The Cost-Effectiveness of Tenofovir Alafenamide for Chronic Hepatitis B Virus in Taiwan

Abstract

Background. Chronic hepatitis B (CHB) is a lifelong disease requiring long-term or indefinite therapy, resulting in substantial economic burden. Thus, careful consideration must be used in the selection of therapies. Aim. This analysis assessed the cost-effectiveness of tenofovir alafenamide (TAF) compared with tenofovir disoproxil fumarate (TDF) and entecavir (ETV) from the perspective of the Taiwan National Health Insurance Administration Healthcare payer for the management of CHB over a lifetime horizon. Methods. An individual patient simulation model assessed the impact of treatment on CHB infection for liver- and safety-related outcomes. Patients could achieve spontaneous or treatment-induced responses, experience a reactivation of the disease, develop long-term liver complications, or experience treatment-related renal or bone complications. Patient population profiles were based on clinical trial and real-world data. Data on clinical parameters (safety, mortality, resistance risk, and flare), health utilities, and costs were sourced from the published literature. Results. TAF was associated with fewer liver disease events and fewer cases of bone and renal complications per 100 person-years. TAF also had higher eAg and sAg seroconversion compared with TDF and ETV. As compared with both treatments, TAF was both more effective and more costly, resulting in incremental cost-effectiveness ratios of USD 3,348 and USD 3,940 per quality-adjusted life-year gained versus TDF and ETV, respectively. Conclusion. TAF leads to better health outcomes at acceptable incremental costs compared with the most commonly used therapies in the management of CHB, thus making it a cost-effective option for the treatment of CHB in Taiwan.

Highlights

The cost-effectiveness of tenofovir alafenamide (TAF) versus tenofovir disoproxil fumarate (TDF) and entecavir (ETV) was assessed in patients with chronic hepatitis B in Taiwan.

TAF was associated with fewer liver disease events, fewer cases of bone and renal complications, and higher eAG and sAG seroconversion compared with TDF and ETV; TAF was found to be cost-effective compared with both treatments.

Keywords

chronic hepatitis B cost-effectiveness analysis tenofovir alafenamide tenofovir disoproxil fumarate

Hepatitis B is an infectious disease caused by the hepatitis B virus (HBV) and is endemic in Taiwan.¹ About 5% of infected adults develop chronic hepatitis B (CHB), defined as hepatic inflammation that persists for more than 6 mo after acute infection with HBV.² CHB can cause progressive liver damage that, when left unmanaged, may lead to serious clinical outcomes such as cirrhosis, liver failure, hepatocellular carcinoma (HCC), or death. Despite the implementation of a universal HBV vaccination program in Taiwan, the incidence of HBV-related HCC remains high.¹ Approximately 20% to 30% of patients who become chronically infected will ultimately develop complications of cirrhosis and HCC,² and CHB is the leading cause of HCC in Taiwan.³ In addition to the clinical burden of CHB, there is equally substantial economic burden. Total inpatient costs for CHB in Taiwan were estimated to account for approximately 1% of the total inpatient expenditure.⁴

As a lifelong disease, CHB requires long-term or indefinite therapy, unless the patient achieves hepatitis B surface antigen (HBsAg) seroclearance, which is considered the “state closest to a cure.”⁵ Goals of treatment include viral suppression and normalization of serum alanine aminotransferase (ALT). Achieving and sustaining low or undetectable viral load levels are associated with improved outcomes and normalization of ALT levels and may decrease the risk for liver disease progression, such as HCC, reducing liver-related mortality.^5–8 While effective treatments exist, antiviral therapies may negatively affect a patients’ bone and renal health, with the risk of developing these comorbid conditions increasing with age.^9–11 Thus, therapies favoring better bone and renal risk profiles should be selected for patients at higher risk of complications.

Treatments for CHB include tenofovir disoproxil fumarate (TDF) or entecavir (ETV) as either first-line nucleotide analogs in treatment-naïve (TN) patients (patients who have never received treatment for CHB) or in patients with cirrhosis.⁵ While both are recommended options for CHB patients, ETV has a resistance rate up to 51% in lamivudine-refractory patients after 5 y.¹² Further, both TDF and ETV have also been associated with adverse renal events and require dose adjustments for patients with creatinine clearance <50 mL/min.^12–14 TDF has also been associated with proximal tubular damage, which can lead to hypophosphatemia, reduced bone mineral density, and potentially osteomalacia.^5,14

Tenofovir alafenamide (TAF) has recently been included as one of the preferred treatment options alongside ETV and TDF for CHB¹⁰ and is reimbursed by the Taiwan National Health Insurance.¹⁵ TAF is a novel nucleotide reverse transcriptase inhibitor and phosphonoamidate prodrug of tenofovir that efficiently targets the hepatocytes when it undergoes conversion to the pharmacologically active metabolite TDF. Based on 2 large phase III trials, TAF has similar viral suppression rates, improved ALT level reduction, and improved safety profile with respect to renal and bone biomarkers compared with TDF.^16,17 Given the substantial economic burden associated with CHB, this analysis assessed the cost-effectiveness of TAF compared with TDF and ETV from the perspective of the Taiwan National Health Insurance Administration Healthcare (NHIA) payer for the management of CHB.

Methods

Model Structure

This cost-effectiveness analysis was developed from the NHIA payer perspective in Taiwan, over a lifetime time horizon (70 y), given the chronic nature of the disease. The individual patient simulation model was created in Microsoft Excel to assess the impact of treatment on CHB infection for liver- and safety-related outcomes. Over the course of a simulation, patients can achieve spontaneous or treatment-induced responses (e.g., HBeAg/HBsAg seroconversion and/or viral suppression), experience a reactivation of the disease (e.g., viral resistance), develop long-term liver complications, or experience treatment-related renal or bone complications (Figure 1). Both costs and effects were discounted at 3.5% per year.¹⁸

Figure 1

Model overview.

