A knowledge-based quantitative approach to characterize treatment plan quality: Application to prostate VMAT planning

Area of Science:

• Radiation Oncology
• Medical Physics
• Cancer Treatment

Background:

• Volumetric Modulated Arc Therapy (VMAT) is a common technique for prostate cancer treatment.
• Ensuring optimal organ at risk (OAR) sparing is crucial for minimizing treatment toxicity.
• Quantitative quality control (QC) methods are needed to assess and improve treatment plan (TP) quality.

Purpose of the Study:

• To develop and validate a quantitative QC tool for characterizing TP quality in prostate cancer radiotherapy.
• To estimate achievable OAR sparing using knowledge learned from prior VMAT plans.
• To identify suboptimal TPs and guide re-optimization for improved OAR protection.

Main Methods:

• A knowledge-based algorithm was developed using 450 prostate cancer VMAT plans (181 homogenous, 269 simultaneous integrated boost - SIB).
• The algorithm related OAR (rectum, bladder) absorbed doses to their proximity to the planning target volume (PTV).
• A metric (M_q,r value) characterized OAR sparing; a 90% probability ellipse defined a threshold for re-optimization.

Main Results:

• Re-optimization improved OAR sparing in 8/11 homogenous and 6/13 SIB plans initially outside the 90% ellipse.
• Further improvements were observed in 3/4 homogenous and 1/9 SIB plans between 80%-90% probability.
• Post-re-optimization, M_q,r values for bladder and rectum decreased, indicating enhanced OAR sparing.

Conclusions:

• The developed QC tool is effective for anatomy-specific assessment of TP quality.
• The tool successfully identifies suboptimal plans and predicts achievable OAR sparing for individual patient anatomies.
• This approach facilitates the optimization of VMAT plans for improved patient safety in prostate cancer radiotherapy.