FormalPara Key Points for Decision Makers

The standard of care for people with haemophilia A includes treatment with clotting factor concentrates (CFCs) or non-factor products, such as emicizumab. Given the need for long-term prophylaxis, it is important to understand the medical expenditures associated with the use of CFCs and non-factor products in real-world clinical practice.

This study analysed real-world data from a hospital-based administrative database in Japan to investigate medical expenditures for patients with haemophilia A without inhibitors receiving plasma-derived factor VIII (pdFVIII), standard half-life (SHL) recombinant FVIII (rFVIII), extended half-life (EHL) rFVIII or emicizumab.

From 2016 to 2020, median annual total healthcare expenditures per patient and median annual drug expenditures per patient increased. Prescribing of EHL rFVIII and emicizumab increased. Total healthcare expenditures and drug expenditures increased for patients who switched from SHL to EHL rFVIII products.

1 Introduction

Factor VIII (FVIII) has a pivotal role in the intrinsic coagulation cascade, promoting and maintaining haemostasis after injury [1]. Haemophilia A results from deficiency or insufficiency in FVIII [2]. According to the World Federation of Hemophilia Annual Global Survey 2020, 165,379 patients worldwide have haemophilia A, with approximately 5500 residing in Japan [3].

Globally, FVIII replacement is still the standard of care for patients with haemophilia A, although use of novel non-factor replacement therapies is increasing [4]. Therapeutic options for patients with haemophilia A in Japan have expanded over recent years from plasma-derived FVIII (pdFVIII) and recombinant FVIII (rFVIII) products with a standard half-life (SHL) to extended half-life (EHL) rFVIII products. EHL rFVIII products approved in Japan include: damoctocog alfa pegol (launched in February 2019 [5]), efraloctocog alfa (launched in March 2015 [6]), rurioctocog alfa pegol (launched in June 2018 [7]) and turoctocog alfa pegol (launched in January 2020 [8]).

Non-factor products, such as antibodies that mimic FVIII activity, have also been developed [4]. For example, emicizumab, a bispecific, monoclonal antibody developed by a Japanese company that bridges activated factor IX and factor X to restore function of missing activated FVIII, was launched in May 2018 [9], offering broad experience in the Japanese population [10,11,12]. In phase 3 studies, emicizumab prophylaxis, administered subcutaneously once-weekly for patients with haemophilia A with FVIII inhibitors [13] or once-weekly or every 2 weeks for those without FVIII inhibitors [14], significantly lowered bleeding rate versus no prophylaxis. Other novel therapeutics—such as molecules that function by inhibiting antithrombin and tissue factor pathway inhibitors [15], as well as gene therapy—are likely to further diversify therapeutic options for patients with haemophilia A [16].

The dosage and associated dosing interval of SHL and EHL rFVIII products vary between patients owing to considerations, such as the patient’s pharmacokinetic profile, bleeding pattern, activity level, prescriber's input as well as the approved indication (e.g. prophylaxis, on-demand, surgery or immune tolerance induction) [17, 18]. Although EHL products enable longer intervals between doses and similar annualized bleeding rates compared with SHL products in studies conducted in several markets to date [19, 20], the expenditures for patients with haemophilia A using EHL versus SHL rFVIII products was consistently higher regardless of the dispensed FVIII dosage [21, 22]. Treatment to resolve breakthrough bleeds adds to healthcare expenditure, regardless of whether an EHL or SHL rFVIII product is used for prophylaxis; however, use of EHL rFVIII products have also been shown to be a cost-effective treatment option in one study conducted in Italy in patients with haemophilia A [23].

With the goal of managing costs, several countries have analysed the direct healthcare expenditures for patients with haemophilia [16, 24,25,26,27,28,29]. Understanding the use of clotting factor concentrates (CFCs) and non-factor products in real-world clinical practice is important given the need for long-term prophylaxis. This study analysed real-world data from a hospital-based administrative database in Japan. The main objective was to investigate the treatment of patients with haemophilia A in Japan, including the amounts of CFC and non-factor product (emicizumab) dispensed and the medical expenditures for patients who had consistent prescription fills of CFCs and without FVIII inhibitors.

