Background: The historical association of acrodermatitis chronica atrophicans (ACA), now known to be a late manifestation of Lyme disease caused by Borrelia afzelii, with cutaneous lymphoma, and several small series of PCBCL with positive Lyme disease borrelial serology initiated a study of this association. Material and methods In the last 9 years, 30 patients with PCBCL have been observed and followed, 22 of them were tested for borrelial serology. The control group consisted of 85 patients with NHL (10 cutaneous T-cell, 25 extranodal B-cell non-PCBCL, 50 nodal B-cell), 30 patients with breast cancer and 60 blood donors. The screening tests were two different ELISA tests for B. burgdorferi sensu lato and sensu stricto, and reactive sera were further tested with the ELISA test for B. garinii, a Western blot (WB) test for Swiss Borrelia strains and a WB test for Bavarian Borrelia strains, since an immunoblot made with local strains was not available. Studies with a differential WB test for B. burgdorferi sensu stricto, B. garinii and B. afzelii was performed afterwards, as well as serological studies ruling out cross-reactions with Leptospiras and Treponema.
Results: Fifteen of 22 patients with PCBCL were positive on the screening tests, three of them falsely. Thus, the incidence of positive borrelial serology was 12/22 (55 per cent) in the PCBCL group. No positives were detected in the cutaneous T-cell lymphoma group; 2/25 patients (8 per cent) were positive in the extranodal B-cell NHL group (the localizations being vestibulum nasi and oral cavity), 2/50 (4 per cent) were positive in the nodal B-cell NHL group, 2/30 (7 per cent) in the breast cancer group and 2/60 (3 per cent) in the blood donor group. The cumulative incidence in the control groups was 8/175 (4,6 per cent). The incidence was significantly higher in PCBCL patients as compared to each of the control groups, p value ranging from 0.004 to <0.0001. Two positive patients had ACA, one arthritis. Borrelia afzelii was most often implied for positive serology in the differential WB. No cross-reactions with Treponema and the Leptospiras were documented.
Conclusion: In conclusion there appears to be a clustering of positive serology for Lyme disease Borrelias in PCBCL patients possibly related to an ethiopathogenic relationship. Mechanisms of Borrelia escape from immunosurveillance mechanisms, persistence of both their mitogenic and antigenic stimuli for B-cells, and SALT formation may be involved in the pathogenesis of a subset of PCBCL.
Copyright 1999 John Wiley & Sons, Ltd.