International Journal of Hematology

, Volume 91, Issue 1, pp 36–45

Thrombotic thrombocytopenic purpura: recognition and management


    • Division of Hematology/Oncology, Department of MedicineUniversity of Pittsburgh School of Medicine
    • Therapeutic Hemapheresis and Blood ServicesThe Institute for Transfusion Medicine
Progress in Hematology Recent advance in thrombotic thrombocytopenic purpura

DOI: 10.1007/s12185-009-0478-z

Cite this article as:
Kiss, J.E. Int J Hematol (2010) 91: 36. doi:10.1007/s12185-009-0478-z


Thrombotic thrombocytopenic purpura is a life-threatening multisystem disorder that represents both a diagnostic and a management challenge to clinicians. Early recognition of the condition coupled with rapid institution of plasma exchange has led to a dramatic improvement in prognosis. Studies performed over the past decade have elucidated the predominant pathophysiology, stemming from a deficiency of ADAMTS13, that accounts for the widespread microvascular deposition of platelet–von Willebrand factor in many sites, including the brain, kidney, and mesenteric vessels. However, in light of the mortality rate of 10–20%, much work remains to be done to translate advances in our understanding of pathophysiology into clinical practice. Improvements in medical management using immunosuppressive and other drugs are being actively explored in clinical trials. Agents that target ADAMTS13 autoantibody production by B-cells, such as anti-CD20 monoclonal antibodies, have the potential to shorten the duration of plasma exchange treatment, reduce relapses, and transform the management of this once enigmatic disorder.


Thrombotic thrombocytopenic purpura (TTP)Thrombotic microangiopathyPlasma exchangeADAMTS13

1 Introduction

A great deal has been learned over the last decade regarding the etiology and pathophysiology of thrombotic thrombocytopenic purpura (TTP), a life-threatening multisystem disorder resulting from platelet and von Willebrand factor deposition in arterioles and capillaries throughout the body. Molecular defects and antibody-mediated deficiency in ADAMTS13 provide the central pathophysiological basis for understanding this disorder. The new paradigm is beginning to change the clinical approach for patients who have thrombotic microangiopathies and is also setting the stage for studies that are designed to answer important questions regarding management. Over the next few years we should have a clearer picture of which patients receive benefit from plasma exchange therapy, which should receive powerful immunosuppressive drugs, and whether to use these agents early in the course of disease (adjuvant) or later as a rescue therapy. This review will focus both on the diagnostic challenges and on current management of these disorders, with a particular emphasis on TTP.

2 Classification

Thrombotic microangiopathies (TMAs) refer to a diverse group of disorders characterized by microangiopathic hemolytic anemia and thrombocytopenia associated with platelet deposition in the microcirculation. More or less distinct categories of TMA’s has been described (Table 1), which differ in causative mechanism, target patient population, and pattern of organ involvement. However, clinical and laboratory features frequently overlap, causing diagnostic uncertainty. TMAs may occur on an idiopathic basis or in association with pregnancy, autoimmune conditions, infections, malignancy, stem cell transplantation, and upon exposure to certain drugs. A number of reference centers and registries have been formed in various countries to better catalog and manage patients with these disorders [15]. Investigation continues as to whether certain secondary disorders should remain in this classification, and whether plasma exchange is efficacious.
Table 1

Primary and secondary thrombotic microangiopathies (TMA)

Thrombotic microangiopathies and thrombotic thrombocytopenic purpura

Diverse clinical presentations

Primary TTP

Secondary TMA


  Autoimmune diseases, vasculitis



Primary HUS

    Hypersensitivity: quinine, ticlopidine, clopidogrel


    Dose related: cyclosporine, cancer chemoRx








  Transplant-related: allogeneic stem cell transplantation


  Shigatoxin-associated (E. coli 0157:H7)

