Randomized clinical trials in thrombotic thrombocytopenic purpura: where do we go from here?

Thrombotic thrombocytopenic purpura (TTP) was one of the most enigmatic disorders in transfusion medicine until 1998 when investigators reported that many patients lacked a key enzyme responsible for regulating the size and function of the von Willebrand factor (VWF) molecule.1, 2 The enzyme, now known as ADAMTS-13, depolymerizes sticky “strings” of ultralarge VWF (ULVWF) secreted by endothelial cells, thereby releasing smaller multimers into the blood stream.3 A deficiency of ADAMTS-13 has been observed in 33 to 100 percent of clinically diagnosed acute TTP patients and is associated with an autoantibody inhibitor in most cases.4, 5 In the absence of ADAMTS-13, the endothelial-bound ULVWF anchors platelets (PLTs) to the vessel wall, causing the characteristic microvascular occlusions in the brain, kidney, heart, and other organs. Decades before the pathophysiologic basis of TTP became known, clinicians observed beneficial effects of transfusion therapy in reversing the fatal course of the disorder.6 Further success with whole blood and plasma exchange (PE)7 heightened interest in the mechanism of action: Was it replacement of a deficient factor or removal of an inhibiting substance in the plasma, and which was better, plasma infusion (PI) or PE? Under the auspices of the Canadian Apheresis Group (CAG), Rock and colleagues8 conducted a multicenter randomized controlled trial (RCT) comparing PE (1.5 plasma volume daily for first 3 days followed by 1 plasma volume daily) with PI (30 mL/kg on first day followed by 15 mL/kg daily).8 Both arms used fresh-frozen plasma (FFP) as the replacement fluid. A crossover design was employed in the event of a failure to respond or if the treatment could not be tolerated (this was due mainly to volume overload, a frequent occurrence in the plasma infusion arm). PE proved superior to PI for both early and late time periods studied: survival at Day 9 was 49 of 51 (96%) versus 43 of 51 (84%) and at 6 months 40 of 51 (78%) versus 32 of 51 (63%; p = 0.036). Because patients treated with PE received three times more plasma, it could not be determined whether the benefit of PE was achieved because more plasma was given or the effect of plasma removal. Henon,9 however, conducted a multicenter RCT in 40 patients comparing PE and PI, with one important difference: Patients in the PE arm received the same amount of plasma (15 mL/kg FFP plus 45 mL/kg albumin) as those in the PI arm (15 mL/kg FFP only). The French trial also strongly favored PE, with 85 percent survival in this group compared to a lower remission rate and only 57 percent survival in the infusion arm. Although retrospective studies with smaller numbers of patients have reported similar outcomes with PE and PI,10, 11 the randomized controlled studies established PE as the standard of care in the management of TTP. With regard to mechanism, we now know that both replacement (of ADAMTS-13) and removal (of autoantibodies) are beneficial. RCTs have also helped to guide the choice of replacement fluid used for PE in TTP. Molecular size is a key determinant of the PLT-binding properties of VWF, with larger multimers being stickier. Therefore, it was suggested that the cryoprecipitate-depleted fraction of plasma (cryosupernatant [CSP]), having only approximately 20 percent of the VWF content as FFP, might be better. Retrospective series in FFP-refractory patients reported favorable outcomes after switching to CSP.12, 13 This observation prompted the CAG to study the use of CSP as initial replacement in 40 untreated patients.14 Compared to historical results with FFP, the response rate of 75 percent and survival of 95 percent at 1 month was improved. Based on these promising results, two small-scale RCTs have been performed. Under the cooperative umbrella of the North American TTP Study Group, Zeigler and coworkers15 reported equivalent clinical and laboratory results in patients randomly assigned to FFP (n = 13) or CSP (n = 14).15 The CAG studied 52 patients and found that 85 percent of the FFP-treated patients and 88 percent of those in the CSP group responded.16 There were no differences in mortality at 1 month. Although their protocol was originally designed to evaluate more than 200 patients for 30-day survival as the primary endpoint, it ended prematurely due to a change in funding and shift in interest from CSP to virally inactivated plasma. The investigators concluded that although the study was seriously underpowered, there was no apparent advantage to the use of CSP. Thus, although equivalency between FFP and CSP has not been definitively proven, many clinicians no longer believe that CSP is inherently the better product. In this issue of TRANSFUSION, Mintz and colleagues17 report the results of their double-blind, multicenter, Phase III trial of standard FFP versus photochemical treatment of plasma with amotosalen and ultraviolet (UV)A light (PCT-FFP).17 Patients who have TTP typically require multiple units of plasma (a mean of 150 units per patient in this study) and relatively few units of red blood cells to obtain remission. The rationale for the use of a pathogen-inactivated plasma product—decreased exposure to pathogens including new epidemic viruses and possibly a decreased incidence of transfusion reactions—may be especially appropriate in this patient population. Previous investigations have confirmed the stability of coagulation factors, anticoagulant proteins, and ADAMTS-13 in amotosalen-treated plasma.18, 19 The primary endpoint was the attainment of remission within 30 days (essentially, normalization of PLT count for 2 consecutive days). PE, targeted at a plasma dose of 1 vol per treatment, was continued for an additional 5 days after remission. Relapses were assessed over a 60-day period. Glucocorticoids (1 mg/kg/day prednisone or equivalent) were used in all patients. Overall, 14 of 17 (82%) TTP patients and 16 of 18 (89%) of the control patients reached the primary endpoint (p = NS). No significant differences were seen in the secondary endpoints of the study, including time to remission (median, 6 days in both groups), relapse rates, time to relapse, number of PEs, and the amount of plasma exchanged. The investigators noted a comparable incidence of adverse events and transfusion reactions in the two groups. This small RCT provides encouraging preliminary evidence for the suitability of