|Home | About | Journals | Submit | Contact Us | Français|
Acquired haemophilia A (AHA) is a rare bleeding disorder caused by the development of specific autoantibodies against naturally occurring factor VIII (FVIII). Although about half of cases are idiopathic, AHA may be associated with several non-neoplastic conditions, autoimmune disorders, as well as haematological malignancies, such as chronic lymphocytic leukaemia and lymphoma. The long-term suppression of inhibitors is one of the mainstays of the treatment of AHA. Apart from standard immunosuppressive treatments, rituximab has been proven to be effective in AHA.
The aim of this review is to provide a systematic description of data available in the literature on this topic. To do so, we performed a search using the indexed online database Medline/PubMed, without temporal limits, matching the words “rituximab” and “acquired h(a)emophilia”. Furthermore, additional published studies were identified in the reference list of the publications found in PubMed.
The review of the literature confirms that rituximab may be a safe and useful treatment for AHA.
Although rituximab is not a standard therapy for AHA, it may be useful in resistant cases. However, the definitive place of this monoclonal antibody in the therapeutic strategy for AHA (first or second-line, alone or in combination with other drugs) remains to be determined more precisely and warrants further investigation.
Rituximab, a chimeric human/murine monoclonal antibody targeting the CD20 antigen on the surface of B cells, is used extensively to treat CD20-positive haematological cancers and is now being increasingly employed to treat several autoimmune disorders1. Acquired haemophilia A (AHA) is a rare bleeding disorder caused by the development of specific autoantibodies (inhibitors) against naturally occurring factor VIII (FVIII)2.
The treatment of AHA is focused on the need to stop potentially life-threatening bleeding and to gain a long-term suppression of the inhibitors3. In detail, the inhibitors are usually eradicated by standard immunosuppressive treatments, such as steroids and cyclophosphamide. More recently, rituximab has also been used, generally as second-line treatment4.
The first experiences with rituximab in AHA were described in 2001 and several patients have been treated with this specific approach since then5. The aim of this review is to provide a systematic description of data available in the literature on this topic. To do so, we searched the indexed online database Medline/PubMed, without temporal limits, matching the words “rituximab” and “acquired h(a)emophilia”. Additional published studies were identified in the reference lists of the publications found in PubMed.
AHA is a rare bleeding disorder. First described in 19406, it is due to the development of autoantibodies against FVIII. The incidence of AHA has been estimated to be 0.2–1.0 cases per million population per year, although it is probably higher because not all affected patients are reported7. The median age of patients at presentation is 60 to 70 years and the condition occurs with similar frequency in males and females. However, females are younger than males, probably because of the risk of post-partum inhibitors.
Bleeding manifestations are often severe and may occur spontaneously or after minor trauma. In addition, patients usually experience soft tissue bleeding other than haemarthroses, which are more commonly seen in the congenital form of haemophilia.
Although the development of inhibitors is idiopathic in approximately half of patients, various conditions associated with FVIII inhibitor development, including connective tissue and inflammatory bowel diseases, puerperium, malignancies, and dermatological disorders, are currently known. In this setting, the database of the European Acquired Haemophilia (EACH2) registry, which has collected data on 501 patients from 117 centres in 13 European countries, showed that among the 482 patients who bled (96.2% of all registered patients), the underlying medical condition was as follows: autoimmune disorder (14.1%), malignancy (11.5%), pregnancy (8.9%), drug-induced (3.3%), transfusion, dermatological, infection, monoclonal gammopathy of undetermined significance or others (7.8%)8. No associated medical condition was identified (the so-called idiopathic forms) in 54.4%, while no information was reported for 22 patients (4.6%) (Figure 1).
Autoimmune phenomena may complicate the clinical course of several haematological malignancies, in particular B-cell lymphoproliferative disorders9.
