ADAMTS13 recovery in acute thrombotic thrombocytopenic purpura after caplacizumab therapy

The advent of caplacizumab, a humanized nanobody that prevents the interaction between von Willebrand factor (VWF) and the platelet glycoprotein Ib-IX-V receptor, has changed the landscape for treatment of immune thrombotic thrombocytopenic purpura (iTTP).¹ iTTP is a rare, potentially life-threatening hematologic disorder caused by autoantibody-induced deficiency of ADAMTS13, which results in the failure of VWF multimer cleavage. The unusually large VWF multimers will then contribute to the accretion of platelet-rich microthrombi, especially in brain, heart, gastrointestinal tract, and kidneys.² The HERCULES and post-HERCULES trials showed that caplacizumab is associated with faster normalization of platelet counts, lower recurrence, and less incidence of thromboembolic events or iTTP-related death.³ Furthermore, these effects are maintained in the long term.⁴ The real-world series confirmed these findings in the settings of either (1) early treatment together with plasma exchange (PEX) and immunosuppression or (2) rescue therapy in scenarios of refractoriness.^5-12 Caplacizumab is also able to reduce the duration of PEX treatment.^5,7-12 The efficacy of this therapy is not achieved at the expense of patient safety, beyond an increased bleeding risk which does not usually result in major hemorrhages. However, a delayed normalization of ADAMTS13 activity subsequent to iTTP management with caplacizumab has recently been reported, which was supported by the observation that once PEX treatment was suspended, those patients with iTTP who were treated with caplacizumab required a longer period of time to recover ADAMTS13 activity than those who had not been administered the drug.¹³ To provide further insight into this potential concern, we have taken advantage of the Spanish Registry of Thrombotic Thrombocytopenic Purpura (REPTT)¹⁴ to assemble 2 cohorts of patients according to whether they had or had not been treated with caplacizumab during the management of their iTTP episodes. The time needed to achieve restored ADAMTS13 levels was determined in both groups either from iTTP diagnosis or from the end of PEX therapy. Patients treated with caplacizumab were also stratified according to when they started treatment–within or beyond 3 days after iTTP diagnosis.

Caplacizumab marketing authorization for the European Union was released at the end of August 2018. In this retrospective study, those patients identified from the REPTT as having had at least 1 iTTP episode between September 2018 and April 2023 were initially recruited. The Spanish hospitals contributing patients are listed in the supplemental Methods, available on the Blood website. When applicable, the decision to treat with caplacizumab depended on clinical criteria and timely access to the drug. Demographic, analytical, and clinical data were retrospectively collected from medical records. The inclusion criteria were compliance with the standards to diagnose an iTTP episode, regardless of whether this was or was not the first-ever episode and availability of data regarding time elapsed between either iTTP diagnosis or end of PEX treatment and recovery of ADAMTS13 activity (or availability of at least 1 ADAMTS13 test in the event of no recovery). ADAMTS13 recovery was considered to occur when activity was ≥20% at least 48 hours after the last PEX because this is associated with clinical remission.¹⁵ iTTP diagnosis criteria as well as definitions of clinical response, clinical remission, exacerbation, and relapse have been agreed by the International Working Group for Thrombotic Thrombocytopenic Purpura^15,16 and are available in supplemental Methods, together with definitions of refractoriness and time to normalization of platelet count.⁹ 

Patients were categorized according to whether they had or had not been treated with caplacizumab, and the collected variables were compared between the 2 cohorts. Patients treated with caplacizumab were further categorized according to whether they had or had not started caplacizumab within 3 days of iTTP diagnosis. Information about statistical procedures is available in supplemental Methods.

All patients gave their written informed consent for their data to be recorded in the REPTT to be used for research purposes. REPTT complies with the Spanish and European legislation on confidentiality of medical data, having been approved by the General Registry of Data Protection of Spain (Registration code: 2161102476) and the Committee of Ethics in Medical Research of the Autonomous Community of Galicia (Registration code: 2010/155).

We recruited a cohort of 108 patients with iTTP from the REPTT, among whom a total of 113 iTTP episodes were reported during the study period. Seventy-five of these were treated with caplacizumab at some point during their management. The flowchart diagram describes the recruitment process (Figure 1). One patient among those treated with caplacizumab and 2 patients who were not treated with caplacizumab died in the first 36 hours after iTTP diagnosis and were only considered for baseline statistics and clinical response, clinical remission, and death rate calculations. Two patients, 1 from each group, who died later and whose information regarding ADAMTS13 level evolution was available, were included in the whole statistical study.

