Description of Studies

Two phase III trials (HAVEN 3, N = 152, and HAVEN 4, N = 41) submitted by the sponsor were included in the systematic review.

HAVEN 3 was a 24-week, open-label, multi-centre, randomized controlled trial (RCT) that aimed to evaluate the efficacy of prophylactic emicizumab compared with no prophylaxis in patients with severe hemophilia A (i.e., intrinsic FVIII levels of less than 1%) without FVIII inhibitors who had received prior treatment with episodic or prophylactic FVIII. Patients who received episodic treatment with FVIII prior to study entry were randomized in a 2:2:1 ratio to the following treatment arms: emicizumab prophylaxis at 3.0 mg/kg weekly for 4 weeks, followed by 1.5 mg/kg weekly; emicizumab prophylaxis at 3.0 mg/kg weekly for 4 weeks, followed by 3.0 mg/kg every 2 weeks; no prophylaxis (control arm). Patients who received FVIII prophylaxis prior to study entry (derived from the non-interventional study [NIS]) were enrolled in a separate, non-randomized, single arm where they received treatment with emicizumab prophylaxis at 3.0 mg/kg weekly for 4 weeks, followed by 1.5 mg/kg weekly. Patients in this arm continued their regular FVIII prophylaxis treatment until the second emicizumab loading dose. In HAVEN 3, the primary outcome was related to the annualized bleeding rate (ABR) ratio for treated bleeds, where treated bleeds were defined as any bleed if coagulation factors were administered for the treatment signs or symptoms of bleeding (e.g., pain, swelling) irrespective of the time between the treatment and the preceding bleed. The ABR was calculated as (number of bleeds divided by the total number of days during the 24-week period) multiplied by 365.25. Secondary outcomes pre-specified in the statistical testing hierarchy included additional bleeding outcomes (ABR ratio for all bleeds, treated joint bleeds, treated spontaneous bleeds, treated target joint bleeds) and health-related quality of life (HRQoL) (i.e., as measured by the Haemophilia Quality of Life Questionnaire for Adults [Haem-A-QoL] physical health score).

HAVEN 4 was a 24-week, open-label, multi-centre, non-randomized, single-arm trial that aimed to investigate the efficacy, safety, pharmacokinetics, and pharmacodynamics of emicizumab prophylaxis in patients with severe congenital hemophilia A without inhibitors or patients with congenital hemophilia A with FVIII inhibitors. All patients in HAVEN 4 received treatment with emicizumab prophylaxis at 3.0 mg/kg weekly for 4 weeks, followed by 6.0 mg/kg every 4 weeks.

Efficacy Results

In HAVEN 3, the primary outcome related to the ABR ratio for treated bleeds demonstrated a statistically significant reduction in bleeding for both emicizumab 1.5 mg/kg weekly and emicizumab 3.0 mg/kg every 2 weeks compared to no prophylaxis (i.e., episodic FVIII) for patients previously treated with episodic FVIII. In HAVEN 3, 55.6% of patients treated with 1.5 mg/kg weekly emicizumab experienced 0 treated bleeds during the efficacy period (Table 2). The ABR ratio between 1.5 mg/kg weekly emicizumab (ABR = 1.5) and no prophylaxis (ABR = 38.2) was 0.04 (95% confidence interval [CI], 0.020 to 0.075; P < 0.0001) in favour of 1.5 mg/kg weekly emicizumab. Sixty percent of patients treated with 3.0 mg/kg emicizumab every 2 weeks experienced 0 treated bleeds during the efficacy period. The ABR ratio between 3.0 mg/kg emicizumab every 2 weeks (ABR = 1.3) and no prophylaxis (ABR = 38.2) was 0.03 (95% CI, 0.017 to 0.066; P < 0.0001) in favour of 3.0 mg/kg emicizumab every 2 weeks. None of the patients in the no-prophylaxis group experienced 0 treated bleeds. The ABR ratios were clinically relevant, according to clinical experts consulted for this review.

Table 2

Summary of Key Results From Pivotal and Protocol Selected Studies.

In HAVEN 3, secondary outcomes related to the ABR ratio for all bleeds, treated joint bleeds, and treated spontaneous bleeds demonstrated a statistically significant reduction in bleeding for both 1.5 mg/kg emicizumab weekly and 3.0 mg/kg emicizumab every 2 weeks compared to no prophylaxis (i.e., episodic FVIII) for patients previously treated with episodic FVIII. The results of the sensitivity analyses for all bleeding outcomes in HAVEN 3 were consistent with the primary results.

For treated bleeds, an ABR ratio of 0.32 (95% CI, 0.195 to 0.514; P < 0.0001) in favour of 1.5 mg/kg weekly emicizumab was reported for the intra-patient comparison of patients treated with FVIII prophylaxis in the NIS (ABR = 4.8) compared to treatment with 1.5 mg/kg weekly emicizumab in HAVEN 3 (ABR = 1.5). The reduction in bleeding was clinically relevant, according to clinical experts consulted for this review. Similar efficacy findings were reported for the ABR ratio for all bleeds (Table 2).

In HAVEN 4, 56.1% of patients treated with 6.0 mg/kg emicizumab every 4 weeks experienced 0 treated bleeds; the ABR was 2.4 (95% CI, 1.38 to 4.28). Similar ABRs were reported for the other bleeding outcomes (i.e., ABR for all bleeds, treated joint bleeds, treated spontaneous bleeds, treated target joint bleeds) (Table 2). Although no statistical hypothesis testing was performed, the clinical experts consulted for this review indicated that the ABR results were clinically relevant.

Haem-A-QoL physical health was a secondary outcome in HAVEN 3 (Table 2). The difference in the adjusted mean Haem-A-QoL physical health subscore at week 25 between patients receiving 1.5 mg/kg weekly emicizumab and those receiving no prophylaxis was 12.51 (95% CI, −1.96 to 26.98; P = 0.891). Given that statistical significance was not achieved, statistical testing according to the pre-specified hierarchy was stopped prior to the assessment of 3.0 mg/kg emicizumab every 2 weeks. Therefore, the potential effect on improved quality of life remains inconclusive.

Harms Results

In HAVEN 3, AEs occurred in 94.4% of patients receiving 1.5 mg/kg emicizumab weekly, 85.7% in those receiving 3.0 mg/kg emicizumab every 2 weeks, 50.0% in the no-prophylaxis arm, and 87.3% in the previous-FVIII-prophylaxis arm (1.5 mg/kg weekly emicizumab) (Table 2). In HAVEN 4, 73.2% of patients treated with 6.0 mg/kg emicizumab every 4 weeks experienced an AE. In both studies, the most common AEs were injection-site reactions. Notable harms identified in the protocol for this review included the following: thrombotic events, injection-site reactions, hypersensitivity reactions, inhibitor development, and blood-borne infections. The only notable harms reported in HAVEN 3 and HAVEN 4 were injection-site reactions, which occurred in 25.0% of patients on 1.5 mg/kg emicizumab weekly, 20.0% on 3.0 mg/kg every 2 weeks, 12.5% in the no-prophylaxis arm, and 31.7% in the previous-FVIII-prophylaxis arm (1.5 mg/kg weekly emicizumab). In HAVEN 4, 22.0% of patients treated with 6.0 mg/kg emicizumab every 4 weeks experienced an injection-site reaction.

Throughout HAVEN 3 and HAVEN 4, no instances of de novo inhibitor development were detected among patients who tested negative for inhibitors (titre less than 0.6 chromogenic Bethesda unit (CBU)/mL) at baseline.

In HAVEN 3, serious adverse events (SAEs) occurred in 2.8% of patients on 1.5 mg/kg emicizumab weekly, 8.6% of patients on 3.0 mg/kg every 2 weeks, 0% of patients in the no-prophylaxis arm, and 12.7% of patients in the previous-FVIII-prophylaxis arm (1.5 mg/kg emicizumab weekly). In HAVEN 4, 2.4% of patients treated with 6.0 mg/kg emicizumab every 4 weeks experienced an SAE. No SAE occurred in more than 1 patient per arm. No patients died in the studies.

The 24-week assessment period of HAVEN 3 and HAVEN 4 was determined to be of sufficient duration to identify harms, according to input provided by the clinical experts consulted in this review. Overall, the safety profiles were likely comparable between the 2 emicizumab treatment regimens (1.5 mg/kg weekly and 3.0 mg/kg every 2 weeks); however, due to the small sample size, further study is warranted.

Critical Appraisal

The key limitations of the body of evidence relate to the open-label study design, the absence of randomized, direct comparative data between emicizumab and FVIII prophylaxis, and generalizability to the Canadian clinical population.

HAVEN 3 and HAVEN 4 were open-label studies. The open-label design may have biased the subjective outcome results in favour of emicizumab. HAVEN 4, a single-arm study, was limited by its non-controlled study design. In HAVEN 4, analyses of all outcomes were descriptive, and no formal hypothesis testing was performed.

There is no direct comparative evidence to support the efficacy of emicizumab compared to FVIII prophylaxis in patients with severe hemophilia A. There is limited evidence from an intra-patient analysis (group D in HAVEN 3) that supports the efficacy of emicizumab in patients previously treated with FVIII prophylaxis.

The majority of patients with severe hemophilia A enrolled in HAVEN 3 (groups A, B, and C) were previously treated with episodic FVIII. This is not reflective of Canadian clinical practice, in which the standard of care for patients with severe hemophilia A is treatment with FVIII prophylaxis. Furthermore, patients in HAVEN 3 and HAVEN 4 who had previously received treatment with episodic FVIII were required to have 5 or more bleeds in the 24 weeks prior to study entry. This is inconsistent with Canadian clinical practice, as it is unlikely for Canadian patients with 5 or more bleeds over a period of 24 weeks to be treated with episodic FVIII.

Results of HAVEN 3 and HAVEN 4 may have limited generalizability to the Canadian patient population. In both studies of patients without inhibitors, the study population was restricted to patients with severe hemophilia A. According to clinical experts consulted for this review, there is a subset of patients with mild or moderate hemophilia who require prophylaxis. The designs of HAVEN 3 and HAVEN 4 did not include these patients; thus, the magnitude of the treatment effect in patients with mild and moderate hemophilia A is unclear.

Description of Studies

One network meta-analysis (NMA) provided by the sponsor was reviewed. The sponsor-submitted NMA compared emicizumab prophylaxis with FVIII prophylaxis for patients with hemophilia A without inhibitors.^12,13 The selection of trials for the NMA was based on a systematic literature review (SLR). The SLR was performed in December 2016 and updated for the non-inhibitor population in May 2018.

Harms Results

Harms were not assessed in the NMA submitted by the sponsor.

Critical Appraisal

Limitations associated with the NMA submitted by the sponsor included the small number of trials with small sample size included in the NMA; the high degree of clinical and statistical heterogeneity across the included studies in terms of disease severity; the use of different comparator FVIII products in different trials; inconsistent or unclear definitions of the bleed outcomes; variable outcome estimation time points across trials; and differences in study design.

These results were aligned with those observed in patients previously treated with FVIII prophylaxis (arm D) in the HAVEN 3 trial. However, due to various methodological limitations, the findings of the sponsor-submitted NMA should be interpreted with caution. In addition, no NMA was performed for safety outcomes, and no evidence for children younger than 7 years old was included in the NMA. No robust conclusions on the comparative clinical efficacy and safety profile of emicizumab prophylaxis regimens versus rFVIII prophylaxis in patients with hemophilia without inhibitors can be drawn.

Description of Studies

One other relevant study submitted by the sponsor was reviewed. The HOHOEMI study provides evidence of the efficacy and safety of SC emicizumab prophylaxis in a pediatric population with hemophilia A without FVIII inhibitors. The HOHOEMI study is a phase III, multi-centre, open-label, non-randomized clinical trial evaluating 2 dosing regimens for emicizumab after the administration of 4 loading doses of 3.0 mg/kg emicizumab per week. Patients received 3.0 mg/kg every 2 weeks or 6.0 mg/kg every 4 weeks for at least 24 weeks. For the purpose of this review, only data from the 3.0 mg/kg cohort are presented. The 6.0 mg/kg dose is not relevant to the Canadian pediatric population because it is beyond the dosing recommended by Health Canada for this patient population. The primary efficacy outcome was bleeding frequency, expressed as an ABR. ABR was estimated using negative binomial regression, with the treatment period as an offset to account for varying follow-up periods. Pre-treatment ABRs were calculated using documented bleed information and standardized for the different age groups over different durations: 12 weeks (▬) for patients less than 2 years of age and 24 weeks (▬) for patients 2 years and older. Calculated ABRs were computed by dividing the number of bleeding events by the evaluation period in days and multiplying this rate by 365.25. Patient preference was also identified in the CADTH review protocol as an outcome of relevance. It was also measured in the HOHOEMI study.

To be eligible for the HOHOEMI study, patients had to be less than 12 years of age, weigh more than 3 kg, and have severe (i.e., less than 1% endogenous FVIII) congenital hemophilia A without FVIII inhibitors (i.e., test negative, with less than 0.6 Bethesda unit (BU)/mL inhibitors within 8 weeks prior to enrolment). A total of 6 Japanese male patients were enrolled in the 3.0 mg/kg emicizumab cohort. All were previously on FVIII prophylaxis treatment.

Efficacy Results

Findings from the HOHOEMI study showed that all patients who received 3.0 mg/kg emicizumab experienced bleeding events. Of these patientsv ▬▬▬▬▬

▬▬▬▬▬

All caregivers preferred emicizumab prophylaxis over previous hemophilia treatments. The main reasons were lower treatment frequency (38% of caregivers) and fewer effects on daily activities and social interactions (23.1% of caregivers).

Harms Results

▬▬▬▬▬

Critical Appraisal

Limitations of the HOHOEMI study include the small sample size, the lack of randomization and blinding, and the lack of a control group. No conclusion can be made regarding the efficacy of emicizumab, given that no statistical testing was conducted against a control group. In addition, the study was conducted in Japanese children at 4 study sites in Japan, and may not be generalizable to Canadian patients. Therefore, the evidence from the HOHOEMI study on the efficacy and safety of emicizumab in pediatric patients is limited by concerns regarding internal validity and generalizability to the Canadian population.

About the Patient Group and Information Gathered

One patient group responded to CADTH’s call for patient input for emicizumab. The CHS is a national voluntary health charity with a mandate to improve the health and quality of life of all people in Canada with inherited bleeding disorders, and ultimately to find cures. The CHS has 10 provincial chapters and is affiliated with the World Federation of Hemophilia along with more than 125 national member organizations around the world. It has approximately 300 active volunteers, family members, individuals affected by bleeding disorders, and health care providers who work in bleeding disorder treatment centres. As well, the CHS collaborates with health care providers from Canada’s 26 inherited bleeding disorder comprehensive care centres, whose physicians make up the Association of Hemophilia Clinic Directors of Canada, Canada Blood Services and Héma-Québec, the Network of Rare Blood Disorder Organizations, the Canadian Organization for Rare Diseases, the hepatitis C community, the AIDS community, and others who share common interests.

A disclosure of any conflicts of interest for the CHS is available on the CADTH website.

In preparing their patient input submission, the CHS received external help from AHCDC physicians and other health care professionals. They provided treatment outcome data and patient-reported outcome comments from 14 of the 15 patients with hemophilia A without inhibitors who were receiving emicizumab through compassionate access. The CHS is a collaborator in the Patient-Reported Outcomes, Burdens, and Experiences (PROBE) study.

The CHS solicited information on patient experiences and perspectives in multiple ways. Because it is in regular contact with its members, it hears about the experiences of patients, caregivers, and physicians through personal meetings or conferences. The CHS also conducted an online survey between May 31, 2019 and June 15, 2019 to collect perspectives from patients with hemophilia A and their caregivers. The objective of was to gain insight into burdens of disease and treatment, satisfaction with current treatments, and improvements that patients and caregivers would like to see in new treatments. The survey was publicized through different CHS and chapter communication tools, such as email, the CHS website, Facebook, Twitter, and Instagram. A total of 52 responses were obtained from 6 provinces. All respondents were affected by hemophilia A without inhibitors: 45 with severe hemophilia, 4 with moderate hemophilia, 2 with mild hemophilia, and 1 with unknown severity.

Disease Experience

According to the patient input received for this review, hemophilia A negatively affects patients’ lives on physical, psychological, and financial levels. The key concerns raised by patients are breakthrough bleeds, venous access challenges, and treatment adherence difficulties due to the complex regimen. Despite prophylaxis therapy, patients reported that they still experience breakthrough bleeds, damage to joints, pain, reduced mobility, and iatrogenic complications due to frequent IV infusions. Chronic joint damage is an insidious consequence of hemophilia. In the survey, 8 out of 52 patients reported damage to joints when asked about key difficulties in an open-ended question. One patient noted that hemophilia A imposes limits on patients’ and families’ daily lives because they worry constantly about bleeds and whether hospital care will be accessible. One family noted that “globally…the majority of decisions about family activities and permissions granted to the children are dictated by hemophilia.” Many families live with anxiety, stress, and depression, which severely affects their quality of life. One parent indicated that their child’s inability to live a normal childhood is the “most difficult aspect of the illness,” while others noted that their children live with anxiety and low self-esteem. Two older patients with severe hemophilia A compared the difficulties they have experienced throughout their lives to having “swords of Damocles hanging over [them],” especially being victims of the HIV and hepatitis C contaminated blood supplies in the 1980s. Both lived with chronic pain and now have severely damaged joints. One of them “[ended] up in a coma for an entire summer” from a fall. Like many others, they were also concerned about passing on the genetic disorder to their children and grandchildren.

Treatment for hemophilia A is “mentally taxing not only on [the patient] but [also] on the person who helps [them].” Given the high frequency of infusions with FVIII, vein access is difficult and time-consuming for many. Parents and patients alike worry about “destroying” veins from frequent IV access. As well, many respondents mentioned difficulties in adhering to prophylactic regimens. Venous access through a port or catheter can be extremely painful and traumatizing. One parent stated that “one of the biggest impacts…is the constant needles…if he has a breakthrough bleed, the frustration and exhaustion is compounded by even more needles.” Parents are often the caregivers. As a result, many have been unable to maintain full-time employment or have had to reduce their work hours to provide care. This puts a significant financial burden on some families, adding to their existing exhaustions.

The PROBE study collected patient perspectives and experiences from 181 Canadian boys and men aged 11 and older with hemophilia A. Almost all respondents (89%) were receiving regular FVIII prophylaxis, and some (5%) were receiving on-demand treatment. Results of the PROBE study show the considerable burden of disease and impact of living of severe hemophilia A across all age groups, despite widespread access to and use of modern prophylactic treatment with FVIII. In the PROBE study, 142 out of 181 respondents reported having a reduced range of motion in at least 1 joint, suggesting that chronic joint damage is very common in older children and adults.

