Enhancing Estimation Accuracy of Prostate Cancer VMAT Planning: A Knowledge-based Approach Using Multiple Collimator...

Area of Science:

• Medical Physics
• Radiation Oncology
• Radiotherapy Planning

Background:

• Volumetric Modulated Arc Therapy (VMAT) is a key technique in modern radiotherapy.
• Accurate dose prediction is crucial for optimizing VMAT plan quality.
• Institutional models (Inst. model) often use fixed collimator angles, potentially limiting plan optimization.

Purpose of the Study:

• To evaluate a knowledge-based planning model incorporating multiple collimator angles (Multi-coll. model).
• To compare the estimation accuracy and plan quality of the Multi-coll. model against a standard Inst. model.
• To assess the performance across various collimator angles for prostate cancer VMAT.

Main Methods:

• Five institutions participated, utilizing VMAT for prostate cancer.
• The Inst. model used fixed collimator angles; the Multi-coll. model used angles of 10°, 30°, 50°, and 70°.
• Estimation accuracy (coefficient of determination, mean squared error) and dosimetric parameters (V20, V50, V90) were compared.

Main Results:

• The Multi-coll. model demonstrated higher estimation accuracy (lower MSE, higher R²) than the Inst. model.
• Mean absolute errors for rectal and bladder V90 were significantly lower with the Multi-coll. model.
• The Multi-coll. model resulted in lower rectal and bladder V20, indicating improved organ-at-risk sparing.

Conclusions:

• The Multi-coll. model offers superior estimation accuracy for VMAT treatment planning.
• This advanced model also provides slightly enhanced plan quality compared to conventional Inst. models.
• The findings support the adoption of multi-collimator angle knowledge-based planning for VMAT.