LBA14 - Neurocognitive Outcome of Conformal Whole Brain Radiotherapy with Bilateral or Unilateral Hippocampal Avoidance Plus Memantine for Brain Metastases: A Phase II Single Blind Randomized Trial

F. M. Hsu^1,2, W. C. Yang^1,2, S. L. Lu^1,2, S. F. Lai^1,2, C. C. Yang³, S. H. Lu², Y. F. Chen⁴, S. H. Kuo², and J. C. H. Cheng²; ¹Department of Radiation Oncology, National Taiwan University Cancer Center, Taipei, Taiwan, ²Division of Radiation Oncology, Department of Oncology, National Taiwan University Hospital, Taipei, Taiwan, ³Department of Psychology, National Chengchi University, Taipei, Taiwan, ⁴Department of Medical Imaging, National Taiwan University Hospital, Taipei, Taiwan

Purpose/Objective(s): Bilateral hippocampal avoidance (HA) in whole brain radiotherapy (WBRT) has been proven to prevent neurocognitive failure compared to conventional WBRT. The literature suggests that the left hippocampus plays a dominant role in short-term verbal memory. However, whether left-sided unilateral HA-WBRT can preserve neurocognitive function as effectively as bilateral HA-WBRT has not been examined in clinical trials. This randomized phase II trial aimed to demonstrate the neuroprotective effect of unilateral HA-WBRT compared to the standard bilateral HA-WBRT. Materials/

Methods: This is a single-blind phase II randomized study. Adult patients with brain metastases were assigned to either left-sided unilateral HA-WBRT or bilateral HA-WBRT, receiving 30 Gy in 10 fractions. Memantine was administered during and after HA-WBRT. A standardized neurocognitive test battery, including the Hopkins Verbal Learning Test-Revised (HVLT-R), Trail Making Test (TMT) A and B, and Controlled Oral Word Association (COWA), was performed at baseline and at 1, 2, 4, 6 months, and every 3 months up to two years. Both participants and examiners were blinded to the treatment assignments. The primary endpoint was the decline in HVLT-R memory score from baseline to 6 months after the start of HA-WBRT with either bilateral or unilateral hippocampal avoidance.
Results: From March 2021 to April 2024, 73 patients were enrolled, with 68 patients (33 in the unilateral HA-WBRT group and 35 in the bilateral HA-WBRT group) completing the protocol treatment. The median follow-up time is 10.5 months. The median age was 61 years old, and the majority (76.7%) had lung cancer. Prior brain surgery or radiosurgery for brain metastasis was performed in 12 and16 patients, respectively. There were no differences in age, gender, or baseline neurocognitive function between the two groups. At the 6-month after WBRT, 20 patients in the unilateral HA-WBRT group and 24 patients in the bilateral HA-WBRT group were compliant with the neurocognitive test battery. Patients receiving unilateral HA-WBRT showed no significant difference in the changes of HVLT-R memory scores compared to those receiving bilateral HA-WBRT (mean ± standard error of mean [SEM] change from baseline: -1.450 ± 1.445 versus -1.583 ± 1.220, p=0.943). Additionally, there were no differences in HVLT-R total recall, delayed recall, and recognition index at any time point of follow-up. TMT parts A and B, as well as the COWA test, also showed no differences between the two arms.
Conclusion: The initial analysis demonstrates that patients with multiple brain metastases receiving left-sided unilateral conformal HA-WBRT have an equivalent effect in neurocognitive function preservation compared to those receiving the standard bilateral HA-WBRT.