Modelling normal tissue toxicity in pancreatic chemoradiotherapy

<p>The relationship between radiation dose and toxicity in pancreatic chemoradiotherapy is not well understood. Clinically, excessive normal tissue toxicity is avoided by placing a constraint on the volume of the organ receiving a dose above a threshold. Dose-volume constraints lack spatial...

Deskribapen osoa

Xehetasun bibliografikoak
Egile nagusia: Witztum, A
Beste egile batzuk: Hawkins, M
Formatua: Thesis
Argitaratua: 2018
Deskribapena
Gaia:<p>The relationship between radiation dose and toxicity in pancreatic chemoradiotherapy is not well understood. Clinically, excessive normal tissue toxicity is avoided by placing a constraint on the volume of the organ receiving a dose above a threshold. Dose-volume constraints lack spatial information which may be important in determining normal tissue response.</p> <p>Spatial dose distribution information can be found in dose-surface maps (DSMs). A dose-surface map is a 2-dimensional virtual unwrapping of the surface dose of on organ. Due to the complex geometry of the duodenum, previous methods for unwrapping tubular organs for spatial toxicity modelling are insuficient. A geometrically robust method for producing dose-surface maps, specifically for the duodenum, was created in order to characterise the spatial dose distribution.</p> <p>This unwrapping methodology was shown to be generalisable to simple organs such as the stomach, and extendable to more complex organs such as the bronchial tree. A graphical user interface to create dose-surface maps from organs in commercial treatment planning systems was also demonstrated.</p> <p>Duodenal and stomach dose-surface maps were created for patients from two pancreatic chemoradiotherapy trials, ARCII and SCALOP, treating locally-advanced pancreatic cancer (LAPC). New spatial features were extracted from dose-surface maps and their correlation with upper-gastrointestinal toxicity quantified and compared to traditional dose-volume metrics. The predictive power of some of these new metrics were found to be superior to dose-volume metrics in the stomach, but no significant correlation was found in the duodenum.</p> <p>Duodenal motion throughout treatment occurs both due to respiratory motion and peristalsis. On board imaging using cone-beam CT (CBCT) during individual treatment fractions shows that interfraction variability is non-systematic and cannot be predicted. While abdominal compression is able to restrict some interfraction motion, it is unable to control peristaltic changes. Accumulated duodenal dose-surface maps from these images were created to investigate the differences between the planned and delivered dose.</p>