2 Materials and Methods

2.1 Study Design

This was a retrospective, non-interventional, observational study of patients with haemophilia A included in the Medical Data Vision Co., Ltd (MDV) database, a large-scale, hospital-based, administrative database. The database contained inpatient hospitalization data, as well as outpatient and prescription data collected after a hospital visit unless the patient was transferred to another hospital. Data collected included outpatients and inpatient health insurance claims, costs, demographics, and limited laboratory data. As of February 2021, the database was the largest administrative database in Japan, with data from 438 facilities and > 35.2 million patients.

2.2 Study Population

Patients who met the following criteria were included: (1) evidence of visiting a healthcare facility in the database during the study period; (2) diagnosis of haemophilia A (ICD-10 code D66) during the study period, excluding an unconfirmed diagnosis; and (3) ≥ 2 prescriptions of the same CFC or emicizumab within the first 6 months of the index date and ≥ 2 weeks apart from each other during the study period, excluding the period during hospitalization. Exclusion criteria included patients who: (1) were diagnosed with congenital haemophilia with a FVIII inhibitor before the index date excluding unconfirmed diagnosis or (2) received bypass agents for treatment of haemophilia A with inhibitors after the index date.

The study period was from 1 January 2016 to 31 December 2020. The index date was defined as the date of the initial prescription fill of the CFC or emicizumab, with a record of the disease name being haemophilia A. Patients could have records pertaining to prescriptions for several CFCs or emicizumab that met this inclusion criterion in the patient’s medical history. Treatment duration was defined as the period from the first prescription fill of the CFC or emicizumab to the day before a switch to a different product, or it was censored at the last prescription of the same product if no switch occurred during the study period.

2.3 Outcome Measures

Treatment duration and annualized dosage in international units (IUs) per kilogram of body weight for CFC (with the calculation for body weight based on the statistical average weight of Japanese individuals in the specified age groups) [30] were summarized. The annual expenditures for CFC and emicizumab and the total healthcare expenditures were summarized in each year and in the overall period for patients receiving treatment with regular replacement therapy, defined as the use of CFC or emicizumab that met all of the following criteria: total treatment period of CFCs or emicizumab ≥ 1 year; prescription of any CFC or emicizumab ≥ two times on different dates during the year and no interval of > 6 months between prescriptions.

2.4 Data Analysis

Patient and treatment setting characteristics were summarized using descriptive statistics. Use of each CFC product (generic name) and emicizumab after the index date were summarized in each year and in the overall period for inpatients, outpatients and the whole population. For outpatient use, the percentage of patients who were prescribed a specific product at least once during the period was calculated in patients who could be followed during the period. Treatment duration and annualized dosage of each CFC (IU/kg) were summarized and, although the treatment duration could be censored, the summary of treatment duration was used for supporting the interpretation of the annualized dosage. All analyses were conducted for age subgroups 0–11, 12–19, 20–64 and ≥ 65 years.

2.5 Healthcare Expenditures

The annual expenditure for overall drug uses in the regular replacement therapy was summarized in each year and the overall period. The expenditures were also summarized separately by CFC and emicizumab. The total healthcare expenditure was summarized using the same method and included product and total medical costs (including medicine, testing/imaging, surgery, medical device/material, rehabilitation and other). Expenditures per product were derived from the Japanese drug price standard.

2.6 Switching from pdFVIII or rFVIII SHL Products to rFVIII EHL Products

Patients with data available for ≥ 3 months before and after switching from an SHL rFVIII (i.e. lonoctocog alfa, octocog alfa, octocog beta, rurioctocog alfa and turoctocog alfa) or pdFVIII product to an EHL rFVIII product (i.e. damoctocog alfa pegol, efraloctocog alfa, rurioctocog alfa pegol and turoctocog alfa pegol; i.e. the first occurrence of a switch to an EHL rFVIII product) and who filled ≥ 2 prescriptions for a EHL rFVIII product were included in the switching analyses.

The expenditures for CFC or emicizumab, total healthcare expenditures, and the summary statistic for product dosage (IU/kg) dispensed in the four quarters (3 months, 4–6 months, 7–9 months and 10–12 months) before and after switching were summarized. Incomplete quarters (i.e. patients with < 90 days’ follow-up) were not included.

