Development of Global and Local Spatial Descriptors of the Radiotherapy Dose: a tool to improve the search for dose-toxicity relationships in hollow organs

Gioscio, Eliana (2024). Development of Global and Local Spatial Descriptors of the Radiotherapy Dose: a tool to improve the search for dose-toxicity relationships in hollow organs. PhD thesis The Open University.

DOI: https://doi.org/10.21954/ou.ro.00101164

Abstract

External beam radiotherapy is a clinical standard for treating tumours, involving approximately 50% of cancer patients. This technique aims to administer a high dose of ionising radiation to the target volume while minimising exposure to surrounding healthy tissues. Technological developments in imaging (CT, MRI, PET) and treatment delivery technology have improved responses to treatments, increasing 5-year survival rates for many tumour sites. Therefore, research focuses heavily on reducing treatment complications (including late ones) and improving patients’ quality of life in cancer survivors.

The benefit of increasing the dose to the tumour target is limited by the presence of organs at risk (OAR). Current treatment plans are based on dose constraints for OARs and predictions of side effects development.

Normal-tissue complication probability (NTCP) models are used in clinical practice to help select the optimal treatment plan that minimises each individual’s risk of complications. However, current dosimetric models have some limitations. They reduce the three-dimensional dose distribution within the organ to a one-dimensional representation (dose-volume histogram, DVH), ignoring the spatial heterogeneity of doses and possible intra-organ radiosensitivity.

This has led to a methodological evolution towards more sophisticated models integrating local spatial descriptors of dose distributions. Recent studies have highlighted the potential of dose-surface histograms (DSH) and dose-surface maps (DSM) to overcome the limitations of DVH. These allow for a better understanding of local dose-response relationships, especially in hollow organs. Although DSM-based analysis is established, it remains fragmented and insufficient for predictive modelling due to the lack of automation in DSM creation, in-depth studies on large patient datasets, and inclusion of patient-specific risk factors.

In response to these challenges, we developed a dedicated software platform that enables automatic computation of DSH/DSM dosimetric metrics and real-world analysis to better understand global and local dose-response relationships for all relevant side effects after radiotherapy, all including modulation from patient specific risk factors.

The research was conducted at Fondazione IRCCS Istituto Nazionale dei Tumori (Milan, Italy), which provided access to data from two important clinical studies on prostate cancer. These studies served as training and use cases for developing the platform.

The document is structured into five chapters based on work presented at international conferences and in preparation for publication, all of which have been co-authored.

Viewing alternatives

Download history

Metrics

Public Attention

Altmetrics from Altmetric

Number of Citations

Citations from Dimensions

Item Actions

Export

About