Use this URL to cite or link to this record in EThOS:
Title: Investigation of intra-tumour heterogeneity to identify texture features to characterise and quantify neoplastic lesions on imaging
Author: Arshad, Mubarik Ahmed
Awarding Body: Imperial College London
Current Institution: Imperial College London
Date of Award: 2017
Availability of Full Text:
Access from EThOS:
Access from Institution:
The aim of this work was to further our knowledge of using imaging data to discover image derived biomarkers and other information about the imaged tumour. Using scans obtained from multiple centres to discover and validate the models has advanced earlier research and provided a platform for further larger centre prospective studies. This work consists of two major studies which are describe separately: STUDY 1: NSCLC Purpose The aim of this multi-center study was to discover and validate radiomics classifiers as image-derived biomarkers for risk stratification of non-small-cell lung cancer (NSCLC). Patients and methods Pre-therapy PET scans from 358 Stage I–III NSCLC patients scheduled for radical radiotherapy/chemoradiotherapy acquired between October 2008 and December 2013 were included in this seven-institution study. Using a semiautomatic threshold method to segment the primary tumors, radiomics predictive classifiers were derived from a training set of 133 scans using TexLAB v2. Least absolute shrinkage and selection operator (LASSO) regression analysis allowed data dimension reduction and radiomics feature vector (FV) discovery. Multivariable analysis was performed to establish the relationship between FV, stage and overall survival (OS). Performance of the optimal FV was tested in an independent validation set of 204 patients, and a further independent set of 21 (TESTI) patients. Results Of 358 patients, 249 died within the follow-up period [median 22 (range 0–85) months]. From each primary tumor, 665 three-dimensional radiomics features from each of seven gray levels were extracted. The most predictive feature vector discovered (FVX) was independent of known prognostic factors, such as stage and tumor volume, and of interest to multi-center studies, invariant to the type of PET/CT manufacturer. Using the median cut-off, FVX predicted a 14-month survival difference in the validation cohort (N = 204, p = 0.00465; HR = 1.61, 95% CI 1.16–2.24). In the TESTI cohort, a smaller cohort that presented with unusually poor survival of stage I cancers, FVX correctly indicated a lack of survival difference (N = 21, p = 0.501). In contrast to the radiomics classifier, clinically routine PET variables including SUVmax, SUVmean and SUVpeak lacked any prognostic information. Conclusion PET-based radiomics classifiers derived from routine pre-treatment imaging possess intrinsic prognostic information for risk stratification of NSCLC patients to radiotherapy/chemo-radiotherapy. STUDY 2: Ovarian Cancer Purpose The 5-year survival of epithelial ovarian cancer is approximately 35-40%, prompting the need to develop additional methods such as biomarkers for personalised treatment. Patient and Methods 657 texture features were extracted from the CT scans of 364 untreated EOC patients. A 4-texture feature ‘Radiomic Prognostic Vector (RPV)’ was developed using machine learning methods on the training set. Results The RPV was able to identify the 5% of patients with the worst prognosis, significantly improving established prognostic methods and was further validated in two independent, multi-centre cohorts. In addition, the genetic, transcriptomic and proteomic analysis from two independent datasets demonstrated that stromal and DNA damage response pathways are activated in RPV-stratified tumours. Conclusion RPV could be used to guide personalised therapy of EOC. Overall, the two large datasets of different imaging modalities have increased our knowledge of texture analysis, improving the models currently available and provided us with more areas with which to implement these tools in the clinical setting.
Supervisor: Aboagye, Eric ; Rockall, Andrea Sponsor: Not available
Qualification Name: Thesis (Ph.D.) Qualification Level: Doctoral