3 Results

3.1 Study Population

Overall, 974 patients with haemophilia A on regular replacement therapy were included in the analysis (Fig. 1). Baseline demographics and characteristics are presented in Table 1. The median (range) age was 30.0 (0.0–91.0) years.

Fig. 1
figure 1

Study cohort selection. CFC clotting factor concentrate, FVIII factor VIII

Full size image
Table 1 Patient demographics and baseline characteristics
Full size table

The median (first to third quartile) duration of follow-up for each patient was 3.7 (2.1–4.8) years. Patients visited university hospitals (41.2%), local public hospitals (27.7%), national hospitals (24.2%) and private hospitals (6.9%). The majority (67.4%) were medical institutions with ≥ 500 beds. A previous history of viral hepatitis and chronic viral hepatitis C was reported in 46.0% and 41.9% of the patients, respectively.

3.2 Usage of Each FVIII Product Dispensed in an Outpatient Setting

Following its launch in 2018, the proportion of patients dispensed emicizumab increased over the subsequent 2 years, with 17% of patients using emicizumab in 2019 and 25% in 2020 (Fig. 2). The percentage of patients who were dispensed EHL rFVIII/emicizumab also increased from 15% in 2016 to 72% in 2020; SHL rFVIII products were still used over the analysis period, with use of octocog beta notably increasing from 1% of patients in 2016 to 21% in 2020, whereas use of octocog alfa and rurioctocog alfa decreased over the analysis period. These trends were observed across all age categories [Fig. S1 in the Electronic Supplementary Material (ESM)].

Fig. 2
figure 2

Outpatient use of treatments for haemophilia A. EHL extended half-life, FVIII factor VIII, pdFVIII plasma-derived factor VIII, SHL standard half-life

Full size image

3.3 Annualized Dosage for Each FVIII Product for Outpatient Use Only

Median annualized dispensed dosage (IU/kg) was lowest for pdFVIII products (901.9 IU/kg). Dispensed dosages of SHL rFVIII products ranged from 2022.4 to 4589.1 IU/kg (Table 2). The median annualized dispensed dosages for rFVIII EHL products ranged from 2189.1 to 2989.7 IU/kg (Table 2). Annualized FVIII dosage and observed treatment duration by age category are included in ESM Table S1. The treatment duration for each product varied (Table 2).

Table 2 Observed treatment duration and annualized dosage of treatment for haemophilia A for outpatients (whole study population)
Full size table

3.4 Prevalence of Joint Damage After the Index Date

Overall, 56.1% of patients had documented joint damage (Table 3), with the proportion of patients with joint damage increasing across age categories from 50.7% in those aged 0–11 years to 60.3% in those aged ≥ 65 years (ESM Table S2). The most frequent joint damage in the overall study population was haemophilic arthritis (Table 3).

Table 3 Prevalence of joint damage after the index date in diseases with a frequency of ≥ 3% among the study population
Full size table

3.5 Expenditures for Haemophilia A Treatments

Median annual total healthcare expenditure per patient increased twofold from ¥9,200,230 ($84,406 USD) in 2016 to ¥19,748,221 ($181,176 USD) in 2020 (Table 4). Annual expenditure per patient for CFC and emicizumab increased twofold from 2016 to 2020 [from ¥8,723,120 ($80,029 USD) to ¥18,051,689 ($165,612 USD)] and for CFC, increased 1.5-fold [from ¥8,642,905 ($79,293 USD) to ¥12,963,480 ($118,931 USD)]. For emicizumab, annual expenditure per patient increased 1.7-fold from 2018 when the product was launched to 2020 [2018: n = 4, ¥26,030,206 ($238,809 USD); 2019: n = 108, ¥29,919,696 ($274,493 USD); 2020: n = 107, ¥45,430,408 ($416,793 USD)]. Median annual total drug expenditure per patient with regular replacement therapy, as well as median annual expenditures for CFC, emicizumab and the total for each age category showed similar trends (ESM Table S3). The expenditures for products accounted for most of the annual total healthcare expenditures. Lower annual total healthcare expenditures were observed for patients without versus with joint damage (Table 4).

