Dosimetric Benefits of Online Adaptive Replanning for Lung Cancer with Multiple Targets
F Liu*, A Tai, E Ahunbay, E Gore, C Johnstone, X Li, Medical College of Wisconsin, Milwaukee, WISU-E-CAMPUS-T-1 Sunday 3:00PM - 6:00PM Room: Exhibit Hall
Purpose: There exist interfraction relative volume variations and/or independent motions between multiple targets in radiotherapy (RT) of lung cancer which cannot be fully accounted for by the current IGRT repositioning or offline adaptive replanning. This work is to study the dosimetric benefits of an online adaptive replanning scheme that accounts for these variations.
Methods:Daily CT data acquired for 5 lung cancer patients treated with IGRT using an in-room CT (CTVision, Siemens) were analyzed. On each daily CT set, contours of the targets [GTV, CTV, and involved nodal regions] and organs at risk (OARs) were generated by populating the planning contours using an auto-segmentation tool based on deformable registration (ABAS, Elekta) with manual editing. For each patient, an IMRT plan was generated based on the planning CT using a planning system (Panther, Prowess Inc.) with a prescription dose of 60 Gy in 2 Gy fractions. Three plans were generated on each daily CT set: an IGRT (repositioning) plan by copying the original plan with the required shift, an online adaptive plan generated rapidly by tailoring the original plan to conform to the daily anatomy, and a new fully re-optimized plan based on the daily CT.
Results:The averages of dose volume quantities are: the PTV-V100=95.3±0.2%, 92.7±4.3%, 95.2±0.2%, 95.2±0.2%, the CTV-V100=99.1±0.6%, 96.8±3.6%, 98.8±0.7%, 98.7±0.7%, the lung V20Gy=1074±437 cc, 1077±394 cc, 1049±371 cc, 1015±715 cc, for the original, repositioning, adaptive, and re-optimized plans, respectively. Paired t-tests show that the adaptive plans are statistically significantly better than the repositioning plans and comparable with the re-optimized plans.
Conclusion:Compared to IGRT repositioning, online adaptive replanning shows dosimetric benefits for lung cancer RT of multiple targets with relative volume variations and independent motion between the targets, ensuring that the original plan quality is delivered.