臺灣博碩士論文加值系統

目標
肺癌是全球性的重要疾病。目前臨床上廣泛使用第七版TNM分期系統來幫助我們決定適當的治療，並提供部分的預後資訊。TNM系統根據原發腫瘤的特性、淋巴結的侵犯及遠端轉移將腫瘤分期。然而，同樣期別的病患應當會因其具有其他獨特的特質而有不同的預後結果，此外，非小細胞肺癌( non-small cell lung cancer, NSCLC) 病患的預後也與其治療關係密切。以上這些因素提供的預後資訊均為TNM系統未考慮到的。第一個評分系統是由弗雷明漢心臟研究(Framingham Heart Study)發展並應用於心臟血管疾病，此研究透過評分系統簡化預測模式，使臨床醫師及病患方便使用。目前有許多相關研究試圖建立更好的非小細胞肺癌預測模式，但這些研究多針對西方族群，且大部分研究為針對特定期別、治療及特質的病人，我們的研究目的是建立一個針對非小細胞肺癌所有期別，且發展於漢人族群的存活預測模式及評分系統。
方法
我們預計納入2011.1.1至2015.12.31被診斷為非小細胞肺癌的病患，並追蹤至病人死亡或至研究結束點(2017.1.21)。研究設計為回顧性世代研究。以隨機分配方式將研究對象三分之二分派至衍生組(derivation set)，另三分之一分派至驗證組(validation set)，衍生組將發展風險分數系統，所建立的分數將以驗證組評估風險分數系統分辨存活風險的能力。採用考克斯比例風險模式 (Cox proportional hazard regression) 建立預測模型，並依照弗雷明漢心臟研究所發展出之步驟建立預測模型之計分系統。我們使用接受者操作特性曲線之曲線下面積(Area Under Receiver Operating Characteristic Curve; AUROC)以及Hosmer-Lemeshow方法來評估模型的鑑別能力和配適度 (goodness of fit)。
結果
在測試集中包含1,045位參與者，其中有551 (52.73%)位死亡，平均追蹤22.1個月。在驗證集中包含522位參與者，其中305 (58.43%)位死亡，平均追蹤21.3個月。死亡危險因子包含診斷年齡、性別、抽菸習慣、ECOG身體功能、組織型態、臨床腫瘤分期、手術、標靶治療以及交互作用項(診斷年齡和性別、診斷年齡和ECOG身體功能)。分數的總分介於-23到57分之間。在驗證集中使用三分位數 (tertiles) 將分數總分切為低風險、中風險、高風險三組，三組的Kaplan-Meier存活曲線有顯著差異(p<0.0001)。一年、三年、五年的AUROC結果在測試集中為0.85、0.83以及0.84，在驗證集中為0.84、0.86以及0.86。
討論
這是第一個在台灣預測全期別非小細胞肺癌死亡風險的預測模型。此模型具有良好的預測一年、三年及五年存活的正確性和鑑別能力。我們研究結果可以提供未來臨床使用。也提供一個簡單使用的工具讓臨床醫生預估台灣非小細胞肺癌病患的存活情況。

Background
Lung cancer is the most common cancer and the leading cause of death by malignancy worldwide. Lung cancer is commonly classified as two main types. Approximately 80 percent of lung cancers are classified as non-small cell lung cancer (NSCLC). The current TNM staging system (The 7th edition) as proposed by the American Joint Committee on Cancer (AJCC) classifies invasion status of NSCLC by tumor characteristics, status of involved lymph nodes, and distant metastasis conditions. The staging of NSCLC determines the main therapeutic plan, and provides prognostic information. However, the heterogeneity of individual patients, new imaging modalities, the number of treatment options including genomics and proteomics approaches lead the TNM staging system is imprecise. The quick and easy-to-use scoring systems for patients with NSCLC based on the easily accessible predictors provide the more precise information of the survival to help physicians and patients about appropriate treatment guide, more realistic outcome prediction, the quality of trials and policy making.
The first risk scoring system was for coronary heart disease which was constructed by the Framingham Heart Study. The study simplified the results of the prediction model through point scoring and charts that can be easily utilized by clinicians and patients. Some prior studies have generated prognostic prediction models with scoring system for NSCLC. However, most of them had focused on Western populations, patients with certain stage NSCLC or patients with particular characteristics or treatments. Our aim is to construct a survival prediction model with the balance of simple-to-use under a variety of medical care settings with easily obtained predictors in routine clinical setting and the good reliability and accuracy of model.
