Tailoring caplacizumab administration using ADAMTS13 activity for immune-mediated thrombotic thrombocytopenic purpura

Immune-mediated thrombotic thrombocytopenic purpura (iTTP) is a rare disease stemming from autoantibodies against ADAMTS13 (a disintegrin and metalloprotease with thrombospondin type 1 motif, member 13), a metalloproteinase that cleaves von Willebrand factor (VWF).¹ The subsequent severe deficiency of ADAMTS13 activity results in the persistence of ultralarge VWF multimers, the formation of platelet-rich microthrombi, and end-organ ischemia.² 

Neurologic manifestations that range from moderate to severe are among the most concerning complications associated with acute iTTP, with more recent evidence underscoring the incidence of long-term chronic sequelae, including depression and neuropsychiatric changes.³ Consequently, interventions aimed at arresting the formation of microthrombi and resultant end-organ damage, particularly to the central nervous system, are essential to limiting acute and long-term adverse outcomes associated with iTTP.

Caplacizumab, a nanobody targeting the A1 domain of VWF, inhibits the interaction between VWF and platelet glycoprotein Ib-IX-V receptors, ultimately restricting the ongoing formation of microthrombi. Therefore, caplacizumab seems to have a rapid disease-modifying effect. In clinical trials, patients with iTTP treated with caplacizumab demonstrated accelerated platelet count recovery, underwent fewer plasma exchange (PLEX) treatments, and had a reduced length of stay.^4-6 

Per manufacturer recommendations, an IV dose of caplacizumab should be administered immediately before PLEX, followed by daily subcutaneous injections up to 30 days after discontinuation of PLEX. However, caplacizumab administration is not without risk, with postmarketing studies reporting an overall bleeding incidence of 18% to 33% of patients, including fatal intracranial hemorrhage.^7-10 Furthermore, a recent before and after study demonstrated that 28% of patients receiving long-term caplacizumab dosing (between 30 and 58 days after PLEX) showed a protracted recovery of ADAMTS13 activity >30% relative to a historical non-caplacizumab cohort.¹¹ Moreover, the financial implications of a fixed-dosing interval should also be considered as recommended by the manufacturer, with a single dose and course of therapy approximating $9510 and $332 850 (estimated 35 doses), respectively, in the United States.

The potential utility of serially monitoring ADAMTS13 activity and VWF activity to individualize therapy and guide caplacizumab treatment was recently described.⁷ Using this approach, caplacizumab was discontinued based on ADAMTS13 activities >10% in 15 patients, with none experiencing an exacerbation or relapse.⁷ Furthermore, 10 patients received nondaily caplacizumab (alternate day up to once weekly), with 6 of the 10 having a VWF activity <30% for >48 hours.⁷ 

Our institutional protocol provides individualized caplacizumab treatment based on frequent ADAMTS13 monitoring and ADAMTS13 recovery as opposed to extended dosing after PLEX. Therefore, this study aimed to compare laboratory and clinical outcomes in consecutive historical cases of iTTP without caplacizumab with those managed with individualized caplacizumab therapy.

This retrospective cohort study was conducted at the University of Texas Southwestern Medical Center in Dallas, TX, at 2 tertiary care hospitals (Clements University Hospital and Parkland Memorial Hospital).

Our treatment protocol for the clinical management of suspected iTTP consists of measurement of ADAMTS13 within 24 hours of suspected iTTP, initiation of caplacizumab, PLEX, and steroids on day 1 of suspected iTTP, and rituximab within 3 to 5 days after the confirmation of an ADAMTS13 autoantibody. The duration of caplacizumab therapy is guided by ADAMTS13 measurements (twice per week on Tuesday and Friday), and it is generally discontinued once an activity >20% is obtained on 2 consecutive occasions. PLEX is generally discontinued when ADAMTS13 is >50%.

Our study population comprised adult patients (aged ≥18 years) with a confirmed diagnosis of iTTP (ADAMTS13 activity <10% with the presence of an autoantibody). We compared consecutive patients treated with PLEX and immunosuppressive therapy before the introduction of caplacizumab in February 2020 (February 2016 to April 2020) with those who received caplacizumab, PLEX, and immunosuppressive therapy over an equivalent time frame after February 2020 (May 2020 to January 2024). This study was approved by our institutional review board in accordance with the Declaration of Helsinki.