Experience With Treatment

The CHS survey reported that the majority of patients with severe hemophilia A (and almost all respondents) were on FVIII prophylactic treatment, with IV infusions administered 2 to 7 times per week. Most respondents reported being “somewhat satisfied” to “quite satisfied” with currently available factor therapies. Although most respondents found that treatment is quite effective in stopping and preventing bleeds, many reported that it provides insufficient protection, given that breakthrough bleeds still occur. One parent stated that they are “so glad that there is some type of treatment available so that [their] son can have some normality in his life… but [they are] frustrated with so many needles…Frustrated with the breakthrough bleeds. Very frustrated.” Additionally, FVIII prophylaxis does not fully protect against joint damage. In the survey, 5 out of 52 patients reported experiencing breakthrough joint bleeding — and some respondents mentioned developing long-term joint damage — despite FVIII prophylaxis. This can lead to a reduction in the range of motion in certain joints, which may affect mobility

The short half-life of FVIII therapy creates many challenges. The frequent treatment administration is not only time-consuming, but physically and emotionally exhausting. The most common challenge reported in the survey is venous access, which is particularly difficult in babies or children. One parent indicated that their son’s veins “were extremely difficult to find and it was taking 4-6 pokes each time to find a vein.” Parents feel as though they are “destroying” their child’s veins and worry about venous scarring. Additionally, the surgeries for the insertion and removal of venous access ports or catheters are “traumatic” and patients have difficulty adjusting. One parent noted that because the port gripper is left in place for a week, their son “cannot have a bath, or go swimming” during this time. Another common challenge is long distances to treatment centres for check-ups, treatments, or to pick up factor supplies for home use. This can make it difficult for patients to adhere to the treatment regimen. One parent expressed that “it is great that we have treatment available to us that allows him some involvement in physical activity” but that the complex schedule, due to frequent administrations, “was the main contributor to our decision to have one parent home to care for him. The costs of travel were significant.” The time-sensitive and short-lasting treatment regimen meant that parents had to coordinate their child’s activities with treatments meticulously “by calculating the approximate level of FVIII in his blood, which is very laborious.” Time lost from school and work resulting from the complicated treatment regimen is another challenge faced by many. One parent indicated that when they are unable to infuse due to venous access issues, her son “misses school” and both parents miss “a full day of work” to go to the hospital, where they also have to worry about parking fees.

Based on input from the patient group, in Canada, access to emicizumab is currently restricted to people with hemophilia A with inhibitors. Approximately 15 people with hemophilia A without inhibitors were granted compassionate access starting in autumn 2019. The Association of Hemophilia Clinic Directors of Canada conducted a survey in June 2020, which was sent to health care providers in 12 bleeding disorder centres caring for 14 patients; 13 surveys were returned. The median age of patients at the time of application to the program was 11 years (range: 13 months to 65 years). Prior to switching to emicizumab, patients were on standard or extended half-life FVIII therapy, with multiple infusions per week or on-demand infusions. With emicizumab, health care providers described the improvement in health outcomes and quality of life as dramatic, especially because the frequency of administration is reduced. The SC mode of administration allowed for greater adherence to treatment regimens, leading to better bleed protection and improved quality of life. Nine out of 13 patients had 0 bleeds on emicizumab. Other improvements included less joint pain and discomfort, fewer hospital visits, better mental health, and more independence. No side effects were reported. The family of a patient with severe hemophilia A, autism spectrum disorder, and severe global developmental delay indicated that while receiving recombinant FVIII treatment, their son had “roughly 100 ER visits in the first 4 years of life, despite being on home treatments” and “showed signs of severe anxiety.” After receiving emicizumab, his bleeds were reduced, but “the most substantial improvements we observed have been to his mental health and development…[He] has become active, learning to run, jump, and climb like other 4-year-olds. He has become extremely social…In the past 6 months, he was able to catch up on almost every developmental milestone and is now exceeding his age level in several areas… We now spend our days just as any other family does.” With emicizumab, “Parents feel less vulnerable, more empowered and in control” and patients say they have less anxiety, more independence, better self-esteem, and an overall better quality of life. For many patients on emicizumab, “family dynamics are better” and hemophilia is no longer “the focus of the day.”

Improved Outcomes

Patient input from the CHS indicated that the vast majority of patients and caregivers want a longer-lasting treatment that requires less frequent administration and an easier mode of delivery (i.e., not IV). A simpler mode of delivery would be especially beneficial to babies and children, for whom venous access is the most challenging. As well, patients noted that being able to self-administer the treatment would greatly improve independence. One patient stated that “treating once a week under the skin is [their] ultimate wish.” A less frequent dosing interval would likely greatly benefit patients who have difficulty adhering to frequent administrations. Additionally, patients hope for more reliable efficacy to achieve better protection from bleeding because many still experience breakthrough bleeds with the current FVIII replacement therapy. These outcomes are expected to improve the quality of life of patients and caregivers by reducing worry and stress; by allowing patients to be more active, more involved in society, and less dependent on health services; and by decreasing absenteeism from school and work. With a more constant factor level, 1 parent noted that they would not have to be “on call 24 hours a day…[and] could decide to go back to work or even commit to activities with greater certainty they would happen.” Another family said that a longer-lasting treatment would result in a “better overall family atmosphere.” Two older patients with severe hemophilia A — grandfathers to grandsons who inherited the disorder — have seen significant improvements in their own health with emicizumab, and wish that everyone with hemophilia A could have access to it so they could “lead more normal lives” and minimize long-term joint morbidity.

The CHS has suggested that patients with severe hemophilia — and those with mild and moderate disease who have a severe phenotype — would benefit most from treatment with emicizumab. As well, the CHS believes that the new treatment would greatly benefit babies and children (for whom venous access is most difficult), patients who suffer from frequent breakthrough bleeds and joint disease despite FVIII prophylaxis, and patients who have difficulties adhering to the current treatment regimen, which requires frequent IV infusions.

Inclusion and Exclusion Criteria

Detailed inclusion and exclusion criteria for HAVEN 3 and HAVEN 4 are presented in Table 5. Both trials included patients 12 years of age and older who weighed 40 kg or more at screening and had adequate hematologic function (platelet count greater than or equal to 100,000/μL and hemoglobin greater than or equal to 8 g/dL). In both trials, patients on episodic regimens were required to have 5 or more bleeds in the 24 weeks prior to study entry. HAVEN 3 included patients with severe congenital hemophilia A (intrinsic FVIII level less than 1%). Patients in HAVEN 3 were required to have a negative test for inhibitor (i.e., less than 0.6 BU/mL) within 8 weeks of enrolment. Conversely, HAVEN 4 included patients with severe congenital hemophilia A (intrinsic FVIII level not specified) without FVIII inhibitors or congenital hemophilia A with FVIII inhibitors.

Baseline Characteristics

The baseline characteristics were generally balanced between the 3 randomized treatment arms in HAVEN 3 (Table 6). All patients in HAVEN 3 and HAVEN 4 were male. In HAVEN 3, patients were 36.4 (standard deviation [SD] = 14.4) years to 40.4 (SD = 11.4) years of age, with no patients under the age of 18 in the arms receiving 1.5 mg/kg emicizumab weekly or 3.0 mg/kg emicizumab every 2 weeks. The mean age in HAVEN 4 was 38.7 (SD = 15.7); 7.3% of patients were under the age of 18. In both studies, White patients represented the largest racial group (HAVEN 3: 57.1% to 66.7%; HAVEN 4: 75.6%). In HAVEN 3, there were fewer Asian patients in the 1.5 mg/kg weekly arm, and fewer Black or African American patients in the arm receiving 3.0 mg/kg every 2 weeks. In HAVEN 4, 12.2% (n = 5) had FVIII inhibitors at study entry. The number of bleeds in the 24 weeks prior to study entry ranged from 15.1 (SD = 6.5) to 21.6 (SD = 16.4) in the randomized arms of HAVEN 3; there were 6.4 (SD = 17.7) in the patients that had previously between treated with FVIII prophylaxis. In HAVEN 4, the number of bleeds in the 24 weeks prior to study was 9.0 (SD = 15.2). The number of patients with more than 1 target joint prior to study entry ranged from 58.8% to 93.3% in the randomized arms of HAVEN 3; it was 69.2% in the patients previously treated with FVIII prophylaxis. In HAVEN 4, 68.0% of patients had more than 1 target joint prior to study entry.

Table 6

Summary of Baseline Characteristics.

Bleeding

The efficacy of emicizumab was assessed using various outcomes related to the frequency of bleeding episodes. The following bleeding outcomes were assessed in both HAVEN 3 and HAVEN 4: treated bleeds, all bleeds, treated joint bleeds, treated spontaneous bleeds, and treated target bleeds. Definitions for bleeds were based on standardized, adapted from criteria defined by the factor VIII, Factor IX and Rare Coagulation Disorders Subcommittee of the International Society on Thrombosis and Haemostasis (ISTH). Definitions for bleeds were consistent in HAVEN 3 and HAVEN 4. Two bleeds of the same type and anatomical location were considered as 1 bleed if the second occurred within 72 hours of the last treatment for the first bleed. The ABR for each outcome was calculated using the following formula: ABR = (number of bleeds divided by total number of days during the efficacy period) multiplied by 365.25.

The primary efficacy outcome in HAVEN 3 was based on the number of treated bleeds over time. An event was considered a treated bleed if coagulation factors were administered for the treatment signs or symptoms of bleeding (e.g., pain, swelling), irrespective of the time between the treatment and the preceding bleed. A primary efficacy outcome was not identified for HAVEN 4.

All bleeds included both treated and non-treated bleeds. Bleeds due to surgery or procedures were excluded.

Treated joint bleeds were defined as bleeds occurring in a joint with at least 1 of the following symptoms: increasing swelling or warmth of the skin over the joint and/or increasing pain, decreased range of motion, or difficulty in using the joint compared with baseline.

Treated spontaneous bleeds were defined as a treated bleed with no known contributing factor, such as trauma, surgery, or procedure.

Treated target joint bleeds were defined as a major joint (e.g., hip, elbow, wrist, shoulder, knee, and ankle) into which repeated bleeds occur (frequency of greater than or equal to 3 bleeds into the same joint over the last 24 weeks prior to study entry).

Data on bleeds (including start date and time, cause, type, location, and associated symptoms of each bleed) were collected through an electronic bleed and medication questionnaire that was developed and validated by the sponsor as part of the emicizumab clinical development program.

Productivity

Days away from school or work were recorded at baseline and week 25 based on the 4-week period preceding each data entry.

Health-Related Quality of Life

HRQoL was assessed using the Haem-A-QoL, Haemophilia-specific Quality of Life Questionnaire for Children Short Form (Haemo-QoL-SF), and the EuroQol 5-Dimensions 5-Levels questionnaire (EQ-5D-5L) index utility score and the EuroQol Visual Analogue Scale (EQ VAS).

The Haem-A-QoL is a disease-specific, self-reported questionnaire used to measure HRQoL in adult patients (18 years or older) with hemophilia.^30,31 It measures the following 10 domains: Physical Health, Feelings, View of Self, Sports and Leisure, Work and School, Dealing With Hemophilia, Treatment, Future, Family Planning, and Partnership and Sexuality. In total, the questionnaire has 46 items, with a 5-point Likert frequency scale (never, rarely, sometimes, often, all the time). Scores are transformed to a value ranging from 0 to 100, with higher scores indicating poorer HRQoL. The validity, reliability, and responsiveness of the Haem-A-QoL have been established.³¹ Construct validity was adequate for 8 out of 10 domains and the total score.³⁰ Convergent validity and internal consistency were determined to be acceptable in patients with hemophilia.³⁰ The Haem-A-QoL was sensitive to detect change over time in patients with hemophilia. A minimal important difference (MID) was not identified in the literature for patients with hemophilia A for the Haem-A-QoL; however a 7-point reduction in total score was found to indicate a threshold for responders which showed a benefit in the HRQoL of individual patients with hemophilia.³² Another study of patients with hemophilia used half an SD of the mean as the MID.³³

The Haemo-QoL-SF is a 35-item, disease-specific, age-related questionnaire for children and adolescents aged 8 years to 16 years with hemophilia.³⁴ It includes the following 9 dimensions: Physical Health, Feelings, View of Self, Family, Friends, Other Persons, Sports and School, Dealing With Hemophilia, and Treatment.³⁵ Responses for the 35 items are scored on a 5-point Likert scale (never, seldom, sometimes, often, and always). The overall score ranges from 0 to 100, with higher scores indicating poorer HRQoL. The Haemo-QoL-SF was modified from the Haemo-QoL long form, which has been validated and shown to be reliable.³⁴ Convergent validity, internal consistency, inter-rater reliability, and test-retest reliability of the Haemo-QoL-SF were determined to be acceptable in patients with hemophilia.³⁵ No literature assessing the responsiveness of the Haemo-QoL-SF in patients with hemophilia was identified. No MID was identified in the literature for the Haemo-QoL-SF.

The EQ-5D-5L is a generic, preference-based HRQoL instrument consisting of a Visual Analogue Scale (VAS) and a composite index score of 5 dimensions: mobility, self-care, usual activities, pain/discomfort, and anxiety/depression.³⁶ Each dimension has 5 levels of function: no problems, slight problems, moderate problems, severe problems, and extreme problems. The EQ VAS is a vertical VAS from 0 to 100 (20 cm), with anchors of the worst imaginable and best imaginable health states, respectively.³⁷ Validity and reliability of the EQ-5D-5L has been demonstrated in patients with hemophilia.^37,38 Item-total correlation and internal consistency of the index score was satisfactory in patients with hemophilia.³⁸ No literature assessing the responsiveness of the EQ-5D-5L utility index in patients with hemophilia was identified. Convergent validity was low to moderate in patients with hemophilia for the EQ VAS.³⁹ No literature assessing reliability and responsiveness of the EQ VAS in patients with hemophilia was identified. No MID was identified for the EQ-5D-5L index or the EQ VAS in populations with hemophilia. The MID was estimated to range from 0.037 to 0.056 in the general population.⁴⁰

Hospitalization

Days hospitalized were recorded at baseline and week 25 based on the 4-week period preceding each data entry.

Pain

Pain was assessed as a domain within the EQ-5D-5L. The 5 levels within the pain domain ranged from “I have no pain or discomfort” to “I have extreme pain or discomfort.”³⁶ Item-total correlation was satisfactory for all domains in the EQ-5D-5L and highest for the “pain/discomfort” domain in patients with hemophilia.³⁷ Construct validity was acceptable for the “pain/discomfort” domain in patients with hemophilia.³⁸ A MID was not identified in the literature for patients with hemophilia A for the pain domain of the EQ-5D-5L.

Patient Satisfaction

Patient preference and satisfaction with treatment were assessed through the Emicizumab Preference (EmiPref) Survey and the Satisfaction Questionnaire – Intravenous Subcutaneous Hemophilia Injection (SQ-ISHI).

The sponsor-developed EmiPref Survey asked patients to report which treatment they would prefer to continue to receive after having been treated with FVIII (episodic or prophylaxis) and emicizumab. No literature was identified that tested the EmiPref Survey for reliability, validity, or responsiveness in patients with hemophilia A. A MID was not identified in the literature for patients with hemophilia A.

Patients who previously received treatment with FVIII prophylaxis completed the SQ-ISHI. This 13-item measure was designed to assess satisfaction with emicizumab or with FVIII prophylaxis by probing the general concepts of discomfort, worry, and difficulty with injections; confidence with treatment; duration and frequency of treatment; ease of treatment; impact on daily activities; adherence; and overall satisfaction. No literature was identified that tested the SQ-ISHI for reliability, validity, or responsiveness in patients with hemophilia A. A MID was not identified in the literature for patients with hemophilia A.

HAVEN 3

In HAVEN 3, an estimated 75 patients were required to achieve at least 90% power at the 2-sided 0.05 level of significance. The sample size was based on randomization of 2:2:1 (i.e., 30 patients in each emicizumab arm, 15 patients in the control arm). It was assumed that patients from each treatment group would be followed for up to 24 weeks. Simulations of varying follow-up times anticipated that power would be greater than 90% at the 2-sided 0.05 level of significance.

In HAVEN 3, the number of bleeds (e.g., treated bleeds, all bleeds) were analyzed using a negative binomial regression model with efficacy period as an offset to account for the difference in follow-up times. The number of bleeds (less than 9 or greater than or equal to 9) in the last 24 weeks prior to study entry were used as a stratification factor. The ABR ratio was estimated from the model and presented with a 95% CI.

The secondary efficacy outcome for the physical health subscore of the Haem-A-QoL at 24 weeks was analyzed using analysis of variance (ANOVA). The model included the treatment group, baseline score, and treatment by baseline interaction as covariates. The primary analysis consisted of a global approach, with a 3-level categorical effect for treatment (1.5 mg/kg weekly, 3.0 mg/kg every 2 weeks, or no prophylaxis) for the comparisons of arm A versus arm C and arm B versus arm C.

Haem-A-QoL total score, the EQ-5D-5L index utility score, and EQ VAS at 24 weeks were analyzed using the same analysis methodology as for the Haem-A-QoL physical health subscore (i.e., through ANOVA).

In HAVEN 3, hierarchical testing was used to account for multiple testing. The first test evaluated the emicizumab 1.5 mg/kg weekly maintenance dose (group A) versus control. This was followed by the second test, which evaluated the maintenance dose of emicizumab 3.0 mg/kg every 2 weeks (group B) versus control. A global approach incorporating a model statement with a 3-level categorical effect for treatment and an appropriate contrast statement was used to allow both tests to be performed.

Secondary efficacy outcomes in HAVEN 3 were tested according to the following hierarchy:

• A versus C randomized comparison: all bleeds
• B versus C randomized comparison: all bleeds
• A versus C randomized comparison: treated joint bleeds
• B versus C randomized comparison: treated joint bleeds
• A versus C randomized comparison: treated spontaneous bleeds
• B versus C randomized comparison: treated spontaneous bleeds
• D intra-patient: all bleeds
• D intra-patient: treated bleeds
• A versus C randomized comparison: Haem-A-QoL physical health at 24 weeks
• B versus C randomized comparison: Haem-A-QoL physical health at 24 weeks

Other outcomes and exploratory outcomes in HAVEN 3 were analyzed using descriptive statistics.

Preplanned subgroup analysis for treated bleeds relevant to the review included the following:

• Age: less than 18, greater than or equal to 18
• Age: less than 65, greater than or equal to 65
• Number of bleeds during 24 weeks prior to study entry: less than or equal to 9, greater than 9
• Number of target joints: no target joint, any target joint

For patients in arm A, B, or D, the end of the efficacy period was defined as the clinical cut-off date or the date of withdrawal from the initial study period (i.e., treatment phase), whichever was earlier. For patients randomized to arm C (no prophylaxis), the end of the efficacy period was defined as the day before the first emicizumab dose was administered (for patients who switched to receive emicizumab after 24 weeks), the date of withdrawal from the initial study period, or the clinical cut-off date. For patients whose dose was up-titrated, the efficacy period on the initial dose ended the day prior to the first day on the higher dose.

Sensitivity analysis in HAVEN 3 was performed using various methods to define bleeds (e.g., analysis without the 72-hour bleed rule) and different statistical models (e.g., alternative negative binomial modelling approach, ANOVA).

Safety data in HAVEN 3 were reviewed by an independent data-monitoring committee consisting of hemostasis/thrombosis experts and a statistician.

