Differences in Immune Indicators and Prognosis Between IgG4-Positive and Negative Lacrimal Gland Benign Lymphoepithelial Lesion

Purpose: The differences in immune indicators and prognosis between IgG4-positive and negative lacrimal gland benign lymphoepithelial lesion (LGBLEL) were analyzed. Methods: This was a single-center retrospective clinical study. Clinical data of 146 patients with LGBLEL were collected from June 2011 to June 2019. Results: The age, preoperative glucocorticoid history, and serum C3, C4, IgG, IgG2, IgG4 had statistical difference between the IgG4-positive and negative groups (P<0.05). The expression levels of IgG and IgG4 in the IgG-positive group with preoperative glucocorticoid therapy were lower than those in the IgG4-negative group without preoperative glucocorticoid therapy (P=0.021 and P=0.013). The 5-year recurrence-free cumulative percentages of IgG4-positive group was 81.85%, and 83.46% in the IgG-negative group, which had no statistical difference (P=0.216). The history of preoperative glucocorticoid therapy, serum C4, IgG1 and IgG2 were the factors affecting IgG4-positive LGBLEL’ recurrence, while the history of preoperative glucocorticoid therapy, serum C4, and IgG1 were the factors affecting LGBLEL’ recurrence (P<0.05). Conclusion: Serum C3, C4, IgG, IgG2, IgG4 had statistical difference between the IgG4-positive and negative LGBLEL. The history of preoperative glucocorticoid therapy, serum C4 and IgG1 were the factors affecting LGBLEL’ recurrence, while the expression level of IgG4 was not the factor affecting LGBLEL’ recurrence. 0.05). Both two groups of lacrimal gland enlargement and eyelid IgG4-positive group mainly as sinus mucosa thickening extraocular muscle thickening and nerve thickening while IgG4-negative group mainly manifested as sinus mucosa thickening (12.2%), and extraocular muscle thickening (14.6%). There were no statistically signicant differences in imaging ndings between the two groups (P > 0.05). The mean course of disease was 18.70 ± 19.16 months in the IgG4-positive group and 21.22 ± 25.87 months in the IgG4-negative group, with no statistical difference (P > 0.05).


Introduction
Lacrimal gland benign lymphoepithelial lesion (LGBLEL) is a kind of benign lesion with diffuse in ltration of lymphocytes into lacrimal gland, leading to atrophy of gland parenchyma and proliferation of myoepithelial basal cells [1][2] . The main manifestations of the disease are painless eyelid swelling and lacrimal gland swelling. Current studies have shown that the occurrence of LGBLEL may be related to the increase of IgG4 level and the disorder of estrogen level [3][4] . IgG4-related disease (IgG4-RD) is an immune-mediated systemic disease characterized by diffuse in ltration of plasma cells expressing positive IgG4, which results in swelling of the diseased tissue accompanied by a signi cant increase in serum IgG4 [5] . IgG4-ROD is a subtype of IgG4-RD, IgG4-related ocular disease (IgG4-ROD) can involve all orbital tissues, the most common of which is the lacrimal gland [6] . The incidence of IgG4-ROD accounts for 4-34% of IgG4-RD and 25% of all instances of orbital lymphoproliferative disease [7][8] .

Laboratory data collection
Peripheral venous blood samples of IgG4-positive and negative LGBLEL were collected using the enzymelinked immunosorbent method (ELISA) to test for related indicators, including complement C3, C4, and rheumatoid factor (RF), c-reactive protein (CRP), anti-streptolysin O (ASO), immunoglobulin A (IgA), IgM, and IgG and its subtypes (IgG1, IgG2, IgG3, and IgG4). The laboratory indicators of the two groups were compared and analyzed.