2.1 Primary TMA

Hereditary TTP, or Upshaw–Schulman syndrome, is the result of a congenital deficiency of ADAMTS13. Over 70 mutations in the ADAMTS13 gene have been identified in this rare autosomal recessive disorder. Patients present during the first decade or early adult life. Neonatal jaundice may be confused with maternal blood type incompatibility and treated by exchange transfusion with improvement on a temporary basis. Some patients have received corticosteroids based on a mistaken diagnosis of Evans syndrome. A precipitating event, such as infection, surgery, or pregnancy frequently triggers the full clinical episode of hemolytic anemia and thrombocytopenia. ADAMTS13 activity is undetectable and inhibitor is not present. Low levels of ADAMTS13, generally 5 to 10%, are sufficient to prevent clinical manifestations. Remission is maintained by giving one or two units of plasma every three weeks. Recombinant ADAMTS13 has also been used [6].

The most common form of TMA, idiopathic TTP, results from an autoantibody that binds to ADAMTS13, inhibiting its VWF multimer cleaving-activity. A few cases have been described in which non-neutralizing antibody accelerates the clearance of ADAMTS13 [7]. Idiopathic TTP comprises 40 to 77% of all TMA cases in large series [2, 4]. The reported frequency of severe ADAMTS13 deficiency varies from 30% to 100% (see Table 2). Inhibitory activity in functional assays is reported in 38–95% of patients, with higher prevalence reported in those with severe ADAMTS13 deficiency. ELISA assays for the measurement of autoantibodies to ADAMTS13 appear to have greater sensitivity than inhibitor measurements. In a cohort of TTP patients with ADAMTS13 activity <10%, 97% had IgG antibodies versus 83% in whom inhibitors were detected [8]. Demographic studies report a preponderance of females, approximately 2–3 to 1 and black race, approximately 9 to 1 compared to whites in US [2]. Intensive plasma exchange remains the standard of care, with immunomodulatory therapy assuming a greater therapeutic role (see below).
Table 2

Prevalence (%) of ADAMTS 13 deficiency in patients clinically diagnosed with thrombotic thrombocytopenic purpura (TTP)



No. of patients

Percentage of pts. w/severe ADAMTS 13 deficiencya

Percentage of pts. w/partial ADAMTS 13 deficiencyb

Furlan et al./Tsai and Lian [63, 64]





Veyradier et al. (prospective cohort) [65, 66]





Mori et al. [28]





Vesely et al. (prospective cohort) [1]





Peyvandi et al. [67]





Coppo et al. [68]





Zheng et al. (prospective cohort) [27]





Matsumoto et al. (retrospective, “idiopathic TTP”) [5]





Scully et al. (retrospective, “idiopathic TTP”) [4]





aSevere deficiency of ADAMTS 13 activity denotes plasma levels of <10%, except Matsumoto <3%

bReduced ADAMTS 13 activity denotes plasma levels below the lower limit of the laboratory reference interval (usually 40% to 140%)

Modified from [26]

HUS presents as acute renal failure in association with microangiopathic hemolytic anemia and thrombocytopenia. A confusing nomenclature has developed. Table 1 lists hereditary and idiopathic forms under the heading of primary HUS (also referred to as “atypical” or “sporadic” HUS, in contradistinction to the “typical,” “epidemic” or diarrhea “D+” secondary form caused by shigatoxins). In patients with hereditary HUS, TPE may be used to replace missing complement regulatory proteins, such as Factor H, Factor I, or CD46/Membrane Cofactor Protein [3]. In HUS, ADAMTS13 levels are generally normal or slightly decreased [9] with only rare HUS patients having severe deficiency [10], and inhibitors are not detected. For this reason, the therapeutic rationale and indications for employing TPE is uncertain in HUS [11]. However, clinical overlap with idiopathic TTP is great enough that some authorities recommend TPE [1]. For example, bloody diarrhea, a feature largely seen in children with epidemic E. coli 0157:H7 infection, may also be seen in adults. Karpac et al. reported on 21 patients, who comprised 6.7% of “TTP/HUS” cases in the Oklahoma registry. They report similar response to TPE as patients with severe ADAMTS13 deficiency (81% vs. 89%), fewer treatments (avg. 10 vs. 20) with lower rate of exacerbation (6% vs. 47%) and no relapses in survivors versus 39%, however, they also observed a greater mortality rate (33% vs. 13%) than those with severe ADAMTS13 deficiency. A few other reports also describe apparent efficacy of TPE in idiopathic HUS [12]. Thus, despite the lack of severe ADAMTS13 deficiency and an uncertain mechanism of action, a therapeutic trial of TPE may be considered in these patients.