The model assigns patients to treatment, upon which active CHB patients (TN or treatment experienced [TE]) are determined to be responders or a nonresponders by week 48. Based on the response status, the model updates the patient’s HBV DNA level and ALT level categories. Responders who achieve viral suppression with an undetectable HBV DNA level by week 48 may achieve HBsAg or HBeAg seroconversion. HBeAg seroconverted patients may move back to active CHB, at which time their HBV DNA and ALT levels are reset to baseline. Patients who achieve HBsAg seroconversion are assumed to discontinue their antiviral therapy after 6 mo. Patients who do not achieve viral suppression at week 48 immediately discontinue to receive best supportive care (BSC). If treatment resistance occurs, it is assumed that patients will have uncontrolled HBV DNA and ALT levels (i.e., HBV DNA viral load >1,000,000 copies/mL and ALT level >44 IU/mL) for 3 mo. A proportion of patients who develop resistance will experience acute flare, some of whom will develop decompensated cirrhosis (DCC).

Patient Population

Patient population profiles were based on clinical trial data and real-world analyses from the National Health Insurance Research Database (NHIRD) in Taiwan between 2018 and 2020.^16,17,19 The model includes Taiwanese adults with CHB infection, of whom 26% of patients are HBeAg positive and 80% of the population have had some treatment for CHB in the past (TE). The distribution of patient ages at baseline according to HBeAg status and treatment experience (TN or TE) is provided in Supplementary Table S1. Within the TE patient population, based on expert opinion, it is assumed that 27% would have existing lamivudine resistance,¹⁹ which increases the risk of developing resistance to ETV. Across all subpopulations in the model, based on NHI claims data, 23.3% of CHB patients were assumed to have liver cirrhosis.^16,17

As there is a strong correlation between reduction in ALT levels and viral load levels with antiviral therapies, 2 patient profiles were created: virally suppressed patients and viremic patients. Based on the treatment outcome (i.e., virally suppressed or viremic), a corresponding HBV DNA-ALT level profile was sampled based on the 320-0108 and 320-0110 trial data.^16,17 Due to the absence of patient-level data for ETV, it is assumed that the HBV DNA-ALT profiles for ETV are the same as the TDF arm of the 320-0108 and 320-0110 trials.^16,17

Clinical Inputs

Active treatments included TAF, TDF, and ETV. The impact of treatment on liver outcomes was determined by a risk score for progression to liver cirrhosis and HCC. Risk scores were calculated based on a patient’s profile and included the variables gender, age, family history of HCC, alcohol consumption, serum ALT levels, and HBeAg/HBV DNA genotype.⁸ While gender, family history of HCC, and alcohol consumption were fixed at patient start in the model, the remaining predictors varied over time. The corresponding 10-y risk was converted into an annual transition probability in the model; 10-y risks were based on Costa et al.²⁰ The risk score is evaluated at each time step to account for change in the variable parameters.

In the model, viral suppression was defined as HBV DNA <300 copies/mL. For each specific patient subgroup, the probability of achieving viral suppression and HBeAg seroconversion for TDF and TAF was estimated from published trial results (Supplementary Table S2).^16,17 Note that the probability of viral suppression for lamivudine-resistant patients was assumed to be equivalent to that of TE patients. In the absence of head-to-head trials between ETV, TDF, or TAF, the efficacy estimates for ETV were derived by applying the relative relations between ETV and TDF as reported in Govan et al.²¹ to the TDF efficacy rates reported in Chan et al.¹⁶ and Buti et al.¹⁷ (Supplementary Table S3). Efficacy in subsequent years is assumed to the same as in the first year.

Disease progression rates excluding the transitions to compensated cirrhosis (CC) and HCC (which are dependent on the risk score) were sourced from Taiwan-specific sources, where possible, and converted to annual transition probabilities (Table 1).

Table 1

Clinical Model Inputs

Parameter	Value	Source
Annual transition probabilities for disease progression
HBeAg seroconversion to HBeAg negative	3%	Shepherd et al. 2006²²
HBeAg seroconversion to HBeAg positive	3%	Shepherd et al. 2006²²
HBeAg seroconversion to HCC^a	0.2%	Hsu et al. 2002²³
HBeAg seroconversion to HBsAg seroclearance	2.9%	Dakin et al. 2010²⁴
HBsAg seroconversion to HCC^b	0.0%	Dakin et al. 2010²⁴
HBV DNA <300 to HBsAg seroclearance	0.2%	Chen et al. 2009²⁵
CC (on treatment) to DCC	2.3%	Liaw et al. 1989²⁶
CC (off treatment) to DCC	3.1%	Veenstra et al. 2008²⁷
DCC (off treatment) to CC	0.0%	Dakin et al. 2010²⁴
DCC (year 1) to CC	13.6%
DCC (year 2) to CC	0.0%
DCC to HCC	2.5%	Crowley et al. 2002²⁸
DCC to LTx	1.4%	Veenstra et al. 2008²⁷
HCC to LTx	0.1%	Chen et al. 2003²⁹
MOF risk over 10 y for TDF-based therapies
Low risk ( <10%)	5%	University of Sheffield³⁰
Medium risk (≥10% to ≤20%)	15%
High risk (≥21%)	25%
MOF hazard ratio versus TDF
TAF (v. TDF)	0.956	320- 0108 and 320-0110 trials^16,17
ETV (v. TDF)	0.956	Assumed equivalent to TAF v. TDF
Mean change in eGFR from baseline
TAF		320- 0108 and 320-0110 trials^16,17
Year 1	−3.1
Year 2	−2.5
Year 3	−2.5
ETV		Wu et al. 2017³¹
Year 1	−2.7
Year 2	−3.7
Year 3	−4.4
TDF		Wu et al. 2017³¹
Year 1	−6.1
Year 2	−5.0
Year 3	−4.9
Excess mortality (hazard ratio) due to active CHB by HBV DNA levels (copies/mL)
<300 (undetectable) to 99,999	1.0	Iloeje et al. 2007³²
100,000 to 999,999	2.0
≥1 million	2.1
Excess mortality (hazard ratio) due to treatment-related complications
Stage III CKD	1.36	Neuhaus et al. 2010³³
ESRD	16.80	USRDS 2014³⁴
MOF	1.26	Bliuc et al. 2013³⁵
Annual mortality risk due to liver-related complications
CC	5.1%	Bermingham 2015³⁶
DCC	39.0%
HCC	37.2%
LTx (first year)	15.0%
Post-LTx (subsequent years)	1.5%

ALT, alanine aminotransferase; BSC, best supportive care; CC, compensated cirrhosis; CKD, chronic kidney disease; DCC, decompensated cirrhosis; eGFR, estimated glomerular filtration rate; ESRD, end-stage renal disease; ETV, entecavir; HCC, hepatocellular carcinoma; LTx, liver transplantation; MOF, major osteoporotic fracture; post-LTx, post–liver transplantation; TAF, tenofovir alafenamide fumarate; TDF, tenofovir diphosphate fumarate.

