Voriconazole用於治療或預防侵入性黴菌感染之9年療劑監測經驗

Abstract

背景 Voriconazole為第二代廣效型抗黴菌藥品，為治療麴菌感染之第一線用藥，可以通過血腦障壁（blood brain barrier，BBB），因此對於中樞之黴菌感染也有重要地位。由於使用病人族群病情複雜、代謝酵素CYP2C19基因多型性、潛在藥品交互作用等等，個體間與個體內血中濃度差異大，過去多數研究顯示voriconazole的療效與副作用皆與血中谷濃度（trough concentration，Ctrough）有關，目前各國治療指引建議使用此藥品之病人需進行療劑監測（therapeutic drug monitoring，TDM），大部分建議濃度落在1-6 mcg/mL之間。由於非線性之藥品動態學特性，臨床上根據血中谷濃度調整voriconazole劑量相當困難，目前對於如何調整劑量還沒有定論與標準，且關於療劑監測與劑量調整的文獻，大多樣本數較少。近期臺灣大學醫學院附設醫院（簡稱臺大醫院）根據過去臨床經驗發展了劑量調整之準則，因此本研究分析過去9年間在臺大醫院進行voriconazole療劑監測之結果，探討其與療效及安全性之相關性，以評估目前臺大醫院之劑量調整準則之適用性。 研究目的 研究voriconazole療劑監測結果與調整劑量方式、劑量調整與濃度變化的關係，並評估目前臺大醫院之調整劑量準則，探討可能影響血中濃度之因子，分析血中濃度與臨床療效及安全性的關係。 研究方法 本研究對象納入18歲以上，於臺大醫院前瞻性收案使用voriconazole同意參加試驗且已簽署同意書之病人，排除條件包括資料不全、檢體錯誤或並非血中谷濃度、病人在納入試驗之療程前12週內有使用voriconazole等。研究收案期間為2009年3月1日至2017年12月31日。資料來源包括病人口述、電子與紙本病歷、本實驗室測得之抽血濃度結果、臺大醫療體系醫療整合資料庫（Integrated Medical Database, National Taiwan University Hospital, NTUH-iMD），蒐集病人基本資料、使用voriconazole前後之實驗室檢驗值、voriconazole處方型態等。 Voriconazole之理想目標濃度訂為1-6 mcg/mL，待病人使用 voriconazole 5-7天達穩定血中濃度狀態後，研究人員在病人服用（或注射）下一劑劑量前量測血中谷濃度；以高效能液相層析儀系統（high-performance liquid chromatography, HPLC）測定病人血中濃度。療效評估分為12週內全原因死亡與感染相關死亡率，安全性則評估在用藥期間是否發生肝臟不良反應。 統計方法以Student’ s t test或Mann-Whitney U Test比較兩組連續性變項間之差異，以ANOVA或Kruskal Wallis U test進行三組連續性變項之比較，以Chi-square test或Fisher’s exact test檢定類別變項，以logistic regression分析可能影響voriconazole血中濃度之因子。以cox proportional hazard regression進行存活分析。檢定以雙尾檢定，若p-value<0.05則視為有統計上顯著意義。 研究結果 本研究排除給藥劑量或方式不符合條件、抽血時間錯誤、非全血檢體、抽血日期未達到穩定狀態（steady-state）、資料嚴重缺失者，共納入670人，2339筆療劑監測資料。病人年齡中位數為歲55.4歲，以血液疾病病人為主（78.2%），主要黴菌感染部位與致病黴菌以肺部及麴菌（Aspergillus spp.）感染為主。每人進行療劑監測次數之中位數為2次，最多達29次。在用藥穩定狀態後第一次進行療劑監測時，共有67.0%達到理想濃度範圍（1-6 mcg/mL）。 針對濃度不在理想範圍內之病人，依照臺大醫院voriconazole劑量調整準則調整，約有61-86%可以達到理想濃度，相較於不調整（僅28-64%達到理想濃度）或是其他方式調整者（43-67%達到理想濃度），有較高之達成率，與不調整者達到統計學上之顯著差異（p <0.01），但與以其他方式調整者則沒有統計上顯著差異。依照濃度分組後，血中濃度愈高，每日減少劑量100 mg造成的影響愈大，濃度4 mcg/mL以下、4-6 mcg/mL、6-8 mcg/mL、以及高於8 mcg/mL之濃度減少分別為0.7-1.1 mcg/mL、1.2-2.0 mcg/mL、2.6-3.4 mcg/mL與2.2-4.3 mcg/mL。在濃度4 mcg/mL以下者，每增加100 mg/day的劑量時，血中濃度平均增加0.8-1.3 mcg/mL。相同劑量頻次下，途徑轉換時，口服與輸注之間濃度變化顯示吸收率遠低於90%。 分析影響血中谷濃度相關之因子，結果顯示在校正其他因子後，體重增加、使用靜脈輸注、每日給藥劑量（mg/kg/day）增加、具有肝臟疾病、肌肝酸清除率（creatinine clearance，CLCr）減少、C-反應蛋白（C-reactive protein，CRP）增加與血中濃度較高者有高度相關性；合併使用降低voriconazole血中濃度之藥品則與血中濃度偏低者有高度相關性。 療效方面，以cox proportional hazard model分析12週內全原因死亡率與感染相關死亡率相關因子，顯示療程濃度、靜脈輸注給藥、腎功能變差、觀察期間內曾休克與死亡風險顯著相關。肝臟不良反應以膽汁淤積型（cholestatic type）或混合型（mixed type）為主，在血中濃度以3或4 mcg/mL為切點時，肝臟不良反應發生率在濃度大於切點時顯著增加，若以其他濃度為切點則沒有顯著差異。 結論 本研究評估之臺大醫院劑量調整準則，與沒有調整之組別（約28-63%達理想濃度）或是依照其他方法調整（約43-67%達理想濃度）相比，依照準則進行調整後，達到理想濃度之比例約61-87%較高。依不同濃度分組時，若每天增加或降低100 mg/day，在濃度小於4 mcg/mL者，濃度變化量約1 mcg/mL左右，隨著濃度上升，濃度變化量也越來越大。血中濃度影響因子，體重增加、使用輸注途徑、每日劑量增加、具有肝臟疾病、腎功能不佳、CRP數值上升，與濃度較高者有顯著相關；併用使voriconazole濃度下降藥品與濃度較低者有顯著相關。肝臟不良反應在濃度以3或4 mcg/mL為切點時有顯著差異。

