Allopurinol相關嚴重皮膚不良反應：臨床結果、風險、預防的成本效果分析研究

Allopurinol-related severe cutaneous adverse reaction: clinical outcomes, risks, and cost-effectiveness of prevention

柯靜華(Jing-Hua Ke)

指導教授 ： 溫燕霞

共同指導教授 ： 許茜甯(Chien-Ning Hsu)

高雄醫學大學/藥學院/藥學系臨床藥學碩士班/碩士(2015年)

摘要

研究背景及目的：
異嘌呤醇(Allopurinol)是目前高尿酸或痛風的第一線治療藥品，但在極少數的病患身上可能會發生嚴重皮膚不良反應。由過去文獻得知人類白血球組織抗原(human leukocyte antigen, HLA) HLA-B*58:01與allopurinol引起的嚴重皮膚不良反應發生有極大的相關性。故此研究有兩個目的：一、評估HLA-B*58:01基因檢測用於臨床上，對於allopurinol引起嚴重皮膚不良反應發生及其他降尿酸藥物處方型態的影響。二、計算allopurinol引起的嚴重皮膚不良反應的直接醫療成本及HLA-B*58:01基因檢測用於allopurinol新使用者的成本效果分析。

研究方法：
(一) 利用橫斷性研究設計，擷取2011年1月至2014年12月期間曾開立allopurinol的新使用者。分析allopurinol引起的嚴重過敏反應的發生率於HLA-B*58:01基因檢測引進(2011年)前後的變化，以及利用時間序列迴歸模式探討新型降尿酸藥物febuxostat的使用前後(1/2010-4/2013和5/2013-12/2014)，其他降尿酸藥物的處方型態的變化。
(二) 利用成本效果分析(cost-effectiveness analysis, CEA)，並透過決策分析模型及醫療系統的觀點，比較服用allopurinol之前接受HLA-B*58:01基因檢測及直接服用其他降尿酸替代藥品的一年的成本及效果，及遞增成本效果比值(incremental cost-effectiveness ratio, ICER)的影響。

研究結果：
(一) 本研究收錄17,532位allopurinol使用者，及其中2,844位病患曾接受過HLA-B*58:01基因檢測。Allopurinol新使用者的百分比由2010年(尚未有基因檢測的期間)23.87%下降至2014年14.28%。在降尿酸藥物處方形態方面，共2,518位病患曾從allopurinol換至其他降尿酸藥物。時間序列迴歸模式的結果顯示，2013年5月之後，病患從allopurinol換至其他降尿酸藥物的百分比明顯大幅度增加 (95% CI, 2.2%, 3.21%)，其中以febuxostat佔最高比例 (77.8%)，而benzbromarone及sulfinpyrazone的百分比則由2010年63.56%及32.98%分別減少至2014年36.44%及11.69%。另外，allopurinol引起的嚴重皮膚不良反應或過敏事件2010-2014年的發生率由0.21%下降至0%。
(二) 在基礎案例(base-case)分析上，有接受基因檢測的嚴重皮膚不良反應的發生率及死亡率是最低的。基因檢測的ICER為每多活一天需多花費1,481,977新台幣，及每多1單位QALY需多花費234,610新台幣。然而，在病患罹患慢性腎臟疾病的同時，基因檢測的ICER低於基礎案例。以台灣的願付費法(willingness-to pay, WTP)為1單位QALY願意付新台幣647,900為例，HLA-B*58:01基因檢測於服用allopurinol之前比起替代藥物是符合成本效果(cost-effective)。

研究結論：
Allopurinol引起嚴重皮膚不良反應的發生率伴隨著基因檢測的使用後逐年減少，同時，allopurinol使用者也逐年降低，且allopurinol使用者轉換至febuxostat的百分比大幅度地增加。在成本效果分析上，比起服用替代藥物，在服用allopurinol之前接受基因檢測是符合成本效果。HLA-B*58:01基因檢測應首先用於有慢性腎臟疾病的病患，或甚至限制allopurinol在有慢性腎臟疾病患者的使用。

