A randomized trial of human C1 inhibitor prophylaxis in children with hereditary angioedema

Abstract Background Patients with hereditary angioedema with C1 inhibitor deficiency or dysfunction have burdensome recurrent angioedema attacks. The safety, efficacy, and health‐related quality of life (HRQoL) outcomes of C1 inhibitor (C1‐INH) prophylaxis (intravenously administered) in patients aged 6‐11 years were investigated. Methods Eligible patients were enrolled in a randomized, single‐blind, crossover, phase 3 trial. After a 12‐week baseline observation period (BOP), patients received 500 or 1000 U C1‐INH, twice weekly, for 12 weeks before crossing over to the alternate dose for 12 weeks. The primary efficacy end‐point was the monthly normalized number of angioedema attacks (NNA). HRQoL was assessed using the EuroQoL 5‐dimensional descriptive system youth version and visual analog scale (EQ‐VAS). Results Twelve randomized patients had a median (range) age of 10.0 (7‐11) years. Mean (SD) percentage reduction in monthly NNA from BOP was 71.1% (27.1%) with 500 U and 84.5% (20.0%) with 1000 U C1‐INH. Mean (SD) within‐patient difference (−0.4 [0.58]) for monthly NNA with both doses was significant (P = 0.035 [90% CI, −0.706 to −0.102]). Cumulative attack severity, cumulative daily severity, and number of acute attacks treated were reduced. No serious adverse events or discontinuations occurred. Mean EQ‐VAS change from BOP to week 9 of treatment (500 U C1‐INH, 10.4; 1000 U C1‐INH, 21.6) was greater than the minimal important difference, indicating a meaningful HRQoL change. Conclusions C1‐INH prophylaxis was effective, safe, and well tolerated in children aged 6‐11 years experiencing recurrent angioedema attacks. A post hoc analysis indicated a meaningful improvement in HRQoL with C1‐INH. Trial registration ClinicalTrials.gov identifier NCT02052141.


| INTRODUC TI ON
Hereditary angioedema with C1 inhibitor deficiency or dysfunction (C1-INH-HAE) is a rare inherited disease with an estimated prevalence of 1:50 000, 1 characterized by episodic swelling of the skin, abdomen, and larynx. [2][3][4][5] C1-INH-HAE's clinical manifestations including age of first symptoms, and attack location, frequency, and severity, are heterogeneous. [6][7][8] Approximately 50% of patients have potentially fatal laryngeal attacks. 9,10 In a European survey of patients aged ≥12 years, recurrent acute angioedema attacks impaired health-related quality of life (HRQoL) to a similar extent to other serious chronic diseases. 11 The HRQoL of patients aged 3-18 years can be impaired relative to healthy controls, especially in school and physical domains, due to the frequency and site of angioedema attacks. 12 Questionnaires completed by/for patients aged 5-18 years in Israel and Hungary found that children and adolescents had higher anxiety traits correlating with HRQoL impairment. 13 International World Allergy Organization/European Academy of Allergy and Clinical Immunology guidelines for managing C1-INH-HAE recommend C1-INH for first-line long-term prophylaxis. 14 Although pediatric patients were included in prior studies of C1-INH, 15 no dedicated randomized clinical trials specifically assessed children aged ≥6 to <12 years. As C1-INH was safely used in children in uncontrolled studies or studies non-specific to a pediatric cohort, C1-INH is recommended for emergencies. 16 Recent consensus guidelines indicated the need for phase 3 clinical trials specifically targeting pediatric populations. 17 In Europe and the United States, C1-INH (Cinryze; Shire) was recently approved for routine prophylaxis in patients aged ≥6 years, but before this study was initiated, it was only approved for adult and adolescent patients. Interim results of this study were previously published. 18 This article describes full study results in 12 patients.

| ME THODS
The randomized, phase 3, single-blind, crossover study involved 10  Efficacy analyses included patients in the safety set with ≥1 post-baseline primary efficacy assessment. The primary efficacy end-point was the monthly normalized number of attacks (NNA) in a 12-week treatment period. Secondary efficacy end-points were cumulative attack severity (sum of the maximum symptom severity score recorded for each attack), cumulative daily severity (sum of the severity scores recorded for each day of symptoms), and number of attacks receiving acute treatment. Primary and secondary end-points were normalized for number of days a patient participated in a given period, expressed as a monthly frequency. C1-INH doses were compared using a 2-sided paired t test conducted at α = 0.1. A post hoc analysis of the patient proportion with a ≥50%, ≥70%, and ≥90% reduction in NNA relative to baseline was performed.
To assess a potential carryover effect, primary and secondary efficacy end-points excluding the first 2 weeks of the second treatment period and then the first 2 weeks of both periods were analyzed. Treatment and sequence effects for all end-points (without removing the first 2 weeks of the second treatment period) also were tested using a mixed-effects model at α = 0.1, with treatment, sequence, and their interactions as fixed effects and assessments for same patient within sequence as repeated measurements.
HRQoL was assessed using the EuroQol 5-dimensional descriptive system youth version (EQ-5D-Y) at screening, weeks 5 and 9 of the BOP, weeks 1, 5, and 9 of both treatment periods, and each day of an