Replaces risk score if patient is in an HBeAg seroconverted state.

Replace risk score if patient is in an HBsAg seroclearance state.

Major Osteoporotic Fracture

Compared with the general population, patients with HBV infection are at a higher risk of osteoporosis; this risk increases with the presence of coexisting cirrhosis.³⁷ Existing antiviral therapies, such as TDF, may expedite bone mineral loss, further increasing the risk of fractures.³⁸ To capture the deterioration of bone density, and the impact of antiviral therapies, patient risk of fracture was based on the fracture risk profile assigned at baseline (low, medium, or high), which was estimated by the FRAX score, as reported by the Foundation for Osteoporosis Research and Education (Table 1).³⁰ The distributions of risk profiles were estimated using real-world data.³⁹

Major osteoporotic fracture (MOF) risk reduction associated with TAF, as compared with TDF, was incorporated into the model by applying a hazard ratio, derived from the trials.^16,17 Due to limited bone measurement data for ETV, it was conservatively assumed that the fracture risk for ETV is the same as that for TAF. In the base case, it was also conservatively assumed that a previous fracture was not a risk factor for a subsequent fracture. Bone damage was assumed to be cumulative; that is, if a patient switches therapies, bone deterioration will continue from where the previous therapy left off.

Renal Disease

CHB patients are also at a higher risk for renal complications compared with the general population, and antiviral therapies such as TDF may also negatively affect renal function.⁴⁰ Estimated glomerular filtration rate (eGFR) in the model was dependent on both treatment and baseline eGFR. Baseline eGFR at model start was determined from the normal distribution of clinical trial eGFR scores prior to treatment. Patients then experience a decline in eGFR for the first 3 y following initiation of therapy, dependent on their treatment (Table 1). eGFR scores are then used to determine whether a patient has stage III chronic kidney disease (CKD) (eGFR of ≥30 to <60) or end-stage renal disease (ESRD; eGFR of <30). Patients with ESRD have an associated incremental cost, utility decrement, and increased risk of mortality.

Mortality

Background mortality was derived from the gender-specific life tables for the general Taiwan population and was adjusted to exclude deaths caused by viral hepatitis infection.⁴¹ In addition to background mortality, excess mortality due to active CHB was obtained from Iloeje et al.³² It was assumed that virally suppressed patients and patients with low viral loads (i.e., 300–9,999 copies/mL) are not subject to excess mortality. Excess mortality due to treatment-related complications (i.e., stage III CKD, ESRD, and MOF), as well as mortality due to liver-related complications, was obtained from the literature (Table 1).

Resistance Risk and Flare

Annual ETV and TDF resistance rates for TE and lamivudine-resistant patients were obtained from Dakin²⁴ (Supplementary Table S4). Due to limited data, based on expert opinion, it was assumed that resistance rates were 0% for TAF and for TN patients receiving TDF.⁴² Treatment resistance risk is applied for the duration of treatment. If resistance occurs, HBV DNA and ALT levels are set to the uncontrolled levels (i.e., HBV DNA viral load >1,000,000 copies/mL and ALT level >45 IU/mL). It is further assumed that about 5% of patients who developed treatment resistance experience a hepatic flare and that 2% of flare cases may develop liver decompensation.⁴²

Health Utilities

Utility values for active CHB, CC, DCC, HCC, and LTx/post-LTx were based on standard gamble utilities, elicited from infected CHB patients in China (Table 2)⁴³; these utility values were validated by Taiwanese clinical experts. It is assumed that the viral suppression and viral failure (without CC) states have the same utility score as active CHB. Patients with a hepatic flare are assigned a utility decrement equivalent to 1 mo of DCC compared with the active CHB condition. Utility values for HBeAg-seroconverted patients are assumed to be 1% lower than that of the HBsAg seroclearance state. Utility values for patients with HBsAg seroclearance are assumed to have the same quality of life as the general Taiwan population.³¹ Patients either with an MOF or who progress to ESRD had an average utility multiplier applied to their health state utility. The average utility multiplier for MOFs was calculated based on the proportion of fractures occurring at the hip, vertebral, forearm/wrist, weighted for the first year after a fracture and the years following.⁴⁴ The average utility multiplier for ESRD was calculated based on the proportion of patients who are treated with dialysis versus kidney transplant.

Table 2

Health Utilities

Health State	Health Utility	Source
Liver disease health states
No cirrhosis—viremic	0.52	Levy et al. 2008⁴³
No cirrhosis—viral suppression	0.71
Compensated cirrhosis—viremic	0.52
Compensated cirrhosis—viral suppression	0.57
Decompensated cirrhosis	0.26
Hepatocellular carcinoma	0.31
Liver transplantation	0.41
Post–liver transplantation	0.55
Flare	0.02	Assumption
HBeAg seroconverted, by age		Assumption of 1% lower than HBsAg seroclearance
18–29	0.98
30–39	0.98
40–49	0.97
50–59	0.97
60–69	0.90
70–79	0.90
≥80	0.90
HBsAg seroclearance, by age		Levy et al. 2008⁴³
18–29	0.99
30–39	0.99
40–49	0.98
50–59	0.98
60–69	0.91
70–79	0.91
≥80	0.91
Major osteoporotic fracture states
Average utility multiplier—year 1	0.864	Calculated based on Si et al. 2016⁴⁵
Average utility multiplier—year 2+	0.929	Calculated based on Si et al. 2016⁴⁵
Renal disease states
Annual stage III CKD	0.899	Wu and Yang 2014⁴⁶
Average ESRD utility multiplier	0.570	Calculated based on Wu and Yang 2014⁴⁶ and Morton et al. 2009⁴⁷

CC, compensated cirrhosis; CHB, chronic hepatitis B; DC, decompensated cirrhosis; HCC, hepatocellular carcinoma; HBeAg, hepatitis B e antigen; HBsAg, hepatitis B surface antigen.