Methods
A retrospective cohort study was conducted in patients with a newly diagnosis of NSCLC during 2011.1.1 to 2015.12.31. These subjects were followed up until January 21, 2017 or death. Participants were randomly allocated into 2:1 ratio for derivation and validation sets. Cox’s proportional hazard regression model was used to identify risk factors of survival in derivation set. We applied the steps of the Framingham Heart Study to derive the scoring system. We used the Area Under Receiver Operating Characteristic Curve (AUROC) to assess the predictive accuracy of the survival prediction model and Hosmer-Lemeshow χ^2 method to assess the goodness of fit of the modelh. Net reclassification improvement (NRI) was used to compare the classification ability with our score and the score developed by Blanchon et al.
Results
Among 1,045 participants in the derivation set, 551 subjects (52.73%) died over a mean 22.1 months of follow-up period from index date to January 21, 2017. Among 522 patients in the validation set, 305 subjects (58.43%) died with a mean 21.3 months of follow-up. We identified age, gender, tobacco use, Eastern Cooperative Oncology Group (ECOG) performance status, histologic type, clinical TNM stage, surgery and targeted therapy, interaction between age at diagnosis and gender and interaction between age at diagnosis and ECOG performance status as significant predictors for survival. The sum risk score ranged from -23 to 57. We categorized subjects in the validation set as low, medium, and high risk groups by tertiles of the total risk scores derived from our scoring system. The results of Kaplan-Meier analysis for 5-year survival were significantly different among risk groups (p<0.0001).
The AUROC of 1-, 3-, and 5-year survival were 0.85, 0.83, and 0.84 in the derivation set and 0.84, 0.86, and 0.86 in the validation set, respectively.
Conclusions
This is the first risk scoring system to predicted survival in patients with all stages NSCLC in Taiwan population based on easily available predictors from routine medical care settings and generated excellent prediction accuracy and calibration. The results of the current study may help the patients and clinicians in Taiwan to predict the survival of patients with NSCLC.

Chapter 1 Introduction - 1 -
1.1 Background - 2 -
1.2 Importance of research - 5 -
1.3 Research purposes - 7 -
1.4 Study hypothesis - 8 -
1.5 Organization of the dissertation - 9 -
Chapter 2 Literature Review - 10 -
2.1 Prediction models in patients with early stage NSCLC - 10 -
2.2 Prediction models in patient with advance stage NSCLC - 12 -
2.3 Prediction models in patient with all stage NSCLC - 14 -
Chapter3 Methodology 25
3.1 Study design 25
3.2 Flowchart of the study 26
3.3 Conceptual framework of the study 28
3.4 Data source 30
3.5 Study subjects 31
3.6 Measurements 34
3.7 Statistical analysis 36
Chapter 4 Results 40
4.1 Baseline characteristics of study subjects in the derivation and validation sets. 40
4.2 Final multivariate Cox’s proportional hazards model and the steps of assigning scores for each risk factor 43
4.3 Discrimination ability, calibration, and sensitivity analysis of the prediction model 46
Chapter 5 Discussion/Conclusion 77
5.1 Major findings of the score system 77
5.2 Strengths and limitations 88
5.3 Conclusion 90
Reference 92

Reference
1.WHO | World Health Organization. WHO (2016).
2.Taiwan Cancer Registry.
3.TNM | UICC. Available at: http://www.uicc.org/resources/tnm/publications-resources.
4.Wilson, P. W. F. et al. Prediction of Coronary Heart Disease Using Risk Factor Categories. (1998).
5.Blanchon, F. et al. 4-Year Mortality in Patients With Non-Small-Cell Lung Cancer: Development and Validation of a Prognostic Index. Lancet Oncol. 7, 829–836 (2006).
6.Schild, S. E. et al. A new scoring system for predicting survival in patients with non-small cell lung cancer. Cancer Med. 4, 1334–1343 (2015).
7.Jie Lin, PhD, M. A Prognostic Model to Predict Mortality among Non-Small Cell Lung Cancer Patients in the U.S. Military Health System. J Thorac Oncol. (2015). doi:10.1097/JTO.0000000000000691.
8.Dehing-Oberije, C. et al. Development and External Validation of Prognostic Model for 2-Year Survival of Non-Small-Cell Lung Cancer Patients Treated With Chemoradiotherapy. Int. J. Radiat. Oncol. Biol. Phys. 74, 355–362 (2009).
9.Dehing-Oberije, C. et al. Development and validation of a prognostic model using blood biomarker information for prediction of survival of non-small-cell lung cancer patients treated with combined chemotherapy and radiation or radiotherapy alone (NCT00181519, NCT00573040, and NCT0. Int. J. Radiat. Oncol. Biol. Phys. 81, 360–368 (2011).