Table 4 Mean (SD) and median (Q1–Q3) annual medical expenditures in Yen in patients receiving regular treatment for haemophilia A (whole study population)
Full size table

3.6 Switch from pdFVIII/SHL rFVIII to EHL rFVIII Products

The breakdown of patients who switched from a pdFVIII or SHL rFVIII to an EHL rFVIII product is shown in Table 5. Overall, 23.9% (233/974) of patients switched from a pdFVIII or SHL rFVIII to an EHL rFVIII product during the study period, with the highest proportion of patients switching from rurioctocog alfa to either rurioctocog alfa pegol [106/233 (45.5%)] or efraloctocog alfa [40/233 (17.2%)]. There was an increase in the median amount of IUs prescribed in the first 3 months after switching product compared with the 3 months prior to switching although overall, there was no noticeable difference in the median IUs prescribed before and after switching (Fig. 3). However, median total healthcare expenditures (Fig. 4) and product expenditures (Fig. 5) increased after switching products.

Table 5 Patients who switched from pdFVIII or SHL rFVIII to an EHL rFVIII product (n = 233)
Full size table
Fig 3
figure 3

Median (Q1, Q3) FVIII IU/kg dispensed before and after switching from pdFVIII or SHL rFVIII to an EHL rFVIII product. EHL extended half-life, FVIII factor VIII, pdFVIII plasma-derived factor VIII, Q quartile, SHL standard half-life

Full size image
Fig. 4
figure 4

Median (Q1, Q3) total healthcare expenditures before and after switching from pdFVIII or SHL rFVIII to an EHL rFVIII product. EHL extended half-life, FVIII factor VIII, pdFVIII plasma-derived factor VIII, Q quarter, SHL standard half-life

Full size image
Fig. 5
figure 5

Median (Q1, Q3) expenditures for FVIII product before and after switching from pdFVIII or SHL rFVIII to an EHL rFVIII product. EHL extended half-life, FVIII factor VIII, pdFVIII plasma-derived factor VIII, Q quarter, SHL standard half-life

Full size image

4 Discussion

This real-world study showed that IUs of EHL rFVIII products dispensed are increasing in Japan, with an accompanying increase in healthcare expenditures associated with switching from a pdFVIII or SHL rFVIII to an EHL rFVIII product. Overall, annual healthcare expenditures per patient with regular replacement therapy, as well as median annual expenditures for CFC, emicizumab and the total increased over the study. Annual expenditure over the analysis period was highest with emicizumab. An increase in overall drug expenditure was observed from 2018 when the emicizumab therapy was launched. On the basis of these results, it is predicted that the number of patients who receive emicizumab will increase, therefore haemophilia-related healthcare expenditure in Japan may also continue to rise. It is possible that the increases in annual healthcare expenditures in this study are related to overall inflation; however, the consumer price index in Japan increased only 2% over the study period (2016–2020) [31], which is much lower than the reported 1.5- to 2-fold increases in annual expenditures. In a real-world, multicentre, observational study in the USA, costs were decreased in patients with haemophilia with and without inhibitors following the initiation of emicizumab; however, the decrease was smaller in patients with haemophilia without inhibitors [32]. The difference between the current study in Japan and the study conducted in the USA could be the result of the high costs associated with the purchase of emicizumab in Japan.

The increase in median expenditures observed in the current analysis was similar to findings from other studies of patients with haemophilia A [21, 33, 34]. A budget impact model was developed to estimate the potential economic impact of switching from a pdFVIII/SHL rFVIII to an EHL rFVIII product in the USA [34]. The introduction of the EHL rFVIII product was estimated to have a budget impact of 1.4% across 2 years for a private payer population of 1,000,000 (estimated 19.7 individuals receiving treatment for haemophilia A) [34]. The introduction of EHL rFVIII was estimated to prevent 124 bleeds across 2 years at a cost of $1891 per bleed avoided [34]. In a real-world study conducted in a smaller population of patients with haemophilia A in the USA, median healthcare expenditures were higher over 2 years after patients switched from an SHL to an EHL rFVIII product. Switching to an EHL product was associated with variable factor IUs dispensed and consistently higher expenditures [21].

