Learning from amyloid trials in Alzheimer's disease. A virtual patient analysis using a quantitative systems pharmacology approach

Reducing brain Aβ burden ameliorates high-fat diet-induced fatty liver disease in APP/PS1 mice

High-fat diet (HFD)-induced fatty liver disease is a deteriorating risk factor for Alzheimer's disease (AD). Mitigating fatty liver disease has been shown to attenuate AD-like pathology in animal models. However, it remains unclear whether enhancing Aβ clearance through immunotherapy would in turn attenuate HFD-induced fatty liver or whether its efficacy would be compromised by long-term exposure to HFD. Here, the therapeutic potentials of an anti-Aβ antibody, NP106, was investigated in APP/PS1 mice by HFD feeding for 44 weeks. The data demonstrate that NP106 treatment effectively reduced Aβ burden and pro-inflammatory cytokines in HFD-fed APP/PS1 mice and ameliorated HFD-aggravated cognitive impairments during the final 18 weeks of the study. The rejuvenating characteristics of microglia were evident in APP/PS1 mice with NP106 treatment, namely enhanced microglial Aβ phagocytosis and attenuated microglial lipid accumulation, which may explain the benefits of NP106. Surprisingly, NP106 also reduced HFD-induced hyperglycemia, fatty liver, liver fibrosis, and hepatic lipids, concomitant with modifications in the expressions of genes involved in hepatic lipogenesis and fatty acid oxidation. The data further reveal that brain Aβ burden and behavioral deficits were positively correlated with the severity of fatty liver disease and fasting serum glucose levels. In conclusion, our study shows for the first time that anti-Aβ immunotherapy using NP106, which alleviates AD-like disorders in APP/PS1 mice, ameliorates fatty liver disease. Minimizing AD-related pathology and symptoms may reduce the vicious interplay between central AD and peripheral fatty liver disease, thereby highlighting the importance of developing AD therapies from a systemic disease perspective.

Analysis of clinical failure of anti-tau and anti-synuclein antibodies in neurodegeneration using a quantitative systems pharmacology model

Misfolded proteins in Alzheimer's disease and Parkinson's disease follow a well-defined connectomics-based spatial progression. Several anti-tau and anti-alpha synuclein (aSyn) antibodies have failed to provide clinical benefit in clinical trials despite substantial target engagement in the experimentally accessible cerebrospinal fluid (CSF). The proposed mechanism of action is reducing neuronal uptake of oligomeric protein from the synaptic cleft. We built a quantitative systems pharmacology (QSP) model to quantitatively simulate intrasynaptic secretion, diffusion and antibody capture in the synaptic cleft, postsynaptic membrane binding and internalization of monomeric and oligomeric tau and aSyn proteins. Integration with a physiologically based pharmacokinetic (PBPK) model allowed us to simulate clinical trials of anti-tau antibodies gosuranemab, tilavonemab, semorinemab, and anti-aSyn antibodies cinpanemab and prasineuzumab. Maximal target engagement for monomeric tau was simulated as 45% (semorinemab) to 99% (gosuranemab) in CSF, 30% to 99% in ISF but only 1% to 3% in the synaptic cleft, leading to a reduction of less than 1% in uptake of oligomeric tau. Simulations for prasineuzumab and cinpanemab suggest target engagement of free monomeric aSyn of only 6-8% in CSF, 4-6% and 1-2% in the ISF and synaptic cleft, while maximal target engagement of aggregated aSyn was predicted to reach 99% and 80% in the synaptic cleft with similar effects on neuronal uptake. The study generates optimal values of selectivity, sensitivity and PK profiles for antibodies. The study identifies a gradient of decreasing target engagement from CSF to the synaptic cleft as a key driver of efficacy, quantitatively identifies various improvements for drug design and emphasizes the need for QSP modelling to support the development of tau and aSyn antibodies.

Modeling Clinical Phenotype Variability: Consideration of Genomic Variations, Computational Methods, and Quantitative Proteomics

Advances in biomedical and computer technologies have presented the modeling community the opportunity for mechanistically modeling and simulating the variability in a disease phenotype or in a drug response. The capability to quantify response variability can inform a drug development program. Quantitative systems pharmacology scientists have published various computational approaches for creating virtual patient populations (VPops) to model and simulate drug response variability. Genomic variations can impact disease characteristics and drug exposure and response. Quantitative proteomics technologies are increasingly used to facilitate drug discovery and development and inform patient care. Incorporating variations in genomics and quantitative proteomics may potentially inform creation of VPops to model and simulate virtual patient trials, and may help account for, in a predictive manner, phenotypic variations observed clinically.

A combined physiologically-based pharmacokinetic and quantitative systems pharmacology model for modeling amyloid aggregation in Alzheimer's disease

Antibody-mediated removal of aggregated β-amyloid (Aβ) is the current, most clinically advanced potential disease-modifying treatment approach for Alzheimer's disease. We describe a quantitative systems pharmacology (QSP) approach of the dynamics of Aβ monomers, oligomers, protofibrils, and plaque using a detailed microscopic model of Aβ₄₀ and Aβ₄₂ aggregation and clearance of aggregated Aβ by activated microglia cells, which is enhanced by the interaction of antibody-bound Aβ. The model allows for the prediction of Aβ positron emission tomography (PET) imaging load as measured by a standardized uptake value ratio. A physiology-based pharmacokinetic model is seamlessly integrated to describe target exposure of monoclonal antibodies and simulate dynamics of cerebrospinal fluid (CSF) and plasma biomarkers, including CSF Aβ₄₂ and plasma Aβ₄₂ /Aβ₄₀ ratio biomarkers. Apolipoprotein E genotype is implemented as a difference in microglia clearance. By incorporating antibody-bound, plaque-mediated macrophage activation in the perivascular compartment, the model also predicts the incidence of amyloid-related imaging abnormalities with edema (ARIA-E). The QSP platform is calibrated with pharmacological and clinical information on aducanumab, bapineuzumab, crenezumab, gantenerumab, lecanemab, and solanezumab, predicting adequately the change in PET imaging measured amyloid load and the changes in the plasma Aβ₄₂ /Aβ₄₀ ratio while slightly overestimating the change in CSF Aβ₄₂ . ARIA-E is well predicted for all antibodies except bapineuzumab. This QSP model could support the clinical trial design of different amyloid-modulating interventions, define optimal titration and maintenance schedules, and provide a first step to understand the variability of biomarker response in clinical practice.

Hallmarks of neurodegenerative disease: A systems pharmacology perspective

Age-related central neurodegenerative diseases, such as Alzheimer's and Parkinson's disease, are a rising public health concern and have been plagued by repeated drug development failures. The complex nature and poor mechanistic understanding of the etiology of neurodegenerative diseases has hindered the discovery and development of effective disease-modifying therapeutics. Quantitative systems pharmacology models of neurodegeneration diseases may be useful tools to enhance the understanding of pharmacological intervention strategies and to reduce drug attrition rates. Due to the similarities in pathophysiological mechanisms across neurodegenerative diseases, especially at the cellular and molecular levels, we envision the possibility of structural components that are conserved across models of neurodegenerative diseases. Conserved structural submodels can be viewed as building blocks that are pieced together alongside unique disease components to construct quantitative systems pharmacology (QSP) models of neurodegenerative diseases. Model parameterization would likely be different between the different types of neurodegenerative diseases as well as individual patients. Formulating our mechanistic understanding of neurodegenerative pathophysiology as a mathematical model could aid in the identification and prioritization of drug targets and combinatorial treatment strategies, evaluate the role of patient characteristics on disease progression and therapeutic response, and serve as a central repository of knowledge. Here, we provide a background on neurodegenerative diseases, highlight hallmarks of neurodegeneration, and summarize previous QSP models of neurodegenerative diseases.

The Beneficial Effects of Combining Anti-Aβ Antibody NP106 and Curcumin Analog TML-6 on the Treatment of Alzheimer's Disease in APP/PS1 Mice

Alzheimer’s disease (AD) is a progressive neurodegenerative disease with a multifactorial etiology. A multitarget treatment that modulates multifaceted biological functions might be more effective than a single-target approach. Here, the therapeutic efficacy of combination treatment using anti-Aβ antibody NP106 and curcumin analog TML-6 versus monotherapy was investigated in an APP/PS1 mouse model of AD. Our data demonstrate that both combination treatment and monotherapy attenuated brain Aβ and improved the nesting behavioral deficit to varying degrees. Importantly, the combination treatment group had the lowest Aβ levels, and insoluble forms of Aβ were reduced most effectively. The nesting performance of APP/PS1 mice receiving combination treatment was better than that of other APP/PS1 groups. Further findings indicate that enhanced microglial Aβ phagocytosis and lower levels of proinflammatory cytokines were concurrent with the aforementioned effects of NP106 in combination with TML-6. Intriguingly, combination treatment also normalized the gut microbiota of APP/PS1 mice to levels resembling the wild-type control. Taken together, combination treatment outperformed NP106 or TML-6 monotherapy in ameliorating Aβ pathology and the nesting behavioral deficit in APP/PS1 mice. The superior effect might result from a more potent modulation of microglial function, cerebral inflammation, and the gut microbiota. This innovative treatment paradigm confers a new avenue to develop more efficacious AD treatments.

Computational Approaches for Supporting Combination Therapy in the Post-Aducanumab Era in Alzheimer’s Disease

With the approval of aducanumab on the “Accelerated Approval Pathway” and the recognition of amyloid load as a surrogate marker, new successful therapeutic approaches will be driven by combination therapy as was the case in oncology after the launch of immune checkpoint inhibitors. However, the sheer number of therapeutic combinations substantially complicates the search for optimal combinations. Data-driven approaches based on large databases or electronic health records can identify optimal combinations and often using artificial intelligence or machine learning to crunch through many possible combinations but are limited to the pharmacology of existing marketed drugs and are highly dependent upon the quality of the training sets. Knowledge-driven in silico modeling approaches use multi-scale biophysically realistic models of neuroanatomy, physiology, and pathology and can be personalized with individual patient comedications, disease state, and genotypes to create ‘virtual twin patients’. Such models simulate effects on action potential dynamics of anatomically informed neuronal circuits driving functional clinical readouts. Informed by data-driven approaches this knowledge-driven modeling could systematically and quantitatively simulate all possible target combinations for a maximal synergistic effect on a clinically relevant functional outcome. This approach seamlessly integrates pharmacokinetic modeling of different therapeutic modalities. A crucial requirement to constrain the parameters is the access to preferably anonymized individual patient data from completed clinical trials with various selective compounds. We believe that the combination of data- and knowledge driven modeling could be a game changer to find a cure for this devastating disease that affects the most complex organ of the universe.

Recent applications of quantitative systems pharmacology and machine learning models across diseases

Quantitative systems pharmacology (QSP) is a quantitative and mechanistic platform describing the phenotypic interaction between drugs, biological networks, and disease conditions to predict optimal therapeutic response. In this meta-analysis study, we review the utility of the QSP platform in drug development and therapeutic strategies based on recent publications (2019-2021). We gathered recent original QSP models and described the diversity of their applications based on therapeutic areas, methodologies, software platforms, and functionalities. The collection and investigation of these publications can assist in providing a repository of recent QSP studies to facilitate the discovery and further reusability of QSP models. Our review shows that the largest number of QSP efforts in recent years is in Immuno-Oncology. We also addressed the benefits of integrative approaches in this field by presenting the applications of Machine Learning methods for drug discovery and QSP models. Based on this meta-analysis, we discuss the advantages and limitations of QSP models and propose fields where the QSP approach constitutes a valuable interface for more investigations to tackle complex diseases and improve drug development.

