Encrypted login | home

Program Information

Multi-Isocenter Breast Treatments On a Prototype Linear Accelerator: A Study of Interplay Effect and Robustness

T Netherton

T Netherton*, S Shaitelman , Y Li , P Nitsch , P Balter , S Gao , M Muruganandham , S Frank , S Hahn , A Klopp , L Court , The University of Texas MD Anderson Cancer Center, Houston, TX


TH-EF-FS1-11 (Thursday, August 3, 2017) 1:00 PM - 3:00 PM Room: Four Seasons 1

Purpose: We have investigated (1) the interplay effect between breathing and dynamic delivery and (2) the robustness of breast plans optimized with multiple couch positions. We are using a prototype linac with increased dose rate, limited field size, and increased MLC and gantry motion.

Methods: To study the dosimetric consequences of the interplay effect, 20 MOSFETs were implanted into a tissue-equivalent breast on a moving stage. Sinusoidal breathing motion was simulated with a 5mm amplitude and 5-sec period. Doses were recorded from four 4-field IMRT plans (200cGy/fraction) ranging in complexity (1.9 MU/cGy-10.5MU/cGy) under motion. Each plan was delivered five times; average standard deviation (σₐ) of dose was calculated for each MOSFET across all deliveries. Dosimetric robustness was evaluated by examining changes in target coverage (V95%>95%) due to couch travel uncertainties and patient motion between isocenters. This was measured for axillary(AX), supraclavicular(SCL), breast/chest wall(WB), and internal mammary(IMN) CTVs for five plans. Shifts of 0.5mm, 1mm, 2mm, and 4mm were incorporated into each plan to simulate uncertainties in superior-inferior directions.

Results: Interplay σₐ of dose increases linearly with plan complexity. The lowest and highest complexity plans had σₐ of 2.7cGy (range =1.7cGy-4.1cGy) and 7.4cGy (3.6cGy-14.2cGy), respectively. All CTVs were robust (V95%>95%) within 0.5mm. When patient motion is simulated by isocenter shifts in the inferior and superior directions, coverage, with exception of the IMN CTV, is robust. WB and IMN targets were the most robust within +/-2mm, on average. For isocenter error > 1mm, V95% fell below 95% for IMN(4/5 plans), AX(3/5 plans), and SCV(1/5 plans) targets.

Conclusion: The quality of breast treatments on this prototype machine are dependent on plan robustness and complexity. The interplay effect is small with low modulation but increases for highly modulated plans. If motion is more than 1mm between isocenter shifts, CTV coverage is negatively affected.

Funding Support, Disclosures, and Conflict of Interest: Funded by Varian Medical Systems

Contact Email: