簡易檢索 / 詳目顯示

研究生: 陳瑋菱
Chen, Wei-Ling
論文名稱: 合併第一期與第二期癌症臨床試驗的藥物療效修正估計
A modified estimation of treatment effect using combined information from Phase I and Phase II oncology clinical trials
指導教授: 杜宜軒
Tu, Yi-Hsuan
學位類別: 碩士
Master
系所名稱: 管理學院 - 統計學系
Department of Statistics
論文出版年: 2009
畢業學年度: 98
語文別: 中文
論文頁數: 34
中文關鍵詞: 3+3設計人體最大耐受劑量截斷二元常態分配最大概似估計
外文關鍵詞: 3+3 design, maximum tolerable dose, truncated bivariate normal distribution, maximum likelihood estimation
相關次數: 點閱:199下載:1
分享至:
查詢本校圖書館目錄 查詢臺灣博碩士論文知識加值系統 勘誤回報
  • 在癌症用藥的臨床試驗中,因為藥的毒性較一般藥物強,因此試驗人數比一般用藥試驗的人數來的少。在第一期試驗中常用3+3設計尋找人體最大耐受劑量,再於第二期試驗探討及估計此劑量之療效。若兩期試驗均有此劑量下的藥物療效記錄,則許多臨床研究者希望合併兩期試驗的資料當作此劑量下的藥物療效估計。然而,3+3設計試驗所觀測的每位受試者對藥物的毒性與有效性反應是一截斷二元常態分配,當療效越大毒性越強時,在第一期試驗所得到的藥物療效通常會低估真正的療效。因此,本論文考慮截斷的影響,利用最大概似估計法修正對藥物療效的估計。

    Cancer drugs are usually more toxic than regular drugs. Therefore, in cancer or oncology dose-finding trial, the sample size is small. 3+3 design is the most common design used in phase I cancer trial to find the maximum tolerable dose (MTD) and followed by phase II trials which aim to evaluate its potential clinical activity. To gain more power, the clinicians attempt to gather the information from two phases to estimate the treatment effect of MTD when the efficacy results in phase I are available. However, the toxicity and efficacy response for each patient collected from 3+3 design follows a truncated bivariate normal distribution. Because of the fact that the efficacy of treatment or drug to cancer usually accompanies with toxicity, the efficacy shows in the phase I trial will underestimate the real one. To overcome this weakness, we construct the maximum likelihood estimate of treatment effect considering the truncation distribution when combine the information from two phases.

    第一章 緒論..........................................1 第二章 文獻回顧......................................3 第一節 3+3設計.......................................3 第二節 截斷二元常態分配的點估計量.......................4 第三章 3+3設計之參數估計..............................7 第四章 模擬分析......................................14 第一節 模擬設計......................................14 第二節 模擬結果比較...................................16 第五章 結論與建議.....................................23 參考文獻..............................................24 附錄一 3+3設計流程圖...................................25 附錄二 推導過程........................................26

    1.Arnold, B. C., Beaver, R. J., Groeneveld, R. A. and Meeker, W. Q. (1993). The nontruncated marginal of a truncated bivariate normal distribution. Psychometrika 58,471-488.
    2.Chow, S. C. and Liu, J. P. (2004). Design and analysis of clinical trials: concepts and methodologies. Wiley-Interscience.
    3.Chevret, S., (2006). Statistical methods for dose-finding experiments. John Wiley and Sons.
    4.Korn, E. L., Midthune, D., Chen, T. T., Rubinstein, L. V., Christian, M. C., and Simon, R. (1994). A comparison of two Phase I trial designs. Statistics in Medicine 13,1799–1806.
    5.Ting, N. (2006). Dose finding in drug development. Springer.
    6.Storer, B. E. (1989). Design and analysis of Phase I clinical trials. Biometrics 45, 925–937

    下載圖示 校內:2013-07-22公開
    校外:2013-07-22公開
    QR CODE