ADAMTS13 Biomarkers in Management of Immune Thrombotic Thrombocytopenic Purpura

Abstract

Context.—

Immune thrombotic thrombocytopenic purpura (iTTP) is a rare but potentially fatal blood disorder resulting from acquired deficiency of plasma ADAMTS13, a metalloprotease that cleaves endothelium-derived ultralarge von Willebrand factor. Standard of care for iTTP including therapeutic plasma exchange, caplacizumab, and immunosuppressives, known as triple therapy, has led to a significant reduction in the disease-related mortality rate. The first International Society of Thrombosis and Haemostasis TTP guidelines stress the importance of having plasma ADAMTS13 activity testing in the algorithm for diagnosis and management of iTTP. However, the predictive role of assessing plasma ADAMTS13 activity and inhibitors or other ADAMTS13-related parameters in patients with acute iTTP and during remission has not been systematically evaluated.

Objective.—

To review and assess the predictive values of testing plasma ADAMTS13 activity, antigen, and inhibitors or anti-ADAMTS13 immunoglobulin G at various stages of disease in outcomes of iTTP.

Data Sources.—

Peer-reviewed publications and personal experience.

Conclusions.—

We conclude that assessing ADAMTS13 biomarkers is not only essential for establishing the initial diagnosis, but also crucial for risk stratification and the early detection of disease recurrence. This may guide therapeutic interventions during acute episodes and for long-term follow-up of iTTP patients.

Thrombotic thrombocytopenic purpura (TTP), a rare blood disorder, occurs at a rate of 4 to 13 cases per million residents.^1–4 TTP is clinically characterized by severe thrombocytopenia (platelet count usually less than 30 000/μL) and microangiopathic hemolytic anemia with various degrees of end organ damage.^1–5 The mechanism of TTP is well understood now. It is associated with severe deficiency of plasma ADAMTS13 (a disintegrin and metalloprotease with thrombospondin type 1 repeats 13) activity.^6,7 ADAMTS13 cleaves newly released endothelial ultralarge von Willebrand factor (VWF)⁸ and reduces the size of circulating VWF under fluidic shear.⁹ An inability to cleave the ultralarge VWF results in an accumulation of highly adhesive VWF strings on activated or injured endothelial surface, which leads to excessive microvascular thrombosis, end organ damage, and death.^10–12

The cause of ADAMTS13 deficiency may be hereditary, resulting from mutations in ADAMTS13,^13,14 or acquired because of the production of autoantibodies against ADAMTS13.^5,15,16 The immunoglobulin (Ig) G–type autoantibodies bind and inhibit plasma ADAMTS13 activity and/or accelerate the clearance of ADAMTS13 protease as immune complexes from circulation.^17,18 As such, 2 forms of TTP are proposed: (1) hereditary or congenital TTP and (2) acquired or immune TTP (iTTP). Regardless of the mechanism underlying ADAMTS13 deficiency, TTP is a medical emergency and can be fatal.^19,20 Thus, early recognition and prompt treatment of TTP greatly improve the survival rate.²⁰ The most common causes of death in patients with acute TTP are myocardial infarction, arrhythmia, and cardiac arrest.¹⁰

Despite marked improvement of TTP survival, the disease exacerbation and relapse remain a challenge, particularly when caplacizumab is not available for treatment in certain parts of the world.^21,22 With a combination of therapeutic plasma exchange (TPE), steroids, and rituximab, most patients may achieve clinical response and/or remission. However, about 50% of iTTP patients experience exacerbation (disease recurs within 30 days following cessation of TPE) or relapse (disease recurs after 30 days following the cessation of TPE).²² These patients require reinitiation of TPE, steroids, and rituximab treatment. It is imperative to identify patients with a high risk of disease recurrence who may benefit from other, more intense therapies and receive a more prolonged or preemptive therapy (such as rituximab or other immunosuppressives).

This review will focus on the roles of plasma ADAMTS13 activity, antigen, and inhibitors or anti-ADAMTS13 IgG in risk stratification and prediction of therapeutic outcomes.

STRUCTURE OF ADAMTS13 PROTEASE

ADAMTS13 is primarily synthesized in the interstitial hepatic stellate cells.²³ It may also be produced in small quantities by other cells, including endothelial cells, megakaryocytes, renal podocytes, and tubular epithelial cells.^24–26 The known function of ADAMTS13 to date is the cleavage of VWF multimers or stringlike structures on the endothelial surface, in circulating blood, and at sites of vascular injury.^27–29 The specific cleavage site (Tyr¹⁶⁰⁵-Met¹⁶⁰⁶) is located at the central A2 domain of VWF.^30–32 This proteolytic cleavage is essential for maintaining normal hemostasis and regulating inflammatory processes.^33,34

Upon secretion, the propeptide of ADAMTS13 protein is removed by furin in the Golgi apparatus.^35,36 Thus, the mature form of circulating ADAMTS13 consists of a metalloprotease domain, a disintegrin domain, the first thrombospondin type 1 repeat (TSP1–1), cysteine-rich domain, and spacer domain, followed by 7 additional TSP1 repeats and 2 CUB domains (Figure).^37,38 The N-terminal MDTCS domains appear to be sufficient for recognition and proteolytic cleavage of VWF under various conditions.^39,40 The distal TSP1 repeats and CUB domains appear to play a regulatory role in maintaining ADAMTS13 enzymatic latency,^41,42 for ligand binding or for membrane association.^43,44 ADAMTS13 activity can be significantly increased at acidic pH 6.0,⁴² by binding of VWF ligand,⁴⁵ and by monoclonal antibodies against the distal C terminus of ADAMTS13,^45–47 likely all through the alteration of ADAMTS13 conformations (Figure).

Domain structure and activation of ADAMTS13. ADAMTS13 consists of a metalloprotease domain (M), a disintegrin domain (D), the first thrombospondin type 1 repeat (TSP1–1), a cysteine-rich domain (C), and a spacer domain (S). Additionally, the C terminus contains 2–8 TSP1 repeats and 2 CUB domains. Under physiological conditions, ADAMTS13 may exhibit a folded or latent conformation, which can be activated by lowering pH to 6.0, binding of VWF-D4 domain, antibodies against the distal CUB domains (anti-CUB), etc.

THE ROLE OF ASSESSING PLASMA ADAMTS13 ACTIVITY IN iTTP

Diagnosis of TTP requires a high index of suspicion. A patient who presents with thrombocytopenia and microangiopathic hemolytic anemia with or without organ damage should be worked up to rule out TTP.^48,49 The International Society of Thrombosis and Haemostasis published guidelines that recommend the use of clinical criteria or a risk assessment score (either French score or PLASMIC score) for assessing the pretest probability for TTP.⁴⁹ As hereditary TTP is extremely rare, most adult patients with clinically suspected TTP should be considered as iTTP until disproved otherwise.^15,50 Thus, if preclinical test probability is high (such as French score >2.0 or PLASMIC score >5), one should consider collecting plasma for an ADAMTS13 test and providing immediate treatment with TPE, corticosteroids, and caplacizumab (if available). Rituximab should also be prescribed to the patient as early as possible.

Commonly used clinical assays for ADAMTS13 activity include the fluorescence resonance energy transfer VWF73 assay,⁵¹ the enzyme-linked immunosorbent assay (ELISA)–based chromogenic assay,^52,53 the Hemosil AcuStar automatic assay,⁵⁴ etc. Each of these assays has its advantages and potential pitfalls, but they all seem to be adequate for establishing the initial diagnosis of TTP. These assays appear to have an analytical sensitivity of 0.1 to 10 U/dL (or 0.1% to 10%).^51,54,55 That is why 10 U/dL (or 10% of normal) was used as a cutoff.

Although undetectable or extremely low plasma ADAMTS13 activity (<10 U/dL) is seen in almost all iTTP patients at the time of diagnosis,^22,56,57 plasma ADAMTS13 activity in some patients achieving clinical response or remission may still be quite low.^5,56 Such an observation is crucial for further management of iTTP. Depending on the amount of inhibitors in plasma, 1 or 2 TPEs or several units of FFP infusion do not normalize plasma ADAMTS13 activity. Zheng et al⁵ and other groups^56,58 have reported that many patients did not recover their plasma ADAMTS13 activity days and weeks following TPE, steroids, and/or rituximab. We recently reported that approximately 15% and 22% of patients had plasma ADAMTS13 activity less than 10 U/dL and 20 U/dL, respectively, despite achieving normalization of their platelet counts.⁵⁶

Interestingly, plasma ADAMTS13 activity of less than 10 U/dL, 3 to 7 days following treatment or when an initial clinical response is achieved, is strongly associated with an increased risk of exacerbation.⁵⁶ These results are consistent with those reported by others^59–62 in which ADAMTS13 activity less than 10 U/dL at clinical response/remission predicts TTP exacerbation or relapse (Table 1).

Table 1.

The Role of ADAMTS13 Activity in Predicting Disease Severity and Outcome in Immune Thrombotic Thrombocytopenic Purpura

Source, y	Sample Size	Sample Time	Outcome	Association
Bettoni et al,59 2012	115	In remission	Recurrence	ADAMTS13 activity <10% is associated with relapse
Kremer Hovinga et al,74 2010	183	On admission	Relapse	ADAMTS13 activity <10% is associated with relapse
Ferrari et al,61 2007	32	In remission	Relapse	ADAMTS13 activity <10% is associated with relapse
Peyvandi et al,60 2008	109	In remission	Recurrence	ADAMTS13 activity <10% is associated with relapse
Jin et al,62 2008	24	In remission	Recurrence	Lower ADAMTS13 activity is associated with higher risk of recurrence within 3 mo
Sui et al,56 2019	48	3–7 d after TPE	Exacerbation	ADAMTS13 activity <10% associated with relapse
	46	Clinical response	Exacerbation	ADAMTS13 activity <10% associated with relapse

Abbreviation: TPE, therapeutic plasma exchange.

THE ROLE OF ASSESSING PLASMA ADAMTS13 ANTIGEN IN iTTP

The levels of ADAMTS13 antigen can be measured by an ELISA-based assay.^56,63,64 Plasma levels of ADAMTS13 antigen vary significantly during the disease process.^56,65 Thus, plasma ADAMTS13 antigen has not been used for the initial diagnosis of TTP. Normal ADAMTS13 antigen but no ADAMTS13 activity strongly suggests the presence of inhibitory anti-ADAMTS13 antibodies that do not cause an accelerated clearance.⁶³

The role of assessing ADAMTS13 antigen in the management of iTTP is not fully established. Alwan et al⁶⁶ reported that iTTP patients who died had lower levels of ADAMTS13 antigen than those who survived; additionally, patients with ADAMTS13 antigen levels in the lowest quartile appeared to have a higher mortality rate than those in the highest quartile (18.4% versus 3.8%). Yang et al⁶⁵ also reported the role of ADAMTS13 antigen in predicting mortality. They demonstrated that severely low ADAMTS13 antigen during an acute episode was statistically associated with increased mortality. In addition, low plasma ADAMTS13 antigen (<10%) during remission is associated with a high risk of disease recurrence.⁵⁹ In the article published by Sui et al,⁵⁶ low ADAMTS13 antigen at clinical response/remission was associated with an increased risk of iTTP exacerbation, although other discrepant results have been reported.^60,65 This may be related to patient population and timing of sampling (Table 2).

Table 2.

The Role of ADAMTS13 Antigen in Predicting Disease Severity and Outcome in Immune Thrombotic Thrombocytopenic Purpura

Source, y	No. of Patients	Sample Time	Outcome	Association
Bettoni et al,59 2012	62	On admission	Recurrence	No association
Peyvandi et al,60 2008	77	In remission	Recurrence	No association
Yang et al,65 2011	40	On admission	Acute disease mortality	Low antigen levels are associated with mortality
Alwan et al,66 2017	312	On admission	Acute disease mortality	Low antigen levels are associated with mortality
Sui et al,56 2019	48	3–7 d after TPE	Exacerbation	No association
	46	At clinical response	Exacerbation	Low antigen levels are associated with exacerbation

Abbreviation: TPE, therapeutic plasma exchange.

THE ROLE OF ASSESSING PLASMA AUTOANTIBODIES AGAINST ADAMTS13 IN iTTP

The plasma levels of autoantibodies against ADAMTS13 or inhibitors can be quantified by a mixing study with a functional assay.^5,15,67,68 The functional assay detects the titer of inhibitory autoantibodies against ADAMTS13 and perhaps some other non–antibody-associated inhibitors, such as bilirubin,⁶⁹ hemoglobin,⁷⁰ and neutrophil peptides.⁷¹ However, an ELISA-based assay detects polyclonal anti-ADAMTS13 igGs,^56,72,73 which appears to be more sensitive than a functional assay. Thus, a negative functional inhibitor test does not rule out the diagnosis of iTTP. Repeating plasma ADAMTS13 activity and inhibitor testing following treatment should be considered.

In addition to its diagnostic role, the presence of ADAMTS13 inhibitors in remission indicated 3 to 4.3 times higher risk for iTTP recurrence.^59,60 The role of detecting inhibitors during an acute episode for predicting mortality and disease severity remains controversial. Kremer Hovinga et al⁷⁴ showed that more patients in the deceased group had ADAMTS13 inhibitors at 2 Bethesda units/mL or higher compared with those who survived. However, this was not confirmed by other groups.^61,75 Sui et al⁵⁶ found that high levels of anti-ADAMTS13 IgG (>75th percentile) 3 to 7 days following therapy, but not on admission, were associated with increased risk of iTTP exacerbation (Table 3).

Table 3.

The Role of ADAMTS13 Inhibitors in Predicting Disease Severity and Outcome in Immune Thrombotic Thrombocytopenic Purpura

Source, y	No. of Patients	Sample Time	Outcome	Association
Bettoni et al,59 2012	62	On admission	Recurrence	No association
	68	In remission	Recurrence	Presence of inhibitors is associated with recurrence
Coppo et al,75 2006	33	On admission	Disease severity	Detectable inhibitors are associated with severe acute episode
Kremer Hovinga et al,74 2010	60	On admission	Acute disease mortality	Inhibitor levels ≥2 Bethesda units are associated with mortality
Ferrari et al,61 2007	35	On admission	Acute disease mortality	No association
Peyvandi et al,60 2008	58	In remission	Recurrence	Presence of inhibitors is associated with recurrence

The role of ADAMTS13 antibodies in predicting mortality has been reported.^60,66,76 The United Kingdom TTP registry has shown that patients who died had a higher level of anti-ADAMTS13 IgG than those who survived (median, 66% versus 49%).⁶⁶ Ferrari et al⁷⁶ demonstrated that IgG4 is the most common subtype (90%), followed by IgG1 (52%). High levels of IgG1 were associated with iTTP mortality. However, other studies⁵⁹ did not find such an association between IgG subtypes and disease recurrence, either in the acute phase or during remission. IgG subtypes may have different capacities to bind Fcγ receptors or activate complement.⁷⁷ Additionally, a higher IgA titer may be associated with an increased mortality rate during the acute episode,^60,76 and the presence of anti-ADAMTS13 IgA was strongly associated with platelet count at presentation⁵⁹ (Table 4). IgA may also activate complement via the mannan-binding lectin pathway, thus enhancing inflammation and thrombosis.⁷⁸ More studies should be conducted on the prognostic roles of various IgG types or IgA against ADAMTS13 in iTTP.

