Session Item

Treatment planning for a brain tumor can be challenging due to the close proximity of numerous critical structures. Plan optimization is usually based on generic dose constraints and manual adjustments. While this may yield an acceptable plan, the full potential is not easily exploited. We investigated the benefit of a semi-automatic process, in which the optimization is driven by patient-specific objectives from a DVH estimation model.

A brain DVH estimation model was configured using the RapidPlan tool within the Eclipse v15.6 environment (Varian, Palo Alto, USA). Dose distribution data was extracted from 100 treatment plans in the department’s database. The resulting model is applicable to the majority of curative brain cases. When applied to a new case, it can generate line objectives for 14 critical structures.

The study concentrated on a separate set of 20 patients previously irradiated for glioblastoma multiforme (GBM): 59.4 or 60 Gy in 33 or 30 fractions. The treatment plans employed 6 MV photons in a dual arc setup on a Varian TrueBeam. Up to 14 critical structures had been contoured (minimum 11), with at least one structure partially overlapping the PTV. Plans had been manually optimized for PTV coverage and maximum OAR doses. Additional mean dose constraints applied to eyes, lenses and pituitary.

All 20 patients were retrospectively replanned using the original beam setup and the DVH estimation model. The resulting dose distributions were evaluated in terms of PTV coverage and mean dose to critical structures. Plan complexity was taken into account by comparing monitor units (MUs).

The knowledge-based replanning resulted in improved dose distributions overall. Most of the critical structures received a lower mean dose (12 structures per patient on average; range 8-14). The remainder received a higher mean dose (2 structures per patient on average; range 0-5). The mean difference in mean dose after replanning was -21%. A large variation between patients was observed here (-4% to -35%). Sorting by structure, the pituitary displayed the largest shift in absolute mean dose (-6.8 Gy), followed by the left optic nerve (-6.1 Gy) and the optic chiasm (-6.0 Gy).

Replanning only had a small effect on PTV coverage. The group mean V95% decreased from 99.1% to 98.7%.

No systematic change was observed for the number of MU which remained around 225 MU/Gy.

The knowledge-based approach is an important step forward in treatment planning for brain tumors. It consistently achieved additional sparing in GBM patients with respect to the conventional method. There was no obvious trade-off in terms of target coverage or plan complexity. The results also indicate that DVH estimation helps to remove some of the inherent variability in manual optimization.

Our department has added the brain model to its set of DVH estimation models for routine clinical use.