After the identification of eligible patients, data were collected by review of the electronic medical records, and the following data were recorded: patient characteristics (Table 1), laboratory results, medication administration records, progress notes, and discharge summaries. Data were recorded onto a standardized electronic abstraction sheet.

ADAMTS13 activities and antibodies before July 2018 for 13 historical patients were measured using a FRETS-VWF73 assay (fluorescence-quenching substrate based on the 73mer polypeptidehormone Von Willebrand Factor; lowest detection limit <5%; Versiti, Milwaukee, WI). All ADAMTS13 activity and antibody testing from August 2018 to the present was performed at the Clements University Hospital. ADAMTS13 activity (normal reference range, 40%-130%) was measured using a commercially available kit from TECHNOZYM ADAMTS13 Activity ELISA (enzyme-linked immunosorbent assay; lowest detection limit <1%; Technoclone, Vienna, Austria), on a QUANTRA-Lyser 3000 analyzer (Werfen, Bedford, MA). ADAMTS13 antibodies (IgG; normal cutoff <15 units per mL) were measured using a commercially available kit, TECHNOZYM ADAMTS-13 inhibitor ELISA (Technoclone), on a BioTek 800 TS absorbance reader (Agilent Technologies, Santa Clara, CA). Blood samples were collected in 3.2% sodium citrate before PLEX initiation and centrifuged, and plasma was stored at −80°C until the performance of the assays. During the hospital stay, plasma samples were collected at the initial presentation from all patients and twice weekly (Monday and Thursday) before PLEX. After the resolution of an acute iTTP episode, ADAMTS13 activities were routinely measured in the outpatient setting, weekly for the first month, followed by every 3 months for the first year, and as needed thereafter.

Clinical outcomes were defined per the most recent International Working Group consensus report.¹² Clinical remission was defined as a sustained normalization of the platelet count ≥150 × 10⁹/L after cessation of PLEX for ≥30 days.¹² Partial remission was defined as a sustained ADAMTS13 activity >20%, whereas complete remission was defined as sustained ADAMTS13 activity >30%.¹² iTTP exacerbation was defined as a fall in the platelet count (<150 × 10⁹/L) <30 days after PLEX discontinuation and required PLEX reinitiation. iTTP clinical relapse was defined as a fall in the platelet count <150 × 10⁹/L after clinical remission ≥30 days after PLEX discontinuation, which required PLEX reinitiation.¹² 

Time to attain a sustained platelet count ≥150 × 10⁹/L after the initiation of PLEX was recorded. Times to attain a sustained rise in ADAMTS13 activities of ≥10%, ≥20%, and ≥30% during acute treatment were also documented. ADAMTS13 activities and platelet counts closest to the discontinuation of caplacizumab and PLEX were documented. The following outcome data were also recorded: length of stay, caplacizumab-associated bleeding events, iTTP-associated mortality, and all-cause mortality.

To summarize the data, descriptive statistical analyses were used. Continuous variables were described using medians along with minimum and maximum values. Frequencies and proportions were used for categorical variables. When evaluating differences between the non-caplacizumab cohort and the caplacizumab cohort, a Mann-Whitney U test was used for continuous variables, whereas a χ² test was used to evaluate categorical variables. Statistical significance was assumed at an alpha value of .05, and all statistical analyses were performed using Wizard statistical software version 1.9.48.

Patient characteristics are summarized in Table 1. Sixteen patients (20 iTTP episodes) were identified in the non-caplacizumab cohort, and 17 patients (20 iTTP episodes) were identified in the caplacizumab cohort. Age and sex were similar for both cohorts. The caplacizumab cohort had a larger proportion of White patients, with no statistical difference in Black or Asian patients in either group. Initial and recurrent iTTP presentations were similar between the 2 groups. On admission, both cohorts demonstrated ADAMTS13 activities <10% with similar ADAMTS13 antibody concentrations and platelet counts.