10.Zhang, K. et al. Modeling the overall survival of patients with advanced-stage non-small cell lung cancer using data of routine laboratory tests. Int. J. Cancer 136, 382–391 (2015).
11.Firat, S., Byhardt, R. W. &Gore, E. Comorbidity and Karnofksy performance score are independent prognostic factors in stage III non-small-cell lung cancer: An institutional analysis of patients treated on four RTOG studies. Int. J. Radiat. Oncol. Biol. Phys. 54, 357–364 (2002).
12.Pfister, D. G. et al. American Society of Clinical Oncology treatment of unresectable non-small-cell lung cancer guideline: Update 2003. J. Clin. Oncol. 22, 330–353 (2004).
13.Buccheri, G. &Ferrigno, D. Prognostic factors in lung cancer: Tables and comments. Eur. Respir. J. 7, 1350–1364 (1994).
14.Jeremiä, B., Miličić, B. &Milisavljevic, S. Clinical prognostic factors in patients with locally advanced (stage III) nonsmall cell lung cancer treated with hyperfractionated radiation therapy with and without concurrent chemotherapy. Cancer 117, 2995–3003 (2011).
15.Brundage, M. D., Davies, D. &Mackillop, W. J. Prognostic factors in non-small cell lung cancer: A decade of progress. Chest 122, 1037–1057 (2002).
16.Hoang, T., Xu, R., Schiller, J. H., Bonomi, P. &Johnson, D. H. Clinical model to predict survival in chemonaive patients with advanced non-small-cell lung cancer treated with third-generation chemotherapy regimens based on Eastern Cooperative Oncology Group data. J. Clin. Oncol. 23, 175–183 (2005).
17.Kawaguchi, T. &Takada, M. Performance status and smoking status are independent favorable prognostic factors for survival in non-small cell lung cancer: a comprehensive analysis of 26,957. J. Thorac. … 5, 620–630 (2010).
18.Louie, A.V. et al. Predicting overall survival after stereotactic ablative radiation therapy in early-stage lung cancer: Development and external validation of the Amsterdam prognostic model. Int. J. Radiat. Oncol. Biol. Phys. 93, 82–90 (2015).
19.Oberije, C. et al. A validated prediction model for overall survival from stage III non-small cell lung cancer: Toward survival prediction for individual patients. Int. J. Radiat. Oncol. Biol. Phys. 92, 935–944 (2015).
20.Wu, C.-Y. et al. Survival Prediction Model Using Clinico-Pathologic Characteristics for Nonsmall Cell Lung Cancer Patients After Curative Resection. Medicine (Baltimore). 94, e2013 (2015).
21.Liang, W. et al. Development and validation of a nomogram for predicting survival in patients with resected non-small-cell lung cancer. J. Clin. Oncol. 33, 861–869 (2015).
22.International Classification of Diseases for Oncology. Available at: http://codes.iarc.fr/. (Accessed: 4thSeptember2017)
23.Surveillance, Epidemiology, and End Results Program. Available at: https://seer.cancer.gov/. (Accessed: 2ndNovember2017)
24.Applied Logistic Regression - David W. Hosmer, Jr., Stanley Lemeshow. Available at: https://books.google.com.tw/books?hl=zh-TW&lr=&id=Po0RLQ7USIMC&oi=fnd&pg=PR5&dq=Hosmer+Jr+DW,+Lemeshow+S.+Applied+logistic+regression.+John+Wiley+%26+Sons%3B+2004.&ots=DobWsa3pIR&sig=_gbSdDIwk46qicRan41vqUUg6og&redir_esc=y#v=onepage&q&f=false.
25.Bendel, R. B. &Afifi, A. A. Comparison of Stopping Rules in Forward "Stepwise" Regression. J. Am. Stat. Assoc. 72, 46 (1977).
26.Mickey, R. M. &Greenland, S. The impact of confounder selection criteria on effect estimation. Am. J. Epidemiol. 129, 125–37 (1989).
27.Sullivan, L. M., Massaro, J. M. &D’Agostino, R. B. Presentation of multivariate data for clinical use: The Framingham Study risk score functions. Stat. Med. 23, 1631–1660 (2004).
28.Pencina, M. J., D’Agostino, R. B., D’Agostino, R. B. &Vasan, R. S. Evaluating the added predictive ability of a new marker: from area under the ROC curve to reclassification and beyond. Stat. Med. 27, 157-72–12 (2008).
29.Kennedy, K. &Pencina, M. A SAS macro to compute added predictive ability of new markers predicting a dichotomous outcome. Sas 1–13 (2011).
