Canine meningiomas treated with three‐dimensional conformal radiation therapy require magnetic resonance imaging to avoid a geographic miss

Abstract

Exact target volume definition is an essential prerequisite for modern radiotherapy treatment planning. Contouring the gross tumor volume of brain tumors on computed tomography (CT) images coregistered with magnetic resonance images is standard practice in human medicine. In this retrospective study, including only cases with an imaging diagnosis of meningioma, we hypothesized that the gross tumor volume contoured from the contrast-enhanced magnetic resonance imaging (MRI) (gross tumor volume-MRI) is larger when compared to the gross tumor volume contoured using contrast-enhanced CT (gross tumor volume-CT). A total of 22 dogs were included in the study. Interestingly, there was a significant statistical difference between the gross tumor volume-CT and the gross tumor volume-MRI (P = 0.001). The gross tumor volume-MRI was larger than the gross tumor volume-CT in all, but two cases. The mean ratio between gross tumor volume-MRI and gross tumor volume-CT was 1.43 (range 0.84-4.00). The mean overlap between gross tumor volume-CT and gross tumor volume-MRI was 68% (range 22-86%), while a mean of 26% of the composite gross tumor volume was defined only by MRI (range 4-76%) and a mean of 6% only by CT (range 0-22%). These findings suggest that CT and MRI are complementary modalities in radiation planning of meningiomas and their composite volume should be used to avoid geographical miss of neoplastic tissue. When the MRI is not available for planning, a margin of 0.3 cm around the gross tumor volume-CT could reduce the probability of a geographical miss. However, such numerical correction cannot be applied to the clinical practice until it is validated in a properly designed treatment planning study.

Use of MRI increases interobserver agreement on gross tumor volume for imaging‐diagnosed canine intracranial meningioma

“…Our finding of larger GTV volume for delineation on CT‐MRI than for delineation on CT is consistent with previous reports for human meningioma 13 and primary brain tumors, 7,22 as well for canine meningioma 10 . Two of the human studies and the veterinary study compared tumor volume delineated using only CT or MRI, while the third human study compared tumor volume delineated using CT or fused CT‐MRI 22 …”

“…Interobserver and intraobserver agreement between canine meningioma GTV contouring on CT and MRI has been studied without using metrics such as DSC and CI that take into account differences in spatial distribution of the tumor. 10 The findings in our study are similar to the aforementioned study 10 for volume, although absolute comparisons cannot be made due to differences in study design. The median absolute volume of contoured GTV was 1.5 times larger on fused CT-MRI than on CT alone, similar to the relative volume difference in the study by Berlato et al of 1.48.…”

“…A 2016 study 9 examined GTV contours for two canine nasal tumors created by four individuals at a single institution, and found considerable variation in GTV volume and unacceptable contour overlap for one of the two dogs. In contrast to studies of human brain tumor patients 7,8 reporting a large interobserver variability, a 2017 study 10 of interobserver variability in GTV volume in nine dogs with imaging‐diagnosed intracranial meningioma reported excellent agreement between three individuals at the same institution. The large interobserver variability in GTV delineation found in human brain tumor patient studies, and the lack of information regarding agreement on canine meningioma GTV delineation between veterinary radiation oncologists working at different institutions led to our first objective: to describe agreement on canine intracranial meningioma GTV between board‐certified veterinary radiation oncologists at multiple institutions.…”

“…21 Our finding of larger GTV volume for delineation on CT-MRI than for delineation on CT is consistent with previous reports for human meningioma 13 and primary brain tumors, 7,22 as well for canine meningioma. 10 Two of the human studies and the veterinary study compared tumor volume delineated using only CT or MRI, while the third human study compared tumor volume delineated using CT or fused CT-MRI. 22 The DSC is a measure of relative spatial overlap between pairs of observers and has been used to assess agreement in numerous human studies of different anatomic sites, 2 although very little data has been published using DSC to assess agreement for intracranial tumors.…”

Treatment outcomes and target delineation utilizingCTandMRIin 13 dogs treated with a uniform stereotactic radiation therapy protocol (16 Gy single fraction) for pituitary masses: (2014‐2017)

“…In this study, target volume contours based on MRI yielded larger volumes than CT alone in the majority of cases (11/13), and the combined CT and MRI contours yielded larger volumes than either modality alone in all cases. This is comparable with a study in which authors utilized both CT and MRI for contouring meningiomas in dogs 16 . In that study, MRI yielded a mean volume that was 26% larger than on CT. Based on these results, in addition to the finding that CT scans alone were subjectively deemed inadequate for PM contouring prior to SRT in >50% of the cases by two ROs in this study, both modalities should ideally be utilized.…”