Background
Immunoglobulin G4-related disease (IgG4-RD) is an immune-mediated condition characterized by dense lymphoplasmacytic infiltrates, storiform fibrosis, frequent elevations of serum IgG4 and good response to steroid therapy [
1]. Almost every organ in the body could be affected in this systemic disease, and salivary glands are among the most commonly involved sites, a condition known as immunoglobulin G4-related sialadenitis (IgG4-RS) [
2‐
4].
Glucocorticoids are regarded as the first-line agents for inducing remission of IgG4-RS. However, the optimal therapeutic strategy has not yet been established [
5]. Relapses are common both during and after treatment, and therefore, maintenance treatment seems essential [
6]. Steroid-sparing agents, such as azathioprine, mycophenolate mofetil and cyclophosphamide, have also been used to treat IgG4-RD, but the optimal timing of their administration is controversial, and their outcomes are unclear [
5,
7]. Xerostomia is commonly reported in IgG4-RS [
8], but the long-term outcomes of the secretory function have not yet been studied. The diagnostic value of serum IgG4 level is controversial, and its role in predicting disease activity is unclear [
9].
To resolve the aforementioned issues, we conducted a study to determine the short-term and long-term outcomes of patients with IgG4-RS treated with glucocorticoids combined with steroid-sparing immunosuppressive agents. We analyzed secretory function and the serological and radiological changes, and evaluated the usefulness of serum IgG4 level as an indicator of disease activity.
Methods
The study protocol was approved by the Ethics Committee for Human Experiments of Peking University School of Stomatology. Informed consent was obtained from all the patients included in this study.
Patients
The study included all patients with confirmed IgG4-RS who were diagnosed, treated and followed up for more than 3 months at the Department of Oral and Maxillofacial Surgery, Peking University School of Stomatology between August 2011 and April 2017. IgG4-RS was diagnosed based on the comprehensive diagnostic criteria for IgG4-RD [
5]: (a) persistent (>3 months) swelling of single or multiple major salivary glands; (b) serum IgG4 concentration >1350 mg/L; (c) histopathological and immunohistochemical examinations of submandibular gland biopsy specimens showing marked lymphocyte and plasmacyte infiltration, fibrosis and infiltration of IgG4+ plasma cells with IgG4+/IgG+ cells >40% and IgG4+ plasma cells >10/high-power field; and (d) exclusion of other diseases that present with glandular swellings, such as sarcoidosis and lymphoproliferative disease. Patients who suffered from contraindications to glucocorticoid treatment or had history of glucocorticoid treatment were excluded from the study. The patients’ basic information and medical history were recorded. Serological tests, salivary gland function assessment and computed tomography (CT) were performed before treatment and during follow up.
Serological tests
Serum IgG4 levels were measured using the Array 360 Immunoassay Assay Protein Serology Chemistry Analyzer system (Beckman Coulter, Fullerton, CA, USA).
Salivary gland function assessments
The degree of subjective oral dryness was assessed using the summated xerostomia inventory (SXI) [
8,
10]. Saliva flow rates were calculated by collecting the whole saliva samples at rest and under stimulation with 2.5% citric acid solution for 5 minutes each [
11]. Scintigraphy with
99mTc-pertechnetate was performed according to a standardized protocol [
12]. Both the parotid and submandibular glands were evaluated and the secretion index (SI) was calculated (Additional file
1).
CT assessments
CT scanning was performed using an eight-slice scanner (BrightSpeed; GE Medical Systems, Waukesha, WI, USA). The CT data in digital imaging and communications in medicine (DICOM) format were imported to iPlan CMF (BrainLAB, AG, Germany). The margins of the right and left parotid and submandibular glands were marked with this software, and then reconstructed using volume rendering [
13]. The images were reconstructed three times, and the mean gland volume and mean CT value of the bilateral glands were recorded.
Treatment regimen
The preferred treatment regimen was intravenous methylprednisolone at a dose of 200 mg/day for 3 days and 40 mg/day for another 3 days, followed by 0.6 mg/kg/day oral prednisone for 2 weeks, tapered by 5 mg every 2 weeks until a daily dose of 30 mg was reached and then tapered by 2.5 mg every 2 weeks to reach a maintenance dose of 5 mg/day. In some patients, only oral prednisone was administered at the initial dose of 0.6 mg/kg/day for 4 weeks before the gradual tapering. The preferred initial immunosuppressive agent for combination therapy was cyclophosphamide at a dose of 400 mg once every 2 to 4 weeks; other immunosuppressive agents, including azathioprine and leflunomide, could also be used at doses of 2 mg/kg/day and 10 mg/day, respectively. Mycophenolate mofetil at 250 mg three times a day (TID) was used as the initial immunosuppressive agent in children.