30.Park, M. J. et al. Prognostic model to predict outcomes in nonsmall cell lung cancer patients treated with gefitinib as a salvage treatment. Cancer 115, 1518–1530 (2009).
31.Mandrekar, S. J. et al. A prognostic model for advanced stage nonsmall cell lung cancer: Pooled analysis of north central cancer treatment group trials. Cancer 107, 781–792 (2006).
32.Hoang, T. et al. Prognostic models to predict survival in non-small-cell lung cancer patients treated with first-line paclitaxel and carboplatin with or without bevacizumab. J. Thorac. Oncol. 7, 1361–8 (2012).
33.Dehing-Oberije, C. et al. Tumor Volume Combined With Number of Positive Lymph Node Stations Is a More Important Prognostic Factor Than TNM Stage for Survival of Non-Small-Cell Lung Cancer Patients Treated With (Chemo)radiotherapy. International Journal of Radiation Oncology Biology Physics 70, 1039–1044 (2008).
34.Chen, Y. M. et al. The impact of clinical parameters on progression-free survival of non-small cell lung cancer patients harboring EGFR-mutations receiving first-line EGFR-tyrosine kinase inhibitors. Lung Cancer 93, 47–54 (2016).
35.Chiu, C. H. et al. Epidermal Growth Factor Receptor Tyrosine Kinase Inhibitor Treatment Response in Advanced Lung Adenocarcinomas with G719X/L861Q/S768I Mutations. J Thorac Oncol 10, 793–799 (2015).
36.Werner-Wasik, M. et al. Recursive partitioning analysis of 1999 radiation therapy oncology group (RTOG) patients with locally-advanced non-small-cell lung cancer (LA-NSCLC): Identification of five groups with different survival. Int. J. Radiat. Oncol. Biol. Phys. 48, 1475–1482 (2000).
37.Grivaux, M. et al. Lung cancer among women in France. Analysis of the 904 French women with lung cancer included in the KBP-2000-CPHG study of the French College of General Hospital-based Pneumologists (CPHG). Lung Cancer 45, 279–287 (2004).
38.Wigren, T. Confirmation of a prognostic index for patients with inoperable non-small cell lung cancer. 44, 9–15 (2000).
39.Cesari, M. et al. Comorbidity and physical function: results from the aging and longevity study in the Sirente geographic area (ilSIRENTE study). Gerontology 52, 24–32 (2006).
40.Ebbert, J. O. et al. Duration of smoking abstinence as a predictor for non-small-cell lung cancer survival in women. (2005). doi:10.1016/j.lungcan.2004.07.045
41.Wisnivesky, J. P., Henschke, C., Mcginn, T., Iannuzzi, M. C. &Cornell, N. Y. H. Prognosis of Stage II non-small cell lung cancer according to tumor and nodal status at diagnosis. (2005). doi:10.1016/j.lungcan.2005.02.010
42.Ferketich, A. K. et al. Smoking status and survival in the national comprehensive cancer network non-small cell lung cancer cohort. Cancer 119, 847–853 (2013).
43.Kogure, Y. et al. Histology and smoking status predict survival of patients with advanced non-small-cell lung cancer. Results of West Japan Oncology Group (WJOG) Study 3906L. J. Thorac. Oncol. 8, 753–8 (2013).
44.Li, C.-T. et al. Smoking and prognostic factors in an observational setting in patients with advanced non-small cell lung carcinoma. J. Cancer 2, 52–61 (2011).
45.Manuscript, A. Cancer Epidemiology. 1–49 (2014). doi:10.1007/978-1-59745-416-2
46.Poullis, M. et al. Smoking status at diagnosis and histology type as determinants of long-term outcomes of lung cancer patients. Eur. J. Cardio-thoracic Surg. 43, 919–924 (2013).
47.DeVita, V. T., Lawrence, T. S. &Rosenberg, S. A. Devita, Hellman, and Rosenberg’s cancer : principles & practice of oncology.
48.Kratz, J. R. et al. A practical molecular assay to predict survival in resected non-squamous, non-small-cell lung cancer: Development and international validation studies. Lancet 379, 823–832 (2012).
49.Xie, Y. &Minna, J. D. Predicting the future for people with lung cancer. Nat. Med. 14, 812–813 (2008).
50.Xie, Y. &Minna, J. D. A lung cancer molecular prognostic test ready for prime time. Lancet 379, 785–787 (2012).
51.Shedden, K. et al. Gene Expression-Based Survival Prediction in Lung Adenocarcinoma: A Multi-Site, Blinded Validation Study: Director’s Challenge Consortium for the Molecular Classification of Lung Adenocarcinoma. Lung 14, 822–827 (2009).
