強化設計標的臨床試驗下處理效應之統計推論

Statistical Inference on Treatment Effects for Targeted Clinical Trials under Enrichment Design

林志榮(Jr-Rung Lin)

指導教授 ： 劉仁沛

國立臺灣大學/生物資源暨農學院/農藝學研究所/博士(2008年)

https://doi.org/10.6342/NTU.2008.10201

摘要

於人類基因體計畫完成後，疾病的分子標的可被鑑別，因此可以發展出分子標的形式的治療方法。在實務上，標的臨床試驗通常是用來評估個別化臨床處置的可能性及可行性。但是鑑定分子標的之診斷試劑通常並非完全準確，所以有些納入標的臨床試驗的陽性診斷病人實際上可能並沒有此分子標的，因此對於真正擁有分子標的之病人族群而言，標的臨床試驗下之標的療法的療效估計值會有偏差。 因此對於真正擁有分子標的之病人，我們則提出標的療法之無偏推論的統計方法。在強化設計的臨床試驗下，考慮鑑定分子標的之診斷試劑的準確度，我們提出利用EM演算法配合拔靴技術與貝氏方法來做處理效應之推論。運用模擬研究以驗證所得之估計值與檢定程序，並提出實例數據以說明方法的應用。對於推論真正擁有分子標的之病人族群的療效，我們所提出的之估計值及檢定程序為適當的統計方法。

關鍵字

標的臨床試驗 ； 強化設計 ； EM演算法 ； 拔靴法 ； 貝氏估計 ； 陽性預測值

並列摘要

After completion of the Human Genome Project (HGP), the disease targets at molecular levels can be identified. As a result, treatment modality for molecular targets can be developed. In practice, targeted clinical trials are usually conducted for evaluation of the possibility and feasibility of the individualized treatment of patients. However, the accuracy of diagnostic devices for identification of such molecular targets is usually not perfect. Therefore, some of the patients enrolled in targeted clinical trials with a positive result for molecular target might not have the specific molecular targets and hence the treatment effects of the targeted therapy estimated from targeted clinical trials could be biased for the patient population truly with the molecular targets. We develop statistical methods for an unbiased inference for the targeted therapy in the patients truly with the molecular targets. Under the enrichment design, we propose using the EM algorithm in conjunction with the bootstrap technique and the Bayesian method to incorporate the inaccuracy of the diagnostic device for detection of the molecular targets on the inference of the treatment effects. The simulation studies were conducted to empirically investigate the performance of the proposed estimations and testing procedures. Numerical example illustrates the application of the proposed method. Our proposed estimations and testing procedures are adequate statistical methods for the inference of the treatment effects for the patients truly with molecular targets.

並列關鍵字

Targeted clinical trials ； Enrichment design ； EM algorithm ； Bootstrap method ； Bayesian approach ； Positive predictive value

參考文獻

Casciano DA, Woodcock J. (2006). Empowering microarrays in the regulatory setting, Nature Biotechnology. 24: 1103.

Dalton WS, Friend SH. (2006). Cancer Biomarkers–an invitation to the table, Science. 312: 1165-8.

Dempster AP, Laird NM, Rubin DB. (1997). Maximum likelihood estimation from incomplete data via the EM algorithm (with discussion), J. Roy. Statist. Soc. B. 39: 1–38.

Fleiss J.L., Levin B. and Paik M.C. (2003). Statistical Methods for Rates and Proportions, 3rd Ed. John Wiley and Sons, New York.

Geman, S. and Geman, D. (1984). Stochastic relaxation, Gibbs distribution and the Bayesian restoration of images. IEEE Transactions on Pattern Analysis and Machine Intelligence. PAMI-6: 721-741.