In this issue of Blood, data presented by Zhou et al1 and Lim et al2 on the use of rituximab (RTX) to treat immune thrombocytopenia (ITP) and thrombotic thrombocytopenic purpura (TTP), respectively, illuminate the importance of considering patient values and preferences in the interpretation of clinical evidence.
Zhou et al randomized 123 adults with persistent or chronic ITP who failed to respond to or relapsed after tapering corticosteroids to open-label RTX 100 mg weekly, alone, or in combination with recombinant human thrombopoietin (rhTPO), for 4 weeks. Treatment was well tolerated in both arms. The rate of complete response (platelet count ≥100 × 109/L) was greater (45.4% vs 23.7%, P = .026) and time to response was shorter (7 vs 28 days, P = .004) in the combination therapy arm. At first blush, these findings suggest important potential advantages of the combination regimen. However, evaluation of evidence in light of patient values and preferences offers a different perspective.
ITP is associated with reduced health-related quality of life.3 Although individual patient experiences vary, important sources of diminished quality of life include fear of bleeding and the unpredictable and disruptive nature of relapse. The addition of 4 weeks of rhTPO to RTX did little to address these patient-important outcomes. A platelet count ≥30 × 109/L is sufficient for prevention of serious bleeding in most patients and situations. Although the rate of complete response was augmented in the combination therapy arm, overall response (platelet count ≥30 × 109/L and doubling from baseline) was not (79.2% vs 71.1%, P = .36). In keeping with this observation, there was no significant difference in bleeding between the two treatment arms. Neither did the addition of rhTPO reduce relapse; sustained response rates were not significantly different in the treatment arms at 6, 12, or 24 months. Only 24.6% and 18.5% of patients remained relapse-free at 24 months in the combination and monotherapy arms, respectively, confirming the disappointing long-term platelet responses with RTX reported by other investigators.4,5
Lim et al provide a timely systematic review on RTX for the management of acquired TTP.2 They evaluated evidence regarding RTX for 3 distinct indications: (1) as initial therapy along with plasma exchange (PEX) and corticosteroids; (2) as supplemental therapy in patients with disease refractory to PEX and corticosteroids; and (3) as prophylactic therapy in asymptomatic patients with severe ADAMTS13 deficiency (<10%) following recovery from a clinical episode of TTP. Evidence is extremely limited and of poor quality for all 3 indications, highlighting the need for well-performed prospective studies in this area. Indeed, only one (historically) controlled study was identified for each indication. Based on their review, the authors suggest that RTX be considered for initial therapy (grade 2C) and recommend its use in refractory patients (grade 1C) along with PEX and corticosteroids.
On the contrary, Lim et al make a strong recommendation against the use of prophylactic RTX in asymptomatic patients with low ADAMTS13 activity (grade 1C). This recommendation is based primarily on a single cohort study comparing outcomes in 30 patients treated with RTX to 18 patients who did not receive RTX.6 Although median ADAMTS13 activity rose to 46% at 3 months and relapse-free survival was longer (P = .049) in the RTX group, Lim et al rightly point out limitations of the data. These include the use of immunosuppressive therapies in addition to RTX in the RTX group and longer follow-up in the control group, where both could have biased results in favor of the RTX group. They also note that ADAMTS13 activity did not increase in all patients after RTX and that activity measurement may vary based on the assay system.2 These are crucial limitations that undoubtedly undermine the strength of the evidence. Nevertheless, one is forced to re-examine the recommendation against prophylactic RTX when the evidence is viewed through the lens of patient values and preferences.
Relapse is not uncommon in patients with acquired TTP. In the Oklahoma registry, 34% of patients with TTP accompanied by severe ADAMTS13 deficiency (<10%) at the time of presentation relapsed over a median of 4.7 years.7 Several studies suggest that the risk of relapse is magnified in patients with severe ADAMTS13 deficiency in remission.8-10 TTP relapses are life-changing events. At best, they are terrifying and highly disruptive to patients’ lives. At worst, they remain fatal in approximately 10% of cases. Emerging evidence suggests that survivors manifest long-term cognitive and physical deficits.11 On the other hand, RTX, while not without toxicity, is safe and well tolerated by most patients. Faced with these facts, many patients (particularly those with multiply relapsed disease), in my experience, are willing to accept the uncertain benefits and potential harms of RTX for a shot at staving off relapse. I cannot say I blame them.
The reports by Zhou et al and Lim et al provide important contributions to our understanding of the management of ITP and TTP, and highlight the need for further studies. They also remind us that such studies should be designed with patient-important outcomes in mind and interpreted in the light of patient values and preferences.
Conflict-of-interest disclosure: The author declares no competing financial interests.