Rituximab for steroid-refractory chronic graft-versus-host disease

Abstract

B cells may be implicated in the pathophysiology of chronic graft-versus-host disease (GVHD), as evidenced by antibody production against sex-mismatched, Y chromosome-encoded minor HLA antigens in association with chronic GVHD. We therefore designed a phase 1/2 study of anti-B-cell therapy with rituximab in steroid-refractory chronic GVHD. Twenty-one patients were treated with 38 cycles of rituximab. Rituximab was tolerated well, and toxicity was limited to infectious events. The clinical response rate was 70%, including 2 patients with complete responses. Responses were limited to patients with cutaneous and musculoskeletal manifestations of chronic GVHD and were durable through 1 year after therapy. The median dose of prednisone among treated subjects fell from 40 mg/day to 10 mg/day, 1 year after rituximab therapy (P < .001). A chronic GVHD symptom score improved in the majority of treated patients. Antibody titers against Y chromosome-encoded minor HLA antigens fell and remained low, whereas titers against infectious antigens (EBV, tetanus) remained stable or rose during the treatment period. We conclude that specific anti-B-cell therapy with rituximab may be beneficial for patients with steroidrefractory chronic GVHD. This trial was registered at www.clinicaltrials.gov as #NCT00136396.

Figures

Figure 1.

Serologic responses to rituximab. (A) Normalized antibody titers. The 2 left panels demonstrate the decreasing antibody titers against 2 H-Y antigens, DBY and UTY, in 4 male patients with female donors. Each subject has 5 measurements, including baseline (week 0) after 1 course of rituximab (week 9), after a second course of rituximab (week 17), 6 months after rituximab (week 27), and 1 year after rituximab (week 52). The 2 right panels demonstrate the stable or increasing titers against 2 non–H-Y antigens, EBV and tetanus. The dashed horizontal line represents the baseline level. (B) Box and whisker plot of EBV and tetanus titers in 12 patients. The boxes represent the 25th and 75th percentiles of the antibody titers measured. The whiskers represent the 5th and 95th percentiles of the measured values. The horizontal line within each box is the median antibody titer measured. All comparisons are P = NS.

Figure 2.

Representative time line of serologic response to rituximab in one patient. The trend in antibody titers against 3 antigens (DBY ♦, UTY ▪, and EBV ▴) before and after administration of rituximab (vertical arrows) is demonstrated by the trend lines (added for clarity). The clinical time course of chronic GVHD is depicted along the x-axis.