PCT-FFP with amotosalen and UVA in the treatment of TTP. Additional study in more patients is needed to establish that it is equivalent to standard FFP. With regard to the need for adequately powered studies, it may be useful to reflect on issues arising from the development of another virally inactivated product, solvent/detergent-treated plasma. This product was compared to standard FFP in two small-scale RCTs (involving a total of 88 TTP patients), with similar reported efficacy and safety.18, 19,20 A retrospective study, however, has raised a question of venous thrombosis in seven of eight patients (mostly involving central catheters), along with slightly decreased levels of protein S.21 Only through trials with adequate power can the exact role of new therapies be fully evaluated. Although there were also no major differences in the two study groups, Mintz and colleagues provide interesting observations on ADAMTS-13 measurements in patients with TTP. Consistent with other studies, nearly 40 percent of their patients had sufficient clinical evidence to diagnose TTP, without having a severe ADAMTS-13 deficiency. Additionally, 38 percent of the patients went into remission despite having undetectable ADAMTS-13 levels; an even higher number (50%) did so with an inhibitor still present. These results underscore the uncertainty of our present state of knowledge in the use of measurements of ADAMTS-13 for the clinical management of TTP.22 All of the recent studies have been hampered by logistic problems leading to low patient accrual. TTP is considered an uncommon disorder, with an estimated incidence of only 4 to 11 annual cases per million-population.22 Most major medical centers see relatively few patients each year, so cooperative studies are essential to accrue sufficient patients. Ideally, they should also have an infrastructure that can enroll and study patients over an extended period of time. The CAG PE/PI study took 7 years to acquire 103 patients to achieve adequate statistical power.8 Second, TTP has multiple forme frustes and secondary associations that require skillful assessment by experienced personnel. Mintz and coworkers screened a total of 190 patients to find 37 patients who met eligibility criteria for the trial. Most of the excluded patients had secondary causes such as malignancy and stem cell transplantation. Another potential limitation that should be recognized is that TTP frequently presents as an acute, life-threatening disorder requiring rapid clinical decision making. Patients and their families may not be receptive to the extensive informed consent process required of clinical trials. Although it is true that PE with plasma replacement has markedly improved the outlook of patients with TTP, there is still considerable room for improvement. In studies reported since 1987, the overall mortality rate ranged from 10 to 22 percent, exacerbation (within 30 days of obtaining remission) occurred in 22 to 45 percent, and late relapses were reported in 13 to 40 percent of cases.4, 23-26 The need for extended PE is resource intensive and may itself cause significant morbidity.27 Additionally, PE does not address the primary autoimmune basis of the disorder. So, where do we go from here? The Transfusion Medicine Hemostasis Clinical Trials Network (TMH CTN)28 has been established by the NHLBI to conduct large-scale studies in disorders that require a multicenter trial format for successful completion. In view of the strong rationale for the use of immunomodulation therapy in TTP, the network has designed an RCT named “STAR” (study of TTP and rituximab), to test whether early use of rituximab will improve the outcome of patients being treated with PE and corticosteroids.29 Rituximab, a chimeric monoclonal antibody directed at the CD20 antigen expressed on the surface B lymphocytes, has been used successfully in the therapy of both refractory and relapsed TTP. Several recent publications describe the experience to date.30, 31 The target population will be patients with idiopathic TTP who will be randomly assigned within a few days after diagnosis to receive four weekly rituximab infusions (375 mg/m2), a schedule similar to patients with other autoimmune hematologic conditions. The primary endpoint is the attainment of a complete remission (PLT count of 150 × 109/L and near normal LDH level) by Day 21 and to maintain remission off PE for an additional 30 days. A number of secondary endpoints will be examined as well, such as defining the behavior of ADAMTS-13 levels and inhibitor levels longitudinally in relationship to remission and relapse in the two study arms. For example, does ADAMTS-13 deficiency predict responsiveness to rituximab or will non-ADAMTS-13–deficient patients also respond as appears to be the case with plasma exchange?4 A number of issues have been considered in the design of the trial. Patients with secondary disease associations will be excluded, because many of these patients appear to have a different pathophysiology and response to treatment.3, 22 For successful recruitment, it was important for the protocol to reflect a reasonable consensus of care in light of the broad institutional variability in the management of TTP.25 Clinical issues regarding choice of replacement fluid (FFP preferred, CSP and thawed plasma permitted), volume processed (1.25 plasma volumes/day), and whether tapering of TPE would be performed (No) were resolved. Another key decision concerned the timing of rituximab: should it be given early or later after failure to respond to standard care? Most deaths in TTP occur early, within one or two weeks after diagnosis. Serious complications including cerebrovascular events, myocardial injury, and renal failure, also tend to occur early and unpredictably. It was agreed that early intervention might have the biggest impact in reducing ADAMTS-13 autoantibody production and the clinical course of the disorder. The study will accrue 236 patients (118 per arm) over a 3-year period. In addition to the 17 trial sites of the TMH CTN, plans have also been formulated to recruit three to five nonnetwork sites to meet the recruitment goal. Looking ahead, it is hoped that this national trial can successfully capitalize on the modern insights into TTP, to determine whether an immunomodulatory approach will improve the outcome of patients with this less enigmatic but potentially devastating disorder.

Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.