摘要:乳癌在近十年來發生率已躍升為女性癌症第一位位居於台灣女性惡性腫瘤死因第四位。目前給予乳癌患者的治療處置是依據臨床分期診斷與組織病理切片結果,臨床分期是以腫瘤大小是否侵犯皮膚與胸壁、腋下淋巴結侵犯程度與遠端器官轉移與否定義,組織病理切片可獲得癌細胞分化程度、賀爾蒙受體(ER、PR)陽性或陰性、人體上皮生長因子受體(HER2)是否過度表現等。根據現行乳癌治療方針,臨床分期為第二期以上之患者,可考慮手術前接受先導性藥物治療(neoadjuvant chemotherapy),而給予哪種藥物是會根據病人停經與否與ER、PR、HER2 反應。先導性化療可能的優點包含:(1)使腫瘤縮小,可施行乳房保留手術的機會,(2)消除淋巴結等微小轉移的範圍,(3)可提早發現藥物能否有效抑制腫瘤生長,(4)及早給予全身性化學治療,以提高預後存活率,(5)提供觀察與分析腫瘤病理、生理、影像相關特徵,對特定化療藥物治療反應的相關性。但可能的缺點:治療效果不佳,腫瘤有變大的風險。過去研究顯示在前導性化療後達到微觀下腫瘤細胞不可見(pCR, pathological complete response)的乳癌患者有較佳的無病存活率(disease-free survival)和整體存活率(overall survival),因此發展預測乳房腫瘤前導性化療治療療效之電腦輔助預測模型非常重要的。本研究目的發展應用於預測乳房腫瘤前導性化療治療療效之電腦輔助預測模型。根據腫瘤在醫學影像上的異質性表徵,本研究計畫跨領域地從臨床影像學的腫瘤組織解剖學與功能性影像於前導性化療的反應及病理學的治療反應診斷、與醫學影像處理技術給予腫瘤的型態學與紋理學之特徵擷取與分類、協同當今的深度學習技術,所發展的預測模型預期能在治療開始前,或是治療前期,根據個體腫瘤特性,提供個人化的治療處置,有效地早期預測腫瘤化療反應,提高乳癌治療療效與預後。在此計畫中,我們將以兩年的時間發展腫瘤及血管增生等病理結構與分子影像間的特徵擷取.並在第三年合併前兩年的動態表徵擷取,以磁振暨正子掃描同步影像PET/MRI 分析結果進行資料探勘.在此最後一年的時間,也將發展多時間點PET/MRI 的病理結構與功能特徵間變化程度之定量分析模型。最終我們將發展結合腫瘤組織解剖學與功能性影像的形態及紋理結構變化程度指標,做為乳癌治療療效評估及預後評估指標的dynamic radiomics。可期待具有下列的創新貢獻 結合 DCE-MRI、PET 和QDS-IR 三種影像模態與其單時間點與多時間之參數動態特徵(parametricdynamics features) 建構乳癌前導性化療療效預測技術; 提出嶄新的參數動態特徵概念將有效的克服傳統特徵擷取時所面對的threshold 選擇問題; 針對 DCE-MRI 影像,將可分割不同型態之乳房腫瘤,並可擷取血管新生之動態表徵,將使得DECMRI 用於治療預測更加完善及靈敏; 針對 QDS-IR 影像,所發展之trajectory analysis 將以非侵入式影像定量分析前導性化療之反應; 針對 PET/CT 影像,所研發之單時間點對位分割技術將提升動態特徵的靈敏性及特異性; 針對 PET/MRI 影像,所結合於PET/CT 與DCE MRI 之特徵擷取技術將應用於多時間點的PET/MRI影像。並將發展多時間點對位後的區域性特徵作為治療反應的影像表徵; 本計畫所研發的深度學習系統,搭配複合式影像特徵,將建構高可靠度的預測模型。
Abstract: Breast cancer is the most frequently diagnosed cancer and remains the fourth leading cause of cancerdeaths in Taiwan women over the past decade. Decisions about the best treatment for breast cancer is basedon the result of estrogen (ER) and progesterone receptor (PR) test, human epidermal growth factor type 2receptor (HER2) test, and TNM staging using biopsy. After evaluation of menopause status and response ofER, PR and HER2, the treatments for stage 2 or above breast cancer may consider neoadjuvant chemotherapy(NAC) for the benefits of (1) converting an inoperable to a surgical resectable cancer, (2) metastasismanagement, (3) shrink the tumor, (4) improved overall survival and recurrence free survival rate (5)histologic parameters predictive. It is known that patients with pathological complete response (pCR) afterNAC are associated with better disease-free survival and improved overall survival. Therefore, it is essentialto develop more effective regimens and stratify patients based on computer assisted prediction model toevaluate the response of NAC.The main purpose of this study is to develop a computer-aided prediction model for NAC treatmentresponse. Based on the heterogeneity of internal parametric tumor composition commonly observed, thisstudy will utilize the histologic characteristics and treatment response to investigate the image features asinput data for predicting treatment response using Deep Learning technology. Using this technique,preoperative treatment evaluation may be facilitated by tumor heterogeneity analysis from developeddynamic radiomics, and the possibility of personal medicine can be realized not far ahead. In the first twoyears of this study using images from DCE-MRI, PET/CT and QDS-IR, we plan to develop the imageprocessing algorithms, including segmenting breast and tumor region, extracting image feature which reflectsangiogenic properties and permeability of tumor, which are highly correlated with NAC treatment response.During the third year of the project, the morphology and texture features from first two years can becombined for PET/MRI and prediction model can be achieved in accordance with the features extracted fromdynamic features extraction using longitudinal images of PET/MRI. The followings are the expectedcontributions: To propose a novel parametric dynamics features for overcoming the issues with traditional thresholdingmethod. To develop segmentation algorithms for breast tissue and tumor region on DCE MRI in order toimprove treatment response prediction. To develop trajectory analysis for non-invasive QDS-IR image To develop segmentation algorithm for metabolic tumor volume by registering PET uptake boundarywith CT tumor boundary in order to improve reliability and reproducibility of morphology feature. By combining preceding techniques developed for PET/CT and DCE MRI, new types of features,namely the parametric dynamics features from PET/MRI can be served as dynamic radiomics forpredicting NAC response. To develop a Deep-Learning algorithm, which is the essential of the project in terms of self-learningmethodology, for exploiting the high dimensional features space in search of the prediction model.