Performance of a knowledge-based planning model for optimizing intensity-modulated radiotherapy plans for partial...

Area of Science:

• Radiation Oncology
• Medical Physics

Background:

• Accelerated partial breast irradiation (APBI) is an effective treatment for early-stage breast cancer.
• Intensity-modulated radiotherapy (IMRT) is a common technique for delivering APBI.
• Optimizing APBI treatment plans can be complex and time-consuming.

Purpose of the Study:

• To evaluate a knowledge-based (KB) planning model for RapidPlan.
• The model was generated using a five-field IMRT beam strategy and strict dosimetric constraints for APBI.
• To assess the model's performance against conventionally optimized plans.

Main Methods:

• The RapidPlan KB model was trained on 64 APBI treatment plans.
• The model was validated using 120 APBI patients not in the training set.
• Dosimetric outcomes of KB plans were compared to clinical plans and trial constraints using Wilcoxon signed-rank tests.

Main Results:

• KB planning generated IMRT-based APBI plans in a single optimization without manual intervention.
• KB plans demonstrated comparable or superior dosimetry to conventionally optimized clinical plans.
• Key organs at risk (heart, contralateral breast, ipsilateral lung) showed no clinically significant dose differences; ipsilateral breast V50% was significantly lower with KB plans.

Conclusions:

• Knowledge-based planning for IMRT-based APBI offers equivalent or improved dosimetry over conventional inverse planning.
• The KB model can be reliably implemented in clinical practice.
• This approach facilitates the transfer of planning expertise, ensuring consistent APBI plan quality.