Costing Inputs

All cost estimates are reported in United States dollars (USD), with the corresponding cost in 2021 Taiwanese dollars (NT$) presented in brackets (using a conversion of 1 USD:31NT$). Cost inputs from previous years were inflated to 2021 costs using the inflation rates from the National Statistics Bureau.⁴⁸

Treatment acquisition costs

Treatment acquisition costs are based on the indicated dosing information from the prescribing information with unit drug cost, obtained from the National Health Insurance Administration of Taiwan in 2022.⁴⁹ A pharmacy dispensing fee of 1.13 USD (NT$35.00) per month was added to total annual treatment costs (Table 3).

Table 3

Cost Inputs

Parameter	Cost^a	Source
Annual treatment acquisition costs
TAF	1,545 USD (NT$47,903)	NHIA⁵⁰
TDF	1,427 USD (NT$44,250)
ETV	1,389 USD (NT$43,063)
Annual liver-related costs
No cirrhosis	4,898 USD (NT$151,832)	NHIRD⁴⁹
Compensated cirrhosis	5,599 USD (NT$173,578)
Decompensated cirrhosis	12,805 USD (NT$396,964)
Hepatocellular carcinoma	13,575 USD (NT$420,833)
Liver transplantation	87,544 USD (NT$2,713,867)
Post–liver transplantation	12,991 USD (NT$402,734)
HBeAg seroconversion	508 USD (NT$15,744)	Spackman and Veenstra 2008⁵¹
HBsAg seroconversion	70 USD (NT$2,176)	Spackman and Veenstra 2008⁵¹
Per event major osteoporotic fracture costs
Hip	2,468 USD (NT$76,513)	NHIRD⁴⁹
Vertebral	2,088 USD (NT$64,739)
Forearm/wrist	961 USD (NT$29,796)
Weighted average cost^b	1,614 USD (NT$50,041)
Annual renal disease costs
Stage III CKD	341 USD (NT$10,566)	NHIRD⁴⁹
Dialysis	17,399 USD (NT$539,378)
Transplant	21,977 USD (NT$681,273)
Per event CHB-related costs
Resistance	4,898 USD (NT$151,832)	Assumption same as no cirrhosis
Flare	89 USD (NT$2,761)	Hung and Chen 2011⁵²
Treatment discontinuation	14 USD (NT$442)	Assumption 16% of flare
Death	41,921 USD (NT$129,954)	NHIRD⁴⁹

CKD, chronic kidney disease; ETV, entecavir; HBeAg, hepatitis B e antigen; HBsAg, hepatitis B surface antigen; NHIRD, National Health Insurance Research Database; NT$, Taiwanese dollars; TAF, tenofovir alafenamide fumarate; TDF, tenofovir diphosphate fumarate; USD, United States dollars.

Converted to NT$ at a rate of 1 USD:31NT$.

Weighted average cost based on the assumption that 16.23% of fractures will be hip, 36.24% will be vertebral, and the remaining 47.53% will be forearm/wrist.

Health state costs

Health state costs were obtained from the NHIRD. Patients with HBV who received ETV, TDF, or TAF between 2018 and 2020 were identified and followed until death or December 31, 2020. Costs were calculated according to the CHB-related health conditions and death. Health state costs included disease management costs for CHB-related health conditions (i.e., DCC, HCC), MOFor CKD (Table 3). MOF costs were incorporated as a weighted average based on the cost of fractures at the hip (2.0% of fractures), vertebral (1.0% of fractures), and forearm/wrist (97.0% of fractures).⁴⁴ Renal disease costs were accounted for stage 3 CKD or ESRD, using a weighted average assuming that 99.6% of patients with ESRD would be treated with dialysis and the remaining 0.14% would receive a transplant.¹⁹ Event costs related to CHB were also included in the model.

Analyses

The proportion of patients experiencing a health event associated with advanced liver disease, along with bone and renal events per 100 patients, were determined for each treatment. Total costs, life-years (LYs), and quality-adjusted life years (QALYs) were calculated for each treatment. The incremental cost-effectiveness ratio (ICER) of TAF versus TDF and ETV was calculated. Deterministic sensitivity analyses are presented for TAF versus TDF and ETV, where parameters were varied individually by ±10%. Probabilistic sensitivity analyses, where all inputs are varied simultaneously over 1,000 simulations, are presented versus TDF and ETV (Supplementary Table S5).

A scenario analysis was explored where it was assumed that all patients would be TN. Additional scenario analyses of time horizons of 3, 10, and 20 y were also explored.

Results

In the base-case analysis, treatment with TAF was associated with fewer liver disease events for CC, DCC, HCC, and LTx along with fewer cases per 100 person-years of stage III CKD, ESRD, and MOF (Table 4). TAF also had higher eAg and sAg seroconversion compared with TDF and ETV. Over the lifetime of a patient, total discounted QALYs were higher with TAF compared with TDF and ETV.

Table 4

Base Case, Health and Effectiveness Outcomes.

	TAF	TDF	ETV
Proportion of patients experiencing liver event
Compensated cirrhosis	34.0%	35.3%	41.9%
Decompensated cirrhosis	7.8%	7.9%	9.3%
Hepatocellular carcinoma	12.8%	16.1%	23.4%
Liver transplants	0.4%	0.4%	0.5%
eAg seroconversion (% of baseline eAG+ve)	62.4%	55.8%	39.7%
sAg seroclearance	6.2%	5.8%	4.5%
Adverse events (per 100 person-years)
Stage III CKD	0.47	1.02	0.67
End-stage renal disease	0.11	0.24	0.16
Major osteoporotic fracture	1.30	1.35	1.31
Effectiveness
Discounted QALYs	9.42	9.00	8.31
Discounted LYs	14.53	14.14	13.63

CKD, chronic kidney disease; ETV, entecavir; HCC, hepatocellular carcinoma; LY, life-year; QALY, quality-adjusted life-year; TAF, tenofovir alafenamide fumarate; TDF, tenofovir diphosphate fumarate.

Liver disease management costs per patient over the lifetime for TAF were higher than those for TDF and for ETV, due to higher disease management costs in the no cirrhosis and HBeAg seroconversion state (Table 5, results in NT$ are presented in Supplementary Table S6). Treatment resulted in lower adverse TAF event costs, mainly driven by the superior efficacy and improved safety profile of TAF.

