Severe muscle bleeds in children and young adults with hemophilia A on emicizumab prophylaxis: Real-world retrospective multi-institutional cohort.

Emicizumab is a humanized bispecific IgG4 antibody that binds factor IXa and factor X mimicking the co-factor function of activated factor VIII (FVIII) in the intrinsic tenase complex to promote thrombin generation.1 Phase 3 clinical trials evaluating the hemostatic efficacy of emicizumab prophylaxis demonstrated significant reductions in annualized bleeding rates (ABR) among pediatric and adult persons with hemophilia A (PwHA) with and without inhibitors.2-6 Although infrequent, breakthrough bleeds can occur and require treatment with FVIII concentrates in PwHA without inhibitors or a bypassing agent in PwHA with inhibitors (i.e., recombinant activated factor VII [rFVIIa] or activated prothrombin complex concentrate [aPCC]). The majority of bleeding events in the HAVEN studies resolved with <24 h of 50–100 units/kg aPCC (HAVEN 1), 90–270 μg/kg rFVIIa (HAVEN 1), or <100 units/kg FVIII (HAVEN 3).2 Thus, most bleeds are anticipated to resolve with 1–2 doses of factor treatment. However, severe bleeding events requiring intensive factor replacement such as muscle bleeds are destined to occur among the general population of PwHA, particularly among active and younger individuals. The main objective of this work is to describe a series of real-world muscle hematomas and their management among pediatric and young adult PwHA on emicizumab. This is a multi-institutional retrospective cohort study of PwHA with and without inhibitors between the ages of 0 and 27 years on emicizumab prophylaxis. Data extraction from electronic medical records was performed between August 1, 2022 to April 4, 2023. PwHA on emicizumab post-U.S. FDA approval in November 2017 with a documented severe muscle bleed (defined as requiring ≥24 h of factor treatment) was included in this report. Data extraction included de-identified demographic information, inhibitor history, emicizumab duration and regimen, muscle bleed details and management, and bleeding/treatment-related serious adverse events. Treatment success was based on clinical resolution and/or radiographic imaging as determined by the treating provider. Details of individual bleed treatment regimens are provided in Table S2. In PwHA without inhibitors, three bleeds received continuous infusion (CI) of FVIII. For these cases, chromogenic FVIII activities using bovine reagents were performed locally every 4–24 h with 4–6 h lab turnaround time. The CI rate was adjusted to maintain chromogenic FVIII activity between 80%–120%. Due to a black box warning on the use of standard dosing aPCC on emicizumab in PwHA and inhibitors, individuals in the inhibitor group received rFVIIa as first-line treatment. Data were collected from the following four medical institutions: Emory University Pediatric Hemophilia Treatment Center and Aflac Cancer and Blood Disorders Center at Children's Healthcare of Atlanta, Vanderbilt University Medical Center, University of Michigan Medical Center, and Children's Hospital Los Angeles. Study approval was obtained from each center's Institutional Review Board and received exempt status. Due to the small sample size and nature of the study, data analysis is primarily presented as descriptive statistics. There was an estimated total of 314 PwHA with and without inhibitors on emicizumab at the 4 institutions (range 41–123 PwHA on emicizumab per institution). We identified 15 PwHA with and without inhibitors who experienced at least one severe muscle bleed (Table 1). All participants were male with severe PwHA and a spectrum of F8 gene variants (Table S1). The cohort predominantly consisted of adolescent males (12–18 years old) who comprised 60% of the cohort. Eleven participants did not have inhibitors while 4 had active inhibitors. The 4 PwHA with inhibitors had peak historical inhibitor titers between 11–256 Bethesda units per milliliter (BU/mL). Three (75%) had prior attempts of immune tolerance induction (ITI) and achieved partial tolerance prior to the transition to emicizumab. One of the PwHA and an inhibitor was on concurrent ITI with recombinant FVIII thrice weekly at 100 units/kg while on emicizumab. For emicizumab prophylaxis, most participants were on every 2-week dosing followed by once weekly dosing as the second most common dosing frequency. There were 41.5 patient years of exposure to emicizumab in the cohort. All patients self-reported adherence to their emicizumab regimen. However, emicizumab levels were not investigated as there is no commercially available assay in the United States. Among the 15 PwHA, there were 19 muscle bleeds. The majority of PwHA had 1 muscle bleed (n = 13), 1 PwHA with an inhibitor had 2 muscle bleeds, and 1 PwHA without an inhibitor had 4 muscle bleeds. There were no muscle bleed recurrences in PwHA with or without inhibitors. One PwHA without an inhibitor had a bleed treatment failure requiring a second course of factor treatment 30 days later (Table S2, patient 9). Fourteen bleeding events occurred among PwHA without inhibitors and 5 occurred among PwHA with inhibitors (Table 1). In the group of 11 PwHA without inhibitors, the median age at the time of bleeding event was 14 years (range 10–21 years). PwHA without inhibitors were on emicizumab for a median of 29 months (range 0.5–50 months) at the time of the bleeding event. One of the muscle bleeds, an iliopsoas hematoma, occurred during the loading dose phase of emicizumab, however, the remainder of muscle bleeds occurred ≥10 months on maintenance dosing. The thigh muscle was the most frequent site for bleeding accounting for 43% of bleeding events in this group. Other muscle sites including 3 iliopsoas muscle bleeds were also documented. Approximately 64% of muscle bleeds in the PwHA without inhibitors group were secondary to trauma and confirmed by radiographic imaging. Management of the muscle bleeds in PwHA without inhibitors consisted of scheduled bolus dosing of a recombinant or plasma-derived FVIII. However, due to bleed severity or lack of rapid improvement, three bleeds (21%) required hospitalization and administration of a CI of FVIII over 48–96 h (Table S2). These three cases were transitioned to intermittent FVIII dosing for 4–12 days before transitioning back to emicizumab alone. PwHA without inhibitors required a median of 12 doses (range 2–42 doses) of FVIII treatment and median of 11 factor exposure days (range 2–37) for resolution. In the 4 PwHA with inhibitors, the median age and duration on emicizumab at the time of muscle bleed were 13 years (range 13–27 years) and 46 months (range 32–62 months), respectively. Although the PwHA and inhibitors were a smaller subset of participants, the iliopsoas and thigh muscles were the most common sites of the 5 muscle bleeds documented. Sixty percent of the muscle bleeds in this group were trauma-related and all the bleeds were confirmed radiographically by ultrasound or computed topography scan. One patient had 2 separate trauma-related thigh muscle bleeds within a 1-month period (patient 12, Table S2). The second thigh muscle bleed was not considered a recurrence as it occurred at a different thigh muscle location based on radiologic imaging. Orthopedic surgery consultation was performed in one PwHA with inhibitor and forearm muscle hematoma to evaluate for concerns of compartment syndrome; however, no surgical intervention was indicated. No other participants required surgical consultation. Another PwHA and an inhibitor who presented with an iliopsoas hematoma had worsening pain and a >2 g/dL decline in their hemoglobin after >24 h on rFVIIa infusions that prompted discontinuation of rFVIIa and initiation of low-dose aPCC infusions (Table S2). The individual ultimately required 18 exposure days of low-dose aPCC for bleed resolution. This patient had a history of poor response to rFVIIa prior to starting emicizumab. Lab testing, including complete blood count with differential, complete metabolic panel, lactate dehydrogenase, and D-dimer was monitored during hospitalization and outpatient follow-up without any clinical signs of thrombotic microangiopathy (Table S3). None of the participants discontinued emicizumab following muscle bleeds, although two participants had alterations of their dosing regimens after bleed resolution at the provider's discretion. One PwHA with inhibitor (patient 14) transitioned from emicizumab 3 mg/kg every 2 weeks to 1.5 mg/kg weekly, while one PwHA without inhibitor (patient 9) switched from emicizumab 6 mg/kg every 4 weeks to 3 mg/kg every 2 weeks and ultimately to 1.5 mg/kg weekly after his 4th muscle bleed. Moreover, there was no clinical evidence (i.e., patient symptomatology or per physical exam) of thrombosis or thrombotic microangiopathy in this young cohort of patients. This study does have its limitations including the small sample size and retrospective design. Additionally, there was no uniform approach in the management of muscle bleeds as 19 different regimens were used for the 19 bleeding events. This is likely attributable to the varying severity of bleeds and provider experience, but also partly due to the lack of standardized guidance on factor treatment for major bleeding events on emicizumab. However, the study is strengthened by the inclusion of a multi-institutional cohort of PwHA with and without inhibitors with up to 62 months of follow-up and 41.5 patient years of emicizumab exposure. In conclusion, the advent of emicizumab has dramatically altered the landscape of PwHA management due to the significant reductions in ABR across the age spectrum in PwHA with and without inhibitors. Yet, bleeding events are inevitable. Most bleeds are expected to resolve in <24 h with 1–2 doses of factor treatment. However, in this real-world, retrospective multi-institutional study, we describe 15 PwHA with and without inhibitors on emicizumab who experienced severe muscle hematomas requiring intensive and prolonged factor treatment. Muscle bleeds, particularly iliopsoas hematomas, can present as severe and potentially life-threatening bleeding events with significant physical symptoms and bleeding-related complications requiring >24 h of treatment. Early recognition and prompt treatment of these types of bleeding events may improve bleeding outcomes and reduce the duration of factor treatment necessary for resolution. G.B. performed research, compiled data, analyzed data, and co-wrote the manuscript. A.P.W., A.C.W., and G.Y. performed research, analyzed data, and co-wrote the manuscript. R.F.S. and G.Y. designed the study, analyzed data, and co-wrote the manuscript. G.B. is supported by the Hemophilia of Georgia Clinical Scientist Development Award. R.F.S. has received investigator-initiated grant funding from Takeda, Octapharma, and Genentech. G.Y. has received investigator-initiated study grants from Genentech and Takeda. G.B. has received honoraria for advisory board participation from Bayer, Genentech, Octapharma, and Kedrion. A.P.W. has received honoraria for consulting work with Sanofi, HEMA Biologics, Takeda Pharmaceuticals, Novo Nordisk, Pfizer, Octapharma, Bioverativ, Bayer, CSL Behring, and Spark Therapeutics. A.C.W. has received honoraria for consulting work with Genentech, Sanofi, Bayer, Takeda, Spark, Novo Nordisk, Kedrion, and Aptevo. R.F.S. has received honoraria for consulting work with Genentech, Octapharma, Sanofi/Sobi, Hema Biologics, Bayer, Takeda, Spark, Pfizer, Novo Nordisk, Guardian Therapeutics, and HEMAb. RFS has investigator-initiated grant funding from Takeda, Octapharma, and Genentech. G.Y. has received honoraria and consulting fees from Bayer, BioMarin, CSL Behring, Genentech/Roche, Novo Nordisk, Octapharma, Pfizer, Sanofi/Genzyme, Spark, and Takeda. Due to the retrospective nature of this work and IRB approval with exempt status, informed consent and assent were waived. The data included in this study are available on request from the corresponding author. Table S1. Summary of patient characteristics. Table S2. Detailed treatment regimens of muscle bleeds for all bleeding events. Table S3. Laboratory parameters monitoring for patient 14. Please note: The publisher is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.