The association of AHA with several haematological cancers has been reported and was first reviewed by Sperr et al.10 More recently, Franchini et al. showed that lymphoproliferative neoplasms, such as chronic lymphocytic leukaemia and lymphomas, are the blood malignancies most frequently associated with AHA (70% of cases). Moreover, muco-cutaneous haemorrhages were the most frequently observed type of bleeding (63% of cases)11 (Figure 2). The relationship between inhibitor production and the underlying haematological disease was demonstrated by the disappearance of the inhibitor when cancer treatment was given successfully. Sallah et al. proposed different pathogenic mechanisms explaining this association: alterations of the FVIII molecule, disturbances in the antigen-presenting cell process and alterations in the function/interaction of B and T cells12.
Finally, immunomodulatory drugs, such as fludarabine and interferon, have been reported to be associated with the appearance of FVIII inhibitors, probably due to the immune disturbance caused by these agents in patients with a disease involving the immune system11.
The presence of FVIII inhibitors must be suspected when an unexplained prolongation of the activated partial thromboplastin time (aPTT) is not corrected by the in vitro addition of normal plasma (mixing test). Several other conditions may cause prolongation of the aPTT: lupus anticoagulant, heparins and other anticoagulants (i.e., vitamin K antagonists and direct oral anticoagulants) and acquired von Willebrand syndrome. For this reason, these conditions must be ruled out. In AHA the FVIII level is reduced, although rarely to less than 2%, and the Bethesda assay demonstrates the presence of an inhibitor. Table I summarises the differential diagnosis of a prolonged aPTT. The diagnostic work-up of AHA was recently reviewed by Tiede et al.13 Finally, an underlying concomitant disorder triggering AHA should be searched for and, if found, removed with appropriate therapeutic measures.
The treatment of AHA is aimed at immediate stopping the acute bleeding and long-term suppression of the inhibitor. In general, bypassing inhibitor agents, such as the recombinant activated FVII (rFVIIa) and activated prothrombin complex concentrate (APCC), are used to control the bleeding. Human FVIII concentrate, recombinant porcine FVIII concentrate and desmopressin are also employed. Furthermore, immunosuppression is required to eradicate the inhibitor. Glucocorticoids combined with cyclophosphamide are frequently given. This approach is able to eradicate inhibitors in approximately 70% of patients with AHA3.
As showed by a very recent report published by Tiede et al., predictors of a positive response are a low inhibitor level and a short interval between the appearance of the inhibitor and the start of therapy14. Moreover, FVIII ≥1 IU/dL and inhibitor level <20 Bestheda units (BU)/mL were independent predictors of overall survival and define a subgroup of patients with better prognosis.
The monoclonal antibody rituximab, a chimeric antibody targeting CD20 antigen on B-cell surface15, was first reported to be effective in eradicating the inhibitors in AHA in 20014. To date, 160 patients with this condition treated with rituximab have been described16–49. Published studies, including several cases reported only in abstract form, describing the use of rituximab in acquired haemophilia have been previously reviewed by Barcellini and Franchini5,7,11 and are updated in the present review.
As shown in Table II, rituximab has been used alone or in combination with other immunosuppressive drugs, such as steroids and cyclophosphamide, as salvage or first line-therapy. Overall, a high rate of response was observed, with 123 patients (77%) achieving complete response. A complete response is usually defined as normalisation of the FVIII level and an inhibitor that is no longer detectable or <1 BU, while a partial response is defined as a >50% decrease in inhibitor titre, a FVIII level >25% and no further bleeding5. However, it should be noted that the criteria for complete or partial response are not identical in all reports. This may explain why the response rate was quite high in this review compared to other studies45 in which more stringent response criteria were used. In addition, several cases were reported only as case reports with a tendency to describe only responding patients. Consequently, the outcome data should be evaluated with caution.
The dose of rituximab infused in almost all cases was 375 mg/m2, the standard dose used in the treatment of lymphoma. However, a lower dose (100 mg) has also been used16. The number of courses of rituximab was very variable, ranging from a single, low dose to 12 standard doses. The time to response was also heterogeneous, ranging from 1 week to more than 1 year. Re-treatment with rituximab was generally effective. Overall, the administration of rituximab in AHA was well tolerated with very few early or late side effects of the infusion; no infectious complications have been reported so far.