Treatment response assessments
Treatment response was assessed every 3 to 6 months using a comprehensive evaluation strategy. The short-term outcomes were collected at 3 months and the long-term outcomes were collected in patients treated for more than 12 months. The following parameters were evaluated:
(a)
Sizes of the salivary glands: physical examinations of the major salivary glands were performed at each follow up as the basic evaluation, and CT assessment of the volumes and mean CT values of the parotid and submandibular glands were performed at 3 months for the first time and then once every 6 to 12 months, or when relapse was suspected during follow up. In comparison to the pretreatment volumes, reductions ≥30%, ≥15% but <30%, and <15% in the short term indicated complete, partial, and poor remission, respectively. Increases ≥30%, ≥15% but <30%, and <15% compared to the short-term value, or the value at 6 months if data from the 3-month visit were absent, indicated complete, partial, and minor relapse of the salivary glands, respectively.
(b)
Function of salivary glands: summated xerostomia inventory (SXI) scores were recorded as the subjective evaluation of xerostomia symptoms. Saliva flow rate was analyzed as the secretion index (SI) for objective evaluation.
(c)
Serological examinations: serum IgG4 levels were tested at each follow up. A sustained increase >20% during follow up as compared with the short-term level was considered to be evident increase.
(d)
Extra-salivary involvement assessment: symptoms such as nasal obstruction or allergic rhinitis, audition decrease, and abdominal pain were noted. Physical and imaging examination of the lacrimal gland, nasal cavity, lymph nodes, and pancreas, etc. was performed.
Clinical remission was defined as complete or partial remission on CT, and improvement in extra-salivary involvement. Clinical relapse was defined by re-enlargement of salivary glands or new appearance or reappearance of extra-salivary involvement after clinical remission. Patients with obvious aggravation of subjective symptoms such as xerostomia or evident increase in serum IgG4 during follow up were suspected to be relapsing and in need of further evaluation. However, isolated increase in serum IgG4 did not indicate clinical relapse, but was defined as a serologically unstable condition.
Statistical analysis
Statistical analyses were performed using IBM SPSS Statistics 22 (IBM, Armonk, NY, USA) and Statistical Analysis System software 9.3 (SAS Institute, Cary, NC, USA). Kolmogorov-Smirnov tests were performed before parametric tests and serum IgG4 concentrations in mg/L were log-transformed to approximate normality. Statistical differences were analyzed using the Student t test and paired t test, Mann–Whitney U test and chi-square test. Multiple linear regression analysis was used to analyze the association between serum IgG4 and parotid and submandibular gland volumes, and Kaplan–Meier analysis was used to evaluate the relapse-free survival rate. P < 0.05 was considered to indicate significant differences.
Discussion
For the treatment of IgG4-RD, a recent systematic review showed that “no therapies” were carried out in 36.6% of patients with IgG4-RS [
7]. Despite the possibility of spontaneous remission, long-term data on the “wait and see” strategy are absent. In our series, spontaneous or temporary remission had not been reported by any patient. Our previous study revealed a lower secretory capacity in patients with higher grades of histopathological fibrosis and inflammation [
11]. Negative correlation between the increase in salivary flow rate after treatment and the disease duration before treatment has also been reported [
14]. It seems that most of the patients with IgG4-RS could not benefit from the “wait and see” strategy in aspect of salivary gland function preservation. On the other hand, remarkable improvement was noted in saliva secretion according to both subjective and objective parameters in our study, and the condition remained stable during the tapering and maintenance periods. Despite there being no statistical difference among patients with different histopathological stage of disease (data not shown), which may be related to the small sample size, we believe that the more severe the fibrosis, the less likely it is that functional glandular tissue exists. Thus, to ensure better recovery of salivary gland function and to minimize the impairment to the whole body, we suggest that treatment be started as soon as the diagnosis is definite.
It is widely accepted that glucocorticoids are the first-line agents for remission induction, and initial remission could be achieved in the majority of patients [
15]. Despite the recommendation of an initial dose of oral prednisone of 30–40 mg/day for 4 weeks, there is no controlled study on the optimal treatment regimen, including the starting dose, the tapering protocol and the timing of glucocorticoid discontinuation [
7,
15]. Although intravenous steroid treatment has seldom been reported in IgG4-RD [
16‐
18], we prefer a full-dose intravenous steroid therapy protocol in all patients without high risk of side effects. With this regimen, all symptoms and signs of IgG4-RS obviously improved after the second day of treatment. In addition, the dose in our regimen was lower than the traditional steroid pulse therapy, and this may reduce the dose-dependent toxic effect on hepatocytes and other side effects. According to our experience in the treatment of other autoimmune diseases, steroid pulse therapy may allow more rapid tapering of oral glucocorticoids and lower daily maintenance dose. This has long-term benefits, which are probably attributable to effects on both genomic and non-genomic pathways during the intravenous pulse therapy [
19,
20]. In our cohort, only four patients were treated without intravenous pulse therapy and the follow-up period was not sufficiently long. As a result, the long-term outcome may not be comparable. A well-designed clinical trial is needed to further evaluate the effectiveness of the full-dose intravenous steroid therapy versus the traditional oral prednisone treatment strategy in IgG4-RD.
