1 Introduction

Immune (idiopathic) thrombocytopenia (ITP) is a disorder mediated by auto-antibodies against the platelet antigens that accelerate their destruction. Antibody-coated platelets are removed by the reticuloendothelial system, particularly in the spleen. ITP is usually an isolated thrombocytopenic disorder without an associated illness when it is called primary or idiopathic. Immune thrombocytopenia can be secondary to a variety of causes and commonly associated with lymphoproliferative disorders, particularly chronic lymphocytic leukemia (CLL) or a systemic immune disease like systemic lupus erythematosus (SLE) [1]. Significant drop in the platelet count can result in mucocutaneous bleeding of variable severity and is usually clinically manifest in many patients when the platelet count drops below 30 × 109/l [1]. Around 20–30% of adult patients are non-responsive to conventional therapy including steroids, intravenous immunoglobulins (IvIg), anti-D and splenectomy [2]. These patients have a chronic, refractory disease unlikely to be cured, although spontaneous remissions may occur in some patients [3]. This group of patients have a significant risk of bleeding with increased morbidity and mortality [1, 4–7]. Further treatment is recommended in these patients if they have a low platelet count and bleeding symptoms.

Because of better awareness and possibility of newer modalities of treatment, many patients are not willing to undergo splenectomy and like to try other forms of medical therapy before this procedure. These options usually include various immune modulating agents. Low response rates and significant side effects make the patients and the physicians hesitant to try these modalities of treatment, particularly in younger patients. In view of this, new treatment strategies are needed which are more effective and less toxic.

Rituximab (Mabthera; Roche Pharmaceuticals, Basel, Switzerland) is an anti-CD20 antibody which was originally developed for B cell lymphomas, and has shown excellent activity in various B cell lymphoproliferative disorders [8]. Rituximab is a chimeric, monoclonal antibody that specifically targets CD20 B cell surface antigen. Rituximab effectively reduces circulating B cell count by at least three distinct mechanisms: complement-mediated cytotoxicity, cell-mediated cytotoxicity, and induction of apoptosis [9, 10]. Apart from being effective in B cell lymphomas, rituximab has shown activity in many autoimmune hematological and non-hematological disorders with a favorable safety profile [11–17]. Rituximab has been reported to produce variable responses in patients with ITP and here we describe our experience with 24 adult ITP patients treated with rituximab.

2 Patients and methods

We retrospectively studied adult ITP (primary or secondary) patients who received rituximab for refractory or relapsed disease at two tertiary care centers in Riyadh. All patients had platelet counts less than 30 × 109/l except one patient who had a relapsed but steroid responsive disease (platelet count 37 × 109/l) and received rituximab because of poorly controlled diabetes. All patients gave informed consent. Rituximab was given at a dose of 375 mg/m2 weekly for a median of 4 doses. Patients received antihistamines, paracetamol and hydrocortisone as premedication prior to rituximab infusion. One patient received rituximab with a presumed diagnosis of ITP but later proved to have May–Hegglin anomaly and was excluded from analysis.

Response to rituximab was assessed weekly for 1 month and 1–3 monthly thereafter. Complete response (CR) was defined as a platelet count of more than 100 × 109/l, partial response (PR) as a platelet count of more than 50 × 109/l but less than 100 × 109/l, minimal response (MR) if the platelet count was >30 × 109/l and <50 × 109/l and no response if platelet count remained unchanged. Durability of responses was assessed at 6 and 12 months as well as at the last follow-up. The response to further therapies in patients refractory to rituximab, type of therapy and the outcome of these patients was also studied.

Additional information was obtained from the patients’ notes and computer records regarding demographic data, age at diagnosis of ITP and the type of previous treatments, age and platelet count at rituximab therapy, timing and type of responses to rituximab, duration of responses and the timing of any relapses.

Data were analyzed using descriptive terms like absolute numbers and percentages, or medians and ranges, as appropriate. Student’s t test was used to study the association of response (CR + PR + MR) with clinical parameters, and that of sustained response of >6 months with the type of response (CR or PR + MR). A p value of ≤0.05 was considered significant.

3 Results

A total of 24 patients received 29 rituximab treatment courses for ITP. Nineteen patients had primary (idiopathic) disease and five patients had secondary ITP. The median number of previous treatments was 3 (range 1–8) and 11 patients had prior splenectomy. Patient characteristics and treatment details are shown in Table 1.

Table 1 Patients’ characteristics
Full size table

Responses were achieved in 19 of 29 (66%) treatment courses and rituximab failed in 10 (34%) cases. A CR was achieved in 13 cases (45%) and a PR or MR in 6 (21%) treatments. The median time to response was 3 weeks (range 1–20) from the start of therapy and median duration of response was 13 weeks (range 1 week–55 months). Responses were mostly short lived and after a median follow-up of 22 months (range 2–70), 10 (34%) responses were sustained after 6 months, 7 (24%) responses sustained after 1 year and only 5 patients continued to have responses without relapse at the last visit after 8, 10, 24, 30 and 54 months of follow-up. Only one of these 5 patients with continued responses had prior splenectomy. All 3 patients with CLL-associated ITP responded and one of these patients achieved a second CR post-relapse after re-treatment with rituximab, lasting more than 6 months. Outcome of rituximab therapy and summary of responses are shown in Table 2.

Table 2 Summary of results
Full size table

Prior splenectomy was associated with a failure of response (p = 0.034) while sustained response of more than 6 months had a borderline correlation (p = 0.055) with achievement of CR but this did not reach statistical significance. Seven out of 10 refractory patients had prior splenectomy. Age at rituximab therapy, gender, duration of disease and number of previous treatments did not correlate with response (Table 3).

Table 3 Comparison of some clinical parameters with response
Full size table

Outcome of patients refractory to rituximab therapy was generally poor in terms of response to subsequent therapies. Two of the 3 refractory patients underwent splenectomy without a response. Other types of treatment used in refractory patients included steroids, IvIg, azathioprine, mycophenolate, vincristine, danazol and romiplostim. These patients received a median of 2 treatments (range 1–4) after rituximab failure. The median platelet count of this group at the time of last follow-up was 16 × 109/l (range 6–128) with 7 out of 10 patients having a platelet count below 30 × 109/l. One of the patients in this group died 2 years after receiving rituximab. The cause of death (renal failure) in this patient was considered to be unrelated to ITP or rituximab therapy.

The adverse events related to rituximab therapy were generally mild. Seven (29%) patients had minor infusion related or subsequent reactions in the form of transient hypotension, palpitations, fever, and skin rashes. One patient with CLL who received rituximab therapy twice, required stopping and delaying the first infusions both the times due to significant hypotension but none of the patients required termination and cancelation of therapy.

Three patients died: one of renal failure-related complications after 24 months and 2 patients with CLL-associated ITP died of progression of the CLL 15 and 18 months post-rituximab treatment. Cause of death was not considered to be related to rituximab therapy or ITP in any of these patients.

4 Discussion

The anti-CD20 monoclonal antibody rituximab has been used to treat chronic ITP patients with variable responses [18–20]. Rituximab produced an initial response in approximately 40–60% of cases and sustained responses were observed in 35–67% of patients [21–25]. The results of our study, largest from the middle east, show that responses were achieved in two-third of the patients, similar to other published reports, but responses were mostly short lived and sustained responses were achieved only in about one-fifth of the patients. Schweizer and colleagues reported low rate of sustained responses in 14 patients. Response duration was short (8 weeks) and only around 20% of patients maintained responses beyond 6 months [26]. The results of long-term responses in our study are similar to those reported by Schweizer et al. and are in contrast with most of the other published reports. It is difficult to comment on the cause(s) of inferior long-term results in our study but this is likely to be due to diverse patient characteristics and disease etiology. It appears that the pathogenesis of ITP is heterogeneous with multiple mechanisms operating in different patients or even in the same patient [27]. The heterogeneity of pathogenesis along with differences in patient characteristics and criteria used for reporting responses and clinical outcomes in ITP, as described by Ruggeri et al. [28], is the possible explanation for conflicting results in some of the published reports.

Factors predictive of initial and long-term response to rituximab treatment in ITP patients have not been clearly identified and no clinical or biochemical parameters were consistently associated with outcome following rituximab therapy. Some of the factors reported to be associated with poor response include older age, failure to achieve a complete response, a delayed response, long standing disease, previous splenectomy, and heavy treatment with multiple regimens prior to rituximab therapy [20, 21, 23, 24, 29]. Splenectomy was associated with a poor response in our patients and 4 out of 5 patients with continued sustained remission at the last follow-up were non-splenectomized. Better responses to rituximab in non-splenectomized patients have been observed by some investigators [24]. One small study found better responses in splenectomised patients [26] while others did not report any correlation between response and splenectomy status [20, 25, 30]. No study has directly compared rituximab to splenectomy in patients with chronic ITP but such a study would be difficult to conduct because of involvement of a surgical procedure.

There is not much information available about heavily pretreated ITP patients who fail rituximab therapy and only few studies have reported the outcome of such patients [31].

In our study, the outcome of patients refractory to rituximab therapy was poor in terms of response to further therapy. It is possible that ITP patients who have long duration of disease and multiple previous therapies including splenectomy and fail rituximab, may be resistant to further therapies. A recent prospective study by Godeau et al. [29] reported results of rituximab use in ITP patients early in the course of the disease as a splenectomy-sparing agent after failure of first line therapy. Sustained responses were achieved in 33% of patients after 2 years and a younger age at the time of therapy was associated with a better response. It is interesting to note that 25 non-responders to rituximab subsequently underwent splenectomy and 15 (60%) of these patients responded to this treatment. Based on these results, it appears that early use of rituximab in the course of chronic ITP may be more effective and spare or delay splenectomy in some patients [21, 29]. Rituximab refractory or relapsed patients can still be expected to benefit from splenectomy. This conclusion is congruent with other lines of evidence suggesting that rituximab may be more effective in patients with a short duration of disease possibly as a result of the reversion of T cell abnormalities following depletion of the CD20+ B cell pool [32]. The observations by Godeau et al. will fuel the debate about the timing of rituximab in chronic ITP, but randomized controlled trials (RCTs) are needed to establish a definitive role for early rituximab administration in non-splenectomized ITP patients [33].

Rituximab also appears to be promising in secondary ITP associated with CLL [24, 34, 35]. Our report included 3 patients with CLL-associated secondary ITP. All the 3 patients responded to rituximab: two patients had complete responses and one patient had a partial response and one of these patients achieved a sustained response beyond 1 year. We believe that these responses were solely due to an independent effect of rituximab on ITP, as the drug was given alone to all 3 patients at the time of ITP diagnosis, although a secondary response due to effect on disease process contributing to platelet count improvement is possible and cannot be ruled out. Peñalver et al. [24] reported 14 patients with CLL-associated ITP treated with rituximab and response rates in these patients were similar to those with idiopathic disease. Our report reiterates the usefulness of rituximab therapy in this setting.

In conclusion, rituximab was well tolerated and achieved responses in two-thirds of relapsed or refractory ITP patients but sustained responses were achieved in only around one-fifth of the patients. Prior splenectomy was associated with a poor response and the outcome of patients refractory to rituximab treatment was poor in terms of further therapy. Rituximab can be useful in some patients with ITP and should be considered in patients with contraindications to, or not willing to undergo splenectomy. Controlled studies are needed to establish a definitive role for early rituximab administration in non-splenectomized ITP patients.