疫苗免疫力銜接性試驗之統計方法的研究

子宮頸癌是常見的女性惡性腫瘤之ㄧ，每年全球約有470,000人罹患子宮頸癌，約有233,000因其而逝世。在台灣，根據2003年的報告指出，大約有4,908名女性被診斷出罹患此疾病，且約有932名女性死於子宮頸癌。根據研究指出，持續性的人類乳突病毒HPV的病毒感染是造成子宮頸癌的主要原因，而近幾年來一種四價HPV疫苗已被台灣核准使用，疫苗處方宣稱適用於9-26歲的年輕女性。然而在此四價HPV疫苗的樞紐臨床試驗包含臨床指標的觀察(CIN2, CIN3, AIS或子宮頸癌的發生)以及抗體濃度的測量，只對16-26歲的女性受試者進行測試，而對9-15歲的年輕女性受試者只進行抗體濃度的測量沒有進行臨床指標的觀察。因為9-15歲的年輕女性受試者的平均抗體濃度比16-26歲的女性受試者的來的高，所以可宣稱其四價HPV疫苗可適用於9-15歲的年輕女性受試者。然而此宣稱並非建構於任何統計推論之上，而只是藉由數據觀察而得到的結論。因此為了以統計推論評估四價HPV疫苗是否可適用於9-15歲的年輕女性受試者，此論文的目標是發展一套統計方法來測試當其疫苗的有效性已在某一特定的族群建立後，如: 16-26歲的女性族群，是否可藉由抗體濃度的銜接性臨床試驗評估 9-15歲的臨床療效。在假設四價HPV疫苗的抗體濃度對於CIN2, CIN3, AIS或子宮頸癌的保護力不會因受試者的年齡大小而有所差異的假設下，我們利用16-26歲的女性受試者的資料，採用邏輯式回歸來建立預測模式，其中自變量為抗體濃度，因變量為臨床指標的發生率。接著可藉由9-15歲的年輕女性受試者所量測的抗體濃度，來預測每個個體的臨床指標的發生率，進而得到9-15歲的年輕女性族群臨床指標的發生率的整體估計值。最後為了評估四價HPV疫苗是否可適用於9-15歲的年輕女性受試者，我們提出一個方法對9-15歲的年輕女性族群臨床指標的發生率是否較16-26歲的年輕女性族群臨床指標的發生率的為少來進行推論。我們亦以一個模擬研究探討所提出方法的型ㄧ誤差之機率即檢定力，且以ㄧ數例介紹所提出方法在實務上的應用。本論文是採用拔靴法進行假設檢定。

關鍵字

子宮頸癌；人類乳突病毒；疫苗效力；邏輯式回歸；拔靴法

並列摘要

4,908 females were diagnosed with cervical cancer and about 932 died from this disease in 2003. It is now recognized that persistent human papillomavirus (HPV) infections are the major cause of cervical cancer. In recent years, the quadrivalent vaccine against HPV has now been licensed for use within Taiwan. The company claims that the HPV vaccine efficacy can be applied to the female 9-26 years of age. Their pivotal trials conducted to evaluate both antibody titer measurement and clinical endpoints such as incidence of HPV infection or related disease only in females 16-26 years of age. On the other hand, for the females 9-15 years of age, the immunogenicity bridging studies is conducted to measure their quantity of antibody titer, not including the clinical endpoints observation. Since the mean titer of the females 9-15 years of age is higher than that of the females 9-26 years of age, the company claimed the vaccine efficacy observed in the females 16-26 years of age can be bridged to the females 9-15 years of age. However, this claim does not have any statistical inference but only with the observations of the average measurements of the quantity of the antibody titers. In order to test the validity of the claimed vaccine efficacy for the females 9-15 years of age in a scientific manner, the main objective of this thesis is to develop a statistical method to test whether the vaccine efficacy established in the females 16-26 years of age, can be applied to the females 9-15 years of age, through immunogenicity bridging studies. Under the assumption that the relationship between the incidence of the clinical end points and the measurement of quantity of anti-HPV antibody titers is the same for both age groups, we fit a logistic regression prediction model to the dataset of the subjects of the females 16-26 years of age. Then based on the estimated prediction model, we can predict the incidence of the clinical endpoints of the subjects of the females 9-15 years of age by their antibody measurement and the estimated the proportions of the occurrence of clinical endpoints in the females 9-15 years of age. We proposed a procedure to test the hypothesis that the proportion of the clinical endpoints in the females of age 9-15 years is less than that of age 16-26 years. The hypothesis testing is based on the bootstrapping method. A simulation study was conducted to empirically examine the size and power of the proposed method. A numerical dataset illustrates the application of the proposed method.

並列關鍵字

Cervical cancer ； HPV ； vaccine efficacy ； logistic regression ； bootstrapping

參考文獻

[1] Agresti, A., Categorical Data Analysis. New York: Wiley-Interscience, 2002.

[2] Department of Health and Human Services, Centers for Disease Control and Prevention. http://www.cdc.gov/vaccines/vpd-vac/hpv/default.htm, 2008.

[3] Department of Health and Human Services, Centers for Disease Control and Prevention. Genital HPV Infection - CDC Fact Sheet, http://www.cdc.gov/STD/HPV/STDFact-HPV.htm, 2008.

[4] FUTURE II Study Group. Quadrivalent vaccine against human papillomavirus to prevent high-grade cervical lesions. N Engl J Med. 356(19):1915-27, 2007.

[5] Hogg, R.V., McKean, J. W., Craig, A. T., Introduction to Mathematical Statistics. Pearson Prentice Hall, 2005.

Google Scholar

國際替代計量

疫苗免疫力銜接性試驗之統計方法的研究

全文下載

主題瀏覽