827 - Anti-TNF Therapy in RA Patients Alters Hepatitis B Vaccine Responses

Rajan Ravikumar, Teresa Owen, Jennifer Barnard, Anagha A. Divekar, Thomas Conley, Emily Cushing, Tim R. Mosmann, Inaki Sanz, John Looney, Jennifer Anolik. University of Rochester, Rochester, NY
Presentation Number: 827

Purpose: Anti-TNF therapy has proven to be extremely efficacious in patients with rheumatoid arthritis (RA). However, we have recently reported disruption of follicular dendritic cell (FDC) organization/function, germinal center (GC) reactions, and B cell mediated humoral immune responses with TNF blockade in humans. In this study, we examined the effect of anti-TNF therapy on vaccine responses of RA patients.
Method: RA patients on methotrexate (MTX) (n=4), etanercept (n=12), or both (n=7) for at least six months duration were enrolled. Hepatitis B surface antibody (Hep B sAb) serum levels were measured at baseline, 1, 2, 3, and 6 months by chemiluminescent ELISA. Vaccine was administered at baseline, 1, and 2 months. Fisher exact probability test was done to assess for statistical significance. Flow cytometry of peripheral blood mononuclear cells was conducted to evaluate B cells (CD19, CD20, IgD, CD27, and CD38) and T cells. Antigen specific T cell responses were analyzed by ELIspot detection of T cell cytokine secretion (IL2, IL4, IFN-gamma) in response to hepatitis B antigen and control antigens (tetanus, influenza).
Results: The average age of the MTX group was 62 years with a 0% response rate at 2 months but 100% response rate by 6 months, compared to the MTX/etanercept group (average age = 49) with a 0% response rate at 2 months and 29% (p=0.057 compared to MTX control) response rate at 6 months. The etanercept only group (average age = 55) had a 0% response rate at 2 months but a 33% response (p=0.057 compared to MTX control) (p=0.023 for combined anti-TNF treated groups compared to MTX control) at 6 months. Because age is known to influence vaccine responsiveness, we compared the response rates in patients over and under age 55. The younger cohort had higher responses (60% compared to 31%), but this did not achieve statistical significance (p=0.22). In accordance with the lack of humoral responses at two months, a subset of patients (n=7) evaluated at 7-10 days post 2nd dose of Hep B vaccine did not show the peripheral blood plasmablast expansion expected in an adequate immune response. The average naïve T cell percentages (CD3, CD4, CD45RA+) were not different between the three treatment groups: 19.09 (MTX), 17.38(MTX/etanercept), and 15.64(etanercept) respectively. Antigen specific T cell responses were largely negative at 1 and 2 months but began to be positive in a subset of patients at 3 months. However, this did not necessarily correlate with a subsequent antibody response, suggesting effects of anti-TNF on T cell priming and other aspects of the immune response.
Conclusions: Overall, the vaccine responses in this RA cohort were surprisingly low compared to previous studies(Looney et al. J Clin Imm 2001). Our data suggests that TNF blockade decreases hepatitis B vaccine responses. Possible mechanisms for these findings include effects on T cells, antigen presentation, and/or GC reactions.

 R. Ravikumar, None.