Page 17 - ARNM-3-3

Advances in Radiotherapy
& Nuclear Medicine Review of image-guided adaptive radiotherapy

precisely monitoring the motion trajectory of the CTV, imaging (DWI), and dynamic contrast-enhanced imaging.
MRI helps reduce the expansion range of the PTV, thereby Leveraging MRI’s quantitative voxel measurements,
minimizing radiation dose exposure to OARs while these functional imaging sequences hold promise as
improving RT precision. In addition, MRI acquisition does potential tools for predicting tumor radiosensitivity and
not involve any additional radiation, protecting patients radioresistant biological characteristics. For example,
from unnecessary exposure to harmful effects. DWI’s apparent diffusion coefficient maps can assess
MRI-guided adaptive radiation therapy (MRIgART) tumor cell density, perfusion imaging can monitor
provides two adaptive modes: adapt to position (ATP), angiogenesis, elastography enables the analysis of tumor
which corrects positioning deviations, and adapt to shape stiffness/matrix, and MRI relaxometry aids in detecting
(ATS), which accounts for deformation-related errors. In tumor hypoxia. 56,57 Despite these advantages, the
the ATP mode, the treatment plan is adjusted by correcting widespread implementation of MRIgART faces challenges,
the position of the target area based on image registration including high equipment costs, longer imaging times,
results, followed by re-optimization using target area and the complexity of integrating MRI into routine RT
contours derived from the planning CT scan to formulate workflows. These limitations hinder its universal adoption,
an updated treatment plan. However, due to the fixed making MRIgART less accessible compared to CT- and
nature of the MR-Linac treatment couch, adjustments are CBCT-guided ART.
vital to account for shifts in the plan’s center point, limiting MRIgART has also proven effective in improving target
its alignment with CBCT-guided treatment modes. On coverage and reducing toxicity. Parikh et al. conducted
58
the other hand, the ATS mode primarily addresses errors a prospective multicenter Phase II trial of 136 borderline
resulting from tumor and OAR deformations. This is resectable or locally advanced pancreatic cancer patients
achieved by transferring the contoured structures from treated with MRI-guided adaptive SBRT. The 1- and 2-year
the planning CT to MR images, followed by recontouring overall survival (OS) rates were 65% and 40.5%, respectively,
or adjusting target areas on the MRI scan. A “virtual CT” with no acute grade ≥3 gastrointestinal toxicities attributed
is subsequently generated by mapping the average CT to ART. Late grade ≥3 toxicities were reported in 4.6% and
values (Hounsfield unit values) from the CT scan onto the 11.5% of cases. Ejlsmark et al. reported a Phase II trial of
59
corresponding anatomical structures on the MRI scan. 28 locally advanced pancreatic cancer patients treated with
The treatment plan is then re-optimized within the MRI MRI-guided adaptive SBRT, showing median progression-
environment to produce an adaptive treatment plan. free survival and OS of 7.8 and 16.5 months, respectively.
However, the implementation of MRIgART involves Six patients experienced grade III toxicities (jaundice,
a series of sequential procedures, including MRI nausea, vomiting, or constipation), and one had a grade IV
60
scanning, image registration, target adjustment, and plan duodenal perforation. Song et al. presented the first case
optimization. Consequently, this leads to a significantly report of using a 1.5T MR-LINAC (Elekta Unity) for gastric
longer patient time on the MR-Linac treatment couch cancer treatment. The patient was administered daily
compared to conventional linear accelerator treatments. scopolamine to suppress gastrointestinal motility, which
The completion of one fraction of MRIgART typically takes enabled the acquisition of high-contrast T2-weighted MRI
approximately 32 min, and the ATP mode demonstrates a images. AI-assisted contouring, combined with manual
relatively faster execution (within 30 min), while the ATS adjustments, facilitated real-time adaptive planning.
mode requires slightly more time (approximately 42 min). However, this process extended the treatment time to
This represents a longer duration compared to CBCT and 45 min per session. Over 25 fractions, compared to the
CT-guided ART. For instance, in treating patients with target coverage of CBCT-guided therapy, as measured
liver metastasis using a 1.5 T MR-Linac system, the median by the Dice similarity coefficient (DSC), improved
time from patient entry into the MRgART treatment room significantly from 0.36 to 0.82 (p<0.01). In 2 fractions,
until completion of RT ranged between 35 and 43 min. traditional methods failed to achieve target coverage
55
In cases where bladder filling is necessary for patients with (DSC = 0), while online ART ensured 100% target coverage
cervical, rectal, or prostate cancer, the extended treatment throughout all fractions. The maximum dose to the
time may potentially impact both the treatment experience duodenum decreased from 54.2 Gy to 47.6 Gy (p=0.03).
and its accuracy. Scopolamine effectively reduced gastric motility by 83%,
Compared to CT and CBCT-guided ART, MRIgART with anatomical displacement <3 mm as observed on MRI.
provides distinct functional imaging capabilities, Hypofractionated stereotactic ART is used for large
encompassing fluid-attenuated inversion recovery brain metastases. The reduction in tumor volume during
(FLAIR), short tau inversion recovery, diffusion-weighted treatment intervals enables replanning with smaller target

Volume 3 Issue 3 (2025) 9 doi: 10.36922/ARNM025110012