What can we expect for adolescents and adults with haemophilia switched to low‐dose prophylaxis from episodic treatment for over 3 years? A real‐world snapshot in China

Abstract

Dear Editor, Prophylactic treatment is widely considered the gold standard in haemophilia care.Due to thedifferential cost constraints inhealth care, the dosages and treatment regimens vary across countries.1,2 Substantial evidence has confirmed the superiority of low-dose prophylaxis over episodic treatment (ET) in children, in both developed and developing countries.3–6 However, data comparing the long-term outcomes ofhaemophilic arthropathy (HA)between these two treatments in adolescents and adults are limited. The Haemophilia Early Arthropathy Detectionwith Ultrasound (HEAD-US) scale7 has beenwidely used for routine monitoring of joint status, due to its wide availability and timeeffectiveness. This letter reports the long-term (≥3 years) effects of low-dose prophylaxis compared with those of ET in Chinese adolescents and adults with haemophilia, in a real-world setting. Weconducted a single-centred, cross-sectional, retrospective study at the Nanjing Drum Tower Hospital from December 2015 to March 2020. Inclusion criteria included diagnoses of moderate or severe haemophilia A and B, age 11–41 years, and previously received ET. Patients in the switch therapy (ST) group initially received ET and then switched to low-dose prophylaxis (< 50 IU/kg/wk of FVIII concentrates for haemophilia A, <30 IU/kg/wk of FIX concentrates for haemophilia B) for ≥36 consecutive months before enrolment. Exclusion criteria included histories of other arthropathies, joint surgeries, FVIII/FIX inhibitors, and acquired haemophilia. All patients with haemophilia A or B were treated with either full-length recombinant FVIII (rFVIII) or plasma-derived prothrombin complex, respectively, throughout the treatment duration. The prophylactic doses administered were 9.1–25.0 IU/kg twice weekly for patients with haemophilia A and 10.0–20.0 IU/kg 1–2 times a week for those with haemophilia B. Patients undergoing ET received infusions of factors whenever bleeding occurred. The studywas approved by the institutional reviewboard of theNanjingDrumTowerHospital. Informed consentswere obtained from either the patients in this study or their parents. The six index joints (bilateral ankles, knees, and elbows) were examined by ultrasound imaging and scored per the HEAD-US system.7 The status of a joint withHEAD-US score>0was defined as HA. The severity of HA was classified into four levels based on the HEAD-US scores: absent HA (score 0), mild HA (scores 1–3), moderate HA (scores 4–6), and severe HA (scores 7–8). Of the 30 Chinese adolescents (n = 10) and adults (n = 20) with moderate or severe haemophilia A or B who met the eligibility criteria, 15 maintained ET, and 15 had switched to low-dose prophylaxis at a median age of 11 (interquartile range, IQR, 25th–75th percentile, 9.5–15) years, and maintained this treatment for a median duration of 6 (range, 3–13.3) years. Ten of these 15 patients started prophylaxis after the age of 10 years, with seven of them starting prophylaxis during adolescence and three during adulthood. A total of 180 index joints were examined and scored. Age at examination, constituent ratios of adolescents and adults, and age at first bleeding were comparable between the two groups (P > 0.05). In the ET group, most patients (13/15) had multiple arthropathies, whereas more than half (8/15) of the patients in the ST group had only one or two affected joints (Table 1). Overall, the ankle was the most commonly affected joint, followed by the knee, and the elbow was the least vulnerable joint in both cohorts (Table 1). Between the ET and ST groups, the ratios of ankle arthropathy and the HEAD-US scores for the ankles were comparable (P = 0.751, 0.485, respectively). The elbow joints of the ST group showed lower susceptibility than those of the ET group (P = 0.004, 0.004, respectively). Lower percentages (63.3% vs. 40.0%) of damage and ultrasound (US) scores [3.0 (0.0, 7.0) versus 0.0 (0.0, 6.0)] were observed for the knees of patients undergoing ST (P = 0.071, 0.143, respectively), but these differences were not significant. In most examined joints, HAs of the ankle, knee, and elbow were detected in patients receiving ET, whereas most knee and elbow joints were normal in patients undergoing ST (Figure 1A). In the ET and ST groups, sonographic changes were absent in 30.0%and 33.3% of the assessed ankles, respectively. Additionally, parallel severity distribution of impaired ankles between the two cohorts was observed. Unlike ankles, more knees and elbows with each specific severity were found among patients undergoing ET than among those undergoing ST, except in those with moderately impaired knees (proportions were 20% vs. 20%). Severe knee arthropathy accounted for most of the overall severe HAs, whereas only a small proportion of cases with elbow arthropathy evolved into those with severe HA. Of the three indicators in the HEAD-US scale (synovium, cartilage, and subchondral bone damage), each of these was detected in at least half (50.0%–60.0%) of the respective index joints in the ET group, except for bone defects of the ankles (43.3%) and synovitis of the elbows (36.7%). Furthermore, osteochondral alterations, which are more common in lateHA,weredemonstrated in 50.0%of the examined