2 research outputs found
Intravoxel incoherent motion diffusion-weighted and amide proton transfer imaging for the prediction of treatment outcome in nasopharyngeal carcinoma
No full textPh.D.INTRODUCTIONPredictive imaging markers obtained from the staging head and neck magnetic resonance imaging (MRI) of patients with nasopharyngeal carcinoma (NPC) have limitations identifying patients at risk of distant metastases, planning treatment and designing post-treatment surveillance programs to detect relapse. Functional MRI techniques can be added to the staging MRI and may provide new predictive imaging markers. This thesis evaluated two new functional MRI techniques. The first technique is an advancement in diffusion-weighted imaging (DWI) called intravoxel incoherent motion DWI (IVIM-DWI), which distinguishes the pure diffusion of water molecules from pseudo-diffusion due to microcirculation. The second technique is amide proton transfer-weighted (APTw) imaging, which detects amide protons in cancer proteins and peptides. The primary objective was to determine whether IVIM-DWI and APTw imaging parameters could predict NPC outcome. The secondary objective was to identify possible correlations between the imaging parameters and NPC staging.METHODSIVIM-DWI and APTw imaging of the primary tumour were performed in patients with NPC. The mean and 90ᵗʰ percentile values of parameters were assessed for IVIM-DWI (pure diffusion coefficient [D], pseudo-diffusion coefficient [D*], perfusion fraction [f] and apparent diffusion coefficient [ADC]) and APT. Staging was divided into early- and advanced-stage groups for tumour (T) staging, node (N) staging, metastasis (M) staging and overall staging. Outcome was assessed in terms of locoregional relapse-free survival (LRRFS), distant metastasis-free survival (DMFS) and disease-free survival (DFS) after 2 years. A Cox regression analysis of significant parameters, with age, sex, treatment and tumour staging as confounding variables, was added to a multivariable model. The area under the curves (AUCs) of significant parameters for disease relapse were compared using the DeLong test. A subgroup of patients underwent pre-treatment and early intra-treatment APTw imaging to correlate the changes in APT values with outcome at 6 months.RESULTSPre-treatment prediction of outcomeIn the univariable analysis, IVIM-DWI (n = 161) and APTw imaging (n = 77) showed that high Dₘₑₐₙ values were correlated with poor LRRFS and poor DFS (p = 0.045 and 0.010, respectively) and high APT₉₀ values were correlated with poor DMFS and poor DFS (p = 0.009 and 0.010, respectively). In the multivariable analysis, Dₘₑₐₙ remained an independent predictor of DFS (p = 0.019) and APT₉₀ remained an independent predictor of DMFS (p = 0.009) and DFS (p = 0.010).Multiparametric imaging of patients who had undergone both imaging techniques (n = 66) showed similar results in univariable analysis. In the multivariable analysis, the independent predictors were high Dₘₑₐₙ for poor LRRFS and DFS (p = 0.014 and 0.012, respectively) and high APT90 for poor DMFS (p = 0.023). For disease relapse, the AUC obtained by incorporating Dₘₑₐₙ and APT₉₀ into T-staging was significantly higher than that obtained without incorporating these parameters into T-staging (AUC = 0.810 vs 0.594; p = 0.002).Early intra-treatment prediction of outcomeAPTw imaging detected changes in all of the tumours (n = 16) after 2 weeks of treatment. There were significant differences in the percentage change in APTₘₑₐₙ and APT₉₀ between the pre- and intra-treatment