A formal method of imputing missing data was not used in HAVEN 3. Responses to questionnaires were reported through a personal electronic handheld device and onsite tablet designed to prevent respondents from leaving questions unanswered or submitting partial (incomplete) data. Therefore, the data for the primary and secondary bleed related end points and the HRQoL questionnaires are considered complete, according to the sponsor.

HAVEN 4

The sample size planned for the expansion cohort in HAVEN 4 was 40 patients. Although no formal sample-size calculation was reported, the sponsor stated that a sample size of 40 was expected to provide robust point estimates with meaningfully narrow CIs in ABR calculations. There was no formal hypothesis testing planned for HAVEN 4. All analyses were descriptive. A primary efficacy outcome was not specified. Bleeding outcomes were analyzed as described for HAVEN 3.

The efficacy period for each patient started on the day of the first emicizumab dose and ended when the last enrolled patient reached 24 weeks of treatment.

In HAVEN 4, preplanned subgroup analysis for treated bleeds and all bleeds relevant to the review were consistent with those previously described for HAVEN 3, with the addition of the following:

• FVIII inhibitor status (with FVIII inhibitor, without FVIII inhibitors)
• Previous treatment regimen (prior episodic, prior prophylactic)

Sensitivity analysis was not performed in HAVEN 4. The study did not use an independent data-monitoring committee or a formal method of imputing missing data. Data were collected using a personal electronic handheld device and onsite tablet, as described for HAVEN 3.

Analysis Populations

Bleeding

Treated Bleeds

In HAVEN 3, 55.6% of patients treated with 1.5 mg/kg emicizumab weekly experienced 0 treated bleeds over the efficacy period (Table 11). The ABR was 1.5 versus 38.2 per year for groups receiving 1.5 mg/kg emicizumab weekly versus no prophylaxis, respectively (ABR ratio = 0.04; 95% CI, 0.020 to 0.075; P < 0.0001) in favour of 1.5 mg/kg emicizumab weekly. Sixty percent of patients treated with 3.0 mg/kg emicizumab every 2 weeks experienced 0 treated bleeds over the efficacy period. Similarly, the ABR was 1.3 versus 38.2 per year for groups taking 3.0 mg/kg emicizumab every 2 weeks and no prophylaxis, respectively (ABR ratio = 0.03; 95% CI, 0.017 to 0.066; P < 0.0001), in favour of 3.0 mg/kg emicizumab every 2 weeks. None of the patients in the no- prophylaxis group experienced 0 treated bleeds over the efficacy period. Results of the sensitivity analyses were aligned with the primary analysis.

Table 11

Bleeding Outcomes.

Compared with the no-prophylaxis group, both of the emicizumab groups (1.5 mg/kg weekly and 3.0 mg/kg every 2 weeks) experienced consistent reductions in treated bleed rates in the subgroup analyses for bleed rate in during the 24 weeks prior to enrolment (less than 9 bleeds, greater than or equal to 9 bleeds) (Appendix 3, Table 39).

An intra-patient analysis was conducted in the patients on prior prophylaxis with FVIII (non-randomized arm D) who entered HAVEN 3 from a previous NIS. In the NIS, an estimated 39.6% of patients treated with FVIII prophylaxis experienced 0 treated bleeds. In HAVEN 3, an estimated 54.2% of patients in the intra-patient analysis (i.e., patients who subsequently received treatment with 1.5 mg/kg emicizumab weekly) experienced 0 treated bleeds over the efficacy period. An ABR of 4.8 versus 1.5 per year was reported for the patients treated with FVIII prophylaxis versus their treatment with 1.5 mg/kg emicizumab weekly, respectively. An ABR ratio of 0.32 (95% CI, 0.195 to 0.514; P < 0.0001) in favour of 1.5 mg/kg emicizumab weekly was reported for the intra-patient comparison of patients treated with FVIII prophylaxis in the NIS compared to their treatment with 1.5 mg/kg emicizumab weekly in HAVEN 3. The results favoured treatment with 1.5 mg/kg emicizumab weekly (Table 12).

Table 12

Bleeding Outcomes for Intra-Patient Comparison (Data From NIS and HAVEN 3).

In HAVEN 4, 56.1% of patients treated with 6.0 mg/kg emicizumab every 4 weeks experienced 0 treated bleeds; the ABR was 2.4 (95% CI, 1.38 to 4.28) (Table 11). ABRs were reported for the following subgroups: bleed rate during the 24 weeks prior to enrolment (less than 9 bleeds, greater than or equal to 9 bleeds), presence or absence of target joints, previous treatment regimen (episodic, prophylactic), and FVIII inhibitor status (Appendix 3, Table 39).

Treated Joint Bleeds

In HAVEN 3, 58.3% of patients treated with 1.5 mg/kg emicizumab weekly experienced 0 treated joint bleeds over the efficacy period (Table 11). The ABR ratio between the groups taking 1.5 mg/kg emicizumab weekly (ABR = 1.1) and no prophylaxis (ABR = 26.5) was 0.04 (95% CI, 0.019 to 0.085; P < 0.0001) in favour of 1.5 mg/kg emicizumab weekly. A total of 74.3% of patients treated with 3.0 mg/kg emicizumab every 2 weeks experienced 0 treated joint bleeds over the efficacy period. The ABR ratio between 3.0 mg/kg every 2 weeks (ABR = 0.9) and no prophylaxis (ABR = 26.5) was 0.03 (95% CI, 0.015 to 0.070; P < 0.0001) in favour of 3.0 mg/kg emicizumab every 2 weeks. None of the patients in the no-prophylaxis group experienced 0 treated joint bleeds. Results of the sensitivity analyses were aligned with the primary analysis.

In HAVEN 4, 70.7% of patients treated with 6.0 mg/kg emicizumab every 4 weeks experienced 0 treated joint bleeds; the ABR was 1.7 (95% CI, 0.82 to 3.68) over the efficacy period (Table 11).

Productivity

Days Away From School

In HAVEN 3, students were away from school in the 4 weeks prior to baseline for an average of ▬▬▬▬▬

During the 4 weeks prior to week 25, patients were away from school ▬▬▬▬▬ (Table 13).

Table 13

Productivity.

In HAVEN 4, students were away from school in the 4 weeks prior to baseline for an average of ▬ During the 4 weeks prior to week 25, patients were away from school ▬▬▬▬▬ (Table 13).

Health-Related Quality of Life

Haemophilia Quality of Life Questionnaire for Adults

In HAVEN 3, the difference in adjusted mean Haem-A-QoL physical health subscore at week 25 between 1.5 mg/kg emicizumab weekly and no prophylaxis was 12.51 (95% CI, – 1.96 to 26.98; P = 0.891) (Table 14). The difference in adjusted mean Haem-A-QoL physical health subscore at week 25 between 3.0 mg/kg emicizumab every 2 weeks and no prophylaxis was 15.97 (95% CI, −1.16 to 30.78; P = 0.0349).

Table 14

Health-Related Quality of Life.

In HAVEN 4, the unadjusted mean change from baseline for the Haem-A-QoL physical health subscore at week 25 was −15.14 (95% CI, −22.44 to −7.83) in the 6.0 mg/kg emicizumab every 4 weeks group (Table 14).

In HAVEN 3, the difference in adjusted mean Haem-A-QoL total score at week 25 between 1.5 mg/kg emicizumab weekly and no prophylaxis was 5.91 (95% CI, −1.72 to 13.55; P = 0.1269) (Table 14). The difference in adjusted mean Haem-A-QoL total score at week 25 between 3.0 mg/kg emicizumab every 2 weeks and no prophylaxis was 8.56 (95% CI, 0.77 to 16.35; P = 0.0317).

In HAVEN 4, the unadjusted mean change from baseline for the Haem-A-QoL total score at week 25 was −13.62 (95% CI, −18.36 to −8.88) in the group receiving 6.0 mg/kg emicizumab every 4 weeks (Table 14).

EuroQol 5-Dimensions 5-Levels Questionnaire

In HAVEN 3, the difference in adjusted mean EQ-5D-5L index utility score at week 25 between 1.5 mg/kg emicizumab weekly and no prophylaxis was −0.13 (95% CI, −0.22 to – 0.04; P = 0.0060) (Table 14). The difference in adjusted mean EQ-5D-5L index utility score at week 25 between 3.0 mg/kg emicizumab every 2 weeks and no prophylaxis was −0.13 (95% CI, −0.23 to −0.04; P = 0.0059).

In HAVEN 4, the unadjusted mean change from baseline for the EQ-5D-5L index utility score at week 25 was 0.06 (95% CI, 0.03 to 0.10) in the group receiving 6.0 mg/kg emicizumab every 4 weeks (Table 14).

In HAVEN 3, the difference in adjusted mean EQ VAS at week 25 between 1.5 mg/kg emicizumab weekly and no prophylaxis was −4.04 (95% CI, −12.43 to 4.35; P = 0.3402) (Table 14). The difference in adjusted mean EQ VAS at week 25 between 3.0 mg/kg emicizumab every 2 weeks and no prophylaxis was – 9.15 (95% CI, −17.74 to −0.55; P = 0.0373).

In HAVEN 4, the unadjusted mean change from baseline for the EQ VAS at week 25 was 5.53 (95% CI, 1.15 to 9.90) in the group receiving 6.0 mg/kg every 4 weeks (Table 14).

Hospitalization

In HAVEN 3, the mean number of days hospitalized within the 24-week period was 0.17 (SD = 1.0) in the arm receiving 1.5 mg/kg emicizumab weekly, 0.43 (SD = 1.77) in the arm receiving 3.0 mg/kg emicizumab every 2 weeks, and 0.11 (SD = 0.47) in the no-prophylaxis arm (Table 15).