2.2 Secondary TMA

2.2.1 Collagen vascular diseases

Collagen vascular diseases frequently serve as the clinical “backdrop” both for TTP and for immune thrombocytopenic purpura (ITP). Differentiation between the two can be particularly vexing in the setting of systemic vasculitis which can also cause red blood cell fragmentation. The clinical picture may be further confused by central nervous system manifestations, such as seizures and renal failure which may be seen both in SLE and in TTP. In addition to corticosteroids and/or other immunosuppressants, plasma exchange should be instituted if there is any question of the diagnosis. The finding of severe ADAMTS13 deficiency and evidence of an inhibitor and/or autoantibody may be of particular help establishing the diagnosis of TTP (see below).

2.2.2 Drug-induced TMAs

Drug-induced TMAs have been classified as immunologic-mediated and dose-dependent. Ticlopidine has been shown to induce an antibody to ADAMTS13, resulting in a syndrome closely resembling idiopathic TTP. Treatment consists of withdrawing the drug and prompt initiation of plasma exchange because survival appears to be less if delayed more than 3 days [13]. The related thienopyridine, clopidogrel, has also been implicated at a much lower frequency. It usually occurs early, within 3–14 days after initiation of therapy [14]. The mechanism does not appear to be immune-related; direct endothelial cell injury has been postulated. Quinine, used for treatment of leg cramps, also has been reported to cause TMA with a high rate of acute renal failure (HUS) that may become chronic [15]. It does not appear to be associated with ADAMTS13 deficiency and efficacy of plasma exchange has been questioned [16]. Dose-dependent drug toxicity may be associated with cancer chemotherapeutic agents including mitomycin C, pentostatin, and gemcitabine. Cumulative toxicity to endothelial cells is the presumptive mechanism with insidious onset, sometimes after the drug is discontinued. The syndrome can also be seen with calcineurin inhibitors including cyclosporin A and tacrolimus. Toxicity may be specific to the drug and not class specific. Therefore, switching to the alternative agent may be beneficial in cases where discontinuing drug is not a good option because of possible organ rejection. If drug substitution is not possible, clinical improvement has also been reported with dosage reduction [13]. The role of plasma exchange is uncertain.

2.2.3 Pregnancy and postpartum

Pregnancy and postpartum are critical periods for autoimmune diseases, in particular TTP, to become manifest. This should be distinguished from a number of pregnancy-associated TMAs, including severe preeclampsia, hemolysis with elevated liver enzymes and low platelets (HELLP) syndrome, acute fatty liver of pregnancy, SLE, and anti-phospholipid syndrome [3]. TTP diagnosed during pregnancy comprises approximately 7% of all TTP cases reported in registry data [2]. Nearly all of these cases are acquired but pregnancy has also been reported as the triggering circumstance in patients with congenital TTP [17]. Accurate diagnosis is paramount particularly in relation to HELLP, because plasma exchange is indicated for TTP and fetal/placental delivery is therapeutic in HELLP syndrome. HELLP syndrome usually occurs during the third trimester of pregnancy or immediately postpartum. Along with signs of severe preeclampsia, including hypertension and nephrotic range proteinuria, evidence of liver injury is present, signified by marked elevation of transaminases, and coagulopathy. A mild decrease in ADAMTS13 levels is noted during the second and third trimester of pregnancy; severe deficiency is not found in HELLP syndrome and strongly favors a diagnosis of TTP.