Treatment characteristics of the non-caplacizumab cohort vs the caplacizumab cohort are compared in Table 2. Daily PLEX (1-1.2 volume, using Octaplas replacement fluid) was initiated immediately after a clinical diagnosis of iTTP. Most patients received steroids in the historical cohort (15/16), whereas all patients received steroids in the caplacizumab cohort. The time from admission to the initiation of PLEX and steroids was comparable between the 2 groups, with a similar number of PLEX procedures performed for each group. Rituximab was given to all patients in the historical cohort and 15 patients in the caplacizumab cohort, with 2 not treated with rituximab due to immunocompromised status. In total, 160 doses of caplacizumab were administered to 17 distinct patients for 20 iTTP events (median, 6 doses; range, 2-30).

Clinical outcomes are summarized in Table 3.

The platelet count between the groups at the time of PLEX discontinuation was similar. After the initiation of acute therapy, the caplacizumab cohort achieved a sustained normalization in platelet count (≥150 × 10⁹/L) at a median of 4 days (range, 2-10) vs a median of 6 days (range, 3-29) in the non-caplacizumab cohort (P = .2).

ADAMTS13 activities at the time of PLEX discontinuation were comparable between the cohorts. Although not statistically significant, patients receiving caplacizumab appeared to show a trend toward an accelerated rate of ADAMTS13 activity recovery with activities ≥10%, ≥20%, and ≥30% achieved by 5 days (range, 1-72), 5 days (range, 1-72), and 9 days (range, 1-230), respectively, compared with 10.5 days (range, 2-929), 14 days (range, 2-929), and 17.5 days (range, 3-929) in the non-caplacizumab cohort (P = .3; P = .1; P = .1, respectively). At the time of caplacizumab discontinuation, the cohort demonstrated a median ADAMTS13 activity of 50% (range, 1%-97%). All non-caplacizumab patients (12/12) and caplacizumab patients (13/13) with ADAMTS13 activities ≥30% at the time of PLEX discontinuation showed sustained ADAMTS13 activities ≥30% over the initial 30-day period after PLEX, as measured by weekly ADAMTS13 activity measurements. The caplacizumab cohort showed a median ADAMTS13 activity in closest proximity to 30 days after PLEX of 70% vs a median ADAMTS13 activity of 84% in the non-caplacizumab cohort.

One patient in the non-caplacizumab cohort demonstrated an exacerbation (1/16 [6%]) 3 days after the discontinuation of PLEX. ADAMTS13 activity and platelet count at the time of discontinuing PLEX were 65% and 166 × 10⁹/L, respectively. The patient was readmitted after outpatient follow-up due to a drop in platelet count and hemoglobin and a rise in LDH with subsequent treatment with a single PLEX treatment, steroids, and rituximab before discharge. ADAMTS13 activity was not collected at this time.

Three patients in the non-caplacizumab cohort demonstrated 4 episodes of iTTP relapse (3/16 [19%]) on days 151, 161, 189, and 203 after discontinuation of PLEX, respectively (median, 175 days). Two of the 3 patients with relapse were female. One of the patients with iTTP recurrence was Black, 1 was White, and 1 was Asian. Two of the 3 patients received rituximab on initial presentation. All patients had an ADAMTS13 activity <5% at the time of recurrence and were treated with PLEX and immunosuppressive therapy.

None of the patients in the caplacizumab cohort had an exacerbation.

Five patients in the caplacizumab cohort demonstrated iTTP relapse (5/17 [29%]) on days 102, 147, 234, 471, and 1053 after discontinuation of PLEX (median, 234 days). Relapses occurred 102, 175, 237, 476, and 1053 days after discontinuation of caplacizumab, respectively (median 237 days). Three of 5 patients with relapse were female. All 5 patients with iTTP relapse were Black, with 4 of 5 patients receiving rituximab treatment on initial presentation. Four patients demonstrated ADAMTS13 activities <1% at the time of relapse, whereas 1 had an ADAMTS13 activity of 13%. All 5 relapses were managed with PLEX and immunosuppressive therapy. Three patients were re-treated with caplacizumab, whereas 2 patients did not receive caplacizumab due to active bleeding.

No deaths occurred in either cohort. Caplacizumab-associated bleeding was observed in 1 patient (1/17 [5.8%]), with 2 episodes of epistaxis, each after the first and second doses of caplacizumab. Caplacizumab was discontinued for 2 days before restarting therapy, with 27 additional doses given without further complication.

