OP0177 Bcl6 identifies ectopic germinal centres in salivary gland biopsies in primary sjÖgren's syndrome patients. (12th June 2018)

• Record Type: Journal Article
• Authors: Nakshbandi, U.; Haacke, E.A.; Bootsma, H.; Spijkervet, F.K.L.; Vissink, A.; van der Vegt, B.; Kroese, F.G.M.
• Is Part Of: Annals of the rheumatic diseases. Volume 77(2018)Supplement 2
• Page Start: 138
• Page End: 138
• Publication Date: 2018-06-12
• Subjects: Rheumatism -- Periodicals; 616.723005
• DOI: 10.1136/annrheumdis-2018-eular.3741
• Languages: English
• ISSNs: 0003-4967

Abstract

Abstract : Background: Primary Sjögren's syndrome (pSS) patients who exhibit germinal centres (GCs) within salivary gland parenchyma have higher focus scores and present with more disease activity than GC negative pSS patients. Moreover, presence of GCs might be of clinical importance for stratification of treatment. However, there is considerable heterogeneity in reported findings concerning the presence of GCs. This can partially be explained by the difficulty in identifying GCs in diagnostic H and E sections and the lack of uniform histopathological criteria. Objectives: The aim of this study was to assess the most appropriate way to identify unequivocally GCs in parotid and labial gland biopsies of pSS patients. Methods: As part of routine diagnostic work-up for pSS, both parotid and labial salivary gland biopsies were taken from 100 consecutive patients suspected for pSS. Forty-two patients were classified as having pSS according to the ACR-EULAR criteria, the remaining 58 patients were classified as non-pSS sicca patients. Diagnostic salivary gland biopsies were formalin fixed, paraffin embedded and serially sectioned at 3–4 µm thickness. Sections were stained with H and E as well as immunohistochemically for CD3, CD20, CD21, CD45, Ki67 and Bcl6. Presence of GCs, the number of GCs/mm 2 salivary gland parenchyma and level of lymphoid organisation were determined in all sections. Results: According to diagnostic H and E staining, in 15% and 2% of pSS patients GCs were present in parotid and labial salivary glands, respectively. Staining for the proliferation marker Ki67 and the GC-B cell associated transcription factor Bcl6, showed higher percentages: 23% and 25%. Much higher percentages of follicular dendritic cell (FDCs) networks were revealed by CD21L (45% and 55%, for parotid and labial glands, respectively) compared to the number of GCs seen in tissue sections stained for Bcl6. Similarly, the median number of CD21 + FDC networks/mm 2 was significantly higher than the number of GCs/mm 2 as revealed by H and E, Ki67 and Bcl6. Careful evaluation of the consecutive sections stained for CD21L and Bcl6, showed that only roughly half of the FDC networks, also harbour GCs. Finally, not all sections that showed clearly defined GCs by Bcl6 staining, also revealed GCs by Ki67 or H and E staining. Conclusions: Due to the difficulty in GC recognition, use of diagnostic H and E leads to an underestimation and incorrect identification of GCs, while using anti-CD21L overestimates GC counts. This suggests that, although CD21 + FDC networks play an essential role in GC development and T/B cell compartmentalization, positivity for CD21 + FDC networks does not always imply presence of ectopic GCs. Furthermore, since Ki67 is an excellent marker for solely the dark zone of GCs, small GCs can be overlooked while other proliferative areas might be mistakenly identified as GCs. This study shows that staining for Bcl6 allows easy and unequivocal identification of GCs and should therefore be implemented in the histopathological evaluation of salivary gland biopsies of pSS patients. Disclosure of Interest: None declared