第一期及第二期臨床試驗之合併設計

林意評(Yi-ping Lin)

指導教授 ： 陳玉英

國立中央大學/理學院/統計研究所/碩士(2015年)

摘要

在新藥研發中，為降低臨床試驗第一期與第二期的成本及其耗費的時間，本文進行合併第一期與第二期臨床試驗設計之研究。傳統上，第一期臨床試驗是在劑量限制毒性(dose-limiting toxicity, DLT)設定下尋求藥品的最大耐受劑量(maximal tolerate dose, MTD)，試驗時間約費半年；第二期臨床試驗則是針對上述MTD進行藥品短期療效的評估，主要的研究終點是二元化資料，試驗花費時間約一年半。本文建議的合併設計則為兩階段設計，其中第一階段以安全為考量，應用貝氏方法，決定暫時性的MTD；第二階段同時考量安全性與有效性，以批次試驗進行，應用貝氏方法彈性調整施予下一批試驗者的藥品劑量。此一合併設計不但考量晚發型毒性反應，也能彈性調整MTD，期望能在較短時間內決定無效藥品或讓有短期效果的藥品進入第三期臨床試驗。最後本文利用模擬方法，在劑量毒性及劑量效應模式及不同的毒性反應與有效反應時間分布下，研究此一合併設計所需的試驗人數及時間。

關鍵字

合併臨床試驗 ； 貝氏方法

並列摘要

In the development of a new drug, to reduce the cost and shorten the time of the phase I and II clinical trials, a study of combining the above two phase trials is investigated. Traditionally, the phase I trial which takes about 6 months is to find the maximal tolerate dose (MTD) of the drug under study under specific dose-limiting toxicity (DLT). The phase II clinical trial, taking about 18 months, is to evaluate the short term effectiveness of the drug at the MTD above, based on binary outcomes. This thesis considers a two-stage combined trial. The first stage concerns with the safety of the drug, where the temporary MTD is determined by applying the Bayesian method. The second stage considers both the safety and effectiveness which is proceeded in batch and the dose applied to next batch is adjusted based on a Bayesian analysis. Note that the two-stage design may take care of the late-onset toxicity since the MTD is adjusted at the second stage. Through an appropriate design, the proposed trial may save time with less patients for futility drugs or effective drugs. Finally, a simulation study is conducted to investigate the required number of patients and the needed time for implement the combined trials under a variety of toxicity-dose and dose-response models under different distributions of reaction times to toxic or effectiveness.

參考文獻

1. Babb J, Rogatko A and Zacks S. Cancer phase I clinical trials: Efficient dose escalation

with overdose control. Statistics in Medicine 1998; 17:1103–1120.

2. Bartroff J, Lai TL and Narasimhan B. A new approach to designing phase I-II cancer trials

3. Chi Y and Chen CM. Curtailed two-stage designs in phase II clinical trials. Statistics in

4. Hastings WK. Monte Carlo sampling methods using Markov chains and their