Patient-Specific Planning for Thermal Magnetic Resonance of Glioblastoma Multiforme

Abstract

Simple SummaryHyperthermia was proven to enhance the efficacy of chemo- and radiation therapy treatment of glioblastoma multiforme, an aggressive brain tumor of poor prognosis. Despite good clinical results in other tumor types and locations, hyperthermia induced by electromagnetic waves in the radiofrequency range is not available so far for the treatment of brain tumors due to the highly sensitive surrounding tissue and lack of non-invasive therapy monitoring. ThermalMR integrates non-invasive diagnosis, therapy, and therapy monitoring in a single RF applicator device by employing radiowaves for magnetic resonance imaging, radiofrequency heating, as well as magnetic resonance thermometry. This work examines three optimization algorithms for hyperthermia treatment planning and up to ten RF applicator configurations for a cohort of nine patient models with glioblastoma multiforme. Clinical diversity is represented in target size and location and the inclusion of post-operative models. Our findings indicate the need and potential for patient-specific treatment planning and RF applicator design when targeting brain tumors.Thermal intervention is a potent sensitizer of cells to chemo- and radiotherapy in cancer treatment. Glioblastoma multiforme (GBM) is a potential clinical target, given the cancer’s aggressive nature and resistance to current treatment options. This drives research into optimization algorithms for treatment planning as well as radiofrequency (RF) applicator design for treatment delivery. In this work, nine clinically realistic GBM target volumes (TVs) for thermal intervention are compared using three optimization algorithms and up to ten RF applicator designs for thermal magnetic resonance. Hyperthermia treatment planning (HTP) was successfully performed for all cases, including very small, large, and even split target volumes. Minimum requirements formulated for the metrics assessing HTP outcome were met and exceeded for all patient specific cases. Results indicate a 16 channel two row arrangement to be most promising. HTP of TVs with a small extent in the cranial–caudal direction in conjunction with a large radial extent remains challenging despite the advanced optimization algorithms used. In general, deep seated targets are favorable. Overall, our findings indicate that a one-size-fits-all RF applicator might not be the ultimate approach in hyperthermia of brain tumors. It stands to reason that modular and reconfigurable RF applicator configurations might best suit the needs of targeting individual GBM geometry.