Pharmacotherapeutic group: antihemorrhagics, other systemic hemostatics, ATC code: B02BX11
Mechanism of action
Marstacimab is a human monoclonal IgG1 antibody directed against the Kunitz domain 2 (K2) of tissue factor pathway inhibitor (TFPI), the primary inhibitor of the extrinsic coagulation cascade. TFPI initially binds to and inhibits the factor Xa active site via its second Kunitz inhibitor domain (K2). The action of marstacimab to neutralise the inhibitory activity of TFPI may serve to enhance the extrinsic pathway and bypass deficiencies in the intrinsic pathway of coagulation by increasing free factor Xa available to increase thrombin generation and promote haemostasis.
Pharmacodynamic effects
Consistent with its anti-TFPI mechanism, marstacimab administration to haemophilia patients causes an increase in total TFPI and downstream biomarkers of thrombin generation such as prothrombin fragments 1+2, peak thrombin, and D-Dimer. These changes were reversible after treatment discontinuation. Sporadic or transient increases in D‑Dimer and prothrombin fragments 1+2 above physiological values were reported in the Phase 3 study with no associated safety concerns.
Clinical efficacy and safety
Clinical studies in adult and adolescent patients with haemophilia A without FVIII inhibitors or haemophilia B without FIX inhibitors
Patients (aged ≥ 12 years old and ≥ 35 kg) with haemophilia A without inhibitors and haemophilia B without inhibitors (Study B7841005)
The pivotal Phase 3 study was a one-way, cross-over, open-label, multi-centre study in 116 adult and adolescent males (aged 12 years and older and ≥ 35 kg) with severe haemophilia A without FVIII inhibitors or severe haemophilia B without FIX inhibitors who previously received “on-demand” (N = 33) or prophylactic (N = 83) treatment with FVIII or FIX. Patients with previous or current treatment for or history of coronary artery disease, venous or arterial thrombosis or ischaemic disease were excluded from the study.
The study population was characterised by a severe bleeding phenotype. The mean annualised bleeding rates (ABRs) were 39.86 and 7.90 in a 6-month Observational Phase for the on-demand and prophylaxis cohorts, respectively, prior to crossing over to weekly marstacimab prophylaxis. All (100%) patients in the on-demand cohort had one or more target joints at study entry and 36% had 3 or more target joints at study entry. In the routine prophylaxis cohort, 56.6% of the patients had one or more target joints at study entry and 15.7% had 3 or more target joints at study entry.
After the 6-month Observational Phase in which patients received either on-demand or routine prophylactic factor-based replacement therapy, patients received an initial 300 mg loading dose of marstacimab followed by maintenance doses of 150 mg of marstacimab once weekly for 12 months. Dose escalation to 300 mg of marstacimab once weekly was allowed after 6 months for patients weighing ≥ 50 kg experiencing 2 or more breakthrough bleeds. Fourteen (12.1%) out of 116 patients who received marstacimab for at least 6 months underwent dose escalation of their maintenance dose.
The mean age across the treatment groups was 32.4 years (min 13, max 66); 16.4% of patients were 12 to < 18 years, and 83.6% were ≥ 18 years, 100% were male. In this study 48.3% of patients were White, 50.0% were Asian, 0.9% were Black or African American, and 0.9% race information missing; 10.3% of patients identified as Hispanic or Latino. All patients were non-inhibitors (78.4% haemophilia A, 21.6% haemophilia B).
The primary efficacy objective of the study was to compare marstacimab prophylaxis during the Active Treatment Phase versus routine prophylactic factor-based therapy in the Observational Phase as measured by the ABR of treated bleeds. Other key efficacy objectives of the study included evaluation of marstacimab prophylaxis in comparison with routine prophylactic factor‑based therapy as measured by the incidences of spontaneous bleeds, joint bleeds, target joint bleeds and total bleeds, as well as assessing patients' health-related quality of life (HRQoL).
Table 2 shows the efficacy results of marstacimab prophylaxis compared with routine prophylactic factor-based therapy. Marstacimab showed non-inferiority and statistical superiority over routine prophylactic factor-based therapy as measured by ABR of treated bleeds.