Treatment and prognosis
All patients were treated with partial surgical excision and glucocorticoid therapy. According to the volume of the lesion, either a subarcuate skin incision or a double eyelid skin incision on the anterior orbital temporal eyebrow was selected for the surgical approach. The subcutaneous tissue was separated to reach the orbital margin, and the orbital partition was cut into the orbit. The lesion was removed and histopathological examination was conducted. The patients were given glucocorticoids 80-120 mg/d for 3 days after the operation, which was then changed to methylprednisolone tablets 24-28 mg/d. The dosage was reduced by 1 tablet for 1-2 weeks until the withdrawal of the drug. The course of treatment was 1.5-3 months.
The effective follow-up time was from the de nitive diagnosis of the rst biopsy to the death of the patient or June 2020. The following observation indicators were collected during follow-up. (1) General conditions: vision, eyelid swelling, proptosis, and others. (2) Imaging ndings: in order to objectively evaluate the therapeutic effect and observe recurrence, preoperative and postoperative MRI examination was necessary. The speci c time points were pre-operation, half a year after operation, 1-3 years after operation, and 5 years after operation. The criteria for recurrence include eyelid swelling and an MRI showing lacrimal gland enlargement. If the patient had contraindications to MRI, CT can be used instead.

Statistical analysis
Graphpad Prism 8.0 and SPSS 25.0 software were used for analysis. Measurement data were tested by One-sample Kolmogorov-Smirnov test. Mean± standard deviation and independent sample t test were used to test the data of two groups consistent with normal distribution. Median and non-parametric rank sum test were used for data that did not conform to normal distribution. The chi-square test or Fisher's exact test were used for counting data. Survival curves for recurrence were created using the Kaplan-Meier method and compared between groups using log-rank tests. The in uencing factors were analyzed by binary logistic regression analysis. A P value <0.05 was considered statistically signi cant.

Results
Higher expressed of age and more preoperative glucocorticoid therapy cases in the IgG4-positive group than IgG4-negative group A total of 146 patients with biopsy-proven LGBLEL were ultimately enrolled in the study, and 105 (71.9%) patients were placed in the IgG4-positive group on the basis of pathological analysis. The results ( Table 1) showed that the male-female ratio of the IgG4-positive group was 1:2.8, the ratio in the IgG4negative group was 1:4.9, and there was no signi cant difference between the two groups (P = 0.283). The mean age of IgG4-positive group was 50.10 ± 14.23 years old, and that of the IgG4-negative group was 44.76 ± 11.43 years old, which was statistically signi cant (P = 0.033). There were 14 cases of right eye lesions, 24 cases of left eye lesions, and 67 cases of binocular lesions in the IgG4-positive group, and 11 cases of right eye lesions, 12 cases of left eye lesions, and 18 cases of binocular lesions in the IgG4negative group, with no statistically signi cant difference between the two groups (P = 0.059). 24 cases of preoperative glucocorticoid therapy, 3 cases of operation history, 3 cases of asthma, 27 cases of sinusitis, 4 cases of history of lymph node enlargement, and 3 cases of other immune system diseases in the IgG4-positive group. 5 cases of sinusitis in the IgG4-negative group. There were statistically signi cant differences in preoperative glucocorticoid therapy between two groups (P = 0.000).
The main clinical manifestations of the IgG4-positive group were eyelid swelling in 96 cases, proptosis in 2 cases, and eyelid mass in 7 cases. The main clinical manifestations of the IgG4-negative group were eyelid swelling in 33 cases, eyelid mass in 2 cases, and proptosis in 6 cases. Eyelid swelling was the main clinical manifestation in both groups. The simultaneous symptoms of the IgG4-positive and negative group included tearing, pain, dry eyes, and decreased vision. There were no statistical differences in main and simultaneous symptoms between the two groups (P > 0.05). Both two groups showed signs of lacrimal gland enlargement and eyelid swelling. IgG4-positive group mainly manifested as sinus mucosa thickening (25.7%), extraocular muscle thickening (15.2%), and nerve thickening (3.8%), while IgG4-negative group mainly manifested as sinus mucosa thickening (12.2%), and extraocular muscle thickening (14.6%). There were no statistically signi cant differences in imaging ndings between the two groups (P > 0.05). The mean course of disease was 18.70 ± 19.16 months in the IgG4-positive group and 21.22 ± 25.87 months in the IgG4-negative group, with no statistical difference (P > 0.05). No recurrence 77 35 Natural death 1 0 Note: "#" represents chi-square test, "&" represents t test, "^" represents non-parametric rank sum test and "*" represents Fisher's exact test. P < 0.05 is considered to indicate statistical signi cance. NA: Not Applicable.
The expression levels of IgG and IgG4 in the IgG4-positive LGBLEL group with preoperative glucocorticoid therapy history were lower than those in the group without preoperative glucocorticoid therapy history As was shown in Table 2, compared with IgG4-positive LGBLEL group without preoperative glucocorticoid therapy history, the expression levels of IgG and IgG4 in IgG4-positive LGBLEL group with preoperative glucocorticoid therapy history were lower (P = 0.021 and P = 0.013), while the expression levels of C3, C4, IgA, IgM, RF, ASO, CRP, IgG1, IgG2 and IgG3 had no difference (P > 0.05). follow-up and natural death cases, we contrastively analyzed the recurrence-free survival curve between IgG4-positive and negative LGBLEL. The results showed that the recurrence-free cumulative percentages at 5 years were about 81.85% and 83.46%, respectively, which had no statistical difference in recurrencefree cumulative percentage between the two groups (P = 0.216) (Fig. 1).
The preoperative glucocorticoid therapy history, serum C4, and IgG1 may be factors in uencing the recurrence of LGBLEL The cases of loss to follow-up and natural death were excluded. Binary logistic regression analysis was performed on 136 patients with LGBLEL. Group (IgG4-positive and negative), gender, age, laterality, preoperative glucocorticoid therapy history, operation history, asthma, sinusitis, lymph node enlargement, other immune system disease history, nerve thickening, extraocular muscle thickening and laboratory indicators were analyzed in Table 3. The results showed that preoperative glucocorticoid therapy history, serum C4, and IgG1 may be factors in uencing the recurrence of LGBLEL (P = 0.003, P = 0.003, P = 0.017, respectively).

