https://orcid.org/0000-0001-7979-7603
In this issue of Blood Advances, Postmus et al1 developed N-glycan variants of the ADAMTS13 CUB domain, termed NGLY-CUB, to block common epitopes targeted by autoantibodies in patients with immune-mediated thrombotic thrombocytopenic purpura (TTP). These data provide new evidence that glycan-shielded recombinant ADAMTS13 (rADAMTS13) may be a viable therapeutic option for patients with TTP.
TTP is a rare thrombotic microangiopathy affecting ∼2 to 6 million individuals annually. It is characterized by hemolytic anemia and thrombocytopenia caused by platelet-rich thrombi that form in the kidney, brain, heart, or pancreas and is fatal if not quickly treated. These thrombi are distinguishable from other thrombotic microangiopathies such as hemolytic uremic syndrome, in which clots are composed of fibrin. As such, anticoagulants such as warfarin are not beneficial in patients with TTP, and the benefit of antiplatelet therapy is not clear.2,3
TTP is caused by genetic or acquired deficiency of the protease ADAMTS13. The only known role of ADAMTS13 is to proteolytically regulate the multimeric length of von Willebrand factor (VWF). Therefore, deficiency in ADAMTS13 leads to the accumulation of large pathogenic VWF multimers that can spontaneously aggregate platelets in the microcirculation. Hereditary TTP, also called congenital TTP (cTTP), accounts for ∼10% of TTP cases and is caused by nonsense or missense mutations in the ADAMTS13 gene that abolish protein production. Acquired TTP, also called immune TTP (iTTP), is caused by autoantibodies against ADAMTS13 and accounts for ∼90% of TTP cases. Treatment of cTTP focuses on replenishing ADAMTS13 with plasma infusion. The treatment of iTTP is more complex and requires immune suppression with corticosteroids along with plasma exchange to simultaneously remove anti-ADAMTS13 antibodies and replenish ADAMTS13 levels. In some refractory cases, patients with iTTP may require B-cell ablation with rituximab to achieve remission. Prophylactic management of patients with TTP is complicated because the risk factors for relapse are poorly understood and the negative impacts of long-term immune suppression or splenectomy limit clinical benefit.
rADAMTS13 was recently granted an approval by the US Food and Drug Administration for the treatment of cTTP. In a phase 3 open-label crossover trial, rADAMTS13 provided 100% protection against relapse compared with 1 relapse event in the standard therapy group and was associated with better platelet counts and lower adverse events.4 These findings support the use of rADAMTS13 for the treatment of cTTP. However, the use of rADAMTS13 for the treatment of patients with iTTP is more tenuous because of the presence of autoinhibitory antibodies. The SOAR-HI trial (ClinicalTrials.gov identifier: NCT03922308) will address questions about the pharmacokinetics, safety, and efficacy of rADAMTS13 in patients with iTTP; however, as of today, these data have not been published. A recent case report described a 28-year-old patient with iTTP who was refractory to therapy after having received daily plasma exchange coupled with immunosuppressants (rituximab and bortezomib), caplacizumab, or intravenous immunoglobulin for >2 weeks with no clinical remission.5 Upon administration of twice daily rADAMTS13 on top of daily plasma exchange, the patient showed immediate improvements in platelet count and reduced evidence of hemolysis, achieving clinical remission and discharge on day 39. This case illustrates the potential benefit of rADAMTS13 for the treatment of iTTP in certain settings. However, the heroic doses of rADAMTS13 used in this patient make the general accessibility of rADAMTS13 to patients with iTTP challenging. Although new bypass agents can target pathogenic VWF multimers in patients with iTTP,6,7 they do not have the favorable pharmacologic properties of ADAMTS13-based therapies.8 An ideal solution is to develop ADAMTS13 variants that are protected against autoantibodies.
Previous work has shown that mutating an immunogenic hotspot in the spacer domain can protect ADAMTS13 from most inactivating iTTP antibodies; however, it also reduces VWF cleaving activity.9 To resolve this dilemma, Ercig et al developed ADAMTS13 mutants that lead to N-glycosylation of residues surrounding the immunogenic hotspot in the spacer domain, which protected against antibody binding while retaining VWF cleaving activity.10 This variant is important because nearly all patients with iTTP have antibodies that target this region on ADAMTS13. Anti-CUB domain antibodies are present in ∼30% of patients with iTTP, making it the second most prominent immunogenic hotspot on ADAMTS13.11
To develop ADAMTS13 variants that are protected against anti-CUB domain antibodies, Postmus et al used structural models of CUB domains to identify 9 residues for modification. As expected, all variants produced novel glycosylation at the expected residues and retained full activity toward FRET-VWF73, a biochemical substrate that does not bind to the CUB domains. Interestingly, all variants adopted a structurally open conformation, suggesting that glycan modifications disrupted the intramolecular interactions between the CUB domains and the spacer domain. Despite disrupting intramolecular interactions, the modifications did not interfere with the ability of ADAMTS13 to cleave multimeric VWF in a microfluidics assay. Given the importance of the CUB domains in VWF cleavage under flow conditions, these assays validate that the antithrombotic activity of NGLY-CUB variants is comparable to that of wild-type ADAMTS13.
They next tested 7 monoclonal antibodies derived from patients with iTTP and showed reduced binding to many of the NGLY-CUB variants. The combination of NGLY-CUB variants (1251N + 1368N and 1255N + 1368N) showed near-complete protection against 6 of 7 monoclonal antibodies. When tested against a panel of 9 patients with iTTP plasma samples known to contain antibodies directed against the c-terminal domains of ADAMTS13, the combination of NGLY-CUB variants displayed improved protection from antibody binding. These encouraging data provide further evidence that novel glycan variants of ADAMTS13 can be engineered to confer protection against autoinhibitory antibodies without interfering with their antithrombotic activity.
Much of the evidence presented in this article shows that NGLY-CUB variants can confer protection against selected antibodies; however, this effect may be less pronounced in an unselected iTTP cohort. Further development of N-glycan shields against the less common epitopes on ADAMTS13 may be necessary before a more fully protected ADAMTS13 variant can be developed. Nonetheless, the early success of the standard rADAMTS13 product in selected iTTP cases provides a roadmap in which even partially protected ADAMTS13 variants may be beneficial. Reducing epitope availability on rADAMTS13 could provide substantial cost savings in the treatment of patients with iTTP, making this work by Postmus et al an important advancement in ADAMTS13-based therapies.
Conflict-of-interest disclosure: C.A.K. declares no competing financial interests.