Table 5

Base Case, Cost Outcomes in USD

Cost Category	TAF	TDF	ETV
Liver disease costs
No cirrhosis	51,852	49,952	46,568
Compensated cirrhosis	11,967	12,062	13,041
Decompensated cirrhosis	1,811	1,823	2,103
Hepatocellular carcinoma	5,399	6,620	9,569
Liver transplantation	181	190	224
Post–liver transplantation	302	312	352
HBeAg seroconversion	502	449	357
HBsAg seroclearance	18	17	14
Resistance	0	185	910
Flare	0	0	1
Treatment switch	2	2	5
Death	2,134	2,191	4,457
Adverse event costs
Stage III CKD	810	1,097	901
End-stage renal disease	536	1,068	753
Major osteoporotic fractures	304	306	288
Drug acquisition costs
Treatment costs	18,496	16,645	12,584
Total costs	94,314	92,921	89,959

CKD, chronic kidney disease; ETV, entecavir; HCC, hepatocellular carcinoma; TAF, tenofovir alafenamide fumarate; TDF, tenofovir diphosphate fumarate; USD, United States dollars.

Compared with both treatments, TAF was more effective and more costly, resulting in incremental cost-effectiveness ratios of USD 3,348 and USD 3,940 per QALY gained versus TDF and ETV, respectively (Table 6, results in NT$ are presented in Supplementary Table S7). When using 1 GDP per capita as a willingness-to-pay (WTP) threshold (which is US$33,059 in 2021 in Taiwan),⁵³ TAF is considered cost-effective versus both TDF and ETV. Based on a fully incremental approach comparing TAF versus ETV and TDF versus ETV, TDF could be ruled out for reimbursement due to extended dominance, with TAF costing $3,940 per QALY gained compared with ETV; the cost-effectiveness plane is shown in Supplementary Figure S2.

Table 6

Cost-Effectiveness Results, TAF versus Other Treatments

	TAF v. TDF	TAF v. ETV
Incremental QALYs	0.42	1.11
Incremental LYs	0.39	0.90
Incremental costs	1,393	4,354
Incremental costs per QALY gained	3,348	3,940

ETV, entecavir; LY, life-year; QALY, quality-adjusted life-year; TAF, tenofovir alafenamide fumarate; TDF, tenofovir diphosphate fumarate; USD, United States dollars.

Results of the deterministic sensitivity analyses for TAF versus TDF and TAF versus ETV are presented in Supplemntary Figure S1. TE progression parameters and drug acquisition costs were among the top drivers for effectiveness and costs, respectively, for both comparisons. The cost-effectiveness acceptability curve showed that treatment with TAF was cost-effective in 92.1% of simulations at a WTP threshold of US$33,059 per QALY (Supplementary Figure S3).

Scenario analyses are presented in Supplementary Table S8. When all patients in the model are TN, the ICER of TAF versus both TDF and ETV increases marginally. Shortening the time horizon decreases the incremental costs of TAF versus TDF, resulting in TAF being cheaper and more effective at time horizons of 3 and 10 y versus TDF.

Discussion

CHB is a lifelong disease that requires long-term or indefinite therapy to prevent liver damage and liver-related complications. While TDF and ETV have reduced liver-related morbidity, both have noted safety limitations. This analysis found that TAF was associated with a lower proportion of patients experiencing CC, DCC, and HCC and requiring LTx along with fewer cases per 100 person-years of stage III CKD, ESRD, and MOF compared with both TDF and ETV. Further, TAF was cost-effective versus both TDF and ETV within WTP thresholds (1 GDP per capita) in Taiwan, despite the availability of ETV at generic prices.

CHB left untreated has a significant clinical burden. Liver cancer is the leading cause of cancer death in Taiwan, accounting for nearly 22% of male cancer deaths, and is the second-leading cause of cancer death in women, accounting for 14% of female cancer deaths.⁵⁴ While Taiwan was the first country to introduce a population-wide vaccination program against hepatitis B, the prevalence of CHB remains high. In addition to vaccination and early detection, proper management of CHB includes early treatment with an antiviral therapy. However, Taiwan reimbursement criteria is more restrictive than international guidelines with duration caps.⁵⁵ Patients who discontinue therapy are at risk of relapse, losing the benefit that treatment conferred.⁵⁶ Maintaining treatment for all patients would move closer to the goal of the World Health Organization of reducing CHB-related mortality by 65% by 2030.⁵⁷

TAF is a relatively recent treatment available for patients with CHB in Taiwan, having been available since 2017 and added to the Taiwanese National Health Insurance Program reimbursement list only in 2019.⁵⁸ As such, there are limited studies evaluating the cost-effectiveness of TAF versus other treatments for CHB. A cost-utility analysis compared front-line TAF and ETV in real-world CHB patients in Taiwan using a decision tree with a Markov model; from a Taiwanese payer’s perspective, the study showed that TAF was dominant versus ETV (lower costs and increased QALYs).⁵⁹ A separate cost-utility analysis of CHB treatments in the United Kingdom using a Markov model demonstrated the cost-effectiveness of front-line TDF versus alternative treatments, including ETV, in nucleos(t)ide-naïve patients with compensated CHB.⁶⁰ By contrast, a patient simulation model was considered appropriate in this study because the impacts of HBV treatments on kidney- and bone-related biomarkers (eGFR, bone mineral density) are known to depend patient baseline characteristics such as age, degree of liver disease, baseline eGFR, and biomarker levels. Notwithstanding, the structure of the current liver disease model remains similar to previously published CHB models. Using a different modeling approach, the current study adds to available evidence and showed that TAF was associated with lower proportions of patients experiencing advanced liver failure (CC, DCC, and HCC and requiring LTx), and demonstrated the cost-effectiveness of TAF compared with both TDF and ETV.