52.Lee, E.-S. et al. Prediction of Recurrence-Free Survival in Postoperative Non-Small Cell Lung Cancer Patients by Using an Integrated Model of Clinical Information and Gene Expression. Clin. Cancer Res. 14, 7397–7404 (2008).
53.Wu, X. et al. Germline genetic variations in drug action pathways predict clinical outcomes in advanced lung cancer treated with platinum-based chemotherapy. Pharmacogenet. Genomics 18, 955–965 (2008).
54.Blackstock, A. W. et al. Outcomes Among African-American/Non-African-American Patients With Advanced Non-Small-Cell Lung Carcinoma: Report From the Cancer and Leukemia Group B. JNCI J. Natl. Cancer Inst. 94, 284–290 (2002).
55.Donovan, M. J. et al. A systems pathology model for predicting overall survival in patients with refractory, advanced non-small-cell lung cancer treated with gefitinib. Eur. J. Cancer 45, 1518–1526 (2009).
56.Sculier, J. P., Chansky, K., Crowley, J. J., VanMeerbeeck, J. &Goldstraw, P. The impact of additional prognostic factors on survival and their relationship with the anatomical extent of disease expressed by the 6th Edition of the TNM Classification of Malignant Tumors and the proposals for the 7th Edition. J Thorac Oncol 3, 457–466 (2008).
57.Taylor, A. E., Olver, I. N., Sivanthan, T., Chi, M. &Purnell, C. Observer error in grading performance status in cancer patients. Support. Care Cancer 7, 332–5 (1999).
58.Brierley, J. D. et al. Accuracy of recorded tumor, node, and metastasis stage in a comprehensive cancer center. J. Clin. Oncol. 20, 413–9 (2002).
59.Ricardi, U. et al. Stereotactic Ablative Radiotherapy for stage I histologically proven non-small cell lung cancer: An Italian multicenter observational study. Lung Cancer 84, 248–253 (2014).
60.Allibhai, Z. et al. The impact of tumor size on outcomes after stereotactic body radiation therapy for medically inoperable early-stage non-small cell lung cancer. Int. J. Radiat. Oncol. Biol. Phys. 87, 1064–1070 (2013).
61.Ou, S.-H. H. I., Zell, J. A., Ziogas, A. &Anton-Culver, H. Prognostic factors for survival of stage I nonsmall cell lung cancer patients : a population-based analysis of 19,702 stage I patients in the California Cancer Registry from 1989 to 2003. Cancer 110, 1532–1541 (2007).
62.Dunlap, N. E. et al. Size matters: A comparison of T1 and T2 peripheral non-small-cell lung cancers treated with stereotactic body radiation therapy (SBRT). J. Thorac. Cardiovasc. Surg. 140, 583–589 (2010).
63.May, M. et al. Association Between the Number of Dissected Lymph Nodes During Pelvic Lymphadenectomy and Cancer-Specific Survival in Patients with Lymph Node–Negative Urothelial Carcinoma of the Bladder Undergoing Radical Cystectomy. Ann. Surg. Oncol. 18, 2018–2025 (2011).
64.Groth, S. S. et al. Determination of the minimum number of lymph nodes to examine to maximize survival in patients with esophageal carcinoma: data from the Surveillance Epidemiology and End Results database. J. Thorac. Cardiovasc. Surg. 139, 612–620 (2010).
65.Vather, R., Sammour, T., Kahokehr, A., Connolly, A. B. &Hill, A. G. Lymph Node Evaluation and Long-Term Survival in Stage II and Stage III Colon Cancer: A National Study. Ann. Surg. Oncol. 16, 585–593 (2009).
66.Putila, J., Remick, S. C. &Guo, N. L. Combining clinical, pathological, and demographic factors refines prognosis of lung cancer: A population-based study. PLoS One 6, 1–13 (2011).
67.Putila, J. &Guo, N. L. Combining COPD with clinical, pathological and demographic information refines prognosis and treatment response prediction of non-small cell lung cancer. PLoS One 9, (2014).
68.Iizasa, T. et al. Preoperative pulmonary function as a prognostic factor for stage I non–small cell lung carcinoma. Ann. Thorac. Surg. 77, 1896–1902 (2004).
69.L??pez-Encuentra, A. et al. Surgical lung cancer: Risk operative analysis. Lung Cancer 44, 327–337 (2004).
70.Ademuyiwa, F. O. et al. Prognostic Factors in Stage III Non–Small-Cell Lung Cancer. Clin. Lung Cancer 8, 478–482 (2007).