Keywords:
Head and neck, Oncology, MR, MR-Diffusion/Perfusion, MR-Functional imaging, Radiation therapy / Oncology, VMAT and Tomotherapy, Cancer, Outcomes, Tissue characterisation
Authors:
C. Rumley1, M. Lee1, L. Holloway1, R. Rai1, M. Min2, D. Forstner1, A. Fowler1, G. Liney1; 1NSW/AU, 2QLD/AU
DOI:
10.1594/ranzcr2017/R-0109
Methods and materials
The study opened for recruitment in May 2016.
Patients receiving radical radiotherapy (with or without chemotherapy) are being prospectively recruited.
Scan setup is demonstrated in Fig. 1.
All patients are scanned on a 3 T wide bore scanner (Skyra,
Siemens,
Erlangen,
Germany) with standard anatomical imaging and functional studies as follows:
- Diffusion weighted imaging (DWI) using RESOLVE sequences with b-values 50,
800 and calculated 1400 s mm-2
- Dynamic contrast-enhanced imaging (DCE) using a TWIST sequence over 60 timepoints with temporal resolution of 5.64 s (Fig. 2)
- Blood oxygen level-dependent imaging (BOLD) using a multiple-echo gradient echo sequence - carried out twice,
eight minutes apart,
at each scan visit
Scans are carried out prior to,
during and after treatment according to the study schema (Fig. 3).
Parametric maps are created using Siemens MapIt and Tissue4D software,
then exported to MIM v6.6.7 (MIM Software,
Cleveland,
OH).
Contours are created for the primary tumour and nodal volumes on the anatomical images then transferred to the parametric maps (Fig. 4),
where they are manually edited for any mismatches and analysed.
Data are extracted from parametric maps as follows:
- Capillary transfer constant (Ktrans) - derived from analysis of contrast dynamics and indicative of perfusion in volume examined [3]
- Apparent diffusion coefficient (ADC) - derived from analysis of diffusion of water molecules and inversely related to cellularity in the volume examined [1,
2]
- Transverse relaxation rate (R2*) - elevated in hypoxia due to local magnetic susceptibilty changes induced by the presence of deoxyhaemoglobin [2]