Main Session
Sep 28
PQA 01 - Radiation and Cancer Physics, Sarcoma and Cutaneous Tumors

2162 - Magnetic Resonance Fingerprinting and IVIM for Differentiating Radiation Necrosis from Tumor Recurrence: A Preliminary Study

02:30pm - 04:00pm PT
Hall F
Screen: 17
POSTER

Presenter(s)

Lan Lu, PhD - Cleveland Clinic Taussig Cancer Center, Cleveland, OH

L. Lu1, M. G. Malkin2, J. H. Suh1, Y. Chen3, and S. T. Chao1; 1Department of Radiation Oncology, Cleveland Clinic Foundation, Cleveland, OH, 2Brain Tumor Center, Cleveland Clinic Foundation, Cleveland, OH, 3Department of Radiology, Case Western Reserve University, Cleveland, OH

Purpose/Objective(s): Magnetic Resonance Fingerprinting (MRF) is a novel imaging technique with potential for distinguishing radiation necrosis from tumor recurrence. This study aims to evaluate MRF-derived T1 and T2 relaxation times in patients with confirmed necrotic brain lesions and compare these values with normal brain tissue from healthy volunteers, as well as previously published data on gliomas and metastases. Additionally, we explore intravoxel incoherent motion (IVIM) imaging to assess structural and perfusion characteristics of necrotic tissues.

Materials/Methods: Two patients with three confirmed regions of radiation necrosis underwent MRF imaging, generating T1 and T2 relaxation maps. These values were compared to data from healthy volunteers with measurements from anatomically similar brain regions. Further comparisons were made with glioblastomas (GBM), low-grade tumors, and metastases from the literature. In addition, IVIM imaging was performed to evaluate diffusion and perfusion parameters. Pathologic confirmation was obtained through surgical resection.

Results: MRF-derived T1 relaxation times for the three necrotic lesions were 3879 ms, 2207 ms, and 2381 ms (± 210 ms). In contrast, normal brain tissue from the same anatomical locations in volunteers had a mean T1 value of 816 ± 57 ms, demonstrating evident difference. Similarly, necrotic lesions exhibited markedly increased T2 values of 145 ms, 401 ms, and 594 ms (± 56 ms), compared to normal brain tissue in volunteers, which had a mean T2 of 36 ± 5 ms.

When compared to previously published data on tumors, necrotic lesions exhibited significantly elevated relaxation times. The literature reports T1 values for glioblastomas, low-grade tumors, and metastases typically below 2000 ms, while T2 values are generally less than 170 ms. The markedly higher T1 and T2 values observed in our study have the potential to differentiate necrotic tissue from tumor recurrence.

Beyond MRF, IVIM imaging provided additional structural and perfusion insights. The apparent diffusion coefficient (ADC) values within necrotic lesions were 2–3 × 10?³ mm²/s, significantly higher than white matter (~0.6 × 10?³ mm²/s). On the other hand, necrotic lesions presented perfusion fraction (f) < 0.02, much lower than normal white matter (f > 0.05).

Conclusion: These preliminary findings suggest that MRF, particularly in combination with IVIM imaging, hold great potential in differentiation of radiation necrosis from tumor recurrence based on quantitative tissue properties. The elevated T1 and T2 values in necrotic tissues, along with distinct diffusion and perfusion characteristics from IVIM, provide a noninvasive imaging biomarker for identifying radiation-induced necrosis. This is an ongoing study, and additional MRI data are being acquired to further validate these findings and will be presented at the Annual Meeting.