Low-dose whole-brain RT misses primary cognitive endpoint in randomized Alzheimer's trial

In a sham-controlled phase II trial, neither 0.24 Gy nor 3 Gy whole-brain radiotherapy significantly improved cognition at six months.

KEY POINTS

  • This multicenter, randomized, participant-blinded, sham-controlled phase II trial (NCT05635968) evaluated whether low-dose whole-brain radiotherapy could improve cognitive outcomes in patients with amyloid-positive mild Alzheimer's disease. Conducted at three academic hospitals in South Korea, the trial randomized 41 participants, of whom 36 were included in the modified intention-to-treat analysis, with 12 patients per group. Enrollment stopped below the planned 60 participants, limiting statistical precision and power.
  • Participants were randomized to sham irradiation, 0.04 Gy × 6 fractions (0.24 Gy total), or 0.5 Gy × 6 fractions (3 Gy total), delivered twice weekly over three weeks using CT-planned, three-dimensional conformal whole-brain irradiation. Eligible patients were 60–85 years old, had confirmed cerebral amyloid positivity, and were receiving stable Alzheimer's medication. The primary endpoint was the change from baseline to six months in the Alzheimer's Disease Assessment Scale–Cognitive Subscale (ADAS-cog), where higher scores indicate greater cognitive impairment.
  • Neither active treatment group achieved a statistically significant improvement in the prespecified primary endpoint. At six months, adjusted mean ADAS-cog changes were +2.33 points with sham, +0.25 with 0.24 Gy, and −0.58 with 3 Gy. Compared with sham, the adjusted differences were −2.08 points for 0.24 Gy (95% CI −7.06 to 2.89; p=0.401) and −2.92 points for 3 Gy (95% CI −6.72 to 0.89; p=0.129). Thus, the randomized trial did not demonstrate cognitive efficacy at its primary assessment.
  • Exploratory 12-month findings favored the 3-Gy regimen, although these analyses were not adjusted for multiple comparisons. Mean ADAS-cog deterioration was 7.11 points with sham, 4.09 with 0.24 Gy, and 1.75 with 3 Gy. The adjusted difference between 3 Gy and sham was −5.36 points (95% CI −10.58 to −0.14; nominal p=0.044), whereas the 0.24-Gy comparison was not significant (−3.02 points; p=0.386). Only 30 participants completed the 12-month assessment, further limiting confidence in these findings.
  • Selected secondary cognitive outcomes showed signals that differed by dose and assessment time. At six months, 0.24 Gy reduced worsening on the Clinical Dementia Rating–Sum of Boxes scale by 2.08 points versus sham (nominal p=0.018); at 12 months, the difference was −1.68 points (p=0.052). The K-MMSE-2 score favored 0.24 Gy by 2.42 points at 12 months (nominal p=0.033), but the 3-Gy group did not demonstrate a corresponding advantage. No meaningful between-group differences were identified for neuropsychiatric symptoms or instrumental activities of daily living, preventing identification of a consistent dose-response relationship.
  • No convincing imaging or molecular evidence of disease modification was demonstrated. Amyloid PET uptake, Centiloid values, hippocampal and ventricular volumes, plasma amyloid measures, and phosphorylated tau-181 did not show significant treatment-related differences. Exploratory TNF-α measurements suggested possible changes in inflammatory signaling at 12 months, but these findings depended partly on statistical transformation and were not supported by consistent changes in other cytokines.
  • Treatment was generally well tolerated in the 41-patient safety population, with one serious and one nonserious adverse event, neither considered related to the study intervention. No clinically meaningful radiation-related toxicity or radiological safety signal was identified during follow-up. Nevertheless, the small study size, incomplete follow-up, participant-only blinding, lack of multiplicity correction, and short observation period prevent conclusions about uncommon or delayed risks, including long-term neurological effects.

CLINICAL TAKEAWAY

Low-dose whole-brain irradiation did not improve the primary six-month cognitive endpoint in this randomized phase II study of mild Alzheimer's disease, despite selected favorable exploratory outcomes at 12 months. These findings are insufficient to establish therapeutic efficacy or an optimal dose, and whole-brain RT should remain investigational for Alzheimer's disease pending larger confirmatory trials with longer follow-up and comprehensive safety assessment.

SOURCE

Radiotherapy and Oncology

Browse more research Suggest a correction