This analysis should be interpreted in the context of its limitations. This model assumed that the probabilities of viral suppression for the second and third year of treatment are similar to those for the first year. However, patients who are virally suppressed are likely to remain suppressed for some time. HBeAg-seroconverted patients were assumed to continue therapy and have a probability of having a viral relapse; as noted above, this is not the current treatment strategy in Taiwan, where patients are no longer reimbursed for CHB treatment after 3 y.⁵⁵ This model assumes that all patients who develop resistance discontinue active therapy and receive BSC. In clinical reality, a patient who fails a given therapy is likely to switch to an alternative therapy or receive combination treatment. While our model may not reflect the current treatment pathway of patients with CHB, it was done to facilitate the individual comparisons of the therapies under consideration. In addition, although aggregated population data are available from the published literature for CHB patients treated with ETV, due to the requirement for patient-level data, this analysis assumed that the HBV DNA-ALT profiles for ETV were the same as those for TDF from published TDF clinical trials and that patient baseline characteristics were generally similar across TAF, TDF, and ETV treatment cohorts. As CHB is a lifelong condition, patients were assumed to adhere to therapy; however, real-life treatment adherence may be suboptimal. Further, there is no consequence of failure of these regimens; thus, the additional treatment sequencing would reflect the same trends in later lines of therapy as in first-line therapy modeled here. Efficacy outcomes were modeled stratified by prior treatment experience with detail on the resulting specific levels of ALT and HBV viral count. Although often preferred, an indirect treatment comparison was not considered in this analysis because such comparisons do not report the detail on treatment efficacy across therapies within specific HBV DNA and/or ALT levels, as was currently done. Additional data on the uncertainty of data inputs is needed to better characterize uncertainty in the cost-effectiveness estimates. Some of the health utilities that were obtained from older published data as sources for HBV-specific patient utilities are limited; this cost-effectiveness analysis should be revisited when additional and newer utility data become available.

TAF leads to better health outcomes at acceptable incremental costs compared with the most commonly used therapies in the management of CHB, thus making it a cost-effective option for the treatment of CHB in Taiwan.

Supplemental Material

sj-docx-1-mpp-10.1177_23814683251328659 – Supplemental material for The Cost-Effectiveness of Tenofovir Alafenamide for Chronic Hepatitis B Virus in Taiwan

Supplemental material, sj-docx-1-mpp-10.1177_23814683251328659 for The Cost-Effectiveness of Tenofovir Alafenamide for Chronic Hepatitis B Virus in Taiwan by Elise Chia-Hui Tan, Alon Yehoshua, Sushanth Jeyakumar, Pongo Peng, Amy Lin, Nathaniel J. Smith and Nandita Kachru in MDM Policy & Practice

Footnotes

Acknowledgements

The authors would like to thank Lianne Barnieh of the Maple Health Group, who contributed to the drafting and revision of this manuscript.

Author Contributions

Elise Chia-Hui Tan: data curation (equal); methodology (equal); writing – review and editing (equal). Pongo Peng: conceptualization (equal); methodology (equal); writing – review and editing (equal). Amy Lin: conceptualization (equal); methodology (equal); writing – review and editing (equal). Nandita Kachru: conceptualization (lead); methodology (equal); data curation (lead); project administration (lead); writing – review and editing (equal). Alon Yehoshua conceptualization (equal); methodology (equal); writing – review and editing (equal). Sushanth Jeyakumar: formal analysis (lead); methodology (equal); validation (equal); visualization (equal). Nathaniel Smith: conceptualization (equal); data curation (equal); methodology (equal); formal analysis (equal); validation (lead); visualization (lead); writing – original draft (lead); writing – review and editing (lead).

The authors declared the following potential conflicts of interest with respect to the research, authorship, and/or publication of this article: Elise Chia-Hui Tan has no conflicts to declare. Pongo Peng, Amy Lin, and Nandita Kachru are employees of Gilead Science and own stocks and shares in Gilead Science. Alon Yehoshua was an employee of Gilead Science at the time of manuscript development. Sushanth Jeyakumar and Nathaniel Smith are employees of Maple Health Group, who received funding from Gilead Sciences for this work. The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: Financial support for this study was provided entirely by Gilead Sciences. The funding agreement ensured the authors’ independence in designing the study, interpreting the data, writing, and publishing the report. The following authors are/were employed by the sponsor at the time of the study: Alon Yehoshua, Pongo Peng, Amy Lin, and Nandita Kachru.

Authorship Statement

Guarantor of article: Nandita Kachru

All authors have approved the final version of the article, including the authorship list.

ORCID iD

Nathaniel J. Smith

References

Lin

Kao

JH.

Perspectives and control of hepatitis B virus infection in Taiwan. J Formos Med Assoc. 2015;114(10):901–9.

World Health Organization. Guidelines for the prevention, care and treatment of persons with chronic hepatitis B infection. [Accessed 28 January, 2016]. https://www.who.int/publications/i/item/9789241549059

Chen

DS.

Hepatocellular carcinoma in Taiwan. Hepatol Res. 2007;37(suppl 2):S101–5.

Hsieh

Kuo

CW.

Cost of chronic hepatitis B virus infection in Taiwan. J Clin Gastroenterol. 2004;38(10 suppl 3):S148–52.

Terrault

Bzowej

Chang

Hwang

Jonas

Murad

MH.

AASLD guidelines for treatment of chronic hepatitis B. Hepatology. 2016;63(1):261–83.

Orr

Thomas

Stow

Eleven year mortality and referral practices associated with abnormal alanine aminotransferase results. Presented at: The International Liver Congress™; April 13–17, 2016; Barcelona, Spain.

Shin

Yeon

Kwon

, et al. Alanine aminotransferase normalization during entecavir treatment and hepatocellular carcinoma risk in hepatitis B virus-associated cirrhosis. Presented at: The International Liver Congress™; April 13–17, 2016; Barcelona, Spain.

Chen

Yang

HI.

Natural history of chronic hepatitis B REVEALed. J Gastroenterol Hepatol. 2011;26(4):628–38. DOI:10.1111/j.1440-1746.2011.06695.x

Gordon

Lim

Liou

, et al. Prevalence of osteoporosis and bone fracture was 2-fold higher in chronic hepatitis B compared to non-chronic hepatitis B patients and continued to increase in the last decade: results of a United States population-based cohort study. Gastroenterology. 2017;152(5):S1198. DOI:10.1016/S0016-5085(17)33991-4

10.

Lampertico

Agarwal

Berg

EASL 2017 Clinical practice guidelines on the management of hepatitis B virus infection. J Hepatol. 2017;67(2):370–98. DOI:10.1016/j.jhep.2017.03.021

11.

