iStent inject? and cataract surgery for mild-to-moderate primary open angle glaucoma in Japan: a cost-utility analysis

INTRODUCTION

Glaucoma is a chronic, progressive disease resulting in a degeneration of the optic nerve. It is a leading cause of irreversible blindness in Japan

. In 2015, a nationwide survey of welfare offices in Japan ranked glaucoma as the first causative disease (29%) among newly certified visually impaired individuals ≥18y

. Primary open angle glaucoma(POAG) is the most common type of glaucoma

. It is characterized by a progressive vision loss due to the loss of retinal ganglion cells and optic nerve damage and by optic neuropathy combined with ocular hypertension (OHT)

.Population-based studies from the early 2000s estimated the prevalence of POAG in Japan to be 4% among adults aged ≥40y

. Due to the asymptomatic nature of mild-tomoderate POAG, 93% of patients with POAG in Japan remains undiagnosed and not treated

. The epidemiological study conducted between 2000 and 2002 found a high rate of newly diagnosed cases of glaucoma (89%)

despite the emphasis of the Japanese Glaucoma Society Guideline

on the importance of an early diagnosis and treatment for avoiding permanent vision loss.

Scenario analysis considering the full caregiver burden from a societal perspective was conducted to test the robustness of results.

In this cost-utility analysis, the iStent

inject

with cataract surgery strategy was found to be cost-effective

cataract surgery alone in patients with mild-to-moderate POAG needing cataract surgery. At the WTP threshold of ￥5 000 000/QALY, the iStent

inject

with cataract surgery strategy was found to have at least a 90% probability of being cost-effective.The results are considered robust based on one-way sensitivity analyses. Considering the loss of productivity of family caregivers and caregiver burden, the iStent

inject

with cataract surgery strategy was found to dominate cataract surgery alone and save costs associated with loss of productivity of working family caregivers and caregiver burden in patients needing cataract surgery. When the cost of iStent

inject

with cataract surgery increased from the original input of ￥279 900 to￥516 300, the ICUR reached the WTP threshold ￥5 000 000/QALY gained.This is the first cost-utility analysis comparing the iStent

inject

with cataract surgery

cataract surgery alone in the Japanese patients with mild-to-moderate POAG. The model inputs were obtained from the Japanese and global data with validation by Japanese clinical experts to represent the local Japanese setting.The results from this model are consistent with previously published studies that used the same model adapted to different settings, including France

, Canada

and Spain

. A similar cost-utility analysis to the present analysis assessed combined MIGS of one or two trabecular micro-bypass stents with cataract surgery in German patients with POAG

. However,the analysis was conducted separately in two subgroups of patients with moderate and advanced POAG. The study found that in the moderate stage, the implementation of two stents during cataract surgery produced the highest effectiveness and the lowest ICUR among cataract surgery combined with three alternative MIGS methods: 1) one trabecular micro-bypass stent, 2) two stents, and 3) intracanalicular scaffold, compared with cataract surgery alone. These findings reinforce the benefit of using iStent

inject

on early stages of POAG.

SUBJECTS AND METHODS

A Canadian health state-transition Markov model

was adapted to estimate the cost-utility of iStent

inject

combined with cataract surgery compared with cataract surgery alone in one eye in patients with mild-to-moderate POAG over lifetime horizon with monthly cycle length from the perspective of Japanese public payer. Health outcomes included quality-adjusted life-year (QALY) as the primary outcome and life years and number of blind eyes as the secondary outcomes. Cost outcomes included a total cost as the primary outcomes, while secondary outcomes were surgery(cataract surgery with or without iStent

inject

, trabeculotomy,trabeculectomy), medication, progression-related medical service (physician consultation, test), and AEs (hyphema,hyperaemia, stent obstruction, medication for AEs). The full caregiver burden was evaluated in the scenario analysis and considered productivity loss of working family caregivers and caregiver burden proxied by long-term care insurance level 1.Costs and health outcomes were discounted at a 2.0% annual rate based on the Japanese guideline

.

Both deterministic and probabilistic sensitivity analyses were conducted to evaluate the impact of assumptions used in the model and the variability surrounding model inputs. The deterministic one-way sensitivity analysis(OWSA) was conducted for the deterministic base case to determine the significant drivers of cost-utility. The 95%confidence interval was used as lower and upper bounds of the one-way sensitivity analysis. When not available, a ±25%variation of the deterministic base value was applied for the low and high values. Probabilistic base case analysis was conducted using 1000 iterations from random draws of the underlying parameter uncertainty. A beta distribution was used for proportion and utility values; a gamma distribution was considered for costs; a lognormal distribution was used for healthcare resource uses; and a normal distribution was considered for clinical data. The probabilistic base case analysis was expressed as ICUR scatterplot and costeffectiveness acceptability curve (CEAC). The willingness-topay (WTP) threshold of ￥5 000 000/QALY in Japan, set by the Central Social Insurance Medical Council (Chuikyo), was used as a marker to judge the cost-utility

.

窗外明媚的秋光從窗簾縫隙里鉆進來，空中懸浮著灰塵。這世界滿是塵埃，漂浮在空氣中或停留在某一處，看似不存在，其實隨處可見。

The model structure has previously been published

and was validated by Japanese clinical experts to reflect current clinical practice in Japan (Figure 1). Patients with mild-to-moderate POAG treated with background ocular hypotensive medications entered the model initiating cataract surgery with or without iStent

inject

. Patients could progress from baseline severity levels to next severity levels, defined according to the visual field (VF) defect (decibels, dB)

as mild glaucoma (0 to 6 dB), moderate glaucoma (6.01 to 12 dB),advanced glaucoma (12.01 to 20 dB), and severe glaucoma or blindness (＜20 dB). Patients could discontinue background medication due to non-adherence, contraindications and intolerable AEs and receive subsequent surgeries in the case of disease progression such as trabeculotomy, followed by trabeculectomy as the last surgery. AEs of background medication, such as dryness, redness, and blurred vision and AEs of cataract surgery and iStent

inject

combined with cataract surgery, including stent obstruction, hyperaemia, and hyphema, were considered.

