2950 - Dosimetric Assessment of an MR-Only Bulk Electron Density Workflow In Stereotactic MR-Guided Adaptive Radiation Therapy (SMART) for Adrenal Metastases
Presenter(s)
T. Bejarano1, N. Bassiri2, E. Ziegler3, M. D. Chuong1, R. Kotecha4, M. P. Mehta5, R. P. Tolakanahalli1, A. Gutierrez6, and K. E. Mittauer1; 1Department of Radiation Oncology, Miami Cancer Institute, Baptist Health South Florida, Miami, FL, 2University of Texas Health Science Center at San Antonio, San Antonio, TX, 3Creighton University, Omaha, NE, 4Miami Cancer Institute, Baptist Health South Florida, Miami, FL, 5Department of Radiation Oncology, Miami Cancer Institute, Miami, FL, 6Department of Oncological Sciences, Herbert Wertheim College of Medicine, Florida International University, Miami, FL
Purpose/Objective(s): MR-guided online adaptive radiotherapy (oART) allows for daily plan adaptation using electron density (ED) assignment for dose calculation. The current standard for low-field MR Linac (MRL) systems uses a deformed CT, requiring time-consuming manual edits during oART. We developed an abdominal MR only bulk ED (Bu-ED) assignment technique for low-field MRL to enable fast oART. We aim to evaluate clinically significant differences between Bu-ED versus deformed CT ED (DCT-ED) for adrenal stereotactic MR-guided oART (SMART).
Materials/Methods: 36 adrenal metastases SMART patients treated to a median dose of 50 Gy in 5 fractions (fx) on a 0.35 T MRL were analyzed retrospectively. Patients were treated with Bu-ED workflow with population ED assignments: vertebral bodies as bone, external (body) as water, and abdominal gas as air. Retrospectively, the simulation CT was deformed to primary MR of fx1, and manual ED corrections were applied to quantify DCT-ED. For each case, dose was calculated using both Bu-ED and DCT-ED assignments on identical daily anatomy (fx1). Mean doses (Gy) were quantified for the target (PTV D95%), spinal cord (Dmax), and kidneys (Dmean). Volume differences (cc) were analyzed for V35Gy/V40Gy of the stomach, duodenum, small bowel, and V38Gy/43Gy for large bowel. A paired-sample t-test (p<0.05) was used to compare the groups. Clinical on-table time for oART density segmentation were acquired for 27 BU-ED adrenal patients (~135 fx) and 39 DCT-ED liver patients (~195 fx).
Results: PTV D95% mean was similar between Bu-ED and DCT-ED (44.6 ± 7.9 Gy vs. 44.0 ± 7.9 Gy, p=0.16). The spinal cord Dmax mean was consistent (15.5 ± 2.7 Gy vs. 15.5 ± 2.7 Gy, p=0.33), as was the kidneys Dmean (4.6 ± 1.9 Gy vs. 4.6 ± 1.9 Gy, p=0.19). Gastrointestinal (GI) OARs: stomach V35Gy had a statistically significant increase (0.09 ± 0.13 cc vs. 0.10 ± 0.15 cc, p=0.03). However, stomach V40Gy had no significant difference (0.01 ± 0.02 cc vs. 0.01 ± 0.03 cc, p=0.20). Other GI OAR metrics were non-significant, including duodenum V35Gy (0.04 ± 0.10 cc vs. 0.04 ± 0.14 cc, p=0.64), small bowel V35Gy (0.05 ± 0.11 cc vs. 0.05 ± 0.10 cc, p=0.78), and large bowel (0.00 ± 0.00 cc for both methods). ED contouring mean ± SD time was statistically significant for adrenal Bu-ED vs. liver DCT-ED (7.8 ± 3.2 vs.15.2 ± 7.4 min, p<0.001).
Conclusion: In adrenal SMART, Bu-ED and DCT-ED did not have clinically significant differences in target coverage, kidney sparing, or high-dose GI OAR volumes. A minor statistical difference for stomach V35Gy (0.02cc) observed and may have potential to impact the plan normalization if an iso-toxic normalization approach to the stomach is used. Adrenal Bu-ED approach is feasible and significantly reduces contouring time (2x reduction), while also eliminating CT simulation resources