Room: Exhibit Hall | Forum 7
Purpose: Although the number of proton therapy centers has increased worldwide, proton treatment slots are a limited resource. As the number of conventional radiotherapy clinics with an integrated single-room proton therapy system is likely to grow, combined proton-photon treatments in which most fractions are delivered with photons and only few with protons may represent an alternative. Therefore, the aim is to demonstrate how to optimally combine both modalities in order to better exploit the protonsâ€™ ability to reduce integral dose to normal tissues.
Methods: Intensity-modulated radiation therapy (IMRT) and proton therapy (IMPT) plans were simultaneously optimized while accounting for fractionation effects through the biologically effective dose (BED) model. The method was investigated for different tumor sites and geometries in which organs at risk (OARs) are located within or near the tumor (liver tumors, spinal metastases, sacral chordomas, atypical meningiomas). To quantify the benefit of optimized proton-photon treatments, IMRT and IMPT plans were first optimized based on the same set of objectives and constraints functions. A reference plan was then generated as a simple proportional combination of the single-modality plans. Finally, a combined plan was optimized using the same planning objectives.
Results: Optimal multi-modality treatments are non-trivial combinations of IMPT and IMRT plans. To protect dose-limiting normal tissues through fractionation, both proton and photon fractions deliver similar doses to serial OARs overlaying the target volume. Meanwhile, parts of the target volume are hypofractionated with protons. Consequently, the total dose delivered with photons is reduced, leading to a reduction of the integral dose to normal tissues.
Conclusion: This work shows that a limited number of proton fractions in a combined treatment is best used if protons hypofractionate parts of the target volume while maintaining near-uniform fractionation in dose-limiting serial OARs.