therefore, 91 patients were evaluated for treatment response. The median interval from treatment initiation to follow-up imaging study was 3 months (IQR: 1.2, 6.5). In total, 43 (47.3%) and 34 (37.4%) patients had complete and partial response, respectively.
Complete response was more common in patients treated with GCs combined with other immunosuppressive agents or surgical resection than in those who received other therapies (P = 0.044).
The proportion of patients who responded to treatment was similar between the single and multiple organ involvement groups or with or without IgG4-RKD groups. In total, 39 (41.5%) of 91 patients achieved remission during the follow-up period, and 18 (19.1%) patients received additional treatments because of relapse. Remission was more common in patients with complete response during the rst evaluation (P < 0.001). The remission and relapse rates did not signi cantly differ according to the presence of IgG4-RKD. Relapse was more frequent in patients with multiple than in those with single organ involvement (28.6% vs. 5.3%; P = 0.005). However, none of the patients died during the follow-up period.

Discussion
There were 94 cases of IgG4-RD in three medical college-a liated hospitals during the 10.5-years study period. In our study, there were only 1-5 new cases per year within the rst 6 years. However, it increased to more than 10 cases per year with the last 4 years. This nding showed that the diagnosis of IgG4-RD made by clinicians is increasing, and more cases of IgG4-RD may be reported in Korea, with consideration of the incidence of IgG4-RD (0.28 -1.08 per 100,000) and the newly diagnosed patients each year of IgG4-RD (336 -1300 patients) in Japan. [17] The clinical features of IgG4-RD in our study were similar to those of previous studies in terms of sex, age, and organs involved. [10][11][12][13]15 Our study differed from previous studies. First, the initial clinical symptoms vary based on organ involvement. However, the most common symptoms in previous studies were salivary gland swelling, lacrimal gland swelling, and nonspeci c abdominal pain. 10,12,16,17 By contrast, in our study, the most common symptom was ank pain, and all patients were initially diagnosed with retroperitoneal brosis (RPF) with hydronephrosis. RPF can occur due to several etiologies, and it is categorized as idiopathic and secondary based on the medications used, autoimmune disease, and malignancy. This study showed the role of IgG4-RD in the development RPF. Second, 10 (10.6%) patients were incidentally diagnosed via imaging studies during health screenings. If patients who incidentally found azotemia were included, 15 (16.0%) patients were not symptomatic at the time of IgG4-RD diagnosis. Third, in our cohort, unlike previous studies, pancreas involvement was signi cantly less common. The pancreas is frequently involved in IgG4-RD. The prevalence rate is approximately 25 6 0%, and that in our cohort is 8.5%. 12,13,[16][17][18] Forth, in previous studies, allergic disease was commonly observed, and eosinophilia was frequently noted particularly among Japanese, Chinese and Caucasian. 10,12,16,17 However, in our cohort, only one patient had bronchial asthma, which was treated with steroid inhaler, prior to IgG4-RD diagnosis. The incidence of eosinophilia was relatively lower in this study than in previous studies. 11,12 In our study, 38 and 56 patients presented with single and multiple organ involvement, respectively. Since IgG4-RD is a systemic disease, whether the clinical features and prognosis differed based on the number of organs involved was evaluated. Results showed no signi cant difference in clinical features and outcomes between patients with single and those with multiple organ involvement, except those with high serum IgG4 concentration and relapse in multiple organ involvement. These results were similar to those of previous studies. 11,16,17,19,20 Although patients had a higher serum IgG4 concentration and relapse rate in multiple organ involvement, the response to treatment and the rate of complete remission did not differ between patient with single and those with multiple organ involvement.
The kidney is one of the relatively common organs involved in IgG4-RD. However, only few studies have investigated the differences in clinical characteristics in patients with or without IgG4-RKD. 11,21 Thus, we compared the clinical features and outcomes between these patients. Results showed that patients with IgG4-RKD were older and had a higher incidence of hypertension. The association between renal and retroperitoneal involvement was not observed in our cohort. This nding may indicate that retroperitoneal involvement is not a secondary reaction to IgG4-RKD. In terms of renal histopathological ndings, all patients had TIN. Glomerular changes were observed in one tissue. However, an accurate diagnosis could not be obtained due to 86% of global glomerular sclerosis. All glomerular and tubulointerstitial disease including TIN, membranous nephropathy, immunoglobulin A nephropathy, mesangial proliferative and membranous proliferative glomerulonephritis can be noted in histopathological ndings, and TIN is known as the most common nding. 22,23 Lymphoplasmacytic in ltration with storiform brosis was the most common nding in our cohort. However, there were speci c histological features that can be used to distinguish IgG4-RKD from other TIN in previous studies. 24,25 Hypocomplementemia was an important laboratory nding, particularly in IgG4-RKD, and this nding is in accordance with previous studies. 4,9,26 In general, serum IgG4 does not to bind to complement component 1q and does not activate the classical complement pathway. 5 Thus, the other IgG subtypes, such as IgG1 and IgG3, may be associated with the pathogenesis of IgG4-RD, with consideration of the incidence of hypocomplementemia.
GCs therapy is the gold standard for treatment. However, there is no the international standard guideline for treatment. IgG4-RD responds well to GCs therapy symptomatically, radiologically, and serologically. 5,6 The most common treatment was according to GCs-based therapy in our cohort. The rate of response to GCs-based therapy was 85.3%. However, the relapse rate was 21.3% when GCs therapy is discontinued or the minimal dose of GCs is maintained, and this is in accordance with a previous report. 27 In addition, approximately 10% of patients received the wait-and-see treatment. Among them, one-third had spontaneous remission based on a previous report. 27 In general, GCs therapy is the gold standard treatment. However, it is not curative. In addition, with consideration of spontaneous remission cases, further studies about the pathophysiology of IgG4-RD have to be conducted.
Our study had several limitations. First, this was a retrospective observational study. Second, the diagnosis and treatment strategy for IgG4-RD were inconsistent. Third, because the follow-up duration was short and there was a lack of standard tool for the assessment of disease outcome, some data about treatment response and outcomes were missing. Despite these limitations, the biopsy rate was 78.7%. Moreover, the serum IgG4 concentration of most patients was measured, and all patients underwent imaging studies for IgG4-RD diagnosis.
In conclusion, kidney or other organ involvement is not signi cantly associated with clinical outcomes. Since IgG4-RD has different clinical features, it should be accurately diagnosed, and all physicians must actively diagnose and treat the condition. Hence, it is necessary to establish international diagnostic criteria and treatment guidelines.

