Personal tools
You are here: Home Research site Dissertations 2017 Tristan van Heijst

2017 Tristan van Heijst

Towards breast cancer radiotherapy on the MRI-linac


With the introduction of the magnetic resonance imaging (MRI)-linear accelerator (linac) at University Medical Center Utrecht, on-line MRI guidance for radiation therapy (RT) treatments is possible. This PhD thesis explores technical implementation of the MRI-linac for local and regional breast cancer RT treatments, for which several aspects are addressed. Dosimetric effects due to the presence of the 1.5 T magnetic field inside the MRI-linac were studied. For whole-breast RT treatments, the skin dose can be increased significantly by a magnetic field, which is a potential disadvantage. However, for partial-breast RT treatments, magnetic-field-induced effects are expected to be negligible. The relevance of motion during RT delivery is increasing, given smaller target volumes and higher fraction dose, thereby making geometric miss more likely. It is demonstrated that for breast RT, this motion is generally limited, although large displacements are sparsely observed. Large shifts should be accounted for in order to ensure safe and adequate RT delivery. For regional RT, MRI was optimized in volunteers to image individual axillary lymph nodes (LNs), and evaluated quantitatively and qualitatively for suitability in RT planning. MRI of LNs was validated in breast cancer patients, who were scanned in supine treatment position, both before and after (axillary) surgery. MRI was feasible and numbers of LNs were highly reproducible before vs. after surgery. On MRI, LN-based target volumes are considerably smaller than standard anatomy-based targets. MRI guidance could enable new RT treatments for breast cancer patients. For instance, additional to conventional loco-regional RT, it is dosimetrically feasible to deliver an ablative RT dose to single LNs, regardless of LN location. In the future, breast-cancer patients may fully benefit from MRI guidance in the MRI-linac.



dissertation - full text