Program Information
A TCP-Based Probabilistic Treatment Planning Framework Proof of Concept
H S Tsang*, P Ziegenhein , S Nill , U Oelfke , Joint Department of Physics at The Institute of Cancer Research and The Royal Marsden NHS Foundation Trust, London, UK SM2 5NG
Presentations
SU-K-FS1-5 (Sunday, July 30, 2017) 4:00 PM - 6:00 PM Room: Four Seasons 1
Purpose: Current practice in radiotherapy treatment planning uses dose prescription as surrogate for TCP, and planning margins to mitigate the effects of geometric uncertainties. We present a proof-of-concept study where dose distributions are optimized to satisfy population-based TCP coverage whilst being robust against geometric and biological uncertainties.
Methods: We generate a pseudo tumor cell density distribution to use as our biological target volume (BTV) by blurring the CTV with a 2mm Gaussian kernel. We modified the objective function to use the relative cell density to guide the optimization, delivering less dose to regions with lower cell density. A baseline TCP (TCPopt) is calculated by assuming the BTV receives a homogeneous prescribed dose. Plans are optimized to satisfy a specified TCP/TCPopt ratio for a stated fraction of the population. For a fixed set of OAR optimization objectives, the optimizer searches for the target’s minimum dose penalty that satisfies the probabilistic TCP objective. Monte Carlo simulation is used to evaluate the TCP distribution after every 25 iterations of fluence optimization, by rigidly shifting the target to model effects of geometric uncertainties, and sampling biological parameters when calculating the TCP. The whole process is repeated for a user-defined range of OAR objectives to generate a plan library for investigation. Five prostate patients were planned using our novel framework and via conventional means. We used 2mm for systematic and random uncertainties isotropically, and biological parameters alpha=0.35Gy⁻¹±0.05Gy⁻¹ and cell density=10⁷cm⁻³±1%
Results: We observed traditional plans to have 73.41%-85.93% of the population achieve a TCP/TCPopt≥0.98, as compared to our framework’s 87.36%-88.35%. Moreover, our probabilistic framework achieved a reduction of 4.61%-31.13% to the rectum’s V60, and 0.96%-31.61% to the bladder’s V60.
Conclusion: Our probabilistic framework produced treatment plans with improved TCP coverage, reduce high dose to OARs and are robust to geometric and biological uncertainties.
Funding Support, Disclosures, and Conflict of Interest: Research at The Institute of Cancer Research is supported by Cancer Research UK under Programme C33589/A19727 and NHS funding to the NIHR Biomedical Research Centre at The Royal Marsden and The Institute of Cancer Research.
Contact Email: