Stereotactic radiotherapy for brain metastases: using computational approaches to predict outcomes and patterns of recurrence
File(s)
Author(s)
Rozati, Hamoun
Type
Thesis
Abstract
Introduction:
Stereotactic radiotherapy (SRT) is commonly used to treat brain metastases (BMs). Issues with SRT include uncertainty in its use for multiple BMs, predicting which BMs will respond to treatment and the early detection of BMs.
Methods:
I performed a systematic review and meta-analysis of published articles analysing ten or more BMs, and a multi-centre retrospective case-series on patients with ten or more BMs treated with SRT alone. I also conducted an analysis of a multi-centre cohort of patients to create a novel radiomic score to predict the likelihood of response of BMs to SRT and an analysis on the same cohort to create a radiomic score to predict the likelihood of BM occurrence. I completed a lobar distribution study to map the distribution of BMs from distinct primary malignancies.
Results:
My meta-analysis has demonstrated overall survival outcomes consistent with many case series of outcomes for limited BMs. My retrospective case series of a contemporary cohort of patients has shown overall survival outcomes above 13 months. My analysis of a large multi-centre, multi-platform cohort of patients with BMs has created a novel radiomic score. A further radiomic analysis has created a radiomic score which predicts the occurrence of a BM better than chance. Finally, my analysis of the lobar distribution of BMs in patients referred for SRT has provided novel insights into the distribution of disease.
Conclusions:
Contemporary patients with ten or more BMs can have prolonged overall survival outcomes following SRT. Radiomics combined with clinical factors can give accurate predictions for which BMs are likely to respond to SRT. The novel application of radiomics to predict BMs early shows promise but is not currently clinically useful. The lobar distribution of BMs in patients referred for SRT can be predicted and may have implications for treatment decisions.
Stereotactic radiotherapy (SRT) is commonly used to treat brain metastases (BMs). Issues with SRT include uncertainty in its use for multiple BMs, predicting which BMs will respond to treatment and the early detection of BMs.
Methods:
I performed a systematic review and meta-analysis of published articles analysing ten or more BMs, and a multi-centre retrospective case-series on patients with ten or more BMs treated with SRT alone. I also conducted an analysis of a multi-centre cohort of patients to create a novel radiomic score to predict the likelihood of response of BMs to SRT and an analysis on the same cohort to create a radiomic score to predict the likelihood of BM occurrence. I completed a lobar distribution study to map the distribution of BMs from distinct primary malignancies.
Results:
My meta-analysis has demonstrated overall survival outcomes consistent with many case series of outcomes for limited BMs. My retrospective case series of a contemporary cohort of patients has shown overall survival outcomes above 13 months. My analysis of a large multi-centre, multi-platform cohort of patients with BMs has created a novel radiomic score. A further radiomic analysis has created a radiomic score which predicts the occurrence of a BM better than chance. Finally, my analysis of the lobar distribution of BMs in patients referred for SRT has provided novel insights into the distribution of disease.
Conclusions:
Contemporary patients with ten or more BMs can have prolonged overall survival outcomes following SRT. Radiomics combined with clinical factors can give accurate predictions for which BMs are likely to respond to SRT. The novel application of radiomics to predict BMs early shows promise but is not currently clinically useful. The lobar distribution of BMs in patients referred for SRT can be predicted and may have implications for treatment decisions.
Version
Open Access
Date Issued
2024-03
Date Awarded
2024-08
Copyright Statement
Creative Commons Attribution NonCommercial Licence
License URL
Advisor
Williams, Matthew
Angelini, Elsa
Glen, Robert
Publisher Department
Department of Surgery & Cancer
Publisher Institution
Imperial College London
Qualification Level
Doctoral
Qualification Name
Doctor of Medicine (Research) MD (Res)