| DC 欄位 |
值 |
語言 |
| DC.contributor | 統計研究所 | zh_TW |
| DC.creator | 吳宥蓁 | zh_TW |
| DC.creator | Yu-Jen Wu | en_US |
| dc.date.accessioned | 2021-8-3T07:39:07Z | |
| dc.date.available | 2021-8-3T07:39:07Z | |
| dc.date.issued | 2021 | |
| dc.identifier.uri | http://ir.lib.ncu.edu.tw:88/thesis/view_etd.asp?URN=108225021 | |
| dc.contributor.department | 統計研究所 | zh_TW |
| DC.description | 國立中央大學 | zh_TW |
| DC.description | National Central University | en_US |
| dc.description.abstract | 在研發分子標靶藥以治療癌症的早期臨床試驗中，目的是估計該藥的最佳生物劑量，亦即服用該劑量的病患發生劑量限制毒性的機率低於設定的目標毒性機率且具有最佳療效。本文提出早期臨床試驗的兩階段設計，第一階段試驗應用目前的連續重評估設計或鍵盤設計，根據二元毒性反應資料進行劑量升降與配置，尋找藥物的最大耐受劑量，藉以決定具有可接受毒性的劑量集合；第二階段進行世代試驗，以疾病無惡化時間為療效指標，根據先前藥物治療的疾病無惡化中位數時間之下的存活函數衡量劑量之藥效。第二階段的世代試驗使用交集-聯集檢定鑑別藥效低於二分之一的無效劑量，因此若最低劑量毒性過高或所有的可接受劑量皆無效，則提早結束試驗；否則在可接受毒性的劑量中估計藥效最高的最低劑量為暫時的最佳生物劑量，然後採用平衡設計隨機配置病人服用最佳生物劑量及其相鄰的可接受劑量。最後當病患人數到達預定數量時，根據所有病患的毒性反應與疾病無惡化時間估計最佳生物劑量，也估計可接受劑量之藥效與最佳藥效的差距在百分之五內的有效生物劑量。本文最後在各種不同的毒性機率與疾病無惡化存活機率之組合情境下進行模擬研究，探討兩階段早期臨床試驗方法的操作特性。模擬研究結果顯示本文建議的兩階段早期臨床試驗方法具有合理的過毒劑量配置機率，也能有效率地鑑別有效生物劑量。 | zh_TW |
| dc.description.abstract | Early clinical trials for molecular target agents (MTA) cancer research aim to estimate the optimal biological dose (OBD) that gives the largest efficacy probability while remaining the dose-limiting toxicity (DLT) probability less than the target toxicity probability (TTP). Two-stage designs are considered in this thesis for the early phase clinical trials. In the first stage, either the CRM or Keyboard design is used to adaptively assign doses to patients based on the binary toxicity responses. The maximum tolerated dose (MTD) and hence doses with acceptable toxicity or the acceptable doses are tentatively determined. In the second stage, cohort trials are employed where the endpoint of interest is the progression-free survival (PFS) time and the efficacy is then evaluated as the survival function at the median PFS time of previous medication. During each cohort trial in the second stage, the intersection-union test is used to screen the acceptable doses with efficacy less than one half. Therefore, the clinical trial is early terminated if the lowest dose is too toxic or all the acceptable doses are ineffective. Otherwise, a tentative OBD is identified to be the lowest dose that maintains the highest efficacy among the acceptable doses. The OBD and acceptable doses adjacent to the OBD are then randomly assigned in balance to the next cohort of patients. When the maximum number of patients is reached, the OBD and the effective biological dose (EBD) as the acceptable dose with efficacy within 5% to the OBD are identified based on all the available toxicity responses and PFS times is also estimated. A simulation study is conducted to investigate the operating characteristics of the two-stage designs over a variety of scenarios of toxicity and efficacy probabilities. The simulation results show that the two-stage designs maintain reasonably on the over-toxic dose assignment and identify efficiently on the EBD selection. | en_US |
| DC.subject | 早期臨床試驗設計 | zh_TW |
| DC.subject | 兩階段設計 | zh_TW |
| DC.subject | 最佳生物劑量 | zh_TW |
| DC.subject | 有效生物劑量 | zh_TW |
| DC.subject | 疾病無惡化存活時間 | zh_TW |
| DC.subject | early clinical trial design | en_US |
| DC.subject | two-stage design | en_US |
| DC.subject | optimal biological dose | en_US |
| DC.subject | effective biological dose | en_US |
| DC.subject | progression free survival time | en_US |
| DC.title | 根據毒性反應與疾病無惡化存活時間 建立之早期臨床試驗設計 | zh_TW |
| dc.language.iso | zh-TW | zh-TW |
| DC.title | Early clinical trial designs based on toxicity response and progression-free survival time | en_US |
| DC.type | 博碩士論文 | zh_TW |
| DC.type | thesis | en_US |
| DC.publisher | National Central University | en_US |