關鍵字

異嘌呤醇 ； 嚴重皮膚不良反應 ； 人類白血球組織抗原 ； 基因檢測 ； 成本效果分析

並列摘要

Objective
Allopurinol is first-line urate-lowering therapy (ULT) for gout or hyperuricemia. The strong association of allopurinol-related severe cutaneous adverse reactions (SCARs) and HLA-B*58:01 allele has been demonstrated in the previous studies. Thus, there were two objective in our study: (1) to assess the uses and effects of genotyping screening for allopurinol on everyday clinical practice and clinical outcomes; (2) to estimate the direct medical costs of allopurinol-related SCARs and cost-effectiveness of HLA-B*58:01 test in allopurinol new users.
Methods
(1) A cross-sectional study was conducted for patients newly prescribed with allopurinol between January 2011 and December 2014 from a large medical institution in Taiwan. We analyzed the incidence of allopurinol-related SCARs and changes in the pattern of switching from allopurinol to other ULT before and after introduction of febuxostat.
(2) The decision analytical model over a time period of 1 year and used the perspective of third-party payer in Taiwan to evaluate the effectiveness and economic outcomes among gout patients with or without HLA-B*58:01 screening test prior to prescribing allopurinol or ULT-naïve patients The results were presented as life-day, quality adjusted life-year (QALY), SCAR event, death attributing to SCAR event within in one year, total medical costs of allopurinol-associated medical expenditures, and the incremental cost-effectiveness ratio (ICER).
Results
(1) A total of 17,532 patients were included and 2,844 were tested for HLA-B*58:01. As a proportion of all new allopurinol users declined from average 23.87% in the pre-HLA B*58:01 screening test introduction period (2010) to 14.28 % in 2014. A total of 2,518 allopurinol new users ever switched from allopurinol to other ULT. The proportion of switching from allopurinol to benzbromarone and sulfinpyrazone among switchers reduced from 63.56%, 32.98% in 2010 to 36.44% and 11.69% in 2014 respectively. The switching rate significantly increased after May 2013, the time that febuxostat was available. Among allopurinol new users, incidence of SCARs or hypersensitivity events reduced from 0.21% to 0%.
(2) In base-case analysis, the incidence of SCAR in screening was the lowest. ICER of screening was incremental NT$ 1,481,977 per life-day and NT$ 234,610 per QALY gained relatively. The ICER was lower in patients with comorbid CKD than base-case cohort. Based on the willingness-to pay (WTP) was NT$ 647,900 per QALY in Taiwan, receiving HLA-B*58:01 screening prior to allopurinol was cost-effective compared to alternatives. The study results were sensitive to the probability of alternatives-related hypersensitivity, utility of gout/hyperuricemia, and negative predicted value of genotyping.
Conclusions
The incidence of allopurinol-related SCAR declined after introducing HLA-B*58:01 screening accompanies by allopurinol new users gradually decreased. The available of febuxostat significantly change the pattern of ULT. HLA-B*58:01 screening is cost-effective compared to alternative ULT. However, the priority of genetic screening should be given to allopurinol user with CKD, or the uses of allopurinol need to be restricted in CKD patients.

並列關鍵字

allopurinol ； severe cutaneous adverse reactions ； HLA-B ； genetic screening ； cost-effectiveness analysis

參考文獻

1. Rodriguez-Monguio, R.,M.J. Otero and J. Rovira. Assessing the economic impact of adverse drug effects. Pharmacoeconomics 2003; 21: 623-50.

2. Tsai, T.F. and T.Y. Yeh. Allopurinol in dermatology. Am J Clin Dermatol 2010; 11: 225-32.

3. Markel, A. Allopurinol hypersensitivity and DRESS syndrome. Am J Med 2008; 121: e25; author reply e27.

4. Roujeau, J.C. and R.S. Stern. Severe adverse cutaneous reactions to drugs. N Engl J Med 1994; 331: 1272-85.

5. Mockenhaupt, M. The current understanding of Stevens-Johnson syndrome and toxic epidermal necrolysis. Expert Rev Clin Immunol 2011; 7: 803-13; quiz 814-5.