Nguyen

Lim

Liou

, et al. Higher and rising prevalence and incidence of renal impairment and chronic kidney disease in chronic hepatitis B patients compared to matched non-chronic hepatitis B controls in the United States: results of a real-world analysis. Gastroenterology. 2017;152(5)(1):S1051–S1052.

12.

Bristol Myers Squibb. Baraclude (entecavir) [package insert]. U.S. Food and Drug Administration website. [Accessed 22 August, 2023]. https://www.accessdata.fda.gov/drugsatfda_docs/label/2010/021797s011lbl.pdf.

13.

Martin

Lau

Nguyen

, et al. A treatment algorithm for the management of chronic hepatitis B virus infection in the United States: 2015 update. Clin Gastroenterol Hepatol. 2015;13(12):2071–2087.e16. DOI:10.1016/j.cgh.2015.07.007

14.

Gilead Sciences, inventor; VIREAD^® (tenofovir disoproxil fumarate) tablets, for oral use and powder, for oral use. US prescribing information. 2017.

15.

Gilead Sciences. Taiwan National Health Insurance Administration of the Ministry of Health and Welfare Approves Vemlidy(R) (tenofovir alafenamide) for reimbursement. Available from: https://www.prnewswire.com/news-releases/taiwan-national-health-insurance-administration-of-the-ministry-of-health-and-welfare-approves-vemlidyr–tenofovir-alafenamide-for-reimbursement-300840369.html [Accessed 22 August, 2023].

16.

Chan

Fung

Seto

, et al. Tenofovir alafenamide versus tenofovir disoproxil fumarate for the treatment of HBeAg-positive chronic hepatitis B virus infection: a randomised, double-blind, phase 3, non-inferiority trial. Lancet Gastroenterol Hepatol. 2016;1(3):185–95. DOI:10.1016/s2468-1253(16)30024-3

17.

Buti

Gane

Seto

, et al. Tenofovir alafenamide versus tenofovir disoproxil fumarate for the treatment of patients with HBeAg-negative chronic hepatitis B virus infection: a randomised, double-blind, phase 3, non-inferiority trial. Lancet Gastroenterol Hepatol. 2016;1(3):196–206. DOI:10.1016/s2468-1253(16)30107-8

18.

National Institute for Health and Care Excellence. NICE health technology evaluations: the manual. NICE Process and Methods [PMG36]. Available from: https://www.nice.org.uk/process/pmg36/chapter/introduction-to-health-technology-evaluation [Accessed 21 May, 2024].

19.

Gilead Sciences. Data on file.

20.

Costa

APM

da Silva

Castro

, et al. PAGE-B and REACH-B predicts the risk of developing hepatocellular carcinoma in chronic hepatitis B patients from Northeast, Brazil. Viruses. 2022;14(4):732. DOI:10.3390/v14040732

21.

Govan

Xin

Hutchinson

Hawkins

Comparative effectiveness of antiviral treatment for hepatitis B: a systematic review and Bayesian network meta-analysis. Eur J Gastroenterol Hepatol. 2015;27(8):882–94. DOI:10.1097/meg.0000000000000376

22.

Shepherd

Jones

Takeda

Davidson

Price

Adefovir dipivoxil and pegylated interferon alfa-2a for the treatment of chronic hepatitis B: a systematic review and economic evaluation. Health Technol Assess. 2006;10(28):iii–iv, xi–xiv, 1–183. DOI:10.3310/hta10280.

23.

Hsu

Chien

Yeh

, et al. Long-term outcome after spontaneous HBeAg seroconversion in patients with chronic hepatitis B. Hepatology. 2002;35(6):1522–7. DOI:10.1053/jhep.2002.33638

24.

Dakin

Fidler

Harper

Mixed treatment comparison meta-analysis evaluating the relative efficacy of nucleos(t)ides for treatment of nucleos(t)ide-naive patients with chronic hepatitis B. Value Health. 2010;13(8):934–45. DOI:10.1111/j.1524-4733.2010.00777.x

25.

Chen

Yang

Iloeje

UH.

Hepatitis B virus DNA levels and outcomes in chronic hepatitis B. Hepatology. 2009;49(5 suppl):S72–84. DOI:10.1002/hep.22884

26.

Liaw

Lin

Chen

Chu

CM.

Natural course after the development of cirrhosis in patients with chronic type B hepatitis: a prospective study. Liver. 1989;9(4):235–41. DOI:10.1111/j.1600-0676.1989.tb00405.x

27.

Veenstra

Spackman

Di Bisceglie

Kowdley

Gish

RG.

Evaluating anti-viral drug selection and treatment duration in HBeAg-negative chronic hepatitis B: a cost-effectiveness analysis. Aliment Pharmacol Ther. 2008;27(12):1240–52. DOI:10.1111/j.1365-2036.2008.03691.x

28.

Crowley

Tognarini

Desmond

Lees

Saal

Introduction of lamivudine for the treatment of chronic hepatitis B: expected clinical and economic outcomes based on 4-year clinical trial data. J Gastroenterol Hepatol. 2002;17(2):153–64. DOI:10.1046/j.1440-1746.2002.02673.x

29.

Chen

Fan

Lee

Makuuchi

Tanaka

Living-donor liver transplantation: 12 years of experience in Asia. Transplantation. 2003;75(3 suppl):S6–11. DOI:10.1097/01.Tp.0000046533.93621.C7

30.

Foundation for Osteoporosis Research and Education. 10-year fracture risk calculator version 2.0. Available from: https://www.sheffield.ac.uk/FRAX/tool.aspx?country=9. Accessed December 14, 2016.

31.

Hung

, et al. Comparison of the efficacy and safety of entecavir and tenofovir in nucleos(t)ide analogue-naive chronic hepatitis B patients with high viraemia: a retrospective cohort study. Clin Microbiol Infect. 2017;23(7):464–9. DOI:10.1016/j.cmi.2017.02.001

32.

Iloeje

Yang

Jen

, et al. Risk and predictors of mortality associated with chronic hepatitis B infection. Clin Gastroenterol Hepatol. 2007;5(8):921–31. DOI:10.1016/j.cgh.2007.06.015

33.

Neuhaus

Angus

Kowalska

, et al. Risk of all-cause mortality associated with nonfatal AIDS and serious non-AIDS events among adults infected with HIV. Aids. 2010;24(5):697–706. DOI:10.1097/QAD.0b013e3283365356

34.

