Neoadjuvant chemoradiotherapy (CRT) followed by total mesorectal excision (TME) is the gold standard treatment for locally advanced rectal cancer with threatened circumferential resection margin (CRM). This strategy considerably improves loco-regional control but fails to improve overall survival (OS) due to distant failure. In addition there is a lack of consensus on optimal timing of surgery and the management of patients with complete clinical response after CRT who could benefit from either less invasive or wait and watch approaches. This shifting paradigm has placed greater recent interest in quantification of imaging biomarkers such as textural analysis (TA) linked to underlying intra-tumour heterogeneity associated with adverse outcomes. This could help to select patients with predicted poor prognosis for personalized intensive therapy. The aims of this thesis were: firstly, to assess the short and long-term effects of delayed surgery after CRT and secondly to investigate the prognostic potential of TA based on conventional magnetic resonance imaging (MRI) in stage II-III locally advanced rectal cancer. Thirdly, the potential of functional parameters (standardized uptake value [SUV], apparent diffusion coefficient [ADC]) quantified on pre-treatment integrated positron emission tomography and MRI (PET/MR) system to predict pathological response to CRT (independent sample t-test) and survival was assessed.
TA using a filtration-histogram technique of MR images was undertaken using TexRAD, a proprietary software algorithm. Regions of interest enclosing the largest cross-sectional area of tumour area were manually delineated on the axial images. Cox-multiple regression analysis determined which univariate features (clinical, textural, radiological and histological) on Kaplan-Meier survival analysis independently predicted OS, disease free survival (DFS) and recurrence-free survival (RFS). The time interval to surgery did not predict the survival outcomes. Male gender independently predicted DFS and RFS while CRM predicted RFS for the entire cohort (n-112). In a subset of the cohort (n-56) pre-treatment TA, extramural venous invasion (EMVI) on MRI independently predicted OS while TA and threatened CRM on MRI predicted DFS. For OS; EMVI on MRI and for DFS; TA and CRM involvement on MRI were the independent post-treatment factors. Only TA independently predicted RFS on pre- or post-treatment analyses. Both SUV and ADC values were not predictive of outcomes.
Delayed surgery after CRT does not lead to worse survival outcomes. Along with local or distant failures, male gender and pathological CRM are associated with worse survival. MR based TA of rectal cancers can predict outcome before undergoing surgery and could potentially select patients for individualized therapy.