Toward Predicting Impact of Common Genetic Variants on Schizophrenia Clinical Responses With Antipsychotics: A Quantitative System Pharmacology Study

CNS disorders are lagging behind other indications in implementing genotype-dependent treatment algorithms for personalized medicine. This report uses a biophysically realistic computer model of an associative and dorsal motor cortico-striatal-thalamo-cortical loop and a working memory cortical model to investigate the pharmacodynamic effects of COMTVal158Met rs4680, 5-HTTLPR rs 25531 s/L and D2DRTaq1A1 genotypes on the clinical response of 7 antipsychotics. The effect of the genotypes on dopamine and serotonin dynamics and the level of target exposure for the drugs was calibrated from PET displacement studies. The simulations suggest strong gene-gene pharmacodynamic interactions unique to each antipsychotic. For PANSS Total, the D2DRTaq1 allele has the biggest impact, followed by the 5-HTTLPR rs25531. The A2A2 genotype improved efficacy for all drugs, with a more complex outcome for the 5-HTTLPR rs25531 genotype. Maximal range in PANSS Total for all 27 individual combinations is 3 (aripiprazole) to 5 points (clozapine). The 5-HTTLPR L/L with aripiprazole and risperidone and the D2DRTaq1A2A2 allele with haloperidol, clozapine and quetiapine reduce the motor side-effects with opposite effects for the s/s genotype. The COMT genotype has a limited effect on antipsychotic effect and EPS. For cognition, the COMT MM 5-HTTLPR L/L genotype combination has the best performance for all antipsychotics, except clozapine. Maximal difference is 25% of the total dynamic range in a 2-back working memory task. Aripiprazole is the medication that is best suited for the largest number of genotype combinations (10) followed by Clozapine and risperidone (6), haloperidol and olanzapine (3) and quetiapine and paliperidone for one genotype. In principle, the platform could identify the best antipsychotic treatment balancing efficacy and side-effects for a specific individual genotype. Once the predictions of this platform are validated in a clinical setting the platform has potential to support rational personalized treatment guidance in clinical practice.

Fluorescent indolizine derivative YI-13 detects amyloid-β monomers, dimers, and plaques in the brain of 5XFAD Alzheimer transgenic mouse model

Alzheimer disease (AD) is a neurodegenerative disorder characterized by the aberrant production and accumulation of amyloid-β (Aβ) peptides in the brain. Accumulated Aβ in soluble oligomer and insoluble plaque forms are considered to be a pathological culprit and biomarker of the disorder. Here, we report a fluorescent universal Aβ-indicator YI-13, 5-(4-fluorobenzoyl)-7,8-dihydropyrrolo[1,2-b]isoquinolin-9(6H)-one, which detects Aβ monomers, dimers, and plaques. We synthesized a library of 26 fluorescence chemicals with the indolizine core and screen them through a series of in vitro tests utilizing Aβ as a target and YI-13 was selected as the final imaging candidate. YI-13 was found to stain and visualize insoluble Aβ plaques in the brain tissue, of a transgenic mouse model with five familial AD mutations (5XFAD), by a histochemical approach and to label soluble Aβ oligomers within brain lysates of the mouse model under a fluorescence plate reader. Among oligomers aggregated from monomers and synthetic dimers from chemically conjugated monomers, YI-13 preferred the dimeric Aβ.

A modeling informed quantitative approach to salvage clinical trials interrupted due to COVID-19

Many ongoing Alzheimer's disease central nervous system clinical trials are being disrupted and halted due to the COVID-19 pandemic. They are often of a long duration' are very complex; and involve many stakeholders, not only the scientists and regulators but also the patients and their family members. It is mandatory for us as a community to explore all possibilities to avoid losing all the knowledge we have gained from these ongoing trials. Some of these trials will need to completely restart, but a substantial number can restart after a hiatus with the proper protocol amendments. To salvage the information gathered so far, we need out-of-the-box thinking for addressing these missingness problems and to combine information from the completers with those subjects undergoing complex protocols deviations and amendments after restart in a rational, scientific way. Physiology-based pharmacokinetic (PBPK) modeling has been a cornerstone of model-informed drug development with regard to drug exposure at the site of action, taking into account individual patient characteristics. Quantitative systems pharmacology (QSP), based on biology-informed and mechanistic modeling of the interaction between a drug and neuronal circuits, is an emerging technology to simulate the pharmacodynamic effects of a drug in combination with patient-specific comedications, genotypes, and disease states on functional clinical scales. We propose to combine these two approaches into the concept of computer modeling-based virtual twin patients as a possible solution to harmonize the readouts from these complex clinical datasets in a biologically and therapeutically relevant way.

