- 學位論文
骨質疏鬆患者接受口服雙磷酸鹽類藥物併發顎骨壞死：流行病學、藥物動力學及藥物基因體學
Oral Bisphosphonate-associated Osteonecrosis of the Jaws in Osteoporotic Subjects: Epidemiology, Pharmacokinetics and Pharmacogenomics
摘要
背景 隨著老年人口不斷攀升，骨質疏鬆及其相關骨折的預防與治療，已是臺灣極重要的健康議題之一。雙磷酸鹽類藥物目前廣泛用於骨質疏鬆的治療與預防，然而對長期使用者，卻有造成顎骨壞死與非典型股骨骨折的風險。 一份來自美國南加州大學的研究指出，因骨鬆服用雙磷酸鹽引發顎骨壞死的病例中，多數是亞裔美國人，暗示在亞洲族群可能有較西方族群高顎骨壞死疑慮。然而，臺灣骨鬆病患接受雙磷酸鹽類藥物治療而併發顎骨壞死的風險仍是未知。要降低顎骨壞死的風險策略之一，便是在族群裡找尋併發顎骨壞死的危險因子。所以，我們此計畫的研究目的，一是分別藉由臺大醫院資料與全民健保資料庫，探查臺灣骨鬆病患接受雙磷酸類藥物治療而併發顎骨壞死的發生率與危險因子；二是找尋不同種族間，有不同雙磷酸鹽類藥物併發顎骨壞死發生率的可能機轉。 材料及方法 研究資料取自臺大醫院處方明細資料與臺灣全民健康保險承保抽樣歸人檔。考慮女性停經平均年齡與男性年老性骨鬆年齡，女性50歲以上或男性60歲以上，於研究區間開始使用阿侖磷酸(alendroante)或雷洛昔芬(raloxifene)的病患，分別收案為實驗組與對照組。在臺灣，此兩種藥物的健保給付規範是相同的，而且阿侖磷酸與雷洛昔芬也是臺灣口服骨鬆藥物的前兩名。此外，雷洛昔芬並未有顎骨壞死的個案被報告。在比較阿侖磷酸與雷洛昔芬兩組顎骨壞死的發生率時，我們只分析比較50歲以上女性的資料，因為雷洛昔芬並沒有治療男性骨鬆的支持證據。 雙磷酸鹽類藥物導致顎骨壞死的定義是，在曾經或正接受雙磷酸鹽類藥物而沒有顎面區放射治療過往史的病患，在顎面區出現骨暴露持續八週以上。在以臺大醫院為基礎的研究中，顎骨壞死的病例，是先搜尋由牙科醫師所下疑似牙槽骨發炎的個案，再逐一比對影像與病理紀錄確診。而在使用全民健保資料的研究中，為克服研究期間尚未有藥物併發顎骨壞死的特定診斷碼，所可能導致的錯誤歸類偏差(misclassification bias)，透過由牙科醫師所下疑似牙槽骨發炎的診斷碼，隨後又有用來處理顎骨壞死的手術，只收集需要手術的顎骨壞死個案。 為探查亞洲與西方種族，在接受相同劑量劑型的雙磷酸鹽藥物後，是否藥物動力學指標(pharmacokinetic parameters)表現有所不同，我們向國際臨床試驗資料服務網申請利用口服伊班磷酸(ibandronate)一期臨床試驗 (Roche藥廠試驗代號BP16331與ML20793) 原始資料進行病人層級的次級分析。為了尋找顎骨壞死之風險對偶基因(risk allele)，我們針對在台大醫院治療過的雙磷酸鹽類藥物導致顎骨壞死病患，在說明與知情同意後，收案進行全基因組關聯分析(genome-wide association study; GWAS)的研究。 結果 我們的研究結果顯示，骨質疏鬆病人服用雙磷酸鹽類藥物，與顎骨壞死的發生是有關聯的，且顎骨壞死的發生率會隨著累積使用時間的增加而增加。阿侖磷酸(alendroante)併發顎骨壞死的發生率，由臺大醫院的資料分析估計是每十萬人年分之283，而在使用全民健保資料的研究中則是每十萬人年分之262。在調整控制可能的干擾因素後，阿侖磷酸(alendroante)依舊是顎骨壞死的獨立危險因子，與雷洛昔芬(raloxifene)相較，風險率為7.42 [信賴區間1.02至54.09]。在骨質疏鬆病人服用阿侖磷酸(alendroante)的情況下，拔牙則會增加顎骨壞死9.6倍的風險，且無論藥物的累積使用時間長短都會增加。其他會增加阿侖磷酸(alendroante)併發顎骨壞死的獨立危險因子，還包括年紀大(65歲以上)、累積使用時間長(大於3年)、糖尿病、風濕性關節炎。 在同樣口服每月一次150毫克的伊班磷酸(ibandronate)進行藥物動力學及藥效分析，與歐洲人種相較，臺灣人有較高的藥物濃度-時間曲線面積(area under the concentration-time curve; AUC)及較大的骨再吸收抑制率-時間曲線面積(area under the antiresorptive effect-time curve; AUEC)。較高的藥物濃度-時間曲線面積(AUC)，可能導因於較高的生體可用率(bioavailability)，而這指標則受體型影響，台灣人的體型身高較西方的英國、比利時人矮小。在全基因組關聯分析研究中，我們最終收案了66位雙磷酸鹽類藥物併發顎骨壞死的骨鬆患者。初步結果顯示，protein kinase C-eta基因多形性(rs17098356)可能與顎骨壞死有關，而protein kinase C-eta與血管內皮生長因子(vascular endothelial growth factor; VEGF)訊息鏈有關。這部分的研究仍在進行，我們很快就將驗證這個發現。 結論 與西方人相較，台灣人服用雙磷酸鹽類藥物後，併發顎骨壞死的發生率較高。雙磷酸鹽類藥物併發顎骨壞死的危險因子，包括治療時間較長(大於3年)、拔牙、老年、糖尿病與風濕性關節炎。發生率高的原因可能包括台灣人體型較小，造成藥物生體可用率高，因此血液藥物濃度高、藥效較高。此外，某些基因也可能影響顎骨壞死的發生。
並列摘要
(Background) With rapidly increasing number of elderly, osteoporosis and its associated fractures have become critical health issues in Taiwan. Bisphosphonates (BPs) are widely used for the treatment of osteoporosis. However, there has been concern about osteonecrosis of the jaw (ONJ) and atypical femoral fracture in the long-term users of BPs. A retrospective study conducted at the University of Southern California demonstrated that most reported ONJ patients on oral alendronate as a treatment for osteoporosis were Asian Americans, raising concern about the higher prevalence of ONJ in the Asians versus Western populations after a bisphosphonate therapy. However, the frequency of ONJ among osteoporotic Taiwanese subjects exposed to BPs remains uncertain. One of the strategies to reduce the risk of ONJ is to identify potential contributing factors especially those unique to this Asian population. Therefore, the aims of our study were (1) to explore the incidence and risk factors of BP-related ONJ in osteoporotic patients in Taiwan by utilizing hospital-based data and nationwide health insurance database; and (2) to investigate the possible mechanisms accounting for the ethnic difference in ONJ incidence. (Methods) Data were retrieved from the pharmacy database at National Taiwan University Hospital and from Taiwan’s Longitudinal Health Insurance Database. Considering the mean age of menopause in women and the usual onset of senile osteoporosis in men, women aged 50 years or older or men aged 60 years or older who began taking alendronate or raloxifene during the study period were identified and as the exposure or control group, respectively. In Taiwan, regulations governing the insurance reimbursement for both agents