INTRODUCTION
Medulloblastoma is the most common childhood malignant central nervous
system (CNS) tumor [1]. Peak incidence occurs at age 7 with slightly
greater incidence in males [1,2]. A large proportion of
medulloblastoma patients have craniospinal fluid (CSF) spread at the
time of diagnosis. The standard of care for medulloblastomas involves
surgical resection, craniospinal irradiation (CSI), and chemotherapy
[3]. For average-risk patients the 5-year survival rate is over 80%
while high-risk patients have a 5-year survival rate of less than 50%
[2,4].
CSI presents challenges due to its large target volume which extends
beyond the 40 cm x 40 cm field size limitation of a commonly used C-arm
linear accelerator collimator opening [5, 6]. Therefore, the use of
multiple plan isocenters provides a solution to this limitation by
dividing the target volume into three fields—the whole brain, the
upper spine, and the lower spine. CSI is commonly performed using 3D
conformal radiation therapy (3D-CRT) technique which is prone to errors
due to the complexity of the planning and the treatment delivery setup
[7-12]. This technique results in dose inhomogeneity and
non-conformality which yields significant dose to the anterior of the
spine target volume. 3D CSI also requires feathering the junctions
resulting in multiple plan pairs, gap calculation and couch rotations,
making the planning and treatment procedures complex and cumbersome and
prolonging treatment times.
Overall, Volumetric Modulated Arc Therapy (VMAT) CSI has been observed
to create plans with better dose conformality, better dose homogeneity,
greater normal tissue sparing, low sensitivity to positioning errors,
and shorter treatment time compared to 3D-CRT CSI [13-17]. While
VMAT can produce clinically favorable plans even with setup errors of up
to 3 mm margin, accurate patient alignment with minimal setup remains
important. A multicenter study conducted by Gram et al. showed daily
image guidance with 6-DoF couch corrections was found optimal in
significantly reducing positioning errors and uncertainties for
pediatric CSI patients [11].
While daily image guidance and 6-degrees of freedom couch corrections
can assist in optimizing patient setup, the inherent risks for
positioning errors and uncertainties cannot be eliminated for VMAT CSI
due to the use of multiple isocenters and field matching.
Helical-delivery radiation treatment techniques such as
TomotherapyTM can reduce these risks associated with
multi-center CSI treatments by using a ring-based gantry to deliver a
single field CSI treatment as the patient moves into the treatment ring
[18-20]. A study by Lee et al. reported Tomotherapy CSI to have
acceptable inter-fractional and intra-fractional errors, and setup
verification based on the measurements and evaluations of treatment
setup for 83 patients [19]. In addition, Tomotherapy CSI techniques
have been demonstrated to produce highly conformal and homogeneous
treatment plans compared to 3D CSI [21,22].
RefleXionTM (RefleXion Medical Inc., Hayward, CA) is a
novel Positron Emission Tomography (PET) treatment modality that
similarly utilizes a ring-based gantry for axial step-and-shoot IMRT
delivery. The first clinical installation of RefleXion X1 was recently
conducted at our institution [23,24]. The RefleXion X1 design
provides potential advantages to CSI treatments using a single isocenter
which can potentially decrease complexity of planning, image guidance
and delivery reducing risk of shift and localization errors. This study
aims to test the feasibility of treatment planning of X1 CSI and compare
the plan quality and beam-on time to the current standard of care at our
institution – VMAT CSI planned in Eclipse and delivered using Trilogy
linear accelerator (Varian Medical Systems, Palo Alto, CA).