In eight patients with AHA treated with rituximab, Aggarwal et al.17 found that the best results to standard treatments were achieved in cases with low inhibitor titres, while patients with high titres had a good response to rituximab. In light of this, Aggarwal et al. proposed an algorithm for the treatment of AHA on the basis of which, patients with a low inhibition titres (<5 BU/mL) and minimal bleeding should be treated only with prednisone, while patients with higher inhibitor titres should also receive rituximab (if titre <30 BU/mL or there is serious bleeding) or rituximab plus cyclophosphamide (if titre ≥30 BU/mL).
Subsequently, Collins et al. reported on 15 patients with AHA identified in 23 Haemophilia Centres in the United Kingdom and reported an overall response to rituximab in 47% of patients, despite only 14% of them obtaining a durable response18. None of the three patients treated with rituximab without adding FVIII responded. The same authors performed a systematic review of the literature and found 28 additional cases treated with rituximab. Inhibitor negativity was achieved in a higher number of patients (56%) than those reported in the English study, while the proportion of patients with a sustained response was similar. Finally, these authors suggested that rituximab combined with FVIII is potentially useful in haemophilia patients with inhibitors resistant to standard immunosuppressive therapy.
Franchini et al. performed a systematic review and a meta-analysis including 49 cases of AHA in which a durable complete remission (i.e., no inhibitor recurrence during follow-up) was obtained in 53% of the patients treated with rituximab19. On the basis of a multivariate analysis, rituximab appeared to be more effective in mild-to-moderate disease than in severe disease and when concomitantly administered with FVIII concentrates and immunosuppressive agents. On this basis, in a more recent paper, Franchini and Mannucci20,21 proposed the treatment algorithm depicted in Figure 3. Briefly, first-line therapy for patients with AHA should be the association of prednisone and cyclophosphamide, while rituximab ± immunosuppressive therapy should be reserved for second-line therapy. Very recently, Tiede et al. reported data from a multicentre, prospective, observational study of the German, Austrian, and Swiss registry (GTH-AH study) in which a uniform immunosuppressive regimen was applied in 102 consecutive patients14. The schedule of immunosuppressive treatment given was as follows: prednisolone alone in weeks 1 to 3 with the addition of cyclophosphamide in weeks 4 to 6 if a partial response was not achieved. Finally, rituximab was added in week 7 and given together with prednisolone. Of the 44 patients requiring second-line therapy, nine (20%) were given rituximab because of contraindications to cyclophosphamide, while 12 patients received the monoclonal antibody as third-line therapy. Unfortunately, no specific data on results obtained in patients treated with rituximab were reported in this paper.
Finally, but not less relevant, rituximab seems to be cost-effective. In fact, Cretel et al.39 reported that the estimated cost of treating their patient with rFVIIIa was 185,000 Euro with respect to the cost of treatment with rituximab of 8,000 Euro. In further support of this, Kain et al.23 reported a patient in which the estimated cost of treating acute bleeding episodes with rFVIIIa was greater than that with rituximab.
In the last years, the anti-CD20 monoclonal antibody rituximab has gained a role in the treatment of autoimmune disorders and its use is rapidly increasing1. Notably, rituximab also seems to be beneficial in AHA. However, its definitive role in inhibitor eradication (first or second-line therapy, high or low inhibitor titres, older and/or frail patients for whom corticosteroid and cytotoxic therapy are unsuitable) still requires further evaluation. Unfortunately, it is difficult to design randomised controlled trials in AHA because of the rarity of this disorder. Notwithstanding, though additional data need to be acquired in order to optimise the use and to compare the efficacy of this drug to that of other treatments, current evidence confirms that rituximab should now be considered as part of the therapeutic armamentarium of AHA, despite reserving it to second line-therapy.
This work is dedicated to the memory of Prof. Giuseppe (Pino) Maiese. He lives on in our hearts.
Disclosure of conflicts of interest
PM received onoraria from Roche, the manufacturer of rituximab. All other Authors declare no conflicts of interest.