Detection of an internal density change in an anthropomorphic head phantom via tracking of charged nuclear fragments in carbon-ion radiotherapy

Abstract

Background

Carbon-ion radiotherapy provides steep dose gradients that allow the simultaneous application of high tumor doses as well as the sparing of healthy tissue and radio-sensitive organs. However, even small anatomical changes may have a severe impact on the dose distribution because of the finite range of ion beams.

Purpose

An in-vivo monitoring method based on secondary-ion emission could potentially provide feedback about the patient anatomy and thus the treatment quality. This work aims to prove that a clinically relevant anatomical change in an anthropomorphic head phantom may be detected via charged-fragment tracking during a treatment fraction.

Methods

A clinically representative carbon-ion treatment plan was created for a skull-base tumor in an anthropomorphic head phantom. In order to imitate an inter-fractional anatomical change — for example, through tissue swelling or mucous accumulation — a piece of silicone was inserted into the nasopharynx. Fragment distributions with and without the silicone insert were subsequently acquired with a mini-tracker made of four hybrid silicon pixel detectors. Experimental irradiations were carried out at the Heidelberg Ion Beam Therapy Centre (HIT, Germany). FLUKA Monte Carlo simulations were performed to support the interpretation of the experimental results.

Results

It was found that the silicone causes a significant change in the fragment emission that was clearly distinguishable from statistical fluctuations and setup uncertainties. Two regions of fragment loss were observed upstream and downstream of the silicone with similar amplitude in both the measurement and the simulation. Monte Carlo simulations showed that the observed signature is a consequence of a complex interplay of fragment production, scattering, and absorption.

Conclusions

Carbon-ion therapy monitoring with charged nuclear fragments was shown to be capable of detecting clinically relevant density changes in an anthropomorphic head phantom under realistic clinic-like conditions. The complexity of the observed signal requires the development of advanced analysis techniques and underscores the importance of Monte Carlo simulations. The findings have strong implications for the ongoing InViMo clinical trial at HIT, which investigates the feasibility of secondary-ion monitoring for skull-base cancer patients.