Advancing beyond the Hippocampus to Preserve Cognition for Patients with Brain Metastases: Dosimetric Results from a Phase 2 Trial of Memory-Avoidance Whole Brain Radiotherapy

Abstract

Recent advances to preserve neurocognitive function in patients treated for brain metastases include stereotactic radiosurgery (SRS), hippocampal avoidance whole brain radiation therapy (HA-WBRT), and memantine administration. However, 23% of patients receiving HA-WBRT and memantine on NRG-CC001 still experienced executive function deterioration at 4 months, with 12% experiencing total recall and delayed recognition deterioration at 6 months. The hippocampus, corpus callosum, fornix, and amygdala are key neurocognitive substructures with a low propensity for brain metastases. Herein, we report our preliminary experience using an advanced "memory-avoidance" WBRT (MA-WBRT) approach sparing these substructures for patients with multiple (>15) brain metastases. Ten consecutive patients treated with MA-WBRT on a phase 2 clinical trial (OSU-21074) were reviewed. In each patient, the hippocampi, amygdalae, corpus callosa, and fornix were contoured. Patients were not eligible for MA-WBRT if they had metastases in these substructures. A memory- avoidance region created using a 5mm volumetric expansion around these substructures. Hotspots were avoided in the hypothalamus and pituitary gland. Dose constraints for these avoidance structures were modeled after NRG CC-001 and include a D100% ≤ 9 Gy and a D0.03 cc ≤ 16 (acceptable to 20 Gy). Coverage of brain metastases was prioritized over memory avoidance dose constraints. Linac-based volumetric modulated arc therapy (VMAT) plans were generated for a prescription dose of 30 Gy in 10 fractions. On average, the memory avoidance structure volume was 37.1 ccs (Range: 25.2-44.6 ccs), occupying 2.5% of the entire whole brain target volume. All treatment plans met the D100% dose constraint, and 8/10 plans met the D0.03cc constraint, with priority given to tumor coverage for the remaining 2 cases. VMAT spared the memory avoidance structures with a median dose range of 10.8-14.2 Gy and a maximum dose (D0.03cc) range of 15.6-22.7 Gy. The mean dose to the memory avoidance structures was 12.7 Gy (Range: 11.5-13.8 Gy). Target coverage (D98% > 25 Gy) and homogeneity (D2% ≤ 37.5 Gy) were achieved for all plans. Modern VMAT techniques allow for sparing of the hippocampus, amygdala, corpus callosum, and fornix with good target coverage and homogeneity. Prospective quality of life and cognitive data including are being collected and include the Functional Assessment of Cancer Therapy - Brain (Fact-Br), Hopkins Verbal Learning Test (HLVT-R), Trail Making Test A/B (TMT-A/B) and Controlled Oral Word Association Test (COWAT). After enrollment is completed, these data will be evaluated to assess the efficacy of MA-WBRT to mitigate declines in quality of life and cognition after whole brain radiation.