Inputs of the model, including patients’characteristics, clinical data, utility, and costs, were drawn from clinical trials, the literature and official Japanese sources.The Japanese data were used whenever available. Japanese clinical experts validated all assumptions and data used in the model.

改革開放前的相當長一段時期，我國新聞出版業重生產輕市場、重出版輕發行的傾向十分明顯。不僅出書品種少，書報刊的流通銷售渠道更是不暢，新華書店作為唯一的合法發行力量，遠遠無法滿足群眾日益旺盛的購書需求。1982年文化部發布《關于圖書發行體制改革工作的通知》，提出我國圖書發行體制改革的總目標：在全國組成一個以國有新華書店為主體，多種經濟成分、多條流通渠道、多種購銷形式、少流轉環節的圖書發行體系，即“一主三多一少”。這項改革打破了新華書店對圖書發行權尤其是批發權的長期壟斷，極大地刺激了民營書商的發展。據統計，到1987年年底，非國有書店從無到有，發展到1萬多家，數量是國有書店的1.18倍。[2]

To reflect the Japanese real-world setting, the model was populated with clinical characteristics collected from the Japanese cross-sectional study

. Patients needing cataract surgery entered the model at a mean age of 64.5y, 60.5% in mild and 39.5% in moderate health state.In the absence of other glaucoma epidemiology in Japan,data were obtained from the Early Manifest Glaucoma Trial(EMGT) conducted in the US

. The relative risk of mortality was obtained from an Australian cohort

.

Three types of clinical data were included in the model: treatment pattern, efficacy, and safety. As no Japanese data were identified, global data from the original model were used. In the absence of treatment pattern for glaucoma in Japan, the VF defect at entry (-3.0 dB for mild patients and -6.0 dB for moderate patients) and during progression (-0.05 dB natural decline rate per month in untreated patients), mean time to receive subsequent surgeries and hazard ratio of receiving subsequent surgeries per unit of IOP reduction compared with no IOP change (0.83) were obtained from the EMGT

and expert opinions. The time to background medication discontinuation (59.53mo) was based on expert opinions. The IOP reduction, utilisation of background medication, and probabilities of AEs caused by cataract surgery and subsequent surgeries were based on an RCT comparing iStent

inject

combined with cataract surgery with cataract surgery alone

. Probabilities of AEs caused by background medication was obtained from a cost-utility analysis investigating the long-term health and economic outcomes of direct pressure-lowering medication for OHT

.

In the absence of Japanese data, global data from the original model were used

. Utility values of patients in different severity levels and disutility values due to subsequent surgeries and background medication for AEs were obtained from a Dutch cross-sectional survey assessing the impact of VF defect on POAG patient utility values

. Health preference was measured by the Health Utilities Index 3 (HUI-3) using tariffs for the Canadian population. In the absence of disutility value for trabeculotomy, the same disutility value for trabeculectomy was assumed for trabeculotomy with confirmation from the clinical experts.

The ICUR scatterplot in probabilistic base case analysis is shown in Figure 2. iStent

inject

with cataract surgery strategy was found to produce higher QALYs in 97.8% of the iterations. All the probabilistic simulations suggest iStent

inject

is associated with an increase in costs.Figure 3 presents the results of CEAC. At the WTP threshold of ￥5 000 000 per QALY gained, the iStent

inject

with cataract strategy was found to have a 90% probability of being cost-effective. Figure 4 presents the results of OWSA.The top key drivers of ICUR were an IOP reduction at 2y due to cataract surgery alone, medication reduction at 2y due to cataract surgery alone and utility value of mild glaucoma.

For background medication, the cost element consisted of actual medication costs and medication service costs, including fees of prescription, basic dispensing, dispensing, and drug management instruction. According to the Japanese Glaucoma Society guidelines

, Japanese published study

and clinical expert opinions, four categories of drugs are currently used as standard medications in Japan: prostaglandin (PG) as the 1

line, beta-blocker (BB) and combination of PG and BB as the 2

line, and more than 2 combinations of carbonic anhydrase inhibitors (CAI)/BB+PG medication as the 3

line.Market shares of medications were obtained from Inoue

and expert opinion. Unit costs of medications were obtained from the MHLW

. Unit cost of medication service per bottle were obtained from the MHLW

. Based on the expert input, medical wastage was not considered. To treat AEs caused by background medication, patients need to consult an ophthalmologist (1 time/mo) and have some tests run,

slitlamp microscopy (1 time/mo) and Goldmann applanation(2 times/mo). The healthcare resource use and frequency of treatment of AEs caused by surgeries are presented in Table 1.

網絡密度指網絡中各節點之間聯系的緊密程度，節點之間聯系越多，網絡密度就越大[12]。經計算分析，三峽旅游流的網絡密度為0.248 3，表明大三峽旅游地區景區存在一定的網絡聯系，但部分節點聯系不緊密。從節點中心性來看，三峽旅游流主要向解放碑、白帝城、小三峽、神女峰、三峽大壩、三峽人家、恩施大峽谷幾個景區聚集，以傳統三峽旅游游線景區為主，與三峽腹地景區聯系較少，與三峽周邊極富吸引力的旅游目的地如九寨-黃龍聯系較少，與其他景區節點的旅游聯系也較……