In order to characterize the autologous mixed lymphocyte response (AMLR) in patients with rheumatoid arthritis (RA) and to define the relationship with disease activity, peripheral blood T lymphocytes were stimulated with either a B lymphocyte-enriched (B cells) or a macrophage-enriched (macrophages) population. A significant reduction (P < 0.01 to P < 0.001) of T cell proliferation stimulated both by B cells and macrophages was observed in patients with active disease. The B lymphocytes were significantly less stimulatory (P < 0.02 to P < 0.001) than macrophages in the patients compared with the controls. In the normal controls, macrophages in higher concentrations were capable of suppressing the B lymphocyte-stimulated AMLR, but macrophages from patients with RA were not excessively suppressive. A significant association (P < 0.02) was observed between disease activity and the AMLR. Using the B-enriched population, the AMLR proliferative response was significantly associated (P < 0.001) with the production of interleukin-2. Defects in proliferation could only be partially restored by the addition of interleukin-2. These data indicate that the defective AMLR observed in patients with RA is related to disease activity and is associated with altered cellular interac- The autologous mixed lymphocyte response (AMLR) is a proliferative response of T lymphocytes to autologous non-T cells (1,2). This response is dependent upon the recognition of class I1 histocompatibility antigens on the surface of the accessory cells (3-6). Macrophages, B lymphocytes, dendritic cells, and null cells are capable of stimulating the AMLR (1,3,(7)(8)(9)(10). Recently, even a subset of la-positive T lymphocytes has been shown to be capable of stimulating an AMLR (11). The AMLR is specific and has been shown to possess memory (12). It is capable of generating both T lymphocyte help and suppression and of augmenting cytotoxicity and natural killer activity (12-
21).The AMLR may represent an in vitro model of in vivo cellular communication capable of generating interleukin-2 (IL-2) in an antigen-independent fashion (5). As such, it may serve as an important model for examining the defective cellular interactions responsible for a number of autoimmune disorders. Although claims have been made that the AMLR is primarily due to xenogenic antigens employed in isolation techniques (22,23), the work of numerous investigators (5,(24)(25)(26), as well as our own unpublished observations, have demonstrated that the AMLR is not due solely to foreign antigens, although such antigens are capable of augmenting the response.A defective AMLR has been described in a Submitted for publication December 12, 1983; accepted in number of autoimmune, immune deficiency, and neoplastic disorders (27-31, reviewed in 32). The data concerning the AMLR in patients with rheumatoid arthritis (RA) are conflicting regarding both the pres-