2.2.4 Human immunodeficiency virus infection TMA

Human immunodeficiency virus infection TMA has been associated with more advanced HIV disease (lower CD4 lymphocyte counts and higher viral load), and clinical AIDS. Fortunately, the era of highly advanced anti-retroviral therapy (HAART) has brought about a decline in incidence of HIV-TMA to well-under 1% of HIV-infected patients [18]. A high response rate is reported to plasma infusion and anti-retroviral therapy, reducing the need for plasma exchange in treating this complication [19, 20]. ADAMTS13 levels are generally not decreased, supporting an alternative mechanism for TMA in these patients [16].

2.2.5 Cancer-associated TMAs

Cancer-associated TMAs may occur as the presenting manifestation of advanced malignancy, including neoplasms of the lung, GI tract, breast, and others. Microvascular tumor emboli may cause microangiopathic hemolysis and thrombocytopenia may be due to both consumption (low grade DIC) and impaired production, the latter resulting from direct marrow infiltration. The key clinical differences include weight loss and the presence of localizing symptoms, such as cough, dyspnea, and bone pain, and a greater time interval from initial symptoms: median 21 days versus 8 days in idiopathic TTP [21, 22]. Bone marrow biopsy is usually diagnostic.

2.2.6 Hematopoietic stem cell transplant-associated TMA

A syndrome resembling TTP may occur in the setting of allogeneic hematopoietic stem cell transplantation, especially using matched unrelated donors. Onset generally occurs within 30–50 days following transplant, usually in the setting of infection (which may be occult) and/or acute graft versus host disease [23]. ADAMTS13 levels are not in the severely deficient range. Mortality is in excess of 80% with or without plasma exchange. The bone marrow transplant clinical trials network has issued a consensus opinion that: (1) the primary intervention should be to withdraw calcineurin inhibitors, (2) there is insufficient data to recommend other immunosuppressive Rx (e.g., corticosteroids) but they are “reasonable” to use, and (3) plasma exchange should not be considered standard of care [24]. The management of this challenging complication at present rests on supportive care and mitigation of triggering stimuli (i.e., graft vs. host disease and infection).

2.2.7 Shigatoxin-associated (E. coli 0157:H7)

Epidemic HUS is associated with shigatoxin-producing bacteria including E. coli O157H:7, or other shigatoxin-producing organisms. Epidemic HUS is the most common cause of HUS in children and recovery is the rule. It accounts for only a few percent of adult cases, mainly in the elderly and is associated with serious neurologic complications and a poor prognosis [11]. ADAMTS13 levels are generally normal. Treatment consists of supportive measures including control of blood pressure and dialysis as needed. There is no established indication for TPE in the management of epidemic HUS in children or adults [11].

3 ADAMTS13 measurements

The increased sophistication of ADAMTS13 assays has been useful in classifying and managing patients with TMAs. However, availability varies and diagnostic sensitivity remains an issue.

3.1 Diagnosis

The normal range of ADAMTS13 differs according to assay, but averages 65 to 150% of normal. Although “severe deficiency” has not been precisely defined, there is a high degree of diagnostic specificity for TTP in patients who have undetectable or very low (<5–10%) levels. The overall proportion of patients with severe deficiency ranges from 33 to 100% (Table 2). Several factors have been proposed to account for this wide variability. Although technical assay differences exist, the most likely explanation involves patient selection. For example, studies that included patients with greater degrees of renal failure or HUS diagnosis reported a lower percentage of patients with severe deficiency than studies which excluded such patients [1, 25]. The same is true for studies which include patients with secondary causes, which have a reduced prevalence of ADAMTS13 deficiency. Rare patients with severe sepsis-induced DIC, transplant recipients, and cancer have also been reported to have very low ADAMTS13 levels [3]; these patients should be distinguishable on clinical grounds. Diagnostic uncertainty increases with mild to moderate ADAMTS13 deficiency because of overlap with many other clinical disorders, including pregnancy, sepsis multiorgan failure, liver disease, and postsurgery [26]. As noted in Table 2, even normal levels of ADAMTS13 may be found in a significant minority (perhaps 10 to 15%) of patients with clinically diagnosed TTP. Therefore, a normal value should not be used as a screening test, in the absence of clinical judgment, to exclude the diagnosis. This task often falls to the consulting hematologist to evaluate alternative diagnoses in the patient with thrombocytopenia and many potential causes of red blood cell fragmentation, including sepsis with or without DIC, malignant hypertension, vasculitis, metastatic malignancy (tumor emboli), severe pulmonary hypertension, and others. The proportion of patients with inhibitors ranges from 44 to 95%, with the highest frequency found in patients with severe ADAMTS13 deficiency [1, 27, 28]. ELISA methodology to detect autoantibodies to ADAMTS13 is reported to be more sensitive than inhibitory assays. Rieger found IgG antibodies in 97% of patients with ADAMTS13 activity <10% [8].

3.2 Management

In patients with severe ADAMTS13 deficiency, some studies have noted the attainment of a “detectable” level during TPE to be associated with clinical responses [27]. This generally occurs in patients with no or low inhibitor activity, in contrast to patients with high titer inhibitors. Other studies note that many patients enter remission despite persistence of severe ADAMTS13 deficiency [29]. Clinical and laboratory parameters, especially platelet count, remain the best measure to assess remission status and the decision to discontinue plasma exchange (Fig. 1).
Fig. 1

General management approach in patients with TTP. Adapted from George JN. Blood. 2000;96:1223–9. TPE therapeutic plasma exchange, PV plasma volume, LDH lactic dehydrogenase

3.3 Prognosis

At the present time, the main clinical value of ADAMTS13 activity assays is as a marker for outcome and the potential for relapse. Compared to patients with non-severe deficiency, severely deficient patients experience higher rates of remission (82–88% vs. 20–75%) and lower mortality (8–18% vs. 18–80%) [1, 27, 28, 30]. The high mortality rate in non-severely deficient patients may be due to the higher proportion of secondary causes and death from underlying diseases, such as patients with hematological malignancies treated using hematopoietic stem cell transplantation. In patients who have a severe deficiency of ADAMTS13, Coppo reported that patients with inhibitory anti-ADAMS13 antibodies take substantially longer to achieve clinical remissions (21 vs. 7 days) and require a higher plasma volume than patients with no measurable inhibitor activity [33]. Both the persistence of ADAMTS13 deficiency and the presence of inhibitory antibodies at clinical remission are also predictive: the relapse rate at 18 months was 38.5% in patients who had undetectable ADAMTS13 activity in remission compared to 5% in those with detectable levels [32] and another study found a 3-fold higher risk of relapse in those with continuing ADAMTS13 deficiency and inhibitor [31]. These observations support the notion that autoantibodies to ADAMTS13 may persist, causing ADAMTS13 deficiency and susceptibility to a “second hit” attack of inflammatory stimuli that may release ULvWF and provoke microvascular thrombosis. Immunosuppressive drugs may be particularly beneficial in these at-risk patients [2, 8, 27, 34].

4 Treatment

4.1 Plasma exchange

Therapeutic plasma exchange is the only therapy demonstrated to be effective in randomized clinical trials (see Table 3). The Canadian Apheresis Study Group compared plasma infusion (30 ml/kg initially, then 15 ml/kg daily) with plasma exchange (1.5 plasma volume/day for 3 days, then 1 plasma volume per day) and demonstrated increased survival in the plasma exchange arm, 78% versus 63% [35]. Plasma exchange delivers a greater ADAMTS13 dose without circulatory overload and removes antibodies to ADAMTS13. Both mechanisms appear to be important, based on a French study that gave similar amounts of replacement plasma in each arm of a randomized trial using TPE, yet found increased survival in the group that was also exchanged using albumin solution [36].
Table 3

Randomized clinical trials in acute TTP






Rock et al. [35]


TPE versus PI

3× more plasma in TPE arm

Survival 78% versus 63%

p = 0.036

TPE superior; volume overload in PI group

Henon [36]


