葡萄糖正子影像預測口咽鱗狀細胞癌病患預後

Prognostic Roles of 18F-Fluorodeoxyglucose Positron Emission Tomography in Oropharyngeal Squamous Cell Carcinoma

鄭乃銘(Nai-Ming Cheng)

指導教授 ： 許靖涵

國立清華大學/原子科學院/生醫工程與環境科學系/博士(2019年)

摘要

口咽鱗狀細胞癌病患有著不佳的預後。本論文研究治療前葡萄糖正子影像對於口咽鱗狀細胞癌的應用。收案病患為無遠端轉移之口咽鱗狀細胞癌。每位病患皆完成同步化學與放射治療。依照正子影像造影日期將病患分組，一組病患正子影像用於訓練預後預測模型，另一組用於驗證模型，兩組之比例為1:1.5。使用接收者操作特徵曲線(receiver operating characteristic curves, ROC curves)與單變量Cox回歸分析，得到與無惡化存活時間顯著相關之正子影像參數。使用這些參數，利用遞迴分割分析(recursive partitioning analysis, RPA)方法，建立病患預後模式。這一個模型接著進入多變量Cox回歸分析，與其他臨床參數比較。本論文發現正子影像參數: 標準攝取值熵值(SUV-entry)和型狀不規則度(asphericity)與無惡化存活時間顯著相關，而且同時具有高標準攝取值熵值和型狀不規則度的病患其預後最差。此一結果可以在驗證組病患可以重現。在全部病患中，此模型之一致性統計量(concordance index)可達0.75，顯著高過其它臨床參數。
更進一步的，我們比較正子影像研究結果與免疫化學組織染色和臨床常用之危險分期系統。洪氏(Ang's)分期是口咽癌常用之臨床系統，包含因子為p16結果，吸菸與否，和口咽癌癌症分期。我們收集第三與第四期口咽癌病患，每位病患皆完成同步化學與放射治療。使用Kaplan-Meier存活曲線與Cox比例風險模式分析無疾病存活率與整體存活率。使用自助法以減少變異。最後採用遞迴分割分析建立預後預測模型。總共收集了113位口咽癌病患，其中28位病患是p16陽性的。多變數Cox比例風險模式分析結果顯示: 洪氏分期高危險群；高EGFR免疫化學組織染色；與高正子掃描標準攝取值熵值(SUV-entry)和型狀不規則度(asphericity)為無疾病存活率與整體存活率預測之獨立因子。這些危險因子使用遞迴分割分析整合成一個模型，可以成功預測p16陽性與陰性之口咽癌預後。在洪氏分期高危險群中(共77位患者)也可以再細分預後。此一新建模型可以更佳分類口咽癌病患預後。
綜合以上，正子影像參數模型可以用於高危險性口咽癌病患預後預測使用。

關鍵字

口咽癌 ； p16陰性 ； 預後 ； 正子葡萄糖斷層影像 ； 亂度 ； 不圓度

並列摘要

Oropharyngeal squamous cell carcinoma (OPSCC) has unfavorable survival outcomes. We investigate the prognostic role of pre-treatment positron emission tomography (PET) in OPSCC. OPSCC patients without distant metastasis were enrolled. All of them completed primary chemoradiotherapy. Patients were classified into training and validation cohorts with a ratio of 1:1.5 according to the PET date. Tumors were segmented. PET heterogeneity and irregularity indices were obtained. PET parameters with significant impact on progression-free survival (PFS) in receiver operating characteristic curves and univariate Cox models were identified and included in recursive partitioning analysis (RPA) for constructing a prognostic model. The RPA-based prognostic model was further tested in the validation cohort using multivariate Cox models. Fifty-eight and 89 patients were in the training and validation groups, respectively. Heterogeneity parameter, SUV-entropy (derived from histogram analysis), and irregularity index, asphericity, were significantly associated with PFS. The RPA model revealed that patients with both high SUV-entropy and high asphericity experienced the worst PFS. Results were confirmed in the validation group. The overall concordance index for PFS of the model was 0.75, which was higher than the clinical stages, performance status, SUVmax, and metabolic tumor volume of PET.
These results were compared with immunohistochemistry (IHC) studies of OPSCC and the clinically used risk stratification system. The Ang's risk profile (based on p16, smoking and cancer stage) is a well-known prognostic factor in OPSCC. Patients with stage III-IV OPSCC who completed primary therapy were eligible. Disease-specific survival (DSS) and overall survival (OS) served as outcome measures. Kaplan-Meier estimates and Cox proportional hazards regression models were used for survival analysis. A bootstrap resampling technique was applied to investigate the stability of outcomes. Finally, RPA-based model was constructed. A total of 113 patients were included, of which 28 were p16-positive. Multivariate analysis identified the Ang's profile, high EGFR expression (IHC result), high SUV-entropy and high asphericity as independent predictors of both DSS and OS. Using RPA, the three risk factors were used to devise a prognostic scoring system that successfully predicted DSS in both p16-positive and -negative cases. In patients showing an Ang's high-risk profile (77 cases), the use of our scoring system clearly identified three distinct prognostic subgroups. It was concluded that a novel index may improve the prognostic stratification of patients with advanced-stage OPSCC.
PET parameters provided useful prediction of survivals for patients with OPSCC.

並列關鍵字

Oropharyngeal cancer ； p16-negative ； prognosis ； FDG PET/CT ； heterogeneity ； asphericity

參考文獻

14. Liu C, Mann D, Sinha UK, Kokot NC. The molecular mechanisms of increased radiosensitivity of HPV-positive oropharyngeal squamous cell carcinoma (OPSCC): an extensive review. J Otolaryngol Head Neck Surg. 2018;47:59. doi: 10.1186/s40463-40018-40302-y.

66. Fang YH, Lin CY, Shih MJ, et al. Development and evaluation of an open-source software package "CGITA" for quantifying tumor heterogeneity with molecular images. Biomed Res Int. 2014;doi: 10.1155/2014/248505.

68. Satoh Y, Nambu A, Ichikawa T, Onishi H. Whole-body total lesion glycolysis measured on fluorodeoxyglucose positron emission tomography/computed tomography as a prognostic variable in metastatic breast cancer. BMC Cancer. 2014;14:doi: 10.1186/1471-2407-1114-1525.

82. Chen RY, Lin YC, Shen WC, et al. Associations of Tumor PD-1 Ligands, Immunohistochemical Studies, and Textural Features in 18F-FDG PET in Squamous Cell Carcinoma of the Head and Neck. Sci Rep. 2018;8:105. doi: 110.1038/s41598-41017-18489-41592.

92. Shen WC CS, Wu KC, Hsieh TC, et al. Prediction of local relapse and distant metastasis in patients with definitive chemoradiotherapy-treated cervical cancer by deep learning from [18F]-fluorodeoxyglucose positron emission tomography/computed tomography. Eur Radiol. 2019;May 27. doi: 10.1007/s00330-019-06265-x.