However, one budget impact analysis model did not predict increased costs with switching from a SHL to an EHL rFVIII product. A budget impact analysis to estimate the economic impact of the introduction of an EHL product, efmoroctocog alfa, for the treatment of haemophilia A in Italy suggested that the introduction of the EHL product did not result in higher expenditures in the Italian National Health System [35]. However, that model estimated treatment costs, whereas the current study explored real-world factor use.

Drug pricing in Japan considers the extended half-life of EHL products, thus EHL products are priced higher per IU than SHL products. For example, the price per IU of the EHL product Rurioctocog alfa pegol is approximately 40% greater than the price per IU of the SHL product Rurioctocog alfa. Typically, a decrease in the median FVIII IU/kg dispensed would be expected after a switch from a SHL to an EHL product; however, a similar or increasing trend in median IU/kg dispensed was observed in this study. Because EHL products are priced higher per IU than SHL products, this corresponded to an increase in treatment and total healthcare expenditures.

These findings suggest that there has been higher and/or more frequent dosing for EHL products than was originally assumed. There are several possible explanations for the observed increase in prescribing: the intensity of treatment for patients may have increased, previously undertreated patients may be adequately treated after switching, and/or patients may have improved adherence to prophylaxis after switching. Additionally, while the approved interval between doses is longer for EHL than SHL products, some patients may choose to follow a more frequent dosing schedule if they are not satisfied with the efficacy associated with the increased average half-life of EHL products. While this study did not examine the clinical outcomes of switching in these patients, more frequent dosing might be associated with improvements in patient quality of life.

There have been inconsistent reports regarding the impact of switching from prophylaxis with an SHL to an EHL rFVIII product on a patient’s health-related quality of life (HRQoL). In a retrospective chart review from a single haemophilia treatment centre in Canada [36], 38 boys (median age, 11.5 years) with moderate/severe haemophilia A who were switched from a SHL to an EHL product (reduction of 1.2 infusions/week) did not show any significant improvements in HRQoL, as measured on the Canadian Hemophilia Outcomes|–Kids’ Life Assessment Tool. However, in a prospective cohort study of persons with moderate or severe haemophilia aged ≥ 6 years, switching from a SHL to an EHL product resulted in short-term meaningful improvement in overall HRQoL and other patient-reported outcome measures, although the observed improvements disappeared by 24 months in most domains examined [37].

In the current study, approximately half of all young patients had some level of joint damage; the prevalence of joint damage then remained constant with ageing. This highlights that despite a relatively high standard of care for patients with haemophilia A in Japan, joint damage in the young population is still prevalent. Joint damage is a major complication associated with haemophilia, with a significant negative impact on quality of life for affected adults [38]. The Joint Outcomes Study demonstrated that joint outcomes in haemophilia are better in young adults if prophylaxis is started before age 2.5 years compared with after age 6 years [38]. This is an important consideration for all therapies to improve long-term joint outcomes.

There are several limitations of this study. First, the retrospective design precluded a more in-depth analysis of potential expenditure savings for various stakeholders, such as a decrease in hospitalizations and associated hospital-related expenditures. Second, treatment-related and broader medical expenditures may have been underestimated as only expenditures for medications dispensed in the hospital setting were included. Third, the diagnostic information in the database only includes data on patients who required in-hospital tests or treatment, and consequently may have under- or over-estimated the presence of comorbidities, such as chronic hepatitis, and the prevalence of joint damage; it therefore may not fully reflect real-world clinical practice. Lastly, the patient population in the database includes only those receiving treatment in the hospital. There may be differences in how patients are managed within a hospital versus a clinic setting, which will influence treatment patterns and expenditures, especially in the context of surgery and trauma. Another limitation could be the change from on-demand use to prophylaxis which may also increase costs, such as prophylactic therapy and hospital costs, over time.

5 Conclusions

This retrospective study found an increase in dispensing of EHL rFVIII versus SHL rFVIII or pdFVIII products among Japanese patients with haemophilia A, along with an increase in associated healthcare expenditures for patients who switched from a pdFVIII or SHL rFVIII to an EHL rFVIII product. Furthermore, healthcare expenditure increased along with the increase of emicizumab use after its launch in 2018. Overall, these results provide insight into the treatment and expenditures associated with haemophilia A in Japan and highlight the disease burden in patients requiring hospital care.