ESTRO 2024 - Abstract Book

S904

Clinical - CNS

ESTRO 2024

Simplifying diagnosis of radiation necrosis and tumour recurrence following SRS for brain metastases

David A Newsome 1 , Phillip Hingley 2 , Helen Truong 1 , Kylie Jung 1,3 , Benjamin J.C. Quah 1,3 , Chaturica Athukorala 2 , Farhan M Syed 1,3 1 Canberra Region Cancer Centre, Radiation Oncology, Canberra, Australia. 2 Canberra Hospital, Medical Imaging, Canberra, Australia. 3 John Curtin School of Medical Research, Australian National University, Division of Genome Sciences & Cancer, Canberra, Australia

Purpose/Objective:

Intracranial radiation necrosis (RN) is likely to be encountered more frequently in clinical practice with technical advancement in radiation treatment delivery, together with clinical data favouring stereotactic radiosurgery (SRS) for an increasing number of patients with brain metastases (BM). Furthermore, with improvement in systemic treatments patients may have a higher likelihood of developing RN following SRS due to longer survival. Up to 30 percent of patients may develop RN following SRS, and a similar proportion of BM will recur locally at 12 months. Radiological diagnosis of RN following SRS, and distinguishing it from local recurrence, remains a clinical challenge in a significant proportion of patients, and invasive histological assessment is regarded as the gold standard. A few MRI-based lesion characteristics favour the presence of RN; namely, the ratios of T1-contrast enhancement (T1c) to T2-hypointensity (T2hypo), and T1c to peri-lesional T2-hyperintensity (T2hyper). Available literature suggests presence of RN with a low lesion quotient (T2hypo:T1c), a high lesion matching ratio (T1c:T2hypo) or a high oedema ratio (T2hyper:T1c), using either linear measurements or 3D volumetric analysis. Our study aims to compare the simpler and more efficient linear calculations to more accurate but time consuming 3-dimensional volumetric assessments of these MRI lesion ratios, with validation against histological confirmation, to assist in the diagnosis of RN or tumour recurrence following SRS in daily clinical practice.

Material/Methods:

A retrospective review of patients who received linear accelerator-based SRS for one or more brain metastasis(es) and had resection for subsequent enlargement of the irradiated lesion at our institution between 2013 and 2021 was conducted. MRI brain performed immediately prior to establishing the histological diagnosis was reviewed, and lesion quotient, lesion matching and oedema ratios were calculated using the two methods. The simple method measured the maximal extent in 2-dimensions for each MRI characteristic and was then multiplied by 0.6 to derive an ellipsoid volume. The 3D method employed manual contouring of each MRI characteristic to generate a volume by the planning software. Two-way ANOVA was then used to compare the means from each ratio group to assess for a statistically significant difference between RN, TR or mixed RN/TR pathology and validated with the histological diagnosis.