TPE versus PI

Plasma dose equal

Survival 85% versus 57%

TPE superior

Bobbio-Pallavincini et al. [69]


Antiplatelet Rx: ASA + dipyridamole 15 days versus SOC, then ticlopidine maintenance 1 year

Survival day 15: 97.2% versus 86.5% (NS)

Bleeding not worse using antiplatelet drugs

Zeigler et al. [70]


FFP versus CSP

Response comparable; survival 23% versus 21% (NS)


Rock [71]


FFP versus CSP

Response: 82% versus 85% (NS)


Mintz et al. [72]


FFP versus PCT FFP

Response: 89% versus 82% (NS)


TPE therapeutic plasma exchange, PI plasma infusion, SOC standard of care, CSP cryosupernatant plasma, PCT photochemical treated

Because of the potential for sudden clinical deterioration, treatment should be initiated as soon as possible after diagnosis. Plasma infusion 15–30 ml/kg should be given in the event that plasma exchange cannot be instituted in a timely manner. The treatment approach consists of a 1 to 1.5 plasma volume exchange with plasma daily until clinical symptoms have resolved and the platelet count has reached a normal level (≥150,000/μl). FFP, 24 h plasma, thawed plasma, and cryosupernatant (AHF, cryoprecipitate removed) plasma are considered equivalent because of comparable levels of ADAMTS13 [37]. Although the reduced levels of von willebrand factor present in cryosupernatant may provide a theoretical benefit, small case series and randomized clinical trials have found no difference in outcome (see Table 3).

Even with TPE performed daily the recovery of platelet levels is frequently delayed or associated with unexpected declines. An “inhibitor rebound” phenomenon has been observed in some responding patients after 7–10 days, which may be associated with refractoriness despite continuing TPE [38]. A single center study grouped platelet responses in four patterns: Group I, platelets rose then declined to <100,000/μl; Group II, platelets declined despite TPE; Group III, platelets rose continuously, and Group IV, platelets decreased to 100,000–150,000/μl plateau after initially rising [39]. The latter has been termed a “pseudo-refractory state,” and the authors suggest that TPE may be safely discontinued with gradual improvement in platelet levels. These investigators also suggest that continued TPE in the face of declines in platelet levels in Groups I and II—“stay the course”—is in most cases sufficient to reverse the decline and achieve eventual remission.

Although platelet counts are the most important measure of clinical response, LDH level should also be monitored since it reflects ongoing tissue ischemia as well as hemolysis [40]. However, clinical experience suggests that less than normal levels (<1.5 × normal) may be an acceptable endpoint to discontinue plasma exchange. Shistocyte measurements are not predictive of ability to maintain remission after cessation of TPE [41]. Although definitions vary, signs and symptoms which recur within 30 days are considered exacerbations and after 30 days, relapses.

The US TTP Apheresis Study Group (USTTPASG) surveyed 20 apheresis centers in the U.S., finding that 8 (40%) routinely utilized a taper regimen after achievement of remission [42]. The exact regimen was not specified and probably varied from center to center (e.g., TPE qod x3, q3d x2, then off). The Canadian Apheresis Study Group employed a routine taper schedule, consisting of 5 plasma exchanges over a two-week period [35]. In the USTTPASG study there were no significant differences in early relapse rates between the taper/no taper groups overall. A tapering schedule of plasma exchanges has not been recommended by any of the professional associations, including the AABB and the American Society for Apheresis. The Society for Hemostasis and Thrombosis (London), recommends continuing plasma exchange for a minimum of two days after attainment of remission [11].

In studies reported from 1987 to 2003, the mortality has ranged between 10% and 22%, exacerbation rate within 30 days of remission occurred in 22 to 45%, and late relapses were reported in 13 to 40% of cases [1, 4244]. More recent studies of patients with severe ADAMTS13 deficiency report mortality in the range of 5–10% and relapses in 30–50% [25]. Although TPE has clear-cut benefit, it does not correct the primary autoimmune defect of the disorder. A number of patients still incur protracted morbidity and mortality due to disease severity. Extended plasma exchange may itself cause adverse outcomes. The 12-year experience of the Oklahoma Registry group reports a 26% risk per patient for major complications of plasma exchange (each receiving an average 8 TPE procedures), including a 2.8% fatality rate [45]. The complications were primarily catheter-related infections, sepsis, bleeding and thrombosis.