In this real-world, before-and-after study, we found that patients with iTTP who were treated with caplacizumab guided by ADAMTS13 activity (median, 6 doses) showed a shorter time to platelet normalization (4 vs 6 days; P = .2) and underwent fewer PLEX procedures (10 vs 11; P = .74) than a historical cohort, although these differences were not statistically different.

With increasing recognition of chronic neuropsychiatric effects and silent cerebral infarction associated with iTTP, using caplacizumab to rapidly disrupt the formation of microthrombi responsible for acute and long-term adverse outcomes is prudent.³ Although caplacizumab acts as a potent disease-modifying agent, it is important to underscore that it does not address the underlying cause of iTTP, the production of autoantibodies that impair ADAMTS13 activity. However, sustained ADAMTS13 levels >10% to 20% should signify adequate control of the underlying autoimmune disease and corresponding antibody production. In cases in which ADAMTS13 activities are sustained at this level, discontinuation of caplacizumab is likely appropriate to mitigate the risk of severe bleeding due to acquired von Willebrand disease.

Interestingly, when measured twice weekly during acute treatment, patients receiving caplacizumab showed a trend toward a shorter time to ADAMTS13 activity recovery >30% relative to the historical non-caplacizumab cohort (median, 9 days vs 18 days; P = .1). Although the rate of ADAMTS13 recovery is predominately influenced by the extent of autoantibody production and early initiation of immunosuppressive therapy, the use, timing, and dosage of corticosteroids were comparable between the 2 groups. Although the initiation of rituximab occurred later in the historical cohort (5 vs 3 days; P = .04), this is unlikely to explain the extent of accelerated ADAMTS13 recovery observed in caplacizumab-treated patients. Findings from the TITAN trial demonstrated that caplacizumab-treated patients showed an increased clearance of caplacizumab-VWF complexes relative to unbound VWF in untreated patients.⁴ Thus, the observed trend toward an accelerated rate of ADAMTS13 recovery during acute treatment may not simply be a product of inhibition of autoantibody production but, additionally, a concurrent fall in VWF concentrations and a corresponding reduction in the extent of ADAMTS13 consumption.

Although caplacizumab patients showed more rapid ADAMTS13 recovery relative to non-caplacizumab patients, patients in both groups with an ADAMTS13 activity >30% at the time of PLEX discontinuation showed sustained activities >30% over the initial 30-day period after PLEX, as measured by weekly ADAMTS13 activity measurements. These findings differ from data reported in other real-world studies, which show a delayed recovery in patients treated with caplacizumab dosed according to manufacturer recommendations compared with non-caplacizumab–treated patients.^8,11 After acute treatment using PLEX, steroids, rituximab, and caplacizumab (used for a median of 33 days; range, 29-39), Coppo et al observed an ADAMTS13 ≥20% at a median of 28 days after discontinuing PLEX.⁸ Similarly, Prasannan et al found that when patients received the aforementioned treatment regimen with caplacizumab used for a median of 35 days (range, 15-130), an ADAMTS >30% was observed at a median of 31 days after PLEX, with 28% of these patients demonstrating protracted recovery with an ADAMTS13 <30% at 58 days after PLEX completion.¹¹ Although the potential underlying cause of delayed ADAMTS13 recovery observed in these 2 studies is uncertain, caplacizumab patients in our study received significantly fewer caplacizumab doses (median 6) while demonstrating a shorter time to ADAMTS13 recovery. These observations suggest a possible relationship between the number of caplacizumab doses and the rate of ADAMST13 recovery.

Despite using an abridged caplacizumab treatment regimen, no exacerbations or patient deaths were observed. Five remote relapse episodes occurred at a median of 234 and 247 days after PLEX and caplacizumab discontinuation, respectively. Findings from the HERCULES trial demonstrated that when ADAMTS13 activities >10% were observed after PLEX, neither exacerbation nor relapse was observed in caplacizumab-treated patients.⁵ Accordingly, only 1 of the 5 patients developing relapse in our study had an ADAMTS13 activity of <10% after caplacizumab discontinuation. Overall, the median ADAMTS13 activity at the time of caplacizumab discontinuation was not significantly different in patients with relapse (41%; range, 1%-73%) vs those without (62%; range, 18%-97%; P = .2).

