Purpose: to evaluate the Varian Ethos online-adaptive platform for prostate cancer including 1.auto-segmentation accuracy for CTV and OARs, 2.achievable dosimetric improvements, and 3.minimal margin necessary with daily adaptation.
Methods: 25 retrospective prostate patients were re-planned using an Ethos emulator, 12-field IMRT, and 2Gy fractions. CTVs included intact prostate and 1cm proximal seminal vesicles. Our clinical asymmetric margin was used: 3mm posterior, 5mm left/right/anterior, and 7mm superior/inferior. Adapted and non-adapted dose distributions for 10 fractions per patient were calculated using their on-treatment CBCTs. No corrections to the auto-segmentation were performed prior to adaptive plan optimization, however all contours were reviewed/corrected for dosimetric evaluation. The auto-segmentation was evaluated by comparing DVH metrics from auto-segmented versus corrected contours. Dosimetric improvements were measured by comparing DVH metrics from adapted and non-adapted doses to corrected contours. Minimum safe CTV-to-PTV margins were determined by replanning each fraction with a clinically-validated RapidPlan™ knowledge-based planning routine and symmetric 2-5mm margins. Then CTV-Dmin with reduced margins and daily adaptation was compared to standard margin without adaptation.
Results: 70% of CTV edits (88% Bladder, 90% Rectum), were less than 10% of the total volumes and resulted in small average changes of: 0.13%±2.9% CTV-D95%, 0.43%±3.3% Bladder-V90%, and 0.12%±5.1% Rectum-V90%. Without adaptation, CTV-Dmin was greater than 95% of prescribed dose for only 45% of fractions; adaptation with clinical margin increased this to 83%. For non-adapted fractions where the OAR-V90% exceeded its clinical threshold, adaptation reduced it on average by 0.6%±4.0%(Bladder) and 2.0%±5.5%(Rectum). For these cases, using 2mm margin preserved an improvement in tumor coverage(65% fractions CTV-Dmin>100%), and further improved the OAR-V90 metrics by 3.5%±6.0%(bladder) and 11.5%±4.7%(rectum).
Conclusion: we established Ethos’s online adaptive workflow without pre-treatment segmentation edits can substantially improve prostate CTV coverage and reduce OAR dose. This approach minimizes the impact on the clinical workflow and dosimetrically benefits patients.
Funding Support, Disclosures, and Conflict of Interest: Dr. Ray has a lab services agreement with Varian Medical Systems. Dr. Moore reports income for personal consulting and speaker's honoraria from Varian Medical Systems. This work was supported in part by the Agency for Healthcare Research and Quality (AHRQ R01HS025440)