As expected from previous series,^5-8,10-12 caplacizumab shortened the time to platelet count normalization and decreased the PEX requirement to achieve this goal, especially when started simultaneously with or shortly after PEX and corticosteroids. Furthermore, caplacizumab reduced the rate of refractoriness, exacerbations, and relapses (Table 1). The AEs experienced by patients treated with caplacizumab were nonserious, with 5% to 10% of patients presenting with mucocutaneous bleeding, metrorrhagia, thrombocytosis, or erythema.

In our cohort, the time to ADAMTS13 recovery (≥20%) from iTTP treatment start was similar regardless of the use of caplacizumab (median [interquartile range], 28.0 [18.0-41.5] vs 29.5 [15.2-45.0] days, P = .579, in caplacizumab-treated and nontreated episodes, respectively) (Figure 2A-B). Differences in this variable were not observed either when a comparison was performed within the caplacizumab cohort after stratifying patients according to whether they had or had not initiated this treatment ≤3 days after PEX and corticosteroid start (28.0 [17.2-47.5] vs 27.0 [19.0-37.5] days, respectively; P = .672), (Figure 2C-D), or when patients who started caplacizumab ≤3 days after PEX and corticosteroids were compared with those who were not administered caplacizumab (P = .835) (Figure 2E-F). When the time required for ADAMTS13 recovery was calculated considering the end of PEX therapy as baseline, no significant differences were seen between caplacizumab-treated and nontreated episodes (14.5 [7.7-27.2] vs 13.0 [8.0-29.0] days, respectively; P = .653) (Figure 3A-B). However, when the former were categorized according to early or late start (≤3 days or >3 days after PEX/corticosteroids, respectively), the time to ADAMTS13 recovery was significantly higher in the first group (20.0 [12.0-43.0] vs 11.0 [3.5-20.0] days, P = .003) (Figure 3C-D). Accordingly, this period of time was also longer when the early caplacizumab start group was compared with the noncaplacizumab group (P = .033) (Figure 3E-F). There was a slight, albeit significant difference in treatment duration between patients who started caplacizumab ≤3 days or >3 days after PEX/corticosteroids (33.0 [22.0-37.5] vs 36.0 [30.0-39.0] days, respectively; P = .045).

Overall, these results are in line with those recently described by Prasannan et al,¹³ and have to be examined in the context of the fact that caplacizumab, especially when started early after iTTP diagnosis, significantly reduces the requirement for PEX (5.5 [4.0-9.0] days) with respect to that observed for episodes either not being treated with caplacizumab (11.0 [6.0-26.0] days, P < .001) or starting this later (15.0 [11.0-21.5] days, P < .001) (Figure 3G-I). The earlier end of PEX therapy in patients who were treated with caplacizumab early contributes to explaining the recently reported findings,¹³ as well as ours. The aforementioned lack of differences between patients treated with caplacizumab, regardless of whether they start early or late after diagnosis, and the patients who were not treated with caplacizumab, when considering the time from iTTP treatment start to ADAMTS13 recovery, invites us to suggest that the potential delay in the restoration of ADAMTS13 activity should not be a concern when caplacizumab use is being considered. Moreover, the fact that the positive correlation observed between duration of PEX therapy and restoration of ADAMTS13 activity in those patients who were not administered caplacizumab was lost in those who received caplacizumab in addition to PEX and corticosteroids (Figure 4A-B) suggests that caplacizumab may reduce PEX dependency in the management of iTTP episodes. This notion is further substantiated by the strong positive correlation observed between the delay in initiating caplacizumab and the PEX requirement (Spearman ρ = .567, P < .001) (Figure 4C).

The other outcomes were consistent with those described in the foregoing paragraphs. Baseline ADAMTS13 levels did not differ between groups (Figure 4D-E). More importantly, the restored levels of ADAMTS13 were not significantly different between caplacizumab-treated and nontreated patients (Figure 4F) or between the early and late caplacizumab groups (Figure 4G). Indeed, the mechanism of action of caplacizumab should not influence the eradication of anti-ADAMTS13 autoantibodies and, accordingly, should not challenge the restoration of ADAMTS13 activity. On the other hand, no association was found between caplacizumab use and duration of corticosteroid therapy (not shown). The similar use of rituximab in both cohorts (Table 1) and the weak correlation observed between the delay in starting rituximab and the time from PEX start to ADAMTS13 recovery when considering all patients treated with rituximab (ρ = 0.144, P = .092, n = 89) do not suggest that this treatment influenced differently the evolution of 1 group or another. Finally, the significant correlation between days on treatment with caplacizumab and time from iTTP treatment start to ADAMTS13 recovery (ρ = 0.233, P = .047) was not surprising because ADAMTS13 level was important to decide when caplacizumab treatment should be suspended.

