The mechanism of tissue injury in chronic GVHD (cGVHD) is unknown, but donor-derived T cells are thought to be the primary effectors. We demonstrated that antibodies against minor histocompatability antigens develop in association with cGVHD, suggesting coordinated B and T cell mediated injury. We therefore began a clinical trial of specific anti-B cell therapy with Rituximab (Rituxan) as therapy for steroid-refractory cGVHD.

Methods: Eligible patients had persistent, clinically extensive cGVHD despite a trial of corticosteroids (at least 0.5 mg/kg/day prednisone for 1 month) and were on stable doses of all immunosuppressants, including steroids, for 4 weeks prior to enrollment. cGVHD was diagnosed clincally in most cases. Rituxan (375 mg/m2/wk x 4) was followed by a second course for non- or partial responders 4 weeks later. All patients underwent thorough oral, ocular, rheumatologic and dermatologic examinations by trained sub-specialists prior to therapy and 8 weeks after Rituxan therapy, where indicated. A validated quality of life (QoL) survey was administered before and after Rituxan therapy as well.

Results: 16 patients are enrolled and 10 (median age, 43 yrs) have undegone thorough response evaluations to date. At the time of enrollment, 9 were on steroids, either alone (n=3) or in conjunction with one (n=5) or two (n=1) immunosuppressive agents. One patient was on tacrolimus alone. End-organ involvement included skin(7), eye(6), musculoskeletal(6) and oral mucosa(4). Fifteen 4-week courses of Rituxan were administered, at a median of 14.4 months from cGVHD diagnosis (range 3.5 to 82 months). Rituxan was well tolerated in all patients, with 5 grade 3 adverse infectious events (infectious colitis(2), Hepatitis B reactivaton(1), gastroenteritis(1) and conjunctivitis(1)) and one infusion reaction. Nine weeks after Rituxan initiation, CD19+ B cells were undetectable in all peripheral blood samples. Median serum IgG levels fell 16%, from a baseline of 772 to 645 mg/dl. Median IgM levels fell 33%, from 84 to 56 mg/dl. IgG antibody titers for tetanus toxoid and EBV remain unchanged. Median follow-up is 130 days. Objective responses were noted in 5 patients (3 cutaneous responses, 4 rheumatologic responses ), no response was noted in 4 patients and 1 patient had progressive cutaneous cGVHD. No ocular or oral responses were noted. Mean rheumatologic VAS pain scores improved from 5 to 1, and VAS fatigue scores improved from 5 to 1, while grip strength was unchanged in all patients. Mean oral pain scores and ocular Schirmer test times were unchanged. Mean QoL scores improved in 4 patients, were unchanged in 3, worsened in 1 and were unevaluable in 2. Of the 4 patients with improved QoL scores, 3 also had clinical responses. At this time, all patients remain on systemic immunosuppression.

Conclusions: Rituxan can be administered safely to patients with cGVHD, although infectious adverse events may occur. Early results suggest that this agent may have activity as therapy for cGVHD, particularly when cutaneous and rheumatologic manifestations are prominent. An evaluation of serologic responses is ongoing.

Author notes

Corresponding author