Galantamine-Memantine combination in the treatment of Alzheimer's disease and beyond

Alzheimer's disease (AD) is the most prevalent form of dementia in the elderly population worldwide. Despite the major unmet clinical need, no new medications for the treatment of AD have been approved since 2003. Galantamine is an acetylcholinesterase inhibitor that is also a positive allosteric modulator at the α4β2 and α7nACh receptors. Memantine is an N-methyl-d-aspartate receptor modulator/agonist. Both galantamine and memantine are FDA-approved medications for the treatment of AD. The objective of this review is to highlight the potential of the galantamine-memantine combination to conduct randomized controlled trials (RCTs) in AD. Several studies have shown the combination to be effective. Neurodegenerative diseases involve multiple pathologies; therefore, combination treatment appears to be a rational approach. Although underutilized, the galantamine-memantine combination is the standard of care in the treatment of AD. Positive RCTs with the combination with concurrent improvement in symptoms and biomarkers may lead to FDA approval, which may lead to greater utilization of this combination in clinical practice.

Simulating the Effects of Common Comedications and Genotypes on Alzheimer's Cognitive Trajectory Using a Quantitative Systems Pharmacology Approach

BACKGROUND

Many Alzheimer's disease patients in clinical practice are on polypharmacy for treatment of comorbidities.

OBJECTIVE

While pharmacokinetic interactions between drugs have been relatively well established with corresponding treatment guidelines, many medications and common genotype variants also affect central brain circuits involved in cognitive trajectory, leading to complex pharmacodynamic interactions and a large variability in clinical trials.

METHODS

We applied a mechanism-based and ADAS-Cog calibrated Quantitative Systems Pharmacology biophysical model of neuronal circuits relevant for cognition in Alzheimer's disease, to standard-of-care cholinergic therapy with COMTVal158Met, 5-HTTLPR rs25531, and APOE genotypes and with benzodiazepines, antidepressants, and antipsychotics, all together 9,585 combinations.

RESULTS

The model predicts a variability of up to 14 points on ADAS-Cog at baseline (COMTVV 5-HTTLPRss APOE 4/4 combination is worst) and a four-fold range for the rate of progression. The progression rate is inversely proportional to baseline ADAS-Cog. Antidepressants, benzodiazepines, first-generation more than second generation, and most antipsychotics with the exception of aripiprazole worsen the outcome when added to standard-of-care in mild cases. Low dose second-generation benzodiazepines revert the negative effects of risperidone and olanzapine, but only in mild stages. Non APOE4 carriers with a COMTMM and 5HTTLPRLL are predicted to have the best cognitive performance at baseline but deteriorate somewhat faster over time. However, this effect is significantly modulated by comedications.

CONCLUSION

Once these simulations are validated, the platform can in principle provide optimal treatment guidance in clinical practice at an individual patient level, identify negative pharmacodynamic interactions with novel targets and address protocol amendments in clinical trials.

Learning from amyloid trials in Alzheimer's disease. A virtual patient analysis using a quantitative systems pharmacology approach

BACKGROUND

Many trials of amyloid-modulating agents fail to improve cognitive outcome in Alzheimer's disease despite substantial reduction of amyloid β levels.

METHODS

We applied a mechanism-based Quantitative Systems Pharmacology model exploring the pharmacodynamic interactions of apolipoprotein E (APOE), Catechol -O -methyl Transferase (COMTVal158Met), and 5-HT transporter (5-HTTLPR) rs25531 genotypes and aducanumab.

RESULTS

The model predicts large clinical variability. Anticipated placebo differences on Alzheimer's Disease Assessment Scale (ADAS)-COG in the aducanumab ENGAGE and EMERGE ranged from 0.77 worsening to 1.56 points improvement, depending on the genotype-comedication combination. 5-HTTLPR L/L subjects are found to be the most resilient. Virtual patient simulations suggest improvements over placebo between 4% and 20% at the 10 mg/kg dose, depending on the imbalance of the 5-HTTLPR genotype and exposure. In the Phase II PRIME trial, maximal anticipated placebo difference at 10 mg/kg ranges from 0.3 worsening to 5.3 points improvement.

DISCUSSION

These virtual patient simulations, once validated against clinical data, could lead to better informed future clinical trial designs.