Table 2. Comparison of ABR with Hympavzi prophylaxis versus previous routine factor‑based prophylaxis in patients ≥ 12 years of age without factor VIII or factor IX inhibitors
| Endpoints in the order of testing hierarchy | Routine factor-based prophylaxis during 6‑month OP (N = 83) | Hympavzi prophylaxis during 12‑month ATP (N = 83) |
| Treated bleeds (Primary) |
| ABR, model-based (95% CI) | 7.90 (5.14, 10.66) | 5.09 (3.40, 6.78) |
| Difference vs. RP (95% CI) | -2.81 (-5.42, -0.20) p-value = 0.0349* |
| Patients with 0 bleeds, n (%) | 33 (39.8) | 29 (34.9) |
| Spontaneous bleeds, treated |
| ABR, model-based (95% CI) | 5.89 (3.57, 8.22) | 3.78 (2.25, 5.31) |
| Difference vs. RP (95% CI) | -2.11 (-4.26, 0.03) Non-inferiority* |
| Joint bleeds, treated |
| ABR, model-based (95% CI) | 5.69 (3.36, 8.02) | 4.13 (2.59, 5.67) |
| Difference vs. RP (95% CI) | -1.55 (-3.73, 0.62) Non-inferiority* |
| Total bleeds, treated and untreated |
| ABR, model-based (95% CI) | 8.90 (6.02, 11.77) | 5.98 (4.14, 7.82) |
| Difference vs. RP (95% CI) | -2.91 (-5.66, -0.17) Non-inferiority* |
| Target joint bleeds, treated |
| ABR, model-based (95% CI) | 3.37 (1.60, 5.15) | 2.51 (1.26, 3.76) |
| Difference vs. RP (95% CI) | -0.87 (-2.42, 0.69) Non-inferiority* |
*Criterion Met (Non-inferiority/p-value if met superiority)
• The protocol specified non-inferiority criterion (upper bound of the 95% CI for the difference) was 2.5 for treated bleeds, spontaneous bleeds, joint bleeds; 1.2 for target joint bleeds; 2.9 for total bleeds. If the non‑inferiority criterion was met, superiority was subsequently tested and established if the confidence interval excluded zero.
• p-value is for the superiority testing.
• The estimated mean, difference, and confidence intervals (CIs) for the ABR come from negative binomial regression model.
• Bleed definitions adapted based on International Society on Thrombosis and Haemostasis (ISTH) criteria.
• Treated bleeds = bleeds treated with FVIII or FIX
• Total bleeds = bleeds treated and not treated with FVIII or FIX
• ABR = Annualised Bleeding Rate; CI = Confidence Interval; OP = Observational Phase; ATP = Active Treatment Phase; RP = Routine Prophylaxis
Study B7841007 interim analysis
In the OLE of the pivotal Phase 3 study, 87 patients received marstacimab at the doses established during participation in the B7841005 study (i.e. 150 mg or 300 mg subcutaneously once weekly) for up to an additional 16 months (mean 7 months) where marstacimab was shown to maintain long-term (> 12 months) efficacy.
Descriptive analyses were conducted to assess marstacimab prophylaxis over time. The model‑based mean and other descriptive summaries for the ABR of treated bleeds are shown in Table 3.
Table 3. ABR with Hympavzi prophylaxis over time in patients ≥ 12 years of age without factor VIII or factor IX inhibitors
| Endpoint | Time interval |
| First 6 months of ATP (N = 116) | Second 6 months of ATP (N = 112) | B7841007* (N = 87) |
| Treated Bleeds |
| Mean ABR (95% CI) | 4.96 (3.67, 6.70) | 3.26 (2.39, 4.44) | 2.79 (1.90, 4.10) |
| Median ABR (IQR) | 2.00 (0.00, 5.99) | 1.91 (0.00, 4.09) | 0.00 (0.00, 4.10) |
*Patients received marstacimab for up to an additional 16 months (mean 7 months) during B7841007.
• The estimated mean and confidence intervals (CIs) for the ABR come from negative binomial regression model.
• The median and the interquartile range (IQR), 25th percentile to 75th percentile, for the ABR comes from the descriptive summary.
• ABR = Annualised Bleeding Rate; CI = Confidence Interval; IQR = Interquartile Range; ATP = Active Treatment Phase (B7841005); N = number of patients who contributed data for analyses at each time interval
Immunogenicity
During the 12-month Active Treatment Phase in the pivotal Phase 3 Study B7841005, 23 of the 116 (19.8%) ADA-evaluable marstacimab-treated patients developed ADAs. ADAs were transient in 61% (14/23) and persistent in 39% (9/23) of the ADA-positive patients, indicative of a transient ADA profile in the majority of the patients. ADA titres resolved in 22/23 (95.7%) patients by the end of the study. Neutralising antibodies (NAbs) developed in 6/116 (5.2%) ADA-evaluable marstacimab-treated patients during the study. The NAbs were transient in all patients and no patients were NAb positive at the end of the study. Although slightly lower mean marstacimab concentrations (approximately 24%‑32% lower) were reported in ADA‑positive patients compared to ADA‑negative patients, concentrations largely overlapped between these 2 groups and there was no identified clinically significant effect of ADAs, including NAbs, on safety or efficacy of marstacimab over the treatment duration of 12 months. Overall, the safety profile of marstacimab was similar between those patients with ADAs (including NAbs) and those without.
In the Phase 3 OLE study, only one of the 44 ADA-evaluable patients continuing to receive marstacimab for at least 6 months was persistently positive for ADAs.
Paediatric population
The European Medicines Agency has deferred the obligation to submit the results of studies with Hympavzi in one or more subsets of the paediatric population in the treatment of congenital haemophilia A and congenital haemophilia B.