1163 - PSMA PET Nodule Discordance Yields Suboptimal Dosimetric Coverage in Patients with High-Risk Prostate Cancer Treated with MRI based SBRT and Intraprostatic Boost
Presenter(s)
Q. Wang, N. Bhargava, G. Kuzmin, R. V. Kavadi, J. A. Aronovitz, D. R. Schmidt, I. D. Kaplan, and N. Aghdam; Department of Radiation Oncology, Beth Israel Deaconess Medical Center, Boston, MA
Purpose/Objective(s): Partial gland intraprostatic boost (IPB) in patients treated with stereotactic body radiation therapy (SBRT) offers an attractive option to enhance treatment efficacy while mitigating toxicity associated with whole gland dose escalation. However, uncertainty remains as to whether the target for dose escalation should be delineated based on MRI or PET. In this study, we evaluate the anatomic, volumetric, and dosimetric differences between MRI-based treatment planning for prostate SBRT with IPB and PET-based segmentation.
Materials/Methods: Between April 2023 and May 2024, 20 patients with high-risk prostate cancer were treated with MRI-based IPB. Of the 20 patients 13 received a treatment regimen consisting of 2 fractions of SBRT boost (19 Gy), while 7 patients received standard 5 fractions regimen (36.25 Gy). PET PSMA scans were retrospectively fused with the planning images, and PET-avid nodules were contoured. We evaluated the anatomic, volumetric, and dosimetric parameters between the PET-defined and MRI-defined nodules. Paired samples Wilcoxon test was used to compare the differences in volume and dosimetric coverage of the nodules. Anatomic discordance was defined as either the absence of a PET nodule on initial staging scans, or in cases where the MR and PET nodules did not co-localize within the gland.
Results: Twenty patients with high-risk prostate cancer were evaluated. MRI-based segmentation delineated intraprostatic nodules with a median ± IQR volume of 1.44 ± 2.22 cm³, whereas PET-based segmentation identified nodules were significantly larger 2.23 ± 3.34 cm³ (p=0.04). Anatomic discordance was observed in 7 (35%) patients; in 2 of these cases there were no corresponding PET nodules, while in 5 patients the lesions identified on MRI and PET did not co-localize within the gland. Dosimetric analysis revealed that the MRI-based monotherapy planning achieved a median D95% of 45.3Gy for MRI-defined nodules compared to 41.7Gy for PET-defined nodules, and 23.5Gy vs. 21.6Gy for the boost (p <0.001). Consequently, in cases of anatomic discordance, the boost optimized to the MRI-defined targets resulted in a median (IQR) coverage of only 115% (1.6%) of the intended dose for PET-defined nodules, versus 124% (0.7%) for MRI-defined nodules.
Conclusion: This retrospective analysis revealed that anatomic discordance exists in a substantial number of patients, with PET-defined nodules exhibiting larger volumes on average compared to MRI-defined nodules. Moreover, optimizing the boost to MRI-defined nodules may result in under-dosing the PET-defined nodules. These findings may inform future dose escalation studies as IPB techniques evolve, highlighting the need to carefully consider the clinical implications of using MRI versus PET for target delineation in high-risk prostate cancer IPB.