Slow responses to standard dose rituximab in immune thrombocytopenic purpura

We read with interest the article by Zaja et al. in which the results of a prospective multicenter Phase II study to assess the response rates of lower dose rituximab in adults with chronic immune thrombocytopenic purpura (ITP) were reported.1 In this single arm study 28 ITP patients received rituximab (100 mg/m) weekly for four weeks. An overall response (platelet count >50×10/L) and complete response (platelet count >100×10/L) was achieved in 21/28 (75%) and 12/28 (43%) of patients, respectively. Interestingly the time to treatment response with lower dose rituximab was longer than in published studies with standard dose (375mg/m weekly for four weeks).2–4 The median time to a complete response was 44 days with a range of 7–90 days.

In a recent prospective Phase II study which assessed the efficacy of standard dose rituximab in chronic adult ITP patients, most patients who responded to rituximab did so early and none of the patients who failed to reach a platelet count of 50×10//L in two weeks achieved a good response at one year.2

Figure 1.Response of chronic ITP patients over time, following first rituximab infusion. Complete response (CR), platelet count >100×10⁹/L partial response (PR), platelet count 50–100×10⁹/L. No response (NR). The arrows and overlying numbers represent patients changing response. Patients who required alternative treatment were classified as having no response.

In our center we have treated 11 patients (6 female and 5 male, mean age 50) with refractory chronic ITP with standard dose rituximab (375 mg/m weekly for four weeks) with similar response rates (6 patients reached a platelet count of >50×10/L, including 3 >100×10/L at six months) but we have noted that delayed responses to standard dose rituximab also occur (Figure 1). The best responses were seen in 2 female patients, aged 32 and 26 years, with baseline platelet counts of 4 and 25×10/L. Their platelet counts at one month were 87 and 84×10/L and at six months were 521 and 230×10/L, respectively without any further treatment. In fact in our experience, at one month, no patient had platelets > 100×10/L (Figure 1). In a systemic review, complete responses usually occurred 3–8 weeks after the first infusion of rituximab.5 However, our data show that delayed responses to standard dose rituximab can occur. These responses are unlikely to reflect spontaneous remission as this rarely occurs in adult chronic ITP.6

Splenectomy has been considered standard second-line therapy for ITP.7 The response rate is about 65% but it is associated with a mortality of 0.2–1% and morbidity of 9.6–12.9% depending on the age of the patient and technique used.8 There is accumulating evidence that rituximab can be a safe and effective way to defer splenectomy particularly in younger patients.9 A lower dose rituximab regimen would result in considerable cost savings compared to a standard dose regimen and may be associated with less transfusion related reactions. While delayed responses may occur in the lower dose regimen they may also occur with standard dose rituximab. Since maximal response to both dosing regimens may be delayed, a decision regarding splenectomy should not be made until at least six months after rituximab therapy.