Discussion
In recent years, the prevalence of IgG4-ROD has been increasing gradually, which has attracted extensive attention from ophthalmologists. According to a study in China, the incidence of IgG4-ROD accounted for 60% of idiopathic orbital in ammatory disease. A study in Japan showed that the incidence of IgG4-ROD is about 61.5%, with 52.4% incidence in the United States, and 45.8% incidence in South Korea [12][13][14][15] .
LGBLEL with IgG4 positive expression was classi ed as IgG4-ROD. The results of this study showed that LGBLEL was more common in female, both in IgG4-positive and negative groups. The lacrimal gland in ammation could be in uenced by the expression level of estrogen. Jiang et al. [16] found that estrogen and its receptor might inhibit the in ammatory response in rat colon tissues, suggesting that a decrease in estrogen may be one of the relevant factors leading to the occurrence of in ammation. In addition, there was a signi cantly difference in age between the two groups, and the mean age of IgG4-positive LGBLEL was older. This may indicate that levels of estrogen in female patients gradually decline with age, increasing the incidence of IgG4-positive LGBLEL.
IgG4-ROD may be associated with immune system diseases, such as allergic rhinitis, asthma, lymph node enlargement, and Sjogren's syndrome [17][18] . In our study, LGBLEL had a history of immune system disease, among which allergic rhinitis, sinusitis, and lymph node enlargement were common. The imaging ndings were primarily lacrimal gland enlargement, accompanied by sinus mucosa, extraocular muscle, and ocular nerve thickening. In this study, about 15.2% of patients in IgG-positive group presented with extraocular muscle thickening and 3.8% of patients presented with ocular nerve thickening, while 14.6% of IgG4-negative group presented with ocular muscle thickening and no case presented with ocular nerve thickening, which showed no signi cant difference between the two groups. These results may suggest that extraocular muscle and ocular nerve thickening are not speci c indicators to differentiate IgG4-positive and negative LGBLEL.
Immunoglobulin has the functions of activating complement, absorbing cells, and extracellular killing.
Compliment is a type of protein that mediates autoimmune and in ammatory responses. Studies have found that IgG can activate complement C3 through the classical pathway and IgG4 can activate complement C3 through the bypass pathway to play an immunomodulation role, while the complement system can play an anti-infection role, regulate in ammatory response and maintain immune homeostasis through the massive consumption of C3 and C4 [19][20][21] . Some studies had found that the expressions of serum C3, C4, IgG1, IgG2, IgG3, IgE, and CRP, were signi cant for the diagnosis of IgG4-ROD [22][23] . Chen [12] et al. found that compared with other orbital idiopathic in ammatory disease, serum C3 of IgG4-ROD were signi cantly decreased. In our study, relevant immune indicators were systematically analyzed, and the results showed that C3, C4, IgG, IgG2, and IgG4 were of great signi cance for the differentiation between IgG4-positive and negative LGBLEL, while there was no signi cant difference in the expressions of RF, ASO, CRP, IgA, IgM, IgG1 and IgG3. Compared with IgG4negative LGBLEL, the expression levels of complement C3 and C4 were decreased, while those of IgG, IgG2, and IgG4 were increased in the IgG4-positive group, which had differential signi cance. And preoperative glucocorticoid therapy could reduce IgG and IgG4 expression in IgG4-positive LGBLEL.