Background Voriconazole, a second-generation broad-spectrum antifungal agent, is first-line antifingal agents for invasive Aspergillus infections. It is able to pass blood brain barrier (BBB), so it’s also an important agent in the treatment of central nervous system infection. Voriconazole has significantly variable intra- and inter-individual plasma concentration because of complicated condition in these patients, including genetic polymorphism of CYP2C19 and potential drug-drug interaction, etc. Previous studies showed the trough concentration (Ctrough) of voriconazole had high correlation with efficacy and safety outcomes. Therefore, therapeutic drug monitoring (TDM) of voriconazole has been strongly recommended. Preferred range of Ctrough is approximately 1-6 mcg/mL, but it’s slightly different in the studies and the treatment guidelines. However, there is also no consensus on dosage modification guidance in response to voriconazole concentration results. Recent studies included relatively small sample size. Therefore, we analyzed our 9-year experience of voriconazole TDM in a large patient population and assessed the effectiveness and safety outcomes to evaluate current protocol of dosage adjustment developed by National Taiwan University Hospital (NTUH). Objectives This study was designed to analyze the results of TDM, the dosage adjustment patterns of voriconazole, and the distribution of concentration before and after dosage adjustment. In addition, the study was aimed to evaluate NTUH dosage adjustment protocol and to explore factors which affected the concentration. The effectiveness and safety outcomes were also assessed. Methods We conducted this study with prospective patient enrollment and retrospective medical data collection in NTUH from March 2009 to December 2017. Patients who were more than 18 years old, received voriconazole, and signed the informed consents with plasma concentration measurements were enrolled. Incomplete records, inappropriate sample collection, and patients had used voriconazole during 12 weeks before the course of study were excluded. Medical data were recorded from medical records and the Integrated Medical Database, National Taiwan University Hospital (NTUH-iMD). Plasma Ctrough of voriconazole were measured at steady state (5-7 days after the initiation of voriconazole) and before the next dose administration. Ctrough were determined by high performance liquid chromatography (HPLC). We used the target range of Ctrough between 1 and 6 mcg/mL. The all-cause mortality and infection-related mortality during 12 weeks after the start of voriconazole therapy were evaluated for effectiveness. All liver adverse events that occurred during voriconazole therapy were recorded as safety outcomes. Continuous variables were compared by Student’s t test or Mann-Whitney U Test. Categorical variables were compared by Chi-square test or Fisher’s exact test. Three continuous variables were compared by analysis of variance (ANOVA) or Kruskal Wallis U test. Factors which influenced the Ctrough was analyzed with a logistic regression model. Survival analysis was analyzed with cox proportional hazard model. Two-sided P-values < 0.05 were considered statistically significant. Results During the 9-year study period, a total of 670 patients with 2339 trough levels were included. Median age of study population was 55.4 years old. Approximately 78.2% patients had hematological diseases. The median of 2 samples for TDM per patient were analyzed. There were 67.0% of the first Ctrough within target level of 1-6 mcg/mL. Dosage adjustment based on NTUH protocol in response to out-of-target levels resulted in approximately 61-87% of target achievement in the subsequent TDM. It had statistically significant high achieving rate compared to the group without dosage adjustment (p<0.01), but no statistically significant difference compared to the group not abiding by protocol. When dose was decreased by 100 mg/day, the mean decrease of Ctrough were 0.7-1.1, 1.2-2.0, 2.6-3.4, and 2.2-4.3 mcg/mL in the groups of Ctrough< 4, 4-6, 6-8, and > 8 mcg/mL, respectively. When dose was increased by 100 mg/day, the mean increase of Ctrough were 0.8-1.3 mcg/mL in the group of Ctrough< 4 mcg/mL. Increased body weight, intravenous infusion, daily dose (mg/kg/day), liver disease, decreased renal function, and increased C-reactive protein (CRP) had statistically significant correlation with higher Ctrough. Concomitant medications which decrease voriconazole level had statistically significant correlation with lower Ctrough. Factors had statistically significant correlation with 12 weeks of all-cause and infection-related mortality were Ctrough¬ of course, intravenous infusion, decreased renal function, and shock during the course. Most of liver adverse events were cholestatic or mixed type hepatotoxicity. The crude risk of liver adverse events were statistically significant in the Ctrough > cut-off concentrations when the cut-off concentration at 3 mcg/mL or 4 mcg/mL. Conclusions Dosage adjustment based on the NTUH protocol in response to previous out-of target range had higher rate to achieve target level (about 61-87%) compared with group without dosage adjustment or by other strategies. Route, daily dose adjusted by body weight, liver disease, CRP, and drug-drug interaction were statistically significant factors related to Ctrough . The risk of liver adverse events were statistically significant high when the concentration > 3 mcg/mL.