Table 15

Hospitalization.

In HAVEN 4, the mean number of days hospitalized within the 24-week period was 0 in the arm receiving 6.0 mg/kg emicizumab every 4 weeks (Table 15).

Pain

In HAVEN 3 and HAVEN 4, the proportion of patients who experienced pain or discomfort at baseline and week 25 according to the EQ-5D-5L is summarized in Table 16.

Table 16

Pain or Discomfort Based on EQ-5D-5L.

Patient Satisfaction

In HAVEN 3, based on results of the EmiPref, 96.4%, 81.0%, and 97.8% of patients in the arms taking 1.5 mg/kg emicizumab weekly, 3.0 mg/kg emicizumab every 2 weeks, and previous FVIII prophylaxis (1.5 mg/kg emicizumab weekly), respectively, preferred treatment with emicizumab compared to their previous hemophilia treatment (Table 17).

Table 17

Patient Satisfaction Outcomes.

In HAVEN 4, all patients (100%) treated with 6mg/kg emicizumab every 4 weeks preferred treatment with emicizumab compared to their previous hemophilia treatment, based on findings from the EmiPref (Table 17).

In HAVEN 3, 73.1% of patients in the previous-FVIII-prophylaxis arm (1.5 mg/kg emicizumab weekly) were “much more satisfied” with their current hemophilia treatment (1.5 mg/kg emicizumab weekly) compared to their pre-study prophylactic treatment (FVIII prophylaxis), as assessed by item 16 of the SQ-ISHI (Table 17).

Adverse Events

In HAVEN 3, AEs occurred in 94.4% of patients in the arm receiving 1.5 mg/kg emicizumab weekly, 85.7% of patients in the arm receiving 3.0 mg/kg emicizumab every 2 weeks, 50.0% of patients in the no-prophylaxis arm, and 87.3% of patients in the previous-FVIII-prophylaxis arm (1.5 mg/kg emicizumab weekly). In HAVEN 4, 73.2% of patients treated with 6.0 mg/kg emicizumab every 4 weeks experienced an AE. In HAVEN 3 and HAVEN 4, the most common AE was injection-site reactions, as described in the Notable Harms section.

Serious Adverse Events

In HAVEN 3, SAEs occurred in 2.8% of patients receiving 1.5 mg/kg emicizumab weekly, 8.6% of those receiving 3.0 mg/kg emicizumab every 2 weeks, 0% of those on no prophylaxis, and 12.7% of those in the previous-FVIII-prophylaxis arm (1.5 mg/kg emicizumab weekly). In HAVEN 4, 2.4% of patients treated with 6.0 mg/kg emicizumab every 4 weeks experienced an SAE.

Withdrawals Due to Adverse Events

One patient (2.9%) treated with 3.0 mg/kg emicizumab every 2 weeks in HAVEN 3 stopped treatment due to AEs.

Mortality

No deaths were reported during HAVEN 3 or HAVEN 4.

Notable Harms

Notable harms identified in the protocol for this review included the following: thrombotic events, injection-site reactions, hypersensitivity reactions, inhibitor development, and blood-borne infections.

The only notable harms reported in HAVEN 3 and HAVEN 4 were injection-site reactions. Injection-site reactions were local AEs that occurred within 24 hours after the study drug was administered and were judged to be related to the study drug injection, based on assessment by the investigator. In HAVEN 3, injection-site reactions occurred in 25.0% of patients receiving 1.5 mg/kg emicizumab weekly, 20.0% of patients receiving 3.0 mg/kg emicizumab every 2 weeks, 12.5% of patients in the no-prophylaxis arm, and 31.7% of patients in the previous-FVIII-prophylaxis arm (1.5 mg/kg emicizumab weekly). In HAVEN 4, 22.0% of patients treated with 6.0 mg/kg emicizumab every 4 weeks experienced an injection-site reaction.

Throughout HAVEN 3 and HAVEN 4, no instances of de novo inhibitor development were detected in patients who tested negative for inhibitors (titre less than 0.6 CBU/mL) at baseline.

Internal Validity

Aspects of the study design in HAVEN 3 and HAVEN 4 may have affected the internal validity of the studies. Both studies were open-label; HAVEN 4 was a single-arm study. The open-label design may have biased the subjective outcome results in favour of emicizumab.

Both studies were 24 weeks in duration, an acceptable length of time to determine bleeding-related efficacy and safety, according to clinical experts.

HAVEN 4, a single-arm study, was limited by its non-controlled design. HAVEN 4 included patients with inhibitors (N = 5, 12.2%), which could have introduced heterogeneity in the patient population, although the impact of this on efficacy and harms outcomes is unknown. While some subgroup analysis was available for these patients, it is unclear how their inclusion may have influenced the results.

The baseline characteristics were generally balanced between the 3 randomized treatment arms in HAVEN 3, indicating adequate randomization. Minor imbalances were observed for the number of bleeds in the 24 weeks prior to study entry. Of note, patients in the no-prophylaxis (control) arm had a mean number of bleeds of ▬ conversely, patients in the emicizumab arms had means of ▬ and ▬ for 1.5 mg/kg emicizumab weekly and 3.0 mg/kg emicizumab every 2 weeks, respectively. A greater number of bleeds in the arm with no prophylaxis at baseline may have introduced bias in the efficacy results in favour of emicizumab.

Definitions for bleeding outcomes were based on standardized definitions, adapted from criteria defined by the FVIII, Factor IX and Rare Coagulation Disorders Subcommittee of the ISTH. The definitions were used consistently across studies (HAVEN 3, HAVEN 4, HOHOEMI). The clinical experts consulted by CADTH for this review indicated that the bleeding outcomes assessed in HAVEN 3 and HAVEN 4 were consistent with those used in clinical practice. Although bleeding outcomes are not associated with formal MIDs, the consulted clinicians were able to assess the clinical relevance of the bleeding outcomes. Several relevant sensitivity analyses performed for the assessment of bleeding outcomes were aligned with the primary analysis, thereby indicating robust findings.

All bleeding outcomes in HAVEN 3 were analyzed using appropriate statistical methods. Hierarchical testing was used to account for multiplicity. While it is important to note that treated target bleeds (an outcome identified as “of interest” to clinicians) were not included in the testing hierarchy, the large magnitude of difference in efficacy between the 2 doses of emicizumab and the control arm were observed consistently for the other bleeding outcomes. Therefore, it is unlikely that the effect on this outcome was due to chance.

One outcome related to HRQoL, the Haem-A-QoL physical health subscore, was analyzed based on a statistical hierarchical approach. Therefore, multiplicity of testing with potential inflated type I error likely does not explain the observed difference. Yet, the statistically significant difference was observed only in the arm receiving 3.0 mg/kg every 2 weeks but not in the arm receiving 1.5 mg/kg weekly, even though the magnitude of reduction in terms of ABR for different types of bleeding was similar between the 2 different treatment regimens. Thus, the effect on improved quality of life remains inconclusive. Of note, the EQ-5D-5L index utility score — a general, non-disease–specific measure — showed a statistically significant difference (P = 0.006) in both treatment regimens in comparison to no-prophylaxis control. However, this outcome measure was out of the statistical testing hierarchy.

In HAVEN 4, analyses of all outcomes were descriptive, and no formal hypothesis testing was performed. HAVEN 4 was the only pivotal study that assessed the maintenance dose of 6.0 mg/kg emicizumab every 4 weeks in patients with and without inhibitors, and its descriptive nature affects the interpretation of results. Input from the Health Canada Reviewers Report indicates sufficient pharmacokinetic and clinical data to support the efficacy of this maintenance dose in adult and adolescent patients.²⁹

Relevant subgroup analyses performed in HAVEN 3 and/or HAVEN 4 included age, number of bleeds, number of target joints, FVIII inhibitor status, and previous treatment regimen. However, subgroup analysis was limited by the small number of patients in each subgroup, which in turn limited the interpretation of the results.

Outcomes related to productivity, hospitalization, pain, and patient preference were reported descriptively in both studies without performing formal statistical testing; this prevents the interpretation of the results for these outcomes.

Missing data were not accounted for statistically in either study. However, the impact of missing data is expected to be minimal in both, given that the amount of missing data were minimal. Discontinuations in HAVEN 3 and HAVEN 4 were low and balanced between treatment arms (0 or 1 patient per treatment arm) and were not expected to affect the validity of the results.

External Validity

The majority of patients with severe hemophilia A enrolled in HAVEN 3 (groups A, B, and C) were previously treated with episodic FVIII. This does not reflect Canadian clinical practice, in which the standard of care for patients with severe hemophilia A is treatment with FVIII prophylaxis. Furthermore, patients in HAVEN 3 and HAVEN 4 who previously received treatment with episodic FVIII were required to have 5 or more bleeds in the 24 weeks prior to study entry. This is also inconsistent with Canadian clinical practice, as it is unlikely for Canadian patients with 5 or more bleeds over a period of 24 weeks to be treated with episodic FVIII.

The impact of including a study population that has greater uncontrolled bleeding may exaggerate the efficacy of emicizumab compared to what would be seen in the Canadian clinical setting. Thus, the magnitude of the treatment effect in patients with better control (consistent with the Canadian clinical population) compared with those included in the trials is unknown.

There is no direct comparative evidence to support the efficacy of emicizumab compared to FVIII prophylaxis in patients with severe hemophilia A. There is limited evidence from an intra-patient analysis (group D in HAVEN 3) that supports the efficacy of emicizumab in patients previously treated with FVIII prophylaxis.

All patients included in HAVEN 3 and HAVEN 4 were male. While most patients in Canada with severe hemophilia A are male, there are a few female patients, and they were not represented in the studies.