Table 4.

The Role of Anti-ADAMTS13 IgG or IgA in Predicting Disease Severity and Outcome in Immune Thrombotic Thrombocytopenic Purpura

Source, y	No. of Patients	Marker	Sample Time	Outcome	Association
Bettoni et al,59 2012	52	Anti-ADAMTS13 IgG or IgA	On admission	Disease severity	High levels of IgG and IgA are associated with severe acute episode
	47	Anti-ADAMTS13 IgG	On admission	Disease severity	High levels of IgG1 and IgG3 are associated with severe acute episode
Ferrari et al,76 2009	58	Anti-ADAMTS13 IgG	On admission	Acute disease mortality	High levels of IgG1 and IgG3 are associated with severe acute episode
Ferrari et al,61 2007	35	Anti-ADAMTS13 IgA	On admission	Acute disease mortality	High levels of IgA isotype are associated with mortality
Peyvandi et al,60 2008	97	Anti-ADAMTS13 IgG	In remission	Recurrence	Presence of autoantibodies is associated with mortality
Jin et al,62 2008	24	Anti-ADAMTS13 IgG	In remission	Recurrence	No association
Alwan et al,66 2017	312	Anti-ADAMTS13 IgG	On admission	Acute disease mortality	High levels of anti-ADAMTS13 IgG are associated with mortality
Sui et al,56 2019	97	Anti-ADAMTS13 IgG	On admission	Exacerbation	No association
	48	Anti-ADAMTS13 IgG	3–7 d after TPE	Exacerbation	High levels of anti-ADAMTS13 IgG are associated with exacerbation
	46	Anti-ADAMTS13 IgG	At clinical response	Exacerbation	High levels associated with exacerbation

Abbreviations: IgG, immunoglobulin G; IgA, immunoglobulin A; TPE, therapeutic plasma exchange.

THE ROLE OF ASSESSING PLASMA ADAMTS13-ANTIBODY COMPLEXES IN iTTP

Anti-ADAMTS13 IgGs may circulate in plasma in a free form and complexed with ADAMTS13.^18,79 The immune complexes may be cleared from circulation through the endoplasmic reticulum system. The remaining immune complexes in plasma can be detected by an ELISA-based assay.⁷⁹ ADAMTS13/anti-ADAMTS13 IgG complexes were found in patients with acute iTTP and during remission. Mancini et al⁸⁰ demonstrated that although the ADAMTS13 immune complexes were not associated with other laboratory markers for disease severity, nor were the immune complexes associated with the number of plasma exchanges required to achieve clinical remission, the presence of ADAMTS13 immune complexes was associated with an increased risk of disease recurrence within 2 years after the initial episode.

CONCLUSIONS

Plasma ADATMS13 activity and inhibitors or anti-ADAMTS13 IgGs are routinely tested in a patient with suspected TTP. In the proper clinical context, plasma ADAMTS13 activity less than 10 U/dL (or 10% of normal) supports the diagnosis of TTP; additionally, assessing plasma ADAMTS13 activity and inhibitors or anti-ADAMTS13 IgG at different time points during the disease course (eg, on admission, during treatment, and at clinical response or in remission) may play a role in stratifying patients with a high risk of death during and acute episode and/or subsequent disease recurrence (eg, exacerbation and/or relapse). This allows more intense therapies, including caplacizumab and rituximab, to be used upfront in these high-risk patients should the cost and drug availability be a concern. Moreover, ADAMTS13 testing, particularly for activity, during clinical response (or remission) may help determine the optimal time for discontinuing these complex or expensive therapies, such as TPE, caplacizumab, etc.