A subset of ocular adnexal marginal zone lymphomas may arise in association with IgG4-related disease

We previously suggested a relationship between ocular immunoglobulin (Ig)G4-related disease (IgG4-RD) and marginal zone lymphomas (MZLs). However, the cytokine background associated with these disorders and whether it differs between ocular adnexal MZLs with (IgG4-associated MZL) and without (IgG4-negative MZL) numerous IgG4+ plasma cells are unknown. In this study, we identified the mRNA expression pattern of Th2 and regulatory T-cell (Treg) cytokines in IgG4-RD and in IgG4-associated MZL and IgG4-negative MZL using real-time polymerase chain reaction analysis. Ocular IgG4-RD and IgG4-associated MZL exhibited significantly higher expression ratios of interleukin (IL)-4/β-actin, IL-10/β-actin, IL-13/β-actin, transforming growth factor (TGF) β1/β-actin, and FOXP3/β-actin than did IgG4-negative MZL (p < 0.05). This finding further supports our prior observations that a significant subset of ocular MZLs arises in the setting of IgG4-RD. Furthermore, the presence of a different inflammatory background in IgG4-negative MZLs suggests that IgG4-associated MZLs may have a different pathogenesis.

a background of IgG4-RD 1 . However, the expression pattern of cytokines in MZL lesions with IgG4 + plasma cells has not been clarified.
In this study of ocular IgG4-RD and MZLs with (IgG4-associated MZL) and without (IgG4-negative MZL) numerous IgG4 + plasma cells, we aimed to identify the mRNA expression patterns of Th2 and Treg cytokines and to determine the inflammatory background associated with benign and neoplastic ocular lymphoplasmacytic proliferations with numerous IgG4 + plasma cells that is distinct from that associated with ocular IgG4-negative MZL.

Material and Methods
Samples and clinical review. Formalin-fixed excisional biopsies from the ocular adnexal region of patients were selected, including 11 patients with IgG4-RD, 11 with IgG4-negative MZL, and 6 with IgG4-associated MZL ( Table 1). All MZL lesions were primary tumors, and there was no other organ involvement. None of the patients were treated prior to the biopsy. Clinical data including serum IgG4 and IgG levels were obtained when available. The IgG4 and IgG levels were measured by routine laboratory blood tests. Informed consent for the use of their samples in research was obtained from all patients. Real-time quantitative polymerase chain reaction (PCR). The following evaluations were carried out in accordance with the approved guidelines. All experimental protocols were approved by the Institutional Review Board at Okayama University. Total RNA was extracted from the paraffin-embedded sections of all samples using an miRNeasy FFPE Kit (QIAGEN, Valencia, CA, USA). cDNA was prepared using a SuperScript VILO MasterMix kit (Invitrogen, Carlsbad, CA, USA). Multiplex real-time PCR was performed for quantitative analysis, according to a standard protocol, using TaqMan Gene Expression Assays (Applied Biosystems, Foster City, CA, USA), a Step One Plus Real-Time PCR System (Applied Biosystems), and specific primers and probes for FOXP3 (Hs01085834_m1), transforming growth factor (TGF)β 1 (Hs00998133_m1), interleukin (IL)-4 (Hs00174122_m1), IL-5 (Hs00174200_m1), IL-10 (Hs00961622_m1), IL-13 (Hs00174379_m1), and β -actin (Hs99999903_m1) (Applied Biosystems). The PCR cycling conditions were as follows: 20 s at 95 °C, 50 cycles of 1 s at 95 °C, and 20 s at 60 °C. The expression of each target was normalized to that of β -actin, which was used as an endogenous control.
Histological examination, immunohistochemistry, and in situ hybridization. Serial sections (4 μ m) were cut from the block of paraffin-embedded tissue, stained with hematoxylin and eosin, and used for the following immunohistochemical stains: CD20 ( Molecular genetic analysis. PCR molecular genetic analysis for immunoglobulin heavy chain gene rearrangements was performed as previously described [18][19][20] . The primers used in this study were:

Discussion
The Th1/Th2 balance is considered important for healthy immune responses, and a Th1/Th2 imbalance can cause immune-mediated disease. Th1-dominant immune responses are noted in patients with rheumatoid arthritis, type 1 diabetes, and multiple sclerosis, whereas Th2-dominant immune responses are observed in patients with type 1 hypersensitivity disorders, such as allergies or asthma 21 . Extranodal MZLs are considered to arise in a background of chronic inflammation. High levels of Th1-type cytokines (interferon [IFN]-γ and IL-2) are typically found in the tumor environment of many mucosa associated lymphoid tissue (MALT) lymphomas 22,23 . Moreover, most extranodal MZLs express high levels of CXCR3, which is the receptor for IFN-γ -induced chemokines [24][25][26] . In contrast, a major subset of cutaneous MZLs typically shows heavy chain class switching and has a Th2 background 14,15 . In addition, 39% of cutaneous MZLs with plasmacytic differentiation are reported to be IgG4 + , whereas an ocular MZL was the only non-cutaneous IgG4 + MZL found in the same study 16 .
Recently, high levels of Th2 cytokines (IL-4 and IL-13) and Treg cytokines (TGFβ 1 and IL-10) have been detected in tissues with IgG4-RD 12 . Costimulation with IL-4 and IL-10 is suggested to cause class switching to IgG4 27 . Furthermore, upregulation of IL-13 induces eosinophil infiltration, and TGFβ 1 is considered to cause fibrosis in IgG4-RD lesions. In this study, we found higher levels of Th2 and Treg cytokines (including FOXP3) in lesions from patients with either IgG4-RD or ocular IgG4-associated MZL than in samples from patients with ocular adnexal IgG4-negative MZL. However, the expression levels of IL-5, which is a Th2 cytokine that promotes eosinophil infiltration, were not significantly different between sample types. These results indicate that IgG4-associated MZL is characterized by the  There is a diffuse proliferation of small-to medium-sized lymphoid cells (A,B) that are CD20 + (C) and CD3 − (D). Numerous Igκ + plasma cells are present (E), but only very few Igλ + plasma cells (F). Some IgG + (G) and IgG4 + (H) cells are present but the IgG4 + /IgG + cell ratio is < 40%. upregulation of Th2 and Treg cytokines, similar to IgG4-RD. Despite sharing a similar inflammatory background with class-switched cutaneous MZLs, and very much unlike most other MALT lymphomas, the ocular cases, with one exception, had IgG4 + plasma cells that were polytypic and not part of the neoplasm. An explanation for this difference remains to be established, and the heavy chain expressed by the ocular lymphomas is uncertain.
The mechanism for the development of ocular adnexal MZL is not clear. However, recent reports indicated that Chlamydia psittaci infection is associated with the development of ocular adnexal MZL [28][29][30] . We previously suggested that ocular adnexal MZLs may arise in the setting of ocular IgG4-RD 1 , and others have reported cases of ocular adnexal lymphoma with IgG4 + cells, suggesting that they arose from IgG4-RD 31 . Moreover, a case involving two types of mass lesions in the same ophthalmic region has also been reported 32 . One of the lesions was typical of IgG4-RD, and the other was an MZL admixed with IgG4 + plasma cells. Because the two lesions were adjacent to each other, the MZL may have arisen in the setting of the IgG4-RD lesion. These reports and the results of the present study suggest that a subset of ocular MZLs may arise in the setting of IgG4-RD.
Patients with IgG4-RD are considered to be at high risk of MZLs as well as other malignancies, with malignancy considered a possible complication of IgG4-RD. Lung and colon cancers, as well as malignant lymphomas, are reported to occur in 10.4% of patients with IgG4-RD, an incidence that is approximately 3.5 times higher than that in the general population 33 . Although the mechanism through which neoplasms arise in a background of IgG4-RD remains unknown, a recent report suggests that IgG4 subclass antibodies impair antitumor immunity against melanomas 34 . The authors reported that IgG4 antibodies did not induce antitumor immunity, unlike IgG1 antibodies; rather, they inhibited the antitumor activity of the IgG1 antibodies. Therefore, although IgG4 subclass antibodies do not specifically induce neoplasms, they may create an environment where induced malignant neoplasms are more likely to grow.
In the current study, upregulation of FOXP3 was detected in cases of IgG4-RD and IgG4-associated MZL lesions. This is consistent with an earlier report describing the presence of numerous FOXP3 + cells in IgG4-RD 12 . FOXP3 is a marker of Tregs and is a central control element in their development and function. Because Tregs have been reported to suppress inflammation, including tumor immunity 35 , FOXP3 + Tregs may infiltrate lesions associated with IgG4-RD and further suppress tumor immunity, also promoting the growth of neoplasms.
Although steroid therapy is effective in IgG4-RD, the most effective treatment for MZLs arising in the setting of IgG4-RD is uncertain. Although evaluation of additional cases of IgG4-associated MZLs is needed to determine the treatment and prognosis, radiation or chemotherapy may be necessary. Therefore, when IgG4-RD is diagnosed, the lesion should be assessed to determine if it includes an MZL component. The possible tumor-promoting environment of IgG4-RD supports treatment for these patients before an MZL arises, in addition to ongoing follow-up.

Conclusion
IgG4-associated MZL is characterized by the upregulation of Th2 and regulatory cytokines, as is the case in IgG4-RD, but unlike the cytokine background of IgG4-negative MZLs. The current results further suggest that a subset of MZLs may arise in an IgG4-RD setting and that this subset of MZLs may have a different pathogenesis than IgG4-negative MZLs.