Treatment of Acquired Hemophilia A (AHA) with Rituximab (R): Clinical Outcomes and Cost of Hemostatic Agents

AHA is a rare bleeding disorder caused by autoantibodies (inhibitors) to coagulation factor VIII (FVIII). Acquired FVIII inhibitors are associated with a high rate of complications, with major bleeding reported in up to 90% of affected individuals and mortality rates as high as 22%. There are currently no formal guidelines for the treatment of AHA, due to the paucity of data from randomized clinical trials and limited number of cases. However, there is general agreement that treatment should consist of a two-pronged approach that includes 1) achieving hemostasis in bleeding patients by administering FVIII bypassing agents, and 2) reducing or eradicating the inhibitor with immunosuppressive treatments such as rituximab (R), cyclophosphamide (C), prednisone (P) or a combination. C and P have been used as first-line treatment for AHA until the late 1990s, when R began to be used off-label for inhibitor eradication. Data from case reports and series suggest that R is effective as both first and second-line therapy for AHA. We hypothesized that the use of R rather than C and P may be cost-effective by reducing the overall use of expensive hemostatic agents. The purpose of this study was to compare clinical course, use of hemostatic agents and costs in patients who received R to historical patients who did not receive R for treatment of AHA.

We performed a retrospective study of patients with AHA, defined by prolonged aPTT (>37 sec), FVIII level <50% and detectable FVIII inhibitor (>0.40 Bethesda Unit (BU)) at Dartmouth-Hitchcock Medical Center (DHMC) using the hospital's electronic medical record system, CIS, to collect pertinent demographic and clinical information. The data collection variables included: age, sex, co-morbidities, FVIII levels, FVIII antibody titers, type of bleeding, amount and total acquisition cost of hemostatic products administered, immunosuppressive regimens administered, time to complete remission (CR) (defined as FVIII level > 30%), and relapses.

We identified 16 patients with AHA treated at DHMC from 4/1997 to 3/2010. The median age at diagnosis was 78 (range 66 to 93), and 56% were women. Co-morbidities included current or prior malignancy (62%) and autoimmune disease (12%). Idiopathic cases accounted for 32%. One patient had both a malignancy and autoimmune disease. The median aPTT at diagnosis was 80 sec (range 53 to >160 sec); FVIII level, 3% (range <1 to 19%); FVIII antibody, 26 BU (range 2 to 220 BU). The types of bleeding included mucocutaneous (94%), intracranial (19%), gastrointestinal (6%) and genitourinary (6%). Two types of bleeding were noted in 19%, and 69% of patients received hemostatic products. R was used in 63% of patients as first-line immunosuppressive treatment. The remaining patients received C and P, P alone or no immunosuppressive treatment. A CR was documented in 63% of patients, of whom 60% received R. Three patients had no documentation of remission and three patients died. The median time to CR for all patients was 33 days. The median time to CR for R-treated patients was 28 days (range 23 to 86) and 66 days (range 28 to 154) for non-R-treated patients. Relapses were documented in two patients who achieved CR. The first patient was treated initially with P alone, and died of a major hemorrhage in relapse without receiving salvage treatment. The second patient was treated with R initially and relapsed twice, achieving a CR after retreatment with R for both relapses. The total acquisition cost of hemostatic agents for the entire AHA population was $2, 211, 262. The median cost of hemostatic agents for R-treated patients was $180, 853 (range $9, 666 to $554, 026), while the median cost for non-R-treated patients was $169, 075 (range $35, 076 to $319, 115).

Our findings suggest that R is an effective first-line immunosuppressive treatment for AHA and appears to shorten the time to CR substantially compared to traditional immunosuppression. The cost for hemostatic agents in our population was similar for both R-treated and non-R-treated patients, suggesting that despite the decreased time to CR, the use of R may not result in decreased use of hemostatic products. A hemostatic agent-sparing effect attributable to R may have been missed due to our small AHA population size, however, and is worthy of further investigation.

Off Label Use: Rituximab for treatment of acquired hemophilia A.