Ding, Boyue; (2025) Characterising Tumour Volume and Growth Dynamics in Non-Small Cell Lung Cancer. Doctoral thesis (Ph.D), UCL (University College London).

Preview

Text Ding_Thesis.pdf - Submitted Version Download (24MB) | Preview

Abstract

Background: This study aims to investigate the factors influencing recurrence and survival in non-small cell lung cancer (NSCLC) from the TRACERx cohort study, optimise surveillance and refine management to enhance patient prognosis using imaging. Method: 200 stage I–III NSCLCs diagnosed between 2013 and 2020 from 13 hospitals in the UK were analysed. Univariable and multivariable Cox models were used to assess associations between clinical characteristics and outcomes. Segmented lesions analysis evaluated metastasis patterns. Surveillance intensity was categorised based on the time from surgery to the first post-operative scan. RECIST criteria were used to evaluate progression-free survival (PFS) and volume changes pre-treatment and post-treatment. ROC curves were used to determine the optimal volume for predicting relapse and progression. Results: Manually contoured primary tumour volume (HR=2.68) was a stronger predictor of relapse site, tumour dynamics and prognosis than diameter-based volume estimates, pT stage or pTNM staging. Larger primary tumours were associated with early relapse, extrathoracic relapse, higher recurrence burden and worse prognosis. Similarly, higher relapse rates were associated with high heterogeneity, larger tumour burden, faster progression, extrathoracic involvement, and poorer survival, highlighting volume and growth rate as better prognostic predictors than lesion count. Pre-treatment growth rate showed a weak correlation with post-treatment growth rate. In pT2N0M0 tumours, exploratory thresholds for high-risk volume (>17,010 mm³) and growth rate (>58 mm³/day) were identified. A tumour volume reduction of more than 65% during initial therapy was associated with improved progression-free survival (AUC=0.81). These thresholds are exploratory and derived from a limited cohort; larger, prospective studies are needed to confirm them before integrating them into clinical practice. Gender, Age and Smoking status were not significantly associated with DFS overall, but current smokers showed higher relapse risk after 2.5 years. Recurrence patterns varied by histology and tumour location. LUSC had larger tumours and peaked at 9–12 and 15–18 months, while LUAD peaked at 12–15 months, with late relapse (>1.5 years) more common in larger LUAD. Intrathoracic relapses, primarily in the lung, correlated with better prognosis, whereas extrathoracic relapses were linked to worse outcomes, particularly in cases with brain involvement. Intrathoracic relapses were likely to progress with new lesions. Conversely, those with extrathoracic relapses often experienced the simultaneous appearance of new lesions and localised expansions, leading to more complex and aggressive progression modes. Relapse rates did not differ in terms of the number of progression events. Surveillance frequency based solely on the TNM stage was insufficient. High-frequency did not improve overall survival, as seen in previous findings, nor did it reduce relapse volume. Site-specific relapse patterns may help to tailor follow-up intensity in the future. Conclusions: Tumour volume and growth rate outperform lesion count and traditional staging in predicting relapse dynamics, patterns and survival. Intrathoracic relapses, primarily in the lungs, are associated with better outcomes, whereas extrathoracic relapses, especially in the brain and multiple organs, indicate faster relapse speed and poorer prognoses. Standard TNM-based follow-up protocols are insufficient; incorporating tumour volume, growth rate, and relapse site could support more personalised follow-up strategies.

Download activity - last month

Export as CSVXMLJSON

Download activity - last 12 months

Export as JSONXMLCSV

Downloads by country - last 12 months

Export as XMLCSVJSON

Archive Staff Only

View Item