A Rapid Hybrid VMAT-IMRT Planning Method Using An Abbreviated Beam Angle Optimization Search
D Schulze*, D Ionascu, N Tyagi, William Beaumont Hospital, Royal Oak, MISU-E-T-622 Sunday 3:00:00 PM - 6:00:00 PM Room: Exhibit Hall
Purpose: To generate hybrid VMAT-IMRT treatment plans by utilizing an IMRT beam-angle-optimality (BAO) search under a commercially available TPS without the aid of custom optimization software or high performance computing.
Methods: The high dose modulation provided by IMRT beams can be exploited to improve the quality of VMAT treatments. To achieve this, a VMAT treatment was created where the gantry pauses at predetermined angles to deliver IMRT segments. To determine IMRT BAO, an IMRT plan ('poly-IMRT') was made with many beams (>30) equally spaced around the patient. For practical reasons and to reduce the total time to approximately 1 hour, BAO was approximated by removing one beam out of the set and noting the new objective score. Determining this 'score penalty' for each of the beams serves as a proxy for true BAO. The hybrid plan was created by combining the VMAT arc with a user-determined number of top-ranked beams from the poly-IMRT set. The BAO from this approach was compared with a more rigorous method ('VMAT+1'), in which a VMAT plan was optimized with 1 IMRT beam at various angles, allowing a direct determination of objective score versus gantry angle.
The overall hybrid planning process was demonstrated by creating separate plans for a SBRT lung patient, with dose normalized to the limiting maximum aorta dose.
Results: Large score penalties from poly-IMRT coincided with large score benefits from VMAT+1, indicating both methods identified the same optimal beams. The VMAT, IMRT, and hybrid plans delivered the prescription dose to 84.3%, 85.6% and 87.7% of the PTV and had homogeneity indices of 1.38, 1.41, and 1.32 respectively. Normal tissue doses were within 0.5%.
Conclusion: The presented method can create hybrid VMAT-IMRT plans which combine delivery efficiency with improved target coverage. The planning process takes about an hour using a standard TPS.