Key Message
Before this study was initiated, intravenous C1-INH (Shire) was approved for routine prophylaxis against HAE attacks in the United States and in the European Union in adolescent and adult patients, but not in pediatric patients <12 years of age. This study shows that patients aged 6-11 years with recurrent and burdensome hereditary angioedema attacks would benefit from prophylaxis with twice-weekly C1-INH (500 or 1000 U) as it is efficacious and safe, and may lead to a meaningful improvement in health-related quality of life. angioedema attack. Patients/caregivers used electronic study diaries to record information on angioedema attacks and complete EQ-5D-Y questionnaires. The EQ-5D-Y is a descriptive system comprising five dimensions in child-friendly language ("mobility," "looking after myself," "doing usual activities," "having pain/discomfort," and "feeling worried, sad, or unhappy"). Each dimension has three problem severity levels ("no problems," "some problems," and "a lot of problems").
The EQ-5D-Y includes a visual analog scale (EQ-VAS) of overall health ranging from 0 (worst imaginable health) to 100 (best imaginable health). EQ-5D-Y responses and EQ-VAS scores were summarized by combining those for the same dose for both periods and scheduled visit using descriptive statistics. EQ-VAS score differences from the BOP (average of all pre-dose visits during the BOP) to week 9 of each treatment period were summarized. The minimal important difference (MID) was estimated using half the SD of the EQ-VAS value during the BOP. 19,20

| Patient demographics and baseline characteristics
Between March 2014 and May 2017, 4 of 16 screened patients failed to meet the randomization criteria and 12 were randomized and completed the study. Table 1 summarizes patient demographics and baseline characteristics. Most patients received on-demand acute treatment for an angioedema attack within 3 months pre-screening.
No patients received long-term prophylaxis for HAE with 3 months before screening or the BOP.

| Attack severity and number of acute attacks treated with rescue medication were reduced
Monthly normalized cumulative attack severity, cumulative daily severity, and number of treated acute attacks were reduced with both doses compared with the BOP (Figure 1B  Using a 2-sided paired t test at α = 0.1, there was a significant difference between either dose for number of angioedema attacks (P = 0.03), cumulative attack severity (P = 0.05), cumulative daily severity (P = 0.04), and number of treated acute attacks (P = 0.07).
These results were consistent when analyzed using a mixed-effects model (Appendix 1), and no significant differences for sequence

| Both C1-INH doses were safe and well tolerated
Breakthrough HAE attacks, captured as treatment-emergent adverse events (TEAEs), occurred in 10/12 (83.3%) patients. Within 24 hours of administration, the most common non-HAE attack TEAEs were fatigue and/or irritability (Table 2). One HAE attack in one patient was considered by the investigator as related to C1-INH (1000 U), and in two patients, mainly mild TEAEs of fatigue (18 events) and irritability (14 events) were considered related to both C1-INH doses.
Eleven severe breakthrough HAE attacks occurred in five patients (six during treatment with 500 U C1-INH, three with 1000 U, and two during follow-up). Furthermore, a severe TEAE (dental caries) occurred in one patient.
No serious TEAEs or discontinuations occurred ( Table 2). All patients tested negative for anti-C1-INH antibodies following 6 months of total exposure to study drug (2.7 person-years), and no thrombotic or thromboembolic TEAEs occurred.

| Patients' HRQoL improved, particularly with 1000 U C1-INH
Most patients experienced no problems in any EQ-5D-Y dimension.
At screening and weeks 5 and 9 of the BOP, ≤10.0% of patients reported problems with mobility and/or feeling worried, sad, or unhappy; ≤14.3% reported problems with self-care and doing usual activities; and ≤33.3% reported problems with pain/discomfort. During treatment with 500 U C1-INH (weeks 5 and 9), no patients reported problems with mobility, self-care, and doing usual activities; 11.1% reported feeling worried, sad, or unhappy; and ≤22.2% reported having pain or discomfort. During treatment with 1000 U C1-INH (weeks 5 and 9), no patients reported problems with any of the five dimensions. For simplicity, the patient proportion with no problems at week 9 is shown in Figure 4A.
The MID for the EQ-VAS was calculated as half the SD at baseline (13.8); therefore, a change in EQ-VAS score >7 was considered the minimal important change, with higher EQ-VAS scores indicating better HRQoL. At week 9, the mean (SD) change in EQ-VAS score from baseline with 500 and 1000 U C1-INH was greater than the MID indicating a meaningful change (ie, 10.4 [19.0] with 500 U C1-INH and 21.6 [13.4] with 1000 U; Figure 4B).  15 In the open-label study, 87% of patients aged 2-17 years experienced <1 monthly attack and 22% had no attacks. 15 The reduction in the number and frequency of attacks with twice-weekly C1-INH is expected to improve the HRQoL of patients with C1-INH-HAE. 12 In this first blinded, randomized, controlled trial of C1-INH for prophylaxis targeting patients aged 6-11 years, the interim analysis in six patients showed that both doses reduced the monthly number of attacks by a mean of 1.89 compared with the BOP. 18 In the complete study described here, mean (percentage) reduction in monthly   Archives of Allergy and Immunology. 18 We are saddened by the recent loss of one of the co-authors of this manuscript, Dr. Dumitru

| D ISCUSS I ON
Moldovan. His dedication to HAE research and to the care of his patients was extraordinary and will be greatly missed.