Overview. Proton Therapy in lung cancer 8/3/2016 IMPLEMENTATION OF PBS PROTON THERAPY TREATMENT FOR FREE BREATHING LUNG CANCER PATIENTS

Transcription

1 IMPLEMENTATION OF PBS PROTON THERAPY TREATMENT FOR FREE BREATHING LUNG CANCER PATIENTS Heng Li, PhD Assistant Professor, Department of Radiation Physics, UT MD Anderson Cancer Center, Houston, TX, 773 8/3/15 Overview Background Understand the motion-induced dose uncertainty in IMPT Implementation of IMPT for selected lung patients Conclusion Proton Therapy in lung cancer Potential to improves therapeutic ratio and allows dose escalation/acceleration By limiting dose to normal tissue Tumor Normal tissue 1

2 Rationale IMPT: Reduce normal tissue dose compared with IMRT in stage IIIB NSCLC (Zhang et al IJROBP 21) IMRT IMPT Motion Induced Dose Uncertainties The motion of the beam could interfere with the motion of target (interplay effect) May result in distortion of the planned dose distribution, local over- and under- dosage One of the major concerns for treating lung cancer with scanning beam proton Understanding Motion Induced Dose Uncertainties Determine the relationship between the delivered dose and the plan dose Statistical quantification of the dose uncertainty with consideration of Fractionation Breathing period Repainting Total delivery time 2

3 Repainting Zenklusen et al. PMB 21 Iso-layer repainting (within each energy) Not necessary helpful repainting could complete within a short time relative to breathing cycle Volumetric repainting (visit all energies, then repeat) To simulate passive scattering beam delivery The total irradiation time would increase considerably Energy change needs to be fast typically ~ 1 to 2 sec; PSI ~ 8 ms E n. E 2 E 1 Proton Beam Target volume Require large number of repainting Scanning motion and target motion are uncorrelated. Motion Induced Dose Uncertainties - Scanning Beam Proton Mean dose difference between 4D dose and delivered dose as a function of number of deliveries (fractions x repainting) 1 1/ =.3 s.8 Mean of D/ D max Number of Deliveries (k) Li et al. Med Phys 212 Delivery Sequence Optimization Minimize the motion-induced dose uncertainty by optimizing delivery sequence Evaluate the efficacy of spot delivery sequence by measurements and compare measurements to analytical model Patient study to validate the technique Reducing Dose Uncertainty for Spot-Scanning Proton Beam Therapy of Moving Tumors by Optimizing the Spot Delivery Sequence Li et. al. IJROBP 215 3

4 T 8/3/216 Delivery Sequence for SSPT Nominal Plan T T5 Nominal Spot Pattern Nominal Target Target at the time of delivery Delivered Spots Optimization of Delivery Sequence Nominal Plan T T5 No rescanning Same delivery time Nominal Spot Pattern Nominal Target Target at the time of delivery Delivered Spots 4D Dynamic Dose T1 T9 Images and deform vectors IMPT treatment plan 4D Dynamic Dose Simulator Y. Li et. al. Med Phys 214 Dose Distribution Breathing Pattern 4

5 Patient Study Optimization of Delivery Sequence Regular sequence (RS) Worst sequence (WS) Optimized sequence (OS) Max dose error in 1 fx 23.4% for WS 14.3% for RS 7.3% for OS Patient Study Li et. al. IJROBP 215 Understanding Motion Induced Dose Uncertainties The delivered dose converges to the 4D dose The dose difference between the delivered dose (4D Dynamic dose) and the planned dose (4D dose) reduces with Fractionation Breathing period Repainting Total delivery time 5

6 Distance [mm] Y[mm] Y[mm] Y[mm] Percent Volume [%] Percent Volume [%] 8/3/216 Summary of Current Techniques (FB) Margin based approach Needed but likely not sufficient Motion assessment 4DCT and Water equivalent thickness (WET) based analysis Optimization based techniques 4D optimization and dynamic dose analysis Robust optimization and analysis Delivery based techniques Scanning direction Rescanning Delivery sequence optimization Patient specific dose evaluation 4D Dynamic dose calculation Repeated CT and adaptive planning Clinical Implementation of Intensity Modulated Proton Therapy for Thoracic Malignancies Chang et. al. IJROBP 214 WET Analysis Example a) Reference b) Target X[mm] X[mm] c) Difference 5 12 d) Histogram X[mm] WET Difference [mm] WET Analysis Example 7 (a) Change in WET (b) Fraction of Surface with change in WET less than 5mm Mean WET change S 5mm passing rate Beam Angles [Degrees] Beam Angles [Degrees] 6

7 D95% (Gy) for ICTV, based CT/5 dose 8/3/216 WET change and Motion Induced Dose Uncertainties WETave (mm) Yu et. al. Med Phys 216 Robust Optimization 5 different plans were needed to decide the best treatment plan for a lung patient: IMRT, PSPT, SFO, MFO and robust MFO. Robust Optimization Original treatment plan for a patient (left, MFO- PTV; right, MFO-RO) Robust optimization in intensity-modulated proton therapy to account for anatomy changes in lung cancer patients Li et. al. Radiotherapy and Oncology 215 7

8 Robust Optimization Dose calculated on verification CT for a patient (left, MFO-PTV; right, MFO-RO) 4D Robust Optimization Interplay effect could be more significant for scanning beam 4D accumulated dose predicts the dynamic dose to patient after multiple fractions and/or repainting and deviates from planned (static) dose generated on single phase 4D robust optimization technique could be adopted to produce a deliverable plan with reduced interplay effect Liu et. al. IJROBP 216 Conclusion Accurate dose calculation is challenging for treating moving targets with proton A (not perfect) system was developed and running Continue developing and improving on the scanning beam proton treatment of moving targets 8

9 Acknowledgments The entire team of Radiation Oncology: Radiation Oncologists Physicists Dosimetrists Physics assistant Therapists Others 9