4.2 Other therapies

4.2.1 Antiplatelet agents

Platelet inhibitors have not been demonstrated to be of benefit in the treatment of TTP and may increase the risk of hemorrhage, particularly with severe thrombocytopenia. An Italian group reported similar response rate and survival at 15 days comparing ASA + dipyridamole to standard of care (notably without excessive bleeding in the anti-platelet treatment group). They also reported a reduced relapse rate using ticlopidine as a maintenance strategy over a 1-year period: 6.25% ticlopidine-treated group versus 21.4% controls (p = 0.018%) (see Table 3). However, both ticlopidine and clopidogrel have been associated with drug-induced TTP [13], limiting their usefulness. The Society for Hemostasis and Thrombosis recommends ASA after the platelet count has risen to greater than 50,000/μl [11]. Novel agents that interfere with vWF–platelet interactions are currently being investigated.

4.2.2 Immunosuppressive treatment

Anecdotal case series and uncontrolled clinical trials in small numbers of patients suggest that immunosuppressive therapies may be effective in TTP. Improved responses have been attributed to early treatment with corticosteroids [46] or vincristine [47]. These agents are also frequently used as a second line therapy after failure to attain sustained remission with plasma exchange. The role of both autoantibody formation and inflammation in idiopathic TTP strengthens the rationale for using corticosteroids in this disorder. However, efficacy remains unclear, with comparable response rates with or without use of corticosteroids in studies reported in the literature [48]. In one retrospective case series, seven of eight patients responded to vincristine 1.4 mg/m2 IV on day one followed by 1 mg on days 4 and 7 [49]. Vincristine has been shown to increase platelet count and ADAMTS13 levels in patients with TTP more rapidly than expected, before changes in autoantibody levels and ADAMTS13 activity [29]. Interference with vWF–platelet binding has been proposed as a potential mechanism.

Other therapies used with variable success include cyclophosphamide, azathioprine, intravenous immunoglobulin, cyclosporine A, staphylococcal protein A immunoadsorption, and/or splenectomy [11]. Splenectomy has been reported to be of particular benefit in patients with a history of relapses, reducing the frequency from 2.3 ± 2 events per year to 0.1 ± 0.1 events per year [50]. Cyclosporine (2-3 mg/kg/day) has been compared to prednisone in two consecutively treated cohorts of patients with clinical TTP [51]. Remission was obtained in 10/12 (83%) patients in the prednisone group; 6/10 (60%) developed in exacerbation within 30 days. In the cyclosporine group, 8/8 (100%) achieved remission. None of these patients exacerbated. The number of TPE treatments needed to induce remission was also reduced, from 16 to 7 (p = 0.029). These investigators have also used cyclosporine alone successfully to treat early recurrences of idiopathic TTP [52]. A direct effect on endothelial cell activation or apoptosis is postulated in light of the fact that clinical improvement occurred despite persistence of ADAMTS13 deficiency and inhibitor activity.

Additional evidence that attacking the immunologic basis of idiopathic TTP may be efficacious involves rituximab, a chimeric monoclonal antibody directed at the CD20 antigen expressed on the surface of B lymphocytes. Efficacy has been reported in different autoimmune disorders [53, 54]; cell depletion occurs by several mechanisms including complement-mediated cell lysis and antibody-dependent cellular cytotoxicity [55].