One limitation of the study was that inhibitor titers were not available; thus precluding the comparison of baseline levels among the different treatment groups. Nevertheless, the inhibitor titer was not the main criterion to decide treatment with capla and there was no difference in ADAMTS13 recovery regardless of the moment of initiating capla. Furthermore, the number of early caplacizumab-treated, late caplacizumab-treated, and non-caplacizumab-treated patients was 37, 38, and 38, respectively, which makes it unlikely that baseline differences in inhibitor titers among groups influenced the time required to ADAMTS13 recovery.

In sum, we have used the REPTT data in order to ascertain whether or not the use of caplacizumab involves a delay in the recovery of ADAMTS13 activity. The delay found when the time at which PEX ended was considered as baseline can be explained by the earlier completion of PEX therapy in those patients who were administered caplacizumab, especially when this was started concomitantly with PEX and corticosteroids. The lack of differences between groups in the periods of time from iTTP diagnosis to ADAMTS13 recovery should be helpful to dispel concerns regarding this topic. Real-world evidence is important to manage rare diseases appropriately. Taken together, the findings described here are in accordance with clinical trials and real-world studies in that caplacizumab should not be seen as a rescue medication for PEX-refractory iTTP cases. Instead, it should be made widely available to be administered upfront to optimize its use and reduce the number of PEX treatments.

The authors thank the Spanish Society of Hematology and Hemotherapy and the Spanish Society of Blood Transfusion for the support required to build and maintain the Spanish Thrombotic Thrombocytopenic Purpura Registry.

Contribution: M.-E.M.-C. and C.P.-I. designed the study, contributed patients to the database, arranged the database, performed the statistical study, analyzed the data, and wrote the paper and F.G.-C., J.M.-N., J.G.-A., J.d.l.R.-C., I.G.-S., M.-L.P.-C., D.V.-F., M.J., J.C., M.L., J.-M.G.-G., S.A.-V., M.V.-P., L.G.-D., L.-F.Á.-I., A.O.-H., S.Z.-M., I.T.-R., S.O.-S., R.G.-V., G.M.-J., L.D.-A., M.A.-R., L.H.-M., E.F.-B., and J.d.R.-G. contributed patients to the database and critically reviewed the article; and all authors had full access to primary clinical data.

Conflict-of-interest disclosure: M.-E.M.-C. participated in advisory boards and speaker sessions for Takeda, Sanofi, Novartis, Novo Nordisk, Amgen, and Sobi. J.d.l.R.-C. participated in advisory boards and speaker sessions for Takeda and Sanofi and got research support from Takeda. I.G.-S. participation in advisory boards for Takeda and speaker sessions for Sanofi and Terumo BCT. D.V.-F. participated in advisory boards and speaker sessions for Agios, Amgen, Bristol Myers Squibb/Celgene, Gilead, Grifols, Janssen, Jazz, Novartis, Sanofi, Sobi, Pfizer, and Takeda. M.J. has received honoraria from AstraZeneca, GlaxoSmithKline, and Sanofi. J.C. received research funding from Cerus, Kawasumi Laboratories, and Sanofi and speaker or advisory fees from Cerus, Fresenius Kabi, Grifols, MacoPharma, Pharm-Olam, Sanofi, and Terumo BCT. M.L., on behalf of his institution, Clinic Research Foundation, has received research support from Terumo BCT, Macopharma, and Sanofi-Genzyme. J.-M.G.-G. is a part of the advisory board for Sanofi. C.P.-I. participated in advisory boards and speaker sessions for Takeda, Sanofi, Amgen, and Novartis. The remaining authors declare no competing financial interests.

A complete list of the members of the Spanish Apheresis Group and the Spanish Thrombotic Thrombocytopenic Purpura Registry appears in the supplemental Appendix.

Correspondence: María-Eva Mingot-Castellano, Hematology Department, Hospital Universitario Virgen del Rocío, Instituto de Biomedicina de Sevilla (IBIS/CSIC), 41013 Seville, Spain; email: mariae.mingot.sspa@juntadeandalucia.es; and Cristina Pascual-Izquierdo, Hematology Department, Hospital General Universitario Gregorio Marañón, Instituto de Investigación Gregorio Marañon, 28007 Madrid, Spain; email: crisizquierdo3@yahoo.es.