The results of HAVEN 3 and HAVEN 4 may have limited generalizability to the Canadian patient population, given that the study population was restricted to patients (without inhibitors) with severe hemophilia A. According to the clinical experts consulted for this review, there is a subset of patients with mild or moderate hemophilia who require prophylaxis. The design of HAVEN 3 and HAVEN 4 did not include them; thus, the magnitude of the treatment effect in patients with mild and moderate hemophilia A is unclear.

Eligibility criteria in both trials excluded patients with thromboembolic disease, at high risk for microangiopathy, and with certain autoimmune diseases, thereby decreasing the trials’ generalizability to the Canadian clinical population. An assessment of baseline characteristics (e.g., gender, age) with the exception of previous bleeds showed reasonable consistency with the Canadian clinical population.

Patients in the randomized arms of HAVEN 3 and all patients in HAVEN 4 were able to up-titrate their emicizumab dose to 3.0 mg/kg if they met the protocol-defined criteria of suboptimal response. In HAVEN 3, 5 patients had their emicizumab dose up-titrated (one in the 1.5 mg/kg emicizumab weekly arm and 4 in the previous-FVIII-prophylaxis arm [1.5 mg/kg emicizumab weekly maintenance]). These patients were sufficiently accounted for in the efficacy period on the initial dose ended the day before the first day on the higher dose. The criterion allowing patients to up-titrate their emicizumab dose after suboptimal response was consistent with methods used in clinical practice.

HAVEN 3 and HAVEN 4 excluded patients under the age of 12; however, given the mechanism of action and body of evidence of emicizumab in children with inhibitors and the absence of direct data on children without inhibitors, the exclusion would not be expected to affect efficacy or safety, according to clinical experts consulted for this review.

Bleeding outcomes assessed in the studies (e.g., treated bleeds, all bleeds) were consistent with those assessed in Canadian clinics. The patient-reported outcomes assessed in the studies (e.g., school and work attendance, pain, HRQoL) are also assessed in clinic during follow-up visits; however, this is done without the use of scales or questionnaires, which differs from the methods used in the studies. Long-term outcomes of clinical interest, such as resolution of target joints, were not assessed in HAVEN 3 or HAVEN 4. However, the sponsor provided additional information pertaining to long-term treatment with emicizumab in its submission from a post hoc pooled analysis across all 4 HAVEN trials (HAVEN 1, HAVEN 2,^a HAVEN 3, HAVEN 4). These data suggest that long-term treatment with emicizumab consistently resolved more than 99% of target joints. However, this conclusion is speculative due to the paucity of methodological detail available to assess the validity of these results.

Adherence to treatment was high in both studies. It was likely inflated compared to what would be seen in the real world, given that the patients were part of a clinical trial. Given the lack of comparative data between FVIII prophylaxis and SC emicizumab, the impact of the SC formulation on adherence in the real-world setting remains unknown.

Surgeries, procedures, and breakthrough bleeds (i.e., bleeding episodes) were treated using episodic treatment with FVIII at the lowest dose expected to achieve hemostasis while continuing the assigned treatment. The use of episodic FVIII to treat breakthrough bleeds is consistent with Canadian clinical practice.

In HAVEN 3 and HAVEN 4, patients received in-clinic training on administering SC emicizumab prior to self-administering in the home setting. This form of training is consistent with training provided in the clinical setting.

Systematic Literature Review

A search strategy was developed based on the Population, Intervention, Comparator, Outcome, Study type framework presented in Table 19 to identify relevant published data investigating the efficacy and safety of pharmacological interventions for hemophilia A. The SLR was performed in December 2016 and updated for the non-inhibitor population in May 2018.

Table 19

Original Study Selection Criteria and Methods for the Sponsor-Submitted ITC (2016).

The original search (in 2016) identified 175 relevant publications. An additional 545 records were identified in the updated search conducted in 2018. Trials with mixed populations (i.e., inhibitor and non-inhibitor patients) were eligible for inclusion in the updated SLR, provided they reported stratified results; however, none of the studies identified in mixed populations reported stratified results. Following screening, a total of 94 studies in patients with hemophilia A without FVIII inhibitors were identified for further feasibility assessment.

NMA Feasibility Assessment

Following the feasibility assessment, single-arm studies, non-randomized studies, observational studies, and post-marketing surveillance studies were excluded. Studies in pediatric populations were excluded to ensure the target population of the trials included in the NMA would match the population in the HAVEN 3 trial as closely as possible. Trials of FVIII products not currently approved were excluded. Only licensed doses of these treatments were of interest in the NMA. Finally, 5 studies on patients with hemophilia A without inhibitors were included in the NMA. The 5 studies included HAVEN3,²⁷ A-LONG, LEOPOLD2,⁴⁴ SPINART,⁴⁵ and Valentino et al. (2012).⁴³ Valentino et al. 2012 was included only for 1 sensitivity analysis.⁷

The sponsor-submitted NMA was only interested in comparisons of prophylaxis versus on-demand treatment/no prophylaxis. Therefore, arm D from HAVEN 3 did not qualify for inclusion and was not included in the NMA.¹²

Summary of Included Studies

Five studies were included in the NMA.

One trial was included for emicizumab: HAVEN 3²⁷ enrolled adults and adolescents (aged greater than or equal to 12 years) with severe hemophilia A without FVIII inhibitors. The 4 included trials on rFVIII comparators were A-LONG (Antihemophilic Factor Recombinant BDD, Fc Fusion Protein, Elocta/Eloctate),⁴⁸ LEOPOLD2 (Antihemophilic Factor Recombinant, Kovaltry),⁴⁴ SPINART (Antihemophilic Factor [Recombinant] – formulated with sucrose, Kogenate),⁴⁵ and Valentino et al. (2012) (Antihemophilic Factor [Recombinant], Advate).⁴³

The characteristics of the 5 included studies are presented in Table 23. The sample size varied in the included studies (n = 89, n = 165, n = 80, n = 84, and n = 135 for HAVEN 3, A-LONG, LEOPOLD 2, SPINART, and Valentino et al. [2012], respectively). Median age was similar in the A-LONG, LEOPOLD 2, SPINART and Valentino et al. (2012) studies (30, 28.5, 29, and 26 years, respectively), but higher in the HAVEN 3 study, in which the median age ranged from 36.5 years to 41.0 years across the 3 included arms. In all studies, the majority of patients were diagnosed as having severe hemophilia A: 100% of patients in HAVEN 3, A-LONG, and LEOPOLD; 86% in the Valentino et al. (2012) study; and 93% and 100% in SPINART (Table 23). The characteristics of 4 FVIII products included in the NMA are presented in Table 24. The efficacy outcomes assessed were total treated bleeds for emicizumab prophylactic and FVIII episodic treatment (i.e., no FVIII prophylaxis), respectively. The definitions of bleed outcomes in each trial are presented in Table 25.

Table 23

Characteristics for the Trials Included in the NMA.

Table 24

Characteristics of Factor VIII Products Included in the NMA.

Table 25

Definitions of Bleeds Outcomes in Each Study.

Assessment of Risk of Bias of Included Trials

The risk of bias in the 4 included RCTs (i.e., LEOPOLD 2, A-LONG, SPINART, and HAVEN 3) was assessed according to the NICE critical appraisal checklist (Table 26). Valentino et al. (2012) was a single-arm study with within-trial comparison; therefore, it was assessed according to the National Institutes of Health Quality Assessment Tool for Case Series Studies (Table 27 and Table 24).

Table 26

Results of Risk of Bias Assessment of RCTs.

Table 27

Results of Risk of Bias Assessment in the Single-Arm Study (Valentino et al. [2012], Part 1).

It was reported that overall, according to the NICE criteria, all RCTs carried at least some risk of bias, most notably the risk introduced by lack of blinding. HAVEN 3, LEOPOLD 2, and SPINART showed low risk of bias for most of the categories assessed. The National Institutes of Health Quality Assessment Tool rated the quality of the Valentino et al. (2012) study as fair to good.¹²

Table 28

Results of Risk of Bias Assessment in the Single-Arm Study (Valentino et al. [2012], Part 2).

Results

The evidence networks of studies included for base-case analysis and sensitivity analysis are presented in Figure 3 and Figure 4, respectively.

Figure 3

Base-Case Analysis Network of Studies Included. ITC = indirect treatment comparison; OD = on demand; prophy = prophylaxis; Q2W = every 2 weeks; QW = per week. Note: Edge width is proportional to the number of inputs for each comparison.

Figure 4

Network of Studies Included in Sensitivity Analyses 2 and 3. ITC = indirect treatment comparison; OD = on demand; prophy = prophylaxis; Q2W = every 2 weeks; QW = per week. Note: Sensitivity analysis where long- and short-acting FVIII treatments were separated (more...)

In addition to arm A (1.5 mg/kg weekly) and arm B (3.0 mg/kg every 2 weeks) of the HAVEN 3 trial, 4 studies (A-LONG for Antihemophilic Factor Recombinant BDD, Fc Fusion Protein [Elocta/Eloctate], LEOPOLD 2 for Antihemophilic Factor Recombinant [Kovaltry], SPINART for Antihemophilic Factor (Recombinant) – formulated with sucrose [Kogenate], and Valentino et al. (2012) for Antihemophilic Factor (Recombinant) [Advate]) comparing FVIII prophylaxis versus on-demand treatment (i.e., no prophylaxis) informed this network. All 4 FVIII products were combined as 1 group (i.e., 1 FVIII node in the networks). Of these, 3 trials ^43–45 compared short-acting FVIII treatments versus on-demand treatment (i.e., no prophylaxis). The A-LONG study⁴⁸ compared long-acting FVIII treatment to on-demand treatment (i.e., no prophylaxis). (See Table 24). The single-arm study (Valentino et al. [2012]) was only included in sensitivity analysis 3.