Evaluation of clinical features
We extracted data on age at diagnosis, sex, follow-up duration, comorbidities (hypertension, diabetes, autoimmune disease, and malignancy), number of organs involved, type of imaging study, whether or not biopsy was required, IgG4-RD disease grade, and treatment strategies and duration. Information on clinical characteristics was collected from the patient's electronic medical records during the rst hospital visit. Organ involvement was de ned as the presence of lesions con rmed via imaging studies or characteristic pathological ndings (e.g., dense lymphoplasmacytic in ltrate with abundant IgG4-positive plasma cells, storiform pattern of brosis, and obliteration of vein with lymphoplasmacytic in ltrate) of IgG4-RD on biopsy. Meanwhile, multiple organ involvement was de ned as two or more affected organs.
To con rm whether organs were affected, all patients underwent imaging studies, including CT scan, magnetic resonance imaging, and 18 F-FDG/CT scan at the time of diagnosis.

Diagnosis of IgG4-related disease
Patients were diagnosed with IgG4-RD according to the CDC and/or other speci c criteria for each organ, such as the presence of type 1 autoimmune pancreatitis, IgG4-RKD, IgG4-related Mikulicz's disease (IgG4related sialadenitis and dacryoadenitis), and IgG4-related sclerosing cholangitis. [14] The CDC comprises three parts, which are as follows: (a) clinical or radiological features showing diffuse/localized swelling or masses in single or multiple organs; (b) serological examination showing elevated serum IgG4 concentrations (≥ 135 mg/dL); (c) histopathological ndings including marked lymphoplasmacytic in ltrate with brosis and in ltrate of IgG4-positive plasma cells with an IgG4+/IgG+ plasma cell ratio of > 40% and IgG4+ plasma cell/high power eld (HPF) of > 10. In addition, patients were classi ed into three based on the number of ful lled criteria: (1) those who ful lled all three criteria were diagnosed with de nite IgG4-RD, (2) those who ful lled the clinical and histopathological criteria but without elevated serum IgG4 concentrations were diagnosed with probable IgG4-RD, and (3) those who ful lled the clinical and serological criteria but without histopathological features were diagnosed with possible IgG4-RD.

Evaluation of treatment response and outcomes
The standard guideline for the evaluation of treatment response has not yet been established. After treatment initiation, the clinical characteristics or follow-up radiologic ndings were identi ed, and treatment responses were classi ed as follows: (1) Complete response: there were no clinical symptoms, and complete disappearance of organ involvement was observed on follow-up imaging studies.
(2) Partial response: there was improvement in clinical symptoms, and > 50% of organ involvement was treated based on follow-up imaging studies.
(3) Stable disease: symptoms still remained and there was no change in organ involvement based on imaging studies, or there was less than 50% regression.
(4) Progression: new symptoms or organ involvement was found, or previous symptoms and organ involvement worsened.
The disease outcomes were complete remission and relapse after treatment completion or minimal maintenance treatment. Complete remission was de ned as persistent state without symptoms and resolution of organ involvement after treatment within the follow-up period. Relapse was de ned as reappearance of organ involvement or occurrence of new organ involvement after treatment completion.

Statistical analysis
Continuous variables (including demographic and clinical characteristics) with a normal distribution were presented as mean ± SD and those with a non-normal distribution as median (IQR). Categorical variables were expressed as frequency and percentage. The chi-square test was performed to compare different groups and categorical variables. Meanwhile, the Fisher's exact test was used if the assumptions required for parametric testing were not met. The student's t-test and the Kolmogorov-Smirnov t-test were utilized to compare continuous variables with a normal distribution. The Mann-Whitney U-test was applied to compare variables with a non-normal distribution. A P value of < 0.05 was considered statistically Patients, n (%)