類風濕性關節炎患者使用抗腫瘤壞死因子α製劑減量及暫緩使用之研究

Abstract

研究背景:類風濕性關節炎（rheumatoid arthritis）的治療過去的二十年有了突破性的發展。在標靶疾病修飾抗風濕病出現後，達標治療(treat-to-target, T2T)的醫療策略變成醫師心中的普遍追求且是容易達到的。但因為標靶疾病修飾抗風濕病藥物藥價昂貴，對國家健康保險支出或個人經濟都是不小的負擔，長期使用也有相關的副作用等疑慮。在患者能長期達到無症狀疾病狀態(remission)後，各治療指引也提出標靶疾病修飾抗風濕病藥物減量的建議。台灣「全民健康保險藥物給付項目及支付標準」也有規定相關的減量及暫緩續用之相關規定。目前低劑量的生物製劑的療效、效果將持續多久尚不明確。因為許多隨機分派試驗納入條件相對嚴格，得出的結論是生物製劑減量是可行的，但暫緩使用會顯著增加疾病復發的風險。因為隨機分配試驗的納入條件和規定與現實世界中病人疾病情形不盡相同，較難反應出到底在常規治療下有多少病人能成功減量或暫緩使用生物製劑。對於最佳的減量時機等研究結果也不一致。台灣也缺乏針對在健保給付限制下生物製劑減量對於病人的疾病的影響。 研究目的:本研究分析類風濕性關節炎患者在抗腫瘤壞死因子α抑製劑減量後能夠維持疾病穩定的時間、探討類風濕性關節炎患者能成功減量抗腫瘤壞死因子α抑製劑及維持較久疾病穩定的預測因子 (predictive factors)，並探討因為減量後疾病復發對於病人恢復疾病穩定的影響。 研究方法 以回溯性世代研究以單一醫學中心病歷資料進行分析，篩選2003/5/2-2017/4/30接受過抗腫瘤壞死因子α抑製劑治療的類風濕性關節炎病人，所有病人以其第一次抗腫瘤壞死因子α抑製劑開方的日期為「進入研究世代日期 (index date)」，並以病人發生死亡、失去後續追蹤或研究結束時間(2020/4/30)為研究終點 (end of study)，且將這些定義為設限 (censor)。在進入研究世代後至病人開始減量為收集病人基本疾病資料的期間則稱為減量前期間 (pre-tapering period)。從病人開始減量至病人有疾病復發為減量期間 (tapering period)。如病人有減量成功並進入停用至復發稱為停用期間 (discontinuation period)。以存活分析抗腫瘤壞死因子α抑製劑減量後疾病可維持穩定的期間中位數。利用比例風險回歸模式分析不同共變項對於類風濕性關節炎維持疾病穩定以及疾病復發的影響。 研究結果:共有89位病人有進行減量，，減量後疾病可維持穩定的期間中位數為40.0 (95% CI=28.6-51.3) 個月。有58位病人在未達停藥前有疾病復發，疾病復發率為65.2 %。有9位病人在減量後可暫緩使用抗腫瘤壞死因子α抑製劑，有1位病人直接暫緩使用抗腫瘤壞死因子α抑製劑，另外有22位病人減量後無疾病復發仍持續用低劑量的抗腫瘤壞死因子α抑製劑。進行單變項比例風險迴歸模式分析發現，沒有併用hydroxychloroquine的組別其在抗腫瘤壞死因子α抑製劑減量後的疾病復發的風險較高，且達統計顯著的差異 （HR=1.85，95% CI=1.08-3.18，p-value=0.03）。針對病人在抗腫瘤壞死因子α抑製劑減量前疾病狀態，相對可以維持血清性緩解維持≥ 2年，沒有維持血清性緩解(serological remission)達≥ 2年會增加疾病復發的風險，且達統計顯著差異（HR=2.42，95% CI=1.40-4.29，p-value<0.01）。在抗腫瘤壞死因子α抑製劑減量前一年，疾病維持穩定且無復發的病人疾病復發的風險也較減量一年前有疾病復發的病人低（HR=0.40，95% CI= 0.23-0.70，p-value<0.01)。在減量前的平均ESR和CRP越高會顯著增加抗腫瘤壞死因子α抑製劑減量後的疾病復發。減量前的平均ESR每增加1 mm/hr時，風險比增加了0.04倍，且達顯著差異（p-value<0.01）。減量前的平均CRP每增加0.1 mg/dL時，風險比增加了1.28倍，且達顯著差異（p-value=0.02）。若進行多變項分析，只有減量前的平均ESR會顯著增加抗腫瘤壞死因子α抑製劑減量後疾病復發的風險。平均ESR每增加1 mm/hr時，風險比增加了0.03倍，且達顯著差異（p-value<0.01）。 研究結論:本論文從真實世界資料分析，發現類風濕性關節炎患者常規使用抗腫瘤壞死因子α抑製劑時，疾病控制狀況和後續是否能成功減量或暫緩使用有相關。給予低劑量抗腫瘤壞死因子α抑製劑維持疾病穩定，將能減少不必要的醫療浪費且能降低副作用。

