Marissa Renardy 3, Camille Vong 1, Alienor Berges 2, Krishnakant Dasika 4, Colleen Witt 4, Lesley Benyon 4, Mike Reed 4, Rianne Esquivel 3, Dave Inman 2, Ben Jones 2, Michael Boss 3, Vivek Gautam 3, Guhan Nagappan 3, Sanjay Kumar 3, Robert Lai 2, Anna Sher 3
1 GSK (Baar, Switzerland), 2 GSK (Stevenage, UK), 3 GSK (Collegeville, USA), 4 Rosa & Co. LLC (San Carlos, USA)
Objectives:
Neuroinflammation plays a critical role in Alzheimer’s Disease (AD) pathogenesis, with multiple therapeutic targets being investigated in the clinic. Historically, QSP models in AD have focused on amyloid and tau pathology, with less emphasis on the role of neuroinflammation and the complex interactions between neurons, microglia, and astrocytes [1-5]. Our objective in this work was to construct a QSP model describing key aspects of neuroinflammation to allow for exploratory simulations of immune-modulating and neuroprotective therapeutic effects and impact on downstream biomarkers and cognitive function.
Methods:
We have constructed a QSP model consisting of a system of ordinary differential equations and algebraic relationships that describe: homeostatic, damaging, and protective microglia and astrocytes; pro- and anti-inflammatory mediators; amyloid and tau aggregation; neuronal damage and death; key clinical biomarkers including glial fibrillary acidic protein (GFAP) and neurofilament light (NfL); and cognitive endpoints including Clinical Dementia Rating – Sum of Boxes (CDR-SB) and Integrated Alzheimer’s Disease Rating Scale (iADRS). The model is calibrated using data from the Alzheimer’s Disease Neuroimaging Initiative (ADNI) [6] to describe disease progression from cognitively unimpaired to early Alzheimer’s Disease dementia. The model is additionally calibrated to clinical/biomarker response data from the TRAILBLAZER-ALZ trial of donanemab [7]. The model is implemented in Matlab SimBiology v2023b. Model code and equations are available upon request.
Results:
The model successfully reproduces population trends in disease progression from cognitively unimpaired to mild cognitive impairment (MCI) to early AD dementia, as well as treatment effects of donanemab on biomarkers and clinical scores. Exploratory sensitivity analyses demonstrate the therapeutic potential for modifying key mechanisms such as lysosomal function, phagocytosis, neuroprotection, and immune cell phenotype conversion. One-at-a-time modification of the model by increasing lysosomal clearance of tau and Aβ species, enhancing phagocytosis of tau and Aβ species, improving neuronal protection, and increasing microglia and astrocyte polarization toward a protective phenotype each resulted in nonsignificant improvements in CDR-SB and iADRS scores. Combined modification of these mechanisms resulted in 20-30% slowing of cognitive decline as measured by CDR-SB, comparable to those observed with approved anti-amyloid therapies, over 10 years of disease progression. These results suggest that pleiotropic effects are necessary for clinically meaningful response. Additionally, the model predicts that if the reduction in neuronal damage due to improved phagocytosis is less than the accumulation of damaged neurons resulting from decreased neuronal death, NfL could increase despite concurrent improvements in cognitive scores, though this hypothesis has yet to be validated biologically.
Conclusions:
We have built a mechanistic model of Alzheimer’s Disease natural history and intervention, describing disease stages from cognitively unimpaired to early Alzheimer’s Disease dementia, for exploring novel targets related to neuroinflammation. This model allows us to simulate in silico scenarios to better understand underlying inflammatory mechanisms of action, optimize combination with anti-amyloids, and quantify the potential changes in biomarkers in early clinical trial readouts.
References:
[1] Lin L, Hua F, Salinas C, et al. Quantitative systems pharmacology model for Alzheimer’s disease to predict the effect of aducanumab on brain amyloid. CPT Pharmacometrics Syst Pharmacol. 2022;11(3):362-372. doi:10.1002/psp4.12759
[2] Madrasi K, Das R, Mohmmadabdul H, et al. Systematic in silico analysis of clinically tested drugs for reducing amyloid-beta plaque accumulation in Alzheimer’s disease. Alzheimers Dement. 2021;17(9):1487-1498. doi:10.1002/alz.12312
[3] Ramakrishnan V, Friedrich C, Witt C, et al. Quantitative systems pharmacology model of the amyloid pathway in Alzheimer’s disease: Insights into the therapeutic mechanisms of clinical candidates. CPT Pharmacometrics Syst Pharmacol. 2023;12(1):62-73. doi:10.1002/psp4.12876
[4] Geerts H, Bergeler S, Walker M, Rose RH, van der Graaf PH. Quantitative systems pharmacology-based exploration of relevant anti-amyloid therapy challenges in clinical practice. Alzheimers Dement (N Y). 2024;10(2):e12474. Published 2024 May 21. doi:10.1002/trc2.12474
[5] Geerts H, Walker M, Rose R, et al. A combined physiologically-based pharmacokinetic and quantitative systems pharmacology model for modeling amyloid aggregation in Alzheimer’s disease. CPT Pharmacometrics Syst Pharmacol. 2023;12(4):444-461. doi:10.1002/psp4.12912
[6] Petersen RC, Aisen PS, Beckett LA, et al. Alzheimer’s Disease Neuroimaging Initiative (ADNI): clinical characterization. Neurology. 2010;74(3):201-209. doi:10.1212/WNL.0b013e3181cb3e25
[7] Pontecorvo MJ, Lu M, Burnham SC, et al. Association of Donanemab Treatment With Exploratory Plasma Biomarkers in Early Symptomatic Alzheimer Disease: A Secondary Analysis of the TRAILBLAZER-ALZ Randomized Clinical Trial. JAMA Neurol. 2022;79(12):1250-1259. doi:10.1001/jamaneurol.2022.3392
Reference: PAGE 34 (2026) Abstr 12123 [www.page-meeting.org/?abstract=12123]
Poster: Drug/Disease Modelling - CNS