Published experience with the use of rituximab in the treatment of TTP has grown to over 100 patients. Gutterman et al. first employed Rituximab in the treatment of three refractory patients with TTP [56] who had plasma exchange-dependent disease for 50 to 180 days, and were also refractory to multiple immunosuppressive treatments. Two of the three patients achieved normal platelet counts within 14 to 20 days after beginning infusions of rituximab, 375 mg/m2 once per week. The third patient improved but did not normalize her platelet count by day 20. Two patients achieved long-term stable remissions. Yomtovian et al. [57] reported a 26-year-old female who failed to achieve a sustained remission despite use of plasma exchange, corticosteroids, vincristine, staphylococcal Protein A immunoadsorption and splenectomy over a period of 268 days. The ADAMTS13 level was severely decreased and an inhibitor detected. After institution of weekly rituximab she gradually entered a stable remission (requiring 7 TPE over a one month period).

Fakhouri et al., studied six patients with acute refractory TTP (defined as unremitting TTP for at least three weeks) and five patients with severe relapsing TTP. All patients had ADAMTS13 levels less than 5% with an inhibitor. Plasma exchange was discontinued because of concerns about removal of the rituximab. Clinical responses were noted in all 11 patients, with normalization of the platelet counts between 5 and 14 days after the fourth infusion in the acute refractory patients. ADAMTS13 levels recovered to 29 to 75% of normal by six months post therapy. Inhibitor to ADAMTS13 was eradicated in between 7 and 24 weeks. B-cell depletion defined as less than 5 CD 19 positive lymphocytes per microliter, was documented in all except one patient (nadir 17 per microliter). B lymphocyte recovery was noted between 6 and 12 months and was frequently associated with the reappearance of inhibitor and decreased levels of ADAMTS13, giving rise to the concept of “minimal residual autoimmune disease” in some patients. This experience led to consideration of using rituximab even earlier. Scully et al. treated 14 patients with acute refractory TTP and 11 patients with acute relapsed TTP. In their protocol rituximab was used exclusively as a replacement for other pharmacologic interventions given if the patient’s platelets were not within normal limits by day seven. Plasma exchange was continued but delayed by ~24 h after each rituximab infusion. The investigators noted complete clinical and laboratory responses approximately 11 days following rituximab, with disappearance of inhibitors and improvement in ADAMTS13 activity in 23 out of 24 patients at three months follow-up. These investigators also noted no relapses at a median of 10 months of follow-up (1–33 +months), however, some patients received up to 8 doses of rituximab for persisting inhibitor levels.

Mild infusion-related reactions were reported in some cases, consisting of fever, chills, headache, nausea and hypotension. Initial concern about serious acute hypotensive reactions like those reported in patients with B-cell malignancies have not been materialized, likely because of the lower B-cell burden. A serious adverse event of an episode of acute cardiogenic shock was reported in a 20-year-old TTP patient who recovered from the episode without residual cardiomyopathy [58].

Of the patients reported in the literature so far, normalization of platelets and LDH has been noted in approximately 95%, however, time to remission has been variable, from one to four weeks after the first dose; a few patients have had partial responses with improvement in platelet count, but not to normal levels. Remissions attained after rituximab treatment appear to be durable. The duration of remission has ranged between 9 months and 4 years, with relapses reported in approximately 10%. ADAMTS13 levels generally increased, though not always to normal levels despite clinical improvement and prolonged remission [3, 27, 57].

Rituximab treatment alone has also induced remission in patients with early relapse, obviating the need for plasma exchange [59, 60]. Based on the reported salutary effects of rituximab in refractory cases and the known morbidity and mortality of acute TTP a phase II single arm trial designed to determine the efficacy of rituximab in the early management of TTP has been launched in Canada [61] and a phase III randomized clinical trial has been initiated in the US [62]. The Study of TTP and Rituximab (“STAR”) trial will evaluate the impact of early treatment with rituximab. In this trial, patients with a clinical diagnosis of idiopathic TTP are randomized to a standard treatment regimen (daily plasma exchange and steroid therapy) or standard treatment plus rituximab (four weekly infusions of 375 mg/m2). These trials should help answer the question of whether an adjuvant immunomodulatory strategy is beneficial in this disorder.

Copyright information

© The Japanese Society of Hematology 2010