United States Renal Data System (USRDS). 2014 annual data report. Volume 1: CKD. Chapter 3: Morbidity and mortality. 2014. https://www.niddk.nih.gov/about-niddk/strategic-plans-reports/usrds/prior-data-reports/2014

35.

Bliuc

Nguyen

Eisman

Center

. Compound risk of high mortality following osteoporotic fracture and refracture in elderly women and men. J Bone Miner Res. 2013;28(11):2317–24. DOI:10.1002/jbmr.1968

36.

Bermingham

Hughes

Fenu

, et al. Cost-effectiveness analysis of alternative antiviral strategies for the treatment of HBeAg-positive and HBeAg-negative chronic hepatitis B in the United Kingdom. Value Health. 2015;18(6):800–809. DOI:10.1016/j.jval.2015.05.007

37.

Chen

Lin

Kao

CH.

Association between chronic hepatitis B virus infection and risk of osteoporosis: a nationwide population-based study. Medicine (Baltimore). 2015;94(50):e2276. DOI:10.1097/md.0000000000002276

38.

Fong

Lee

Tien

, et al. Improvement of bone mineral density and markers of proximal renal tubular function in chronic hepatitis B patients switched from tenofovir disoproxil fumarate to tenofovir alafenamide. J Viral Hepat. 2019;26(5):561–7. DOI:10.1111/jvh.13053

39.

Liou

Nguyen

Lim

, et al. Longitudinal changes in renal function and comorbidities in treatment-naive and treatment-experienced chronic hepatitis B patients in an US insured population: analysis of a real-world cohort with long-term follow-up. Gastroenterology. 2017;152(5):S1087–8. DOI:10.1016/S0016-5085(17)33665-X

40.

Jung

Jang

Park

, et al. Effect of tenofovir on renal function in patients with chronic hepatitis B. Medicine (Baltimore). 2018;97(7):e9756. DOI:10.1097/md.0000000000009756

41.

Ministry of the Interior Republic of China (Taiwan). Abridged life table of Republic of China. Available from: https://ws.moi.gov.tw/

42.

Gilead Sciences. Panel of national and international CHB key opinion leaders in the United States.

43.

Levy

Kowdley

Iloeje

, et al. The impact of chronic hepatitis B on quality of life: a multinational study of utilities from infected and uninfected persons. Value Health. 2008;11(3):527–38. DOI:10.1111/j.1524-4733.2007.00297.x

44.

Wang

Yang

Shen

Hsiao

FY.

Age- and gender-specific epidemiology, treatment patterns, and economic burden of osteoporosis and associated fracture in Taiwan between 2009 and 2013. Arch Osteoporos. 2017;12(1):92. DOI:10.1007/s11657-017-0385-5

45.

Winzenberg

Chen

Jiang

Neil

Palmer

AJ.

Screening for osteoporosis in Chinese post-menopausal women: a health economic modelling study. Osteoporos Int. 2016;27(7):2259–69. DOI:10.1007/s00198-016-3502-1

46.

Yang

The quality of life and work ability in patients with chronic kidney disease in urban china. Value Health. 2014;17(3):A142.

47.

Morton

Howard

Webster

Wong

Craig

JC.

The cost-effectiveness of induction immunosuppression in kidney transplantation. Nephrol Dial Transplant. 2009; 24(7):2258–69. DOI:10.1093/ndt/gfp174

48.

National Statistics Republic of China (Taiwan). Consumer Price Index Medical Care Services, February 2021. May 1, 2021. Available from: https://eng.stat.gov.tw/

49.

Hsieh

Shao

, et al. Taiwan’s National Health Insurance Research Database: past and future. Clin Epidemiol. 2019;11:349-358. DOI:10.2147/clep.S196293

50.

National Health Insurance. Annual medical expenses report. Available from: https://www.nhi.gov.tw/english/Content_List.aspx?n=1D1ECC54F86E9050%26topn=1D1ECC54F86E9050

51.

Spackman

Veenstra

DL.

A cost-effectiveness analysis of currently approved treatments for HBeAg-positive chronic hepatitis B. Pharmacoeconomics. 2008;26(11):937–49. DOI:10.2165/00019053-200826110-00006

52.

Hung

Chen

HH.

Cost-effectiveness analysis of prophylactic lamivudine use in preventing vertical transmission of hepatitis B virus infection. Pharmacoeconomics. 2011;29(12):1063–1073. DOI:10.2165/11586470-000000000-00000

53.

World Health Organization, Baltussen RMPM, Adam

, et al. Making choices in health: WHO guide to cost-effectiveness analysis. Geneva (Switzerland): World Health Organization; 2003.

54.

Wait

Chen

D-S.

Towards the eradication of hepatitis B in Taiwan. Kaohsiung J Med Sci. 2012;28(1):1–9. DOI:10.1016/j.kjms.2011.10.027

55.

Lim

Amarapurkar

Chan

, et al. Reimbursement policies in the Asia-Pacific for chronic hepatitis B. Hepatol Int. 2015;9:43–51. DOI:10.1007/s12072-014-9593-x

56.

Fan

Hou

Stopping oral antiviral treatment to cure chronic hepatitis B, is it in sight?

Hepatol Int. 2020;14(3):302–304. DOI:10.1007/s12072-020-10039-9

57.

World Health Organization. Global Health Sector Strategy on Viral Hepatitis, 2016-2021: Towards Ending Viral Hepatitis. Geneva (Switzerland): World Health Organization; 2016.

58.

Cheng

Hung

Lee

, et al. Converting from tenofovir disoproxil fumarate to tenofovir alafenamide in patients with hepatitis B following liver transplantation. Ann Transplant. 2023;28:e938731. DOI:10.12659/aot.938731

59.

Lai

Shi

Tsai

Yang

Chiu

SF.

Cost-utility analysis of tenofovir alafenamide and entecavir in chronic hepatitis B patients: a Markov decision model. Cancers (Basel). 2024;16(4):813. DOI:10.3390/cancers16040813

60.

Dakin

Bentley

Dusheiko

Cost-utility analysis of tenofovir disoproxil fumarate in the treatment of chronic hepatitis B. Value Health. 2010;13(8):922–33. DOI:10.1111/j.1524-4733.2010.00782.x

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.16 MB