Background:With the emergence of biologic disease-modifying antirheumatic drugs (bDMARDs), the treatment options of rheumatoid arthritis (RA) have improved tremendously since these 20 years. Treat-to-target strategy, treating the patient aggressively to achieve measured goals, such as remission or low disease activity is now realistic. However, the costs of bDMARDs are high and healthcare resources are under growing economic pressure. Drug reimbursement and pricing policy in Taiwan regulates the use of bDMARDs in RA. Besides, RA patients with symptom-free disease states continuous use of life-long, full dose bDMARD may provide more harms than benefits. Therefore, many treatment guidelines have proposed dose reduction if RA patients reached sustained remission. There are a lot of studies that are dedicated to find out relapse rate if bDMARDs are being tapered or discontinued, also predictive factors of successfully tapered bDMARDs. However, there are no consistent findings. Objective This study aims to analyses time to disease relapse of RA patients if TNF-α inhibitors are being tapered, the predictive factors that are associated with successful taper of TNF-α inhibitors. Also, investigate the effects associated with dose tapering of TNF-α inhibitors. Methods This is a retrospective cohort study with patients who were diagnosed with RA and were treated with TNF-α inhibitors between May 2, 2003 and April 30, 2017 at medical center. All patients are followed the first dose of TNF-α inhibitors (index date) loss follow up or study end (2020/4/30). Baseline characteristics are collected during pre-tapering period. I calculate the disease relapse rate and time to disease relapse using survival analysis (Kaplan-Meier survival curve). Cox-proportional hazards regression model is applied to examine any predictive factors associated with successful taper of TNF-α inhibitors. Repeated measured ANOVA and paired t test are used to measure the effect associated with the tapering of TNF-α inhibitors. Results There are 90 patients included in this cohort study, 89 patients underwent tapering of TNF-α inhibitors. The median of tapering maintenance period is 40.0 (95% CI 28.6-51.3) months. Fifty-eight patients had disease relapse before they can discontinue TNF-α inhibitors. Disease relapse rate of tapering of TNF-α inhibitors is 65.2%. There are 9 patients succeeding in tapering and discontinuing TNF-α inhibitors, and 1 patient directly discontinued TNF-α inhibitors. There are 22 patients maintaining low dose TNF-α inhibitors. Using univariate cox proportional regression model, no concomitant hydroxychloroquine (HR=1.85, 95% CI=1.08-3.18, p-value=0.03), and no sustained serological remission for ≥ 2 years (HR=2.42, 95% CI 1.40-4.29, p-value<0.01) are associated with increased risk of disease relapse, while no pre-taper flare within 1 year before TNF-α inhibitors being tapered (HR=0.40, 95% CI 0.23-0.70, p-value<0.01) is associated with decreased risk of disease relapse. Higher ESR and CRP before TNF-α inhibitors being tapered are associated with increased risk of disease relapse. However, the multivariate cox proportional regression model was performed, only increase of ESR before TNF-α inhibitors being tapered was associated with increase risk of disease relapse. The hazard rate increases by 0.03 with the increase of ESR by 1 mm/hr. Fifty-eight patients with disease relapse had significant increase of ESR even after TNF-α inhibitors dose was escalated for rescue. Conclusion This is the first time to show mandatoy dose reduction from real-world evidence that the disease control status under standard use of TNF-α inhibitors is associated with successfully tapering or discontinuing TNF-α inhibitors in the RA patients. Low dose TNF-α inhibitors which can sustain disease stable would have the benefit of decreasing unwanted adverse effects and reducing unnecessary health resources.