Lymphoproliferative Lesions of the Ocular Adnexa Analysis of 112 Cases Sarah E. Coupland, MBBS, PhD, 1 Lothar Krause, MD, 2 Henri-Jacques Delecluse, MD, 3 Ioannis Anagnostopoulos, MD, 1 Hans-Dieter Foss, MD, 1 Michael Hummel, PhD, 1 Norbert Bornfeld, MD, 2 William R. Lee, MD, 4 Harald Stein, MD 1 Objective: Lymphoproliferative lesions of the ocular adnexa were analyzed to examine (1) the suitability of the Revised European–American Lymphoma (REAL) classification for the subtyping of the lymphomas in these sites; (2) the predictive value of the REAL classification for the evolution of these tumors; and (3) the frequency and prognostic impact of tumor type, location, proliferation rate (Ki-67 index), p53, and CD5 positivity and the presence of monoclonality within these tumors. Design: Retrospective review. Methods: The clinical, histomorphologic, immunohistochemical, and molecular biologic (polymerase chain reaction [PCR]) features of lymphoid proliferations of the ocular adnexa were studied. Study Materials: The ocular adnexal lymphoproliferative lesions were located as follows: orbit in 52 patients (46%), conjunctiva in 32 patients (29%), eyelid in 23 patients (21%), and caruncle in 5 patients (4%). Results: Reactive lymphoid hyperplasia was diagnosed in 12 cases and lymphoma in 99 cases; 1 case remained indeterminate. The five main subtypes of lymphoma according to the REAL classification were extranodal marginal-zone B-cell lymphoma (64%), follicle center lymphoma (10%), diffuse large cell B-cell lymphoma (9%), plasmacytoma (6%), and lymphoplasmocytic lymphoma (5%). Age, gender, and anatomic localization of the lymphomas did not have prognostic significance during a follow-up period of 6 months to 16.5 years (mean, 3.3 years). Extent of disease at time of presentation was the most important clinical prognostic factor: advanced disease correlated with increased risk ratios of having persistent disease at the final follow-up and with lymphoma-related death (P 0.001). Histomorphologic features and immunohistochemical markers positively correlating with disseminated disease at presentation, stage at final follow-up, and occurrence of lymphoma-related death included cytologic atypia (P 0.001), MIB-1 proliferation rate (P 0.001), and tumor cell p53 positivity (P 0.001). The MIB-1 proliferation rates greater than 20% in extranodal marginal-zone B-cell lymphoma corresponded to at least stage II lymphoma (P 0.05). Conclusion: The REAL classification is suitable for the subdivision of the ocular adnexal lymphomas. The MIB-1 proliferation rate and p53 positivity may aid the prediction of disease stage and disease progression, whereas PCR can support the diagnosis and reduce the number of histologically indeterminate lesions. Ophthalmology 1998;105:1430 –1441 Lymphomas of the ocular adnexa (i.e., the conjunctiva, orbit, and lacrimal gland) represent approximately 8% of all extranodal lymphomas. 1 They form one end of the spectrum of lymphoproliferative lesions that are seen at these sites; the other end of the spectrum is represented by reactive lymphoid hyperplasia (RLH), also previously termed benign lymphoma, 2 pseudolymphoma, 3–5 and in- flammatory pseudotumor. 6 The majority of lymphomas of the ocular adnexa are composed of small lymphocytic cells with a relatively bland cytologic appearance and are, therefore, not easily distinguished from their benign counterparts on conventional histology. To further com- pound diagnostic difficulties, a “gray zone” lies between the two poles of the spectrum and consists of a group of borderline cases in which the diagnosis of the lesion cannot be determined with any certainty using conven- tional histologic techniques. Lesions allocated to this group consist of diffuse or follicular lymphoid prolifer- ating cells that either manifest borderline maturity or contain a subpopulation of atypical cells with large hy- Originally received: July 24, 1997. Revision accepted: January 12, 1998. Manuscript no. 97417 1 Department of Pathology, Universita ¨tsklinikum Benjamin Franklin, Freie Universita ¨t, Berlin, Germany. 2 Department of Ophthalmology, Universita ¨tsklinikum Benjamin Franklin, Freie Universita ¨t, Berlin, Germany. 3 Department of Clinical Molecular Biology, GSF–Environment and Health Research Centre, Munich, Germany. 4 Department of Pathology, Western Infirmary, Glasgow, Scotland. Presented in part at the Combined Meeting of the International Society for Ocular Pathology and American Association of Ocular Pathology, Chi- cago, Illinois, October 1996. Reprint requests to Sarah E. Coupland, MBBS, PhD, Institute of Pathology, Universita ¨tsklinikum Benjamin Franklin, FU-Berlin, Hindenburgdamm 30, D-12200 Berlin, Germany. 1430