At present, partial surgical resection combined with glucocorticoid therapy is the main treatment for LGBLEL, and more patients still relapse after treatment. Chen [12] et al. reported that the recurrence rates of IgG4-positive and IgG4-negative idiopathic orbital in ammatory disease were 32.2% and 19.1%, respectively. A French study showed that about 2/3 of IgG4-ROD cases relapsed, and Suimon et al. found that the recurrence rate was about 33.3% [22,24] . Owing to differences in race, sample size, follow-up time, and sensitivity to glucocorticoids, the recurrence rate of IgG4-ROD differed across some research centers, but mostly ranged from 18-58% [25][26][27] . In our study, the recurrence rate of 5-year in IgG4-positive and negative LGBLEL patients was 18.15% and 16.54%, respectively. There was signi cant difference in recurrence rates between the two groups.
All patients were treated via the partial surgical resection combined with glucocorticoid therapy. The prognostic factors of LGBLEL were statistically analyzed, and lost to follow-up and natural death cases were excluded. The results showed that preoperative glucocorticoid therapy history, serum C4, and IgG2 were the factors affecting the recurrence of LGBLEL. Studies have found that C3 and C4 levels are signi cantly decreased during active in ammation, while IgG and its subtypes are increased [12,22−24] . In patients with active IgG4-ROD, a large number of immune complexes will be produced, leading to elevated levels of IgG and its subtypes, thus activating the complement system. However, complement C3 and C4 can be consumed through classical and bypass pathways to eliminate immune complexes, leading to a decreased level of C3 and C4. Therefore, patients with increased C4 expression and decreased IgG2 expression may have a lower risk of recurrence.
Kubota [28] et al. found that extraocular muscle and trigeminal nerve thickening could affect the therapeutic effect of glucocorticoids and the prognosis of IgG4-ROD. Previous studies had shown that serum IgG4, RF, and male gender were risk factors for recurrence of IgG4-ROD treated with glucocorticoid therapy [29][30][31] . Our study showed that gender, extraocular muscle thickening, nerve thickening, sinusitis, IgG4 and RF were not risk factors for recurrence of LGBLEL. And the preoperative glucocorticoid therapy history, serum C4, IgG1 and IgG2 were the factors affecting the recurrence of IgG4-positive LGBLEL.
There were some shortcomings in this study. First, this was a retrospective study, which presented di culties to the complete collection of research data. Second, there were no standardized criteria for treatment, and the dose of glucocorticoid depended on the disease condition, which had an impact on the prognosis of LGBLEL. In addition, the follow-up time also in uenced prognosis of LGBLEL. Some LGBLEL with IgG4 positive expression were treated with glucocorticoid before operation, which had a certain effect on the expression of immunological indexes.
In summary, this study analyzed the clinical characteristics, immune indicators, and prognosis of IgG4positive and negative LGBLEL. The expression of C3 and C4 was decreased, while the expression of IgG, IgG2, and IgG4 was increased, which had signi cance for distinguishing between IgG4-positive and negative LGBLEL. Preoperative glucocorticoid therapy, serum C4, IgG1 and IgG2 could in uence the recurrence of LGBLEL, and the expression of IgG4 was not a major prognostic factor.

Authors' contributions
Rui L analyzed and wrote the manuscript; Nan W, Jinjin W and Jing L helped collect data; Jianmin M, Xin G and Jingxue Z read and criticized the manuscript. All authors critically read and edited the manuscript.
All authors read and approved the nal manuscript.

Figure 1
Comparative analysis of prognosis between IgG4 positive and IgG4 negative LGBLEL. The recurrence-free cumulative percentages at 5 years in IgG4 positive and IgG4 negative dacryoadenitis was 81.85% and 83.46%, respectively (P=0.216).