The number of bleeds and the total exposure (patient-year) in each treatment arm reported in the 5 included studies are presented in Table 29.

Table 29

Number of Bleeds and Total Exposure Reported in Included Studies.

Efficacy Outcomes NMA Results

Base-case NMA results

The NMA base-case analysis results are presented in Table 30. In terms of total treated bleeds, emicizumab prophylaxis (1.5 mg/kg weekly) showed a reduction compared with FVIII prophylaxis (rate ratio = 0.36; 95% credible interval [CrI], 0.13 to 0.95). Emicizumab prophylaxis (3.0 mg/kg every 2 weeks) also showed a reduction compared with FVIII prophylaxis (rate ratio = 0.31; 95% CrI, 0.11 to 0.84). These findings were in line with those reported in HAVEN 3.

Table 30

Base-Case NMA Results for Total Treated Bleeds.

Sensitivity analyses NMA results

The results of the sensitivity analyses are presented in Table 31.

Table 31

Sensitivity Analysis NMA Results for Total Treated Bleeds.

In sensitivity analysis 1, as expected, the CrIs for the fixed-effects model were narrower than for the base-case RE model. The results were similar to those of the base-case analysis for both emicizumab prophylaxis (1.5 mg/kg weekly) versus FVIII prophylaxis (rate ratio = 0.46; 95% CrI, 0.32 to 0.63) and emicizumab prophylaxis (3.0 mg/kg every 2 weeks) versus FVIII prophylaxis (rate ratio = 0.39; 95% Crl, 0.27 to 0.56) (Table 31).

Sensitivity analysis 2, which separated FVIII treatments into short- or long-acting FVIII, suggested that emicizumab prophylaxis (1.5 mg/kg weekly and 3.0 mg/kg every 2 weeks) was associated with a reduction in total treated bleeds compared with long-acting FVIII prophylaxis, but not compared with short-acting FVIII prophylaxis (Table 31).

Sensitivity analysis 3 also included the single-arm study (Valentino et al. [2012] on rFVIII [Advate]) in addition to the base-case studies. The results suggested that emicizumab prophylaxis (3.0 mg/kg every 2 weeks) was associated with a reduction in treated bleeds compared with FVIII prophylaxis (rate ratio = 0.35; 95%Crl, 0.13 to 0.96). Whether there was a difference in the reduction of treated bleeds between emicizumab prophylaxis (1.5 mg/kg weekly) and FVIII prophylaxis remains uncertain (rate ratio = 0.40; 95% Crl, 0.15 to 1.10) (Table 31).

The results from sensitivity analysis 4 suggested ▬▬▬▬▬ (Table 31).

Methods

The HOHOEMI study was a phase III, multi-centre, open-label, non-randomized clinical trial to evaluate the efficacy, safety, and pharmacokinetics of emicizumab in Japanese pediatric patients aged less than 12 years with hemophilia A without FVIII inhibitors. The study was conducted at 4 sites in Japan from September 2017 to July 2019. All patients received emicizumab administered SC at a dose of 3.0 mg/kg every 2 weeks or 6.0 mg/kg every 4 weeks following 4 loading doses of 3.0 mg/kg per week. Patients received emicizumab for at least 24 weeks or until study withdrawal. Patients were enrolled sequentially in each cohort and were not randomized to the treatment arms. For the purpose of this review, only data from the 3.0 mg/kg arm will be presented. The 6.0 mg/kg dose is not relevant to the Canadian population, given that it is beyond the dosing recommended by Health Canada for the pediatric population.

Populations

To be eligible for the HOHOEMI study, patients needed to be less than 12 years of age, weigh more than 3 kg, and have severe congenital hemophilia A without FVIII inhibitors (i.e., an endogenous FVIII level of less than 1%). Patients had to test negative for inhibitors (less than 0.6 BU/mL) within 8 weeks prior to enrolment. The main exclusion criteria were having a bleeding disorder other than hemophilia A, previous or current thromboembolic disease, or other conditions that may increase risk of bleeding or thrombosis, and being at high risk of thrombotic microangiopathy.

A summary of the baseline characteristics of patients in the cohort receiving emicizumab every 2 weeks is shown in Table 32. With regard to baseline demographics, all patients were Japanese male children. The median age was 6.6 years (range = 1.5 years to 10.7 years), the median weight was 19.50 (range = 10.9 kg to 35.6 kg), and the median height was ▬▬▬▬▬ As for the medical history of hemophilia A, all patients received at least 1 treatment for hemophilia A, and all patients but 1 were previously on FVIII prophylaxis treatment; the patient who had never received prophylaxis was aged 4 months in the 6.0 mg/kg cohort. The prior FVIII prophylaxis was administered approximately 2 to 3 times per week in most patients. Four patients also received short-acting FVIII episodic treatment, and 2 received long-acting FVIII. One patient had target joints at baseline. One patient had been previously treated with immune tolerance induction therapy.

Table 32

Summary of Baseline Characteristics.

Interventions

Patients in the cohort receiving emicizumab every 2 weeks were administered the drug SC, with loading doses of 3.0 mg/kg every week for the first 4 weeks followed by maintenance doses of 3.0 mg/kg every 2 weeks. The minimum follow-up period was 24 weeks. Patients with insufficient bleeding control after 12 weeks were eligible to receive up-titration of the maintenance dose to 3.0 mg/kg weekly. Patients were initially treated at the study site, and patients aged 7 years or older (or their caregivers) were able to administer treatment at home if the investigator deemed it feasible. Patients showing sustained clinical benefits with emicizumab during the first 24 weeks were eligible to continue receiving emicizumab prophylaxis. Concomitant medications were permitted, but authors did not list which medications were permitted. Breakthrough bleeds were managed with episodic FVIII treatment.

Outcomes

The primary efficacy outcome was bleeding frequency, expressed as an ABR. The efficacy end points included various definitions of bleeds — namely, treated bleeds, treated spontaneous bleeds, treated joint bleeds, treated target joint bleeds, and all bleeds consistent with the definitions of the FVIII, Factor IX and Rare Coagulation Disorders Subcommittee of the ISTH used in the HAVEN 3 and HAVEN 4 studies. Another evaluated efficacy outcome was patient preference. Safety outcomes were reported in the form of AEs.

Statistical Analysis

The efficacy analysis was performed using the efficacy analysis set, which included all enrolled patients who received at least 1 dose of emicizumab. The efficacy analysis evaluated bleeding rates and was conducted once all patients had reached week 24 or withdrawn from the study. ABRs were estimated using negative binomial regression with the treatment period as an offset to account for varying follow-up periods. Pre-treatment ABRs were calculated using documented bleed information and standardized for the different age groups over different durations: 12 weeks ▬ for patients less than 2 years of age and 24 weeks ▬ for patients 2 years of age and older. ABRs in the post-treatment period were calculated using duration in days. A model with different data elements was used to calculate the ABR ratio between pre- and post-treatment ABRs. Calculated ABRs were computed by dividing the number of bleeding events by the evaluation period in days and multiplying this rate by 365.25.

Patient-reported outcomes were estimated at week 17 using the patient-reported outcomes analysis set, which included all patients who received at least 1 dose of emicizumab and completed the EmiPref Survey. Summary statistics were computed for the patient-reported outcomes.

The safety analysis was performed using the safety analysis set, which included all patients who received emicizumab at least once. Safety was assessed by summarizing the number of AEs and patients with AEs by system organ class, preferred term, and severity.

Patient Disposition

Patient disposition in the HOHOEMI study is summarized in Table 33. In this study, 6 pediatric patients with hemophilia A without FVIII inhibitors were enrolled sequentially in the cohort receiving emicizumab every 2 weeks. All patients were treated with emicizumab and completed the study. Patients continued to receive emicizumab until it was available for commercial use. No patients required up-titration. No patients were excluded from the analysis.

Table 33

Patient Disposition.

Exposure to Study Treatments

Exposure to the study treatment is summarized in Table 34. All patients in the cohort receiving emicizumab every 2 weeks received 3.0 mg/kg emicizumab every 2 weeks and continued their allocated dosing regimen. ▬▬▬▬▬ No patients had a dose up-titration during the study period. After study termination, all patients continued to receive emicizumab prophylaxis until it was available for commercial use. The median treatment duration for the cohort receiving emicizumab every 2 weeks was ▬▬▬▬▬

Table 34

Exposure to Study Treatment.

Efficacy

▬▬▬▬▬ (Table 35). ▬▬▬▬▬

Table 35

Bleeding Outcomes.

Pre- and post-treatment intra-patient bleeding outcomes are summarized in Table 36. ▬▬▬▬▬

Table 36

Bleeding Outcomes (Pre- and Post-Treatment).

At week 17, all caregivers (n = 13) completed the EmiPref. All caregivers preferred emicizumab prophylaxis over previous hemophilia treatments. The main reasons for this preference were a lower treatment frequency (38%) and fewer effects on other activities, such as work, school, and social interactions (23.1%) (data not shown).

Harms

The AEs are summarized in Table 37. All patients in the cohort receiving emicizumab every 2 weeks experienced at least one AE. A total of 112 AEs were reported. Of these, 1 was listed as having grade 3 severity, while the others were grade 1 or 2. The most common AEs reported were contusion (100%), ligament sprain (50%), scratch (50%), and arthralgia (50%). One patient experienced local injection-site reaction, which resolved without treatment. One patient experienced an SAE (post-traumatic pain). No AEs led to treatment discontinuation, and no deaths were reported.

Table 37

Summary of Harms.

Critical Appraisal