All 5 caplacizumab-treated patients with relapse were Black, which is notable as a recent study showed a significantly shorter relapse-free survival time in Black patients, despite the use of rituximab compared with White patients.¹³ Relative to patients evaluated in clinical trials and postmarketing studies, the proportion of Black patients in our study population (65%) was higher (clinical studies 11%-21%; postmarketing 10%-41%).^4,5,8,11,14 Moreover, the time to follow-up of patients evaluated in our study was longer, with a median patient follow-up of 624 days (range, 30-1364) relative to those reported in clinical trials (follow-up 28-365 days) and postmarketing studies (median 80-127 days).^4,5,8,14 Thus, the incidence of recurrence may reflect the differences in patient populations and length of observation relative to other studies. Nonetheless, the 12-month incidence of relapse observed in our study was equivalent to those reported in the TITAN trial (3/17 [18%] vs 11/36 [31%]; P = .51).⁴ Finally, the incidence of caplacizumab-associated bleeding events in our study was notably low (1/17 [6%]) compared with the rates reported in other real-world, postmarketing studies (range, 18%-33%), a finding possibly explained by the significant reduction in caplacizumab treatments administered to our patients.^7-10 

In our study, 160 doses of caplacizumab were administered to 17 distinct patients for 20 iTTP events. If managed according to manufacturer recommendations, ∼840 doses of caplacizumab would have been administered. Using the reported wholesale caplacizumab price of $9510 per dose, the cost of administering caplacizumab to our patients was $1 521 600. In contrast, the cost of administering caplacizumab per manufacturer recommendations would have been $7 988 400, a difference of $6 466 800 (81% cost reduction). If caplacizumab were discontinued based on an ADAMTS13 activity >10%, 36 fewer doses of caplacizumab would have been administered to our patients, translating to an additional cost savings of $342 360 and a cost of $1 179 240 (85% cost reduction). Further savings would be realized if dosing were individualized based on the routine monitoring of VWF activity levels. Overall, the financial savings reported in our study are significant, especially given recent research findings that question the cost-effectiveness of caplacizumab using the manufacturer-recommended dosing schema relative to PLEX and immunosuppression alone.¹⁵ 

The frequency of ADAMTS13 activity monitoring to guide caplacizumab and PLEX therapies throughout the disease course and optimize patient care is a strength of this study. Moreover, a strength of this study is the uniformity of treatment and monitoring used to manage the acute and longitudinal care of our consecutively treated patients in a real-world setting. Our study has inherent limitations due to the retrospective observational design, which warrants caution when interpreting our data. The retrospective approach relied upon the availability and accuracy of recorded data and is thus prone to selection bias. In addition, this study was limited to 2 tertiary care hospitals, which may limit the generalizability of our study findings. Lastly, the small sample size of this study, an inherent limitation frequently observed with studies of rare diseases, may lead to difficulties in detecting true effects and increase the likelihood of false-negative study findings. Consequently, this study aimed to identify associations rather than causality due to the nature of the retrospective approach being assessed.

In conclusion, this study suggests that an abridged, individualized caplacizumab dosing regimen guided by ADAMT13 activity is safe and effective. Such an approach can potentially maximize caplacizumab’s benefits with reduced bleeding risks and health care costs. A prospective study using ADAMTS13 and VWF activities to guide the frequency and duration of caplacizumab therapy is warranted.

The authors thank Natalie Newsom for her assistance with collecting the clinical and laboratory data.

Contribution: S.G.Y. conceptualized and designed the research, collected and analyzed data, and wrote the manuscript; S.L.H., I.F.I., Y.-M.P.S., and A.P.G. reviewed the manuscript; and R.S. conceptualized and designed the research and reviewed the manuscript.

Conflict-of-interest disclosure: R.S. is a consultant for Octapharma and has been an adviser to Sanofi and Takeda. The remaining authors declare no competing financial interests.

Correspondence: Ravi Sarode, Division of Hematology, Division of Transfusion Medicine and Hemostasis, Department of Pathology and Internal Medicine, UT Southwestern Medical Center, 5323 Harry Hines Blvd, Dallas, TX 75390; email: ravi.sarode@utsouthwestern.edu.