Usefulness of Flow Cytometry in Diagnosis of IgG4-Related Ophthalmic Disease and Extranodal Marginal Zone B-Cell Lymphoma of the Ocular Adnexa

Discussion

It is important to know that IgG4-ROD can simulate lymphoid malignancies. Indeed, malignant lymphoma in the ocular adnexa possibly arose from IgG4-ROD (2). We have reported a coincidence of IgG4-ROD and EMZL proved by histological as well as FCM analyses (3). Therefore, to compare FCM findings in IgG4-ROD and EMZL is critical for differentiation of these disorders. In this study, the B-cell linages, CD19, 20 and CD23-positive cell population were significantly greater in EMZL than IgG4-ROD, whereas CD2, 3, 4 and CD7-positive T cells were significantly lower in the former than the latter. Taken together, FCM could show significant differences regarding B-cell and T-cell linages quantitatively in EMZL and IgG4-ROD. Therefore, FCM is a useful method to differentiate IgG4-ROD from EMZL.

Recently, Ueda et al. demonstrated that T-cell markers (CD3, 4, 8, 10, 25) and B-cell markers (CD19, 20, 23) tested by FCM were useful in differentiating benign lymphoproliferative disorders from malignant lymphoma (6). This study was consistent with the previous report. Moreover, this study further showed other T-cell markers including CD2 and CD7, both of which were more characteristic in benign IgG4-ROD/IOI than in EMZL. Rabinowich et al. reported that CD2 and CD7 were expressed not only in T cells, but also in NK cells (9). In addition, CD7, but not CD2, -positive NK cells play a critical role in the various immune responses through cell adhesion molecules, integrins (9). Further studies are needed to elucidate localization of CD2 and CD7 and their functions in the immune cell surfaces in IgG4-ROD.

In this study, according to the FCM findings, benign IgG4-ROD/IOI demonstrated that cells having both T-cell and B-cell linages were moderately detected in the tumor tissues. These results clearly show that benign lymphoproliferative disorders in the ocular adnexa are made up of polyclonal populations in lymphoid cells. However, there was no statistical difference between definitive IgG4-ROD and IOI. These results suggest that FCM findings might not contribute to the differentiation of IgG4-ROD from IOI. It has been reported that Th2 cells play a pivotal role in germinal center formation and IgG4 production in IgG4-related disease (10). Moreover, CD19(+) CD27(+) CD20(−) CD38(hi) plasmablasts take part in the active inflammation in IgG4-related disease (11). Although this study examined representative T-cell and B-cell linages, further cell surface markers such as CD38 and CD27 in the tumor tissues would contribute to differentiation in both disorders.

The presence of B-cell monoclonality is an important proof of B-cell malignancy. Ways to prove B-cell monoclonality comprise IgH gene rearrangement by PCR method/Southern blot analysis, and/or immunoglobulin light chain restriction by FCM/immunohistochemistry. Indeed, examination of IgH gene rearrangement is useful to confirm the B-cell monoclonality; however, it may be challenging to gain reliable results if an insufficient number of cells was collected from the isolated tissues. Sometimes local resected tumor tissues may not have enough cells to be examined in ophthalmology practice. Therefore, further data should be collected to validate true diagnosis of ocular adnexal lymphoproliferative disorders. This study further examined cell surface immunoglobulin light chains in ocular adnexal lymphoproliferative disorders. Indeed, kappa-positive cell number was significantly greater in EMZL than IgG4-ROD/IOI. And then kappa/lambda deviation was determined based on FCM analysis. False positive was noted in 3 (7.5%) benign IgG4-ROD/IOI and false negative was observed in 10 (34.5%) EMZL cases. Samoszuk et al. reported that a false-positive rate was less than 10%, whereas a false-negative rate was approximately one fourth (8). The current data showed that a false positive rate was similar to the previous report; however, false negative was slightly higher. In fact, this study demonstrated that the sensitivity and specificity of kappa/lambda deviation were 65.5 and 92.5%, respectively, indicating that analyses of B-cell clonality using FCM revealed relatively low sensitivity and high specificity in ocular adnexal lymphoproliferative disorders.

Interestingly, the number of B cell markers CD19/20 was lower as shown above, whereas CD23 was significantly greater in IgG4-ROD than in EMZL, the latter of which was not observed between IOI and EMZL. It has recently been reported that CD23-immunopositive cells were clearly detected in the tissues of IgG4-ROD (6). CD23 is a cell surface marker for activated B cells which plays an important role in the pathophysiology of allergic diseases. Celiksoy et al. demonstrated reduction of the activated B-cell population in patients with allergic rhinitis/bronchial asthma following immunotherapy (12), suggesting that CD23 population is useful in not only diagnosis but also evaluation of therapeutic effects. The limitation of this study is that the tumor tissues were isolated only at an initial presentation; however, the tumor tissues were not available after treatments. Therefore, it was impossible to look into CD23 alteration before and after treatments in ocular adnexal lymphoproliferative disorders. Another limitation is that this study only contained definitive IgG4-ROD whilst probable/possible IgG4-ROD was not examined. Therefore, further studies are required to clarify the difference of cell surface markers between definitive and probable/possible IgG4-ROD.

In conclusion, quantification of cell surface markers by FCM is useful to differentiate benign lymphoproliferative disorders from EMZL in the ocular adnexa. Sensitivity of immunoglobulin light chain restriction was not high in this method.