Seminar khoa học về chủ đề: Các tiến bộ trong lĩnh vực điều trị bệnh Alzheimer

Vào 14h00, ngày 09/01/2026 Viện IAST tổ chức buổi trao đổi học thuật tại Tầng 3, 13 ngõ Hàng Bột, Hà Nội (cơ sở Hà Nội) và phòng họp lầu 5 Thư viện (cơ sở Tân Hưng) với nội dung chi tiết như sau:

1/ ThS. Hoàng Thị Mến báo cáo về chủ đề: Value-based prices of emerging disease-modifying therapies for Alzheimer’s Disease in 174 countries: A cost-effectiveness and threshold analysis
Abstract:
INTRODUCTION: Countries have varying limited healthcare budgets for emerging disease-modifying therapies. Cost-effectiveness analysis, combined with country-level cost-effectiveness thresholds, can be used to estimate value-based prices (VBPs) for lecanemab and donanemab across 174 countries.
METHODS: The cost-effectiveness of lecanemab and donanemab was estimated by the incremental cost and Quality-Adjusted Life Years (QALYs) compared to usual care. Published cost-effectiveness thresholds were used to estimate value-based prices of these drugs across 174 countries.
RESULTS: Compared to usual care, lecanemab and donanemab, respectively, increased average QALYs by 0.38 and 0.51. By country income status, VBPs for lecanemab and donanemab (respectively) ranged between $254-$9,434 and $387-$13,964 (high income), $90-$1,025 and $137-$1,507 (upper middle income), $11- $623 and $21-$956 (lower middle income) and $4-$18 and $9-$32 (low income). Incorporating spillover costs and health effects of informal caregivers substantially increased the VBPs of lecanemab and donanemab by up to 82%, and 109%.
DISCUSSION: VBPs indicate what 174 countries should be willing to pay. This framework can also be adapted and refined in the negotiation of country pricing.

2/ TS. Nguyễn Hoàng Long báo cáo về chủ đề: Accuracy of Artificial Intelligence Across Diagnosis, Prevention, Treatment, and Disease Progression in Alzheimer’s Disease: A Systematic Review
Abstract:
This systematic review synthesizes current evidence on the accuracy of artificial intelligence applications across diagnosis, prevention, treatment, and disease progression in Alzheimer’s disease, a neurodegenerative disorder marked by long preclinical stages and complex clinical trajectories. We reviewed studies employing machine learning and deep learning models based on clinical characteristics, neuroimaging, biomarkers, genetic data, digital markers, and multimodal inputs, focusing on performance indicators such as accuracy, sensitivity, specificity, and area under the receiver operating characteristic curve. Overall, AI models demonstrate high accuracy in diagnostic tasks, particularly in differentiating Alzheimer’s disease from cognitively normal individuals and identifying conversion from mild cognitive impairment, with multimodal approaches consistently outperforming single-data-source models. Predictive accuracy for disease progression is moderate to high, whereas evidence supporting AI-driven prevention strategies and treatment response prediction remains limited, inconsistent, and methodologically heterogeneous. Variations in datasets, model development, validation procedures, and reporting standards substantially affect comparability and clinical interpretability. In conclusion, while AI shows strong potential in enhancing diagnostic precision and monitoring disease progression in Alzheimer’s disease, further standardized, externally validated, and clinically grounded research is required before its broader application in prevention and treatment decision-making.