are identical. Alendronate and raloxifene were the top two most commonly prescribed osteoporosis drugs in oral form. Besides, ONJ has not yet been reported with raloxifene. To compare the incidences of ONJ between alendronate and raloxifene groups, we analyzed only female patients aged 50 years or older due to the paucity of evidence to treat male osteoporosis with raloxifene. BP-related ONJ was defined as the presence of exposed bone in the maxillofacial region for more than 8 weeks in persons treated with a BP without radiotherapy to the jaws. ONJ cases were identified by a possible diagnosis codes for ONJ made by dentists and followed by confirmation with radiographic and pathological evidences in our hospital-based study. At the time of the analysis in the nationwide health insurance database, we ascertained only advanced ONJ cases required operative therapy through any dental surgery for ONJ following the diagnosis codes for ONJ made by dentists to overcome a potential misclassification bias resulting from no specific disease code available for drug-related ONJ To investigate whether the pharmacokinetic parameters are different between Asian and Western populations after receiving BP with a similar dosing formula, we utilized anonymised patient level data retrieved from two oral ibandronate phase 1 studies for secondary analysis (Roche sponsored studies, BP16331 & ML20793) through a consortium of clinical study data providers. To discover risk alleles/genes responsible for ONJ through genome-wide association study (GWAS), we prospectively enrolled osteoporotic patients with BP-related ONJ treated at National Taiwan University Hospital after obtaining informed consent. (Results) Our results support the association of ONJ with oral BP in osteoporotic patients. The incidence of ONJ increases with the duration of therapy. The estimated incidence rates of ONJ associated with alendronate were respectively 283 and 262/100,000 person-years in our hospital-based and nationwide population studies. After adjusting for the potential confounders, alendronate remains an independent predictor for ONJ occurrence (hazard ratio, 7.42 [1.02-54.09]) compared with raloxifene. Tooth extraction was associated with a 9.6-fold increased risk of ONJ among patients taking oral alendronate, independent of the duration of BP use. Other risk factors to the development of oral alendronate-related ONJ include advanced age (>= 65 years), drug duration (>= 3 years) and coexisting diabetes and rheumatoid arthritis. In pharmacokinetic study, Taiwanese subjects have higher pharmacokinetic area under the concentration-time curve (AUC) and subsequent pharmacodynamic area under the antiresorptive effect-time curve (AUEC) compared with European subjects after dosing with oral ibandronate of 150 mg per month. The higher pharmacokinetic area AUC may result from greater bioavailability, which was related to body height. For genetic study, we so far enrolled 66 BP-related ONJ osteoporotic patients to investigate the ONJ pathophysiology through GWAS. The preliminary data suggested a polymorphism (rs17098356) of the Protein Kinase C-eta in vascular endothelial growth factor (VEGF) signaling pathway as a risk allele for ONJ. The work is in progress and we will validate this finding in the near future. (Conclusion) The incidence of oral BP-related ONJ in Taiwanese population is higher than those reported in the Western populations. The risk factors to the occurrence of alendronate-related ONJ include longer drug duration (>= 3 years), antecedent tooth extraction, advanced age (>= 65 years), and coexisting diabetes and rheumatoid arthritis. The mechanisms of a higher incidence of alendronate-related ONJ in our Taiwanese population than the Europeans may include higher serum BP concentrations with higher subsequent antiresorptive effects resulting from greater bioavailability, which may be determined by shorter height of Taiwanese compared to the Europeans. In addition, risk genetic loci may give some contribution to the development of ONJ in this Asian population.