scans (p = 0.015 and 0.015, respectively) in patients with good vs poor outcome. Poor outcome was associated with an increase in APT values and good outcome with a decrease in APT values.StagingIVIM-DWI (n = 202) and APTw imaging (n = 97) showed that high Dₘₑₐₙ and ADCₘₑₐₙ were correlated with advanced N-staging (p = 0.005 and 0.045, respectively) and high Dₘₑₐₙ with advanced overall staging (p = 0.025). No associations were found between IVIM-DWI and T-staging or M-staging and between APT parameters and staging.CONCLUSIONPre-treatment IVIM-DWI and APTw imaging, especially when used with a multiparametric approach, have the potential to improve outcome prediction of the current NPC staging system. IVIM-DWI showed correlations with N and overall staging, but neither IVIM-DWI nor APTw imaging was found to be correlated with T-staging or distant metastases at diagnosis.簡介鼻咽癌（NPC）在治療前會接受頭頸磁力共振常規掃描進行腫瘤分期，并用該信息挑選有高遠處轉移風險的病人進行全身正電子掃描以探測遠處轉移，以及挑選具有治療后復發高風險的人進行更合理的治療以及治療后的密切監控。然而這些信息的預估效能往往不佳。功能性磁力共振可加入常規序列，並有可能增加常規序列對疾病復發的預估效能。本論文分析了兩種新的功能性磁力共振序列。第一個是IVIM-DWI。它是改進后的彌散加權序列，可以將真實的彌散效能從由於微循環所產生的偽彌散效能中區分出來。第二個是質子交換加權序列（APTw）。它可以探測腫瘤中的蛋白質以及肽段的改變。本論文主要目的有兩個：1. 研究治療前原發NPC中的IVIM-DWI 和APTw 係數是否能夠預估治療后疾病復發風險；2. 研究治療前原發NPC中的IVIM-DWI 和APTw數值是否與NPC分期有關。方法本研究將IVIM-DWI 和APTw 序列成像于原發腫瘤，並算出相關純擴散係數（D），偽擴散係數 (D*)，灌注分數 (f) 和表觀擴散係數（ADC）以及APT的平均值（mean）以及第90個百分位數。腫瘤分期包括原發腫瘤（T），淋巴結轉移（N），遠處轉移（M）以及綜合分期被分為早期和進展期。治療后長期預後評估使用2年後的無原發淋巴復發生存期（LRRFS），無遠處轉移生存期 （DMFS）以及無復發疾病生存期 （DFS）進行評估。上述功能磁力共振係數以及年齡、性別、治療、腫瘤分期添加到多變量Cox 回歸分析中與治療后長期預後相關聯。 針對DFS有顯著意義的變量用曲線下面積（AUC）表示診斷效能，並用Delong 檢驗探測不同變量的診斷效能。另外，一小組患者接受了治療前和治療中（開始治療后2周）APTw成像。兩次APT 參數的變化與病人治療后6個月的預後相關聯。結果治療前IVIM-DWI 和APTw 參數與治療后長期預後在IVIM-DWI （n=161）和APTw（n=77）兩個獨立研究中，單變量分析顯示高Dₘₑₐₙ 預估較差的LRRFS 和DFS （p值 分別為0.045 and 0.010）以及高APT₉₀預估較差的DMFS和DFS （p值 分別為0.009 and 0.010）。多變量分析顯示Dₘₑₐₙ是預估DFS 的獨立因素以及APT₉₀ 是預估DMFS 的獨立因素。在評估聯合技術的研究中（n=66），單變量研究結果如上述，而多變量分析發現高Dₘₑₐₙ 預估較差的LRRFS 和DFS （p值 分別為0.014 and 0.012）以及高APT₉₀預估較差的DMFS（p=0.023）。針對復發疾病的診斷效能，將Dₘₑₐₙ和APT₉₀結合到T分期后，AUC 從0.594 上升到0.81（p=0.002）。治療中APTw參數改變與短期治療后預後研究中，所有病例（n=16）的治療中APTw的參數相對治療前都發生改變。APTₘₑₐₙ（p=0.015）和APT₉₀（p=0.015）的改變在短期預後好的病人中是下調，而預後差的病人中是上調。分期在IVIM-DWI （n=202）和APTw（n=97）兩個研究中發現高Dₘₑₐₙ和ADCₘₑₐₙ跟進展N分期有關（p值 分別為0.005 and 0.045），而且高Dₘₑₐₙ于進展綜合分期有關 （p=0.025）。IVIM-DWI 係數與T和M分期沒關係。APTw係數與任何分期都沒關係。總結治療前的IVIM-DWI和APTw的原發腫瘤相關係數在聯合分析時具有能夠提升當前NPC分期的預估治療后長期預後效能的潛能。IVIM-DWI係數與N分期和綜合分期有關，然而IVIM-DWI 和APTw與治療前是否出現的遠處腫瘤無關。Qamar, Sahrish.Thesis Ph.D. Chinese University of Hong Kong 2020.Includes bibliographical references (leaves 104-122).Abstracts also in Chinese.Title from PDF title page (viewed on November 11, 2021)
Substantial and sustained reduction in under-5 mortality, diarrhea, and pneumonia in Oshikhandass, Pakistan : Evidence from two longitudinal cohort studies 15 years apart
No full textFunding Information: Study 1 was funded through the Applied Diarrheal Disease Research Program at Harvard Institute for International Development with a grant from USAID (Project 936–5952, Cooperative Agreement # DPE-5952-A-00-5073-00), and the Aga Khan Health Service, Northern Areas and Chitral, Pakistan. Study 2 was funded by the Pakistan US S&T Cooperative Agreement between the Pakistan Higher Education Commission (HEC) (No.4–421/PAK-US/HEC/2010/955, grant to the Karakoram International University) and US National Academies of Science (Grant Number PGA-P211012 from NAS to the Fogarty International Center). The funding bodies had no role in the design of the study, data collection, analysis, interpretation, or writing of the manuscript. Publisher Copyright: © 2020 The Author(s).Peer reviewe