The application of steroid-sparing immunosuppressive agents in IgG4-RS is controversial. Some specialists disagree with the addition of immunosuppressive agents at the beginning, while other experts think it is acceptable in some patients [
15]. Several studies have reported that steroid-sparing agents seem unable to significantly reduce the relapse rate of IgG4-RD [
21], while Fei et al. and Gupta et al. reported favorable responses in patients with IgG4-RD treated with glucocorticoids and steroid-sparing immunosuppressive agents [
22,
23]. In our series, no difference was observed between the groups treated with or without steroid-sparing agents. However, the size of the group treated with steroid alone was rather small and the duration of follow up was not sufficiently long. The annual relapse rate was 8.35% in regularly treated patients, lower than the rate of 11.5% reported by Yamamoto et al. [
6]. However, our low annual relapse rate could also be related to the dose adjustments made in response to increased serum IgG4 levels. Additionally, the use of immunosuppressive agents helped to minimize the maintenance dose of prednisone, which helps reduce the long-term toxicity of glucocorticoids [
19]. Yamamoto et al. reported that only 52.8% patients were treated with < 5 mg/day prednisolone for the maintenance of clinical remission [
6], while in our study, the percentage was 74.1%. Cyclophosphamide was chosen as the preferred immunosuppressive agent as it is a classic and widely used immunosuppressive agent that is mainly related to B cell activation [
24]. A recent randomized controlled trial (RCT) on IgG4-RD also showed that glucocorticoids combined with cyclophosphamide treatment had a better effect and lower relapse rate than steroid monotherapy during the observation period of 12 months [
22]. The long-term toxicity of cyclophosphamide had not been observed in our study. However, further clinical trials and longer-term follow up are necessary, and the efficacy of different immunosuppressive agents needs to be compared. It has been suggested that B cell depletion, such as rituximab, could be effective in many refractory conditions, and XmAb5871, a specific plasmablast-targeted monoclonal antibody, is currently in phase II development for IgG4-RD treatment [
25‐
27]. Targeted medications might be used for the treatment of IgG4-RD in the future. However, as all the patients in our series had acceptable outcomes without severe side effects, and because of financial considerations, rituximab has not yet been recommended for our patients.
In our study, six patients stopped medication by themselves, and relapse was observed in all of them, indicating the high risk of relapse after glucocorticoid discontinuation. Similar results were also reported in autoimmune pancreatitis and IgG4-related cholangitis [
28,
29]. Thus, the therapeutic course of IgG4-RS is controversial, and the discontinuation of glucocorticoids is questionable. Several clinicians recommend maintenance therapy for up to 3 years. However, the relapse rate still seems to be high [
6,
15]. Furthermore, whether or not immunosuppressive agents or monoclonal antibodies can replace glucocorticoids remains unknown.
Because of the lack of an established standard for the evaluation of therapeutic outcomes, the responses to treatment in different studies may not be comparable [
7]. Although an IgG4-RD responder index has been developed for systemic evaluation and has good consistency with physician global assessment [
30], the specific standard for assessing each single organ remains unclear. Mikulicz’s Disease Assessment Questionnaire has been used for outcome evaluation, but it is based on subjective judgment [
6]. In this study, we came up with a criterion mainly based on the variation in salivary gland volume on CT, which is quantitative, and much more sensitive and objective than physical examination. A similar method has been used in the evaluation of the outcomes in the lacrimal glands in IgG4-RD, though it did not involve CT volume rendering [
31]. We set 15% and 30% as the cutoff values for partial and complete remission/relapse, respectively, for both the parotid and submandibular glands, and found good consistency with clinical observations in the submandibular glands and in parotid glands with high CT values. Therefore, we believe that therapeutic evaluation should be systematic and organ specific.
Though being an important feature of IgG4-RD, the significance of serum IgG4 concentration is ambiguous. On one hand, serum IgG4 is an item for consideration in the diagnostic criteria for IgG4-RD; on the other hand, several studies have demonstrated its poor diagnostic utility [
9,
32]. Higher baseline serum IgG4 could be related to wider systemic involvement, greater risk of relapse and shorter time to relapse [
4,
33,
34], indicating its role in revealing the general condition of the body and predicting the prognosis of the disease. In this study, multiple linear regression revealed strong association between serum IgG4 and parotid and submandibular gland volume, suggesting the significance of serum IgG4 concentrations in reflecting volumes in the salivary glands, and subsequently the disease activity. The variation in salivary gland size could be slight or undetectable by physical examination or traditional radiological imaging, but the serological parameter seems to be more sensitive. In the four patients with obvious serum IgG4 elevation, a further increase in serum IgG4 and salivary glands volumes on CT occurred during follow up. Based on these results, we suggest that serum IgG4 concentration may be useful in predicting the activity of the disease, and dose adjustments could be considered when serum IgG4 increases significantly, even if clinical relapse is absent.