Purpose: To investigate the potential of knowledge-based planning (KBP) in producing consistent, protocol-compliant treatment plans in the context of hypofractionated intensity-modulated radiotherapy (IMRT) for left-sided breast cancer patients on the Hypofractionated Irradiation At Regional Nodal Area for Breast Cancer Vs Existed Standard Treatment(HARVEST) clinical trial.
Methods: A KBP model was developed for left-sided breast cancer patients receiving postmastectomy radiotherapy involved with regional nodal irradiation. Patients were prescribed to 40.05Gy/15fx. Internal validation of the KBP model was applied to 39 patients. Plans were re-optimized using the KBP model without manual modification. The initial and KBP plans were compared via their protocol compliance, target coverage, and dose to the surrounding normal tissues.
Results: The KBP model generated plans met all protocol objectives in a single optimization when tested on internal cases. KBP resulted in comparable target coverage, superior organ sparing as compared to initial plans. Quality improvement with regard to the normal tissue sparing was observed in more than 64% of the cases. KPB plans outperformed the initial plans for the heart, mean dose ( 293 ± 94.5 cGy vs 314.7 ± 92.4 cGy), V8 (7.6 ± 3.5% vs 8.16 ± 3.4%). The sparing was modest for ipsilateral lung, mean dose (966.3 ± 98.6cGy vs 979.5 ± 97.9cGy), V16 (38.6 ± 3.4% vs 39.1 ± 3.8%). KBP plans were also found to be more consistent in several metrics, including target uniformity and dose to the surrounding normal tissues.
Conclusion: Incorporation of KBP models into the clinical trial setting may have a profound impact on the quality of trial results, owing to the increase in consistency and standardization of planning, especially for the treatment strategy that are nonstandard.
Quality Control, Clinical Trials
TH- External Beam- Photons: Treatment planning using machine learning/Knowledge Based Planning/automation