Exploring Precision Medicine in Behavioral Research

There are no “one size fits all” treatments in some fields of medicine. Although patients may share the same broad diagnosis, their disease characteristics can differ on the molecular, pathological, and behavioral levels.

In addition, patients usually have drastically different genetic variants, biochemistry, lifestyle, and environment. Consequently, some patients may respond well to a particular therapy whilst others do not see any benefits. The need for tailor-made therapies is where precision medicine comes in.

In this article, we will explore what is precision medicine and how progress is being made in animal models for dementia, Rett syndrome, and cancer.

Precision medicine aims to be the solution for offering each patient with the most appropriate treatment. It is often synonymous with targeted therapy, especially in the context for cancer therapies.

In simple terms, precision medicine aims to provide “the right treatment at the right dose at the right time for the right patient.”^[1] To achieve this goal, a treatment considers an individual’s specific genetic makeup, biomarkers, phenotypes, and psychosocial characteristics.^[2] The targeted treatment is then tracked closely and adapted to the patient’s response over time.

What Are the Benefits of Precision Medicine?

Precision medicine may ultimately enable researchers, clinicians, patients, and technology providers to collaborate toward developing individualized treatments. The benefits of precision medicine include:

• More accurately predict which treatments are safe and most effective
• Shift healthcare professionals’ response to diseases from reaction to prevention
• Improve disease detection and prevent disease progression
• Avoid prescribing medications with predictable side effects or poor outcomes
• Reduce the cost and improve the efficiency when developing new medications by eliminating trial and errors
• Reduce side effects in patients who are less likely to respond to a treatment

How Did Precision Medicine Begin?

Precision medicine has an imprecise history but was made possible largely by the rise of technology and data. In the 19th century, pioneers in biometry and statistics were interested in how genetics and disease may interact to shape individual differences in therapeutic responses.^[2] With the rapid advances in genetic testing, notably the successful completion of the Human Genome Project, precision medicine took off.

Commercial DNA Testing Perpetuates Precision Medicine

Today, commercial DNA testing companies (such as 23andMe) have enabled roughly 30 million people to test their genetic variants at an affordable price. This enables people to take their health right in their hands.

In addition, there are currently more than two dozen third-party genetic analysis websites where users can upload their raw DNA data (sometimes for free). Within a day or two, users can access personalized health, nutrition, fitness, ancestry, and even personality reports. These reports are then taken to genetic counselors or health practitioners to design personalized treatments or preventative strategies.^[3] Moreover, wearable biosensors and artificial intelligence can collect and interpret a variety of user health data that can be shared with health professionals.^[4]

Technological advancements like this which occur on a general public level increase the relevance and demand for precision medicine.

Where Is Precision Medicine Heading?

Precision medicine is also getting attention on a national level from the government. In 2015, President Obama launched the Precision Medicine Initiative, which involved the key program, All of Us Research Program (AoURP). It aims to create a research repository that consists of Electronic Health Records (EHRs), biospecimens, physical and lifestyle data from one million U.S. individuals.^[5]

The interest of the government in precision medicine shows that it can have potential applications that go beyond the academic realm. As new and more data are being collected in the health system in similar programs, precision medicine may be accessible to everyone in the future. This raises the possibility that precision medicine may even become standard medical practice where appropriate.

Today, there are still many challenges posed to precision medicine. For example, there is an urgent need for encouraging new animal modeling approaches that recognize individual differences. Then, different drug regimens can be tested on those animal models that best mimic specific patient groups, consequently improving translational therapies.^[6]

The impact of precision medicine will be greatly accelerated if it is integrated into model organisms from the outset.^[7] Therefore, the next section will provide examples of how rodent models have been used to drive discoveries in precision medicine for various diseases.

Establishing a Genetically Diverse Model for Alzheimer’s Disease (AD)

Alzheimer’s is one of the diseases that researchers are trying to target using precision medicine. AD has a high burden on the elderly with 1 in 3 seniors dying with AD or another dementia. Yet our current understanding and treatment of the disease are inadequate. This is partly because AD is highly heterogeneous, but individual patient variability is largely ignored in preclinical studies.

Traditional AD animal models are predominantly transgenic mice harbouring high-risk AD genetic mutations. However, most of these models are maintained on only a single or a few genetic backgrounds, thereby lacking genetic diversity. These rigid, low-utility AD animal models may also be a reason why no successful treatments have been developed over the years. Considering that the genetic heritability of AD is estimated to be 50%-80%, traditional models miss how individual genetics affect AD trajectory.^[8]

To create precision medicine tailored to individual variability, researchers first need to improve the genetic diversity of the disease mouse models. To do this, Neuner et al. introduced the AD-BXD model as a new preclinical resource that helps identify genetic determinants of individual susceptibility to AD.^[9] They essentially established a new model that can display high individual variability while fulfilling all of the characteristics behavioral and cognitive impairments seen in human AD.

The authors crossed a well-established mouse model of AD (the 5XFAD transgenic mouse strain)  with a genetically diverse reference panel (the BXD genetic reference panel). This way, the resulting F1 hybrids carry the identical high-risk human AD mutations, but the rest of their genomes are different.^[9] This novel model therefore mimics human patients from a diverse genetic background who carry the familial AD mutations. Studying the phenotypic outcomes in mice may help understand the different clinical presentations by patients.

As the authors expected, the presence of genetic variation in the background genome profoundly modified the impact of human AD mutations on cognitive phenotypes.

Through a variety of behavioral assays detailed in Table 1, AD-BXD mice were associated with the loss of working memory and contextual fear memory deficits. The extent of cognitive decline also varied across individual genetic backgrounds, mimicking the variations among human patients. As expected, non-cognitive traits such as anxiety were not affected in AD-BXD mice.

Table 1: Behavioral assays that demonstrate cognitive impairment in the genetically diverse model for AD, the AD-BXD model.^[9]

Overall, the novel AD mouse model appears to be a promising and reproducible resource for preclinical AD research. Besides parallel cognitive findings, the AD-BXD model demonstrated high degrees of genetic, transcriptomic, and phenotypic overlap with human AD. Such models with genetic diversity have high translational value for:

• Understanding the etiology of AD
• Understanding how genetic variability predispose risk or resilience to AD
• Understanding the phenotypic variation in AD onset and severity
• Efficiently evaluating therapeutic responses in different patient subgroups

In summary, this is an example of how genetic complexity and individual variations in the phenotypes can be integrated into preclinical animal models. Basing research in such a model where genetically different mice all display varying extents of AD is likely to accelerate the development of precision medicine.

Precision Medicine for Frontotemporal Dementia

Similar to AD, frontotemporal dementia (FTD) is another neurodegenerative disorder with high individual variability. Patients can present with a variety of behavioral phenotypes, differences in neuroimaging, and pathological biomarkers.^[10] FTD affects the frontal and temporal lobes, in particular, causing early-onset dementia.^[11] Identification of genetic risk factors, diagnostic biomarkers, and therapeutic targets are crucial for diagnosing the disease before symptoms arise and potentially halting disease progression. This is where precision medicine and targeting come in.

Thanks to new techniques including next-generation sequencing, scientists now know that tau pathology is related to FTD. Tau pathology includes mutations in the microtubule-associated protein tau (MAPT) gene are major contributors to FTD.^[12] Tau protein’s function is to bind and stabilize microtubules, thereby ensuring normal cellular morphology and function. However, MAPT mutations cause tau to dissociate from microtubules and aggregate inside neurons and glia, leading to axonal transport,synaptic defects, and observable cognitive defects.

Knowing that tau pathology leads to the loss of microtubule-stabilizing function, alleviating tau aggregation and strengthening microtubule stability through precision medicine may be helpful.

Matsuoka et al. tested microtubule interacting agent, NAPVSIPQ (NAP), in a triple transgenic mouse model (3xTg-AD) that has both Aβ and tau pathology.^[13] The researchers found that three months of intranasal NAP treatment at an early-to-moderate stage of disease significantly alleviated tau pathology. However, do these physiological changes also translate to better cognitive functions?

Next, the researchers used the Morris water maze (MWM) to examine spatial learning and memory. The test is based on the animal’s innate desire to escape a stressful situation (like being placed in water) by finding the hidden platform. It turns out that mice treated for six months with NAP performed significantly better in the MWM task, compared to control mice.

In addition, the researchers used a habituation & dishabituation test to determine whether a mouse can distinguish between new and old odors. Generally, mice habituate quickly to odors that they have encountered before and remembered, which would be reflected by reduced sniffing times. In the study, only mice that were treated with NAP showed significantly more interest when presented with a new odor, indicating superior olfactory recognition memory. These findings demonstrate the targeted tau interventions can significantly alter physiological and cognitive outcomes in FTD.

In general, tau-targeted therapeutics focus on intervening on tau pathology through tau-targeted immunotherapy, microtubule-stabilizing drugs, tau aggregation inhibitor administration, and vaccines. [a] The study by Matsuoka et al. showed promising results of a microtubule-interacting agent in mice with tau pathology, paving the way for future clinical trials and studies. This is an example of how preclinical evidence using transgenic animals, behavioral assays, and targeted approaches could fuel further discoveries and research on FTD.

Precision Medicine for Rett Syndrome

Rett syndrome is a neurodevelopmental disorder that is most often caused by mutations in the methyl-CpG binding protein 2 (MECP2). Patients develop a host of cognitive and other behavioral defects. Traditionally, pre-clinical investigations of how a gene affects brain development would involve generating transgenic knockout models. In fact, Mecp2-null mice have driven initial mechanistic studies for Rett syndrome.

However, human patients with Rett syndrome can have different mutations in the Mecp2 gene that produce different versions of defective MECP2 proteins. This peculiarity opens the doors for precision medicine. Individual genetic mutations have effects that are likely different from a complete lack of Mecp2 in the genome. This poses the question: could different mutations have different disease progression and require different treatments?

A good way to answer this question and make treatment more precise for patients is to generate different Mecp2 mouse models carrying a specific human pathogenic mutation. For example, Gandaglia et al. generated a novel knock-in model (Mecp2Y^120D) that carries the human pathogenic mutation Y120D.^[14]

Using a battery of behavioral assays, the researchers demonstrated the typical deficits in motor coordination, gait, strength, learning, and memory (Table 2).

Table 2: Assays used to demonstrate various behavioral deficits in Mecp2Y^120D mouse model.^[14]

Despite developing similar Rett-like behavioral impairments, the molecular phenotypes of Mecp2^Y120D mice were different to the Mecp2-null mice. Specifically, the Y120D mutation impairs a tight binding between Mecp2 and the chromatin, thereby preventing it from recruiting corepressors on heterochromatin. In contrast, the Mecp2-null model is characterized by a more closed chromatin structure.

Due to the variability in molecular phenotypes, effective therapeutic strategies are expected to differ depending on the specific mutation. Precision medicine can offer the best solution knowing a patient’s specific mutation profile for Mecp2. The Mecp2Y^120D mouse mode is a step in the right direction.

Precision Medicine for Cancer

Cancer is another excellent example for demonstrating the importance of precision medicine. It is a disease with a high global burden but varies tremendously from individual to individual.

Since there is no one magic pill, a multitude of strategies have been explored to combat cancer, including:

• Recombinant proteins
• Antibodies
• Checkpoint inhibitors
• Vaccines

From all of these approaches, the fast-growing field of immunotherapy based on patient-specific immune profiles is one of the most promising approaches for treating cancer.

Glioblastoma

For example, glioblastoma is a highly heterogeneous brain tumor that is associated with early cognitive decline.^[15] Interestingly, Lopes et al., demonstrated in a mouse study that a specific peptide (HOP_230-245) could block the interaction between the HOP protein and cellular prion protein, thereby inhibiting glioblastoma growth and improving cognitive decline. These findings warrant further development of HOP_230-245 as a precision medicine for patients with glioblastoma showing high levels of biomarkers indicating HOP-prion interaction.^[15]

In the study by Lopes et al., mice inoculated with glioblastoma cells were tested for object recognition, a popular test for cognitive changes in memory. The mice were trained with two identical objects to gain familiarity. On the test day, one object is replaced by a novel one. Normally, mice have an innate preference for the novel object, which should be reflected in a longer exploratory time.

When tumor-inoculated mice were treated with HOP_230–245 peptides, they could successfully distinguish between novel and familiar objects 12, 17 and 25 days after the treatment. By contrast, mice treated with a control peptide that does not disrupt the interaction showed an inability to recognize new objects by day 12.^[15]

Similarly, another subgroup of glioblastoma patients (20-30% of all patients) have an Epidermal growth factor receptor (EGFR) deletion mutation, whilst other patients have different mutations in the same gene or different genes.^[16] This further exemplifies the need for precision medicine. For treating patients with EGFR deletion mutation specifically, a precision vaccine (rindopepimut) targeting the mutated sequence was developed. The rationale is that since the patients have a cancer cell-specific version of EGFR that acts as an antigen. Delivering a specific antibody may induce an effective immune response that eradicates the tumor cells expressing the target antigen.

In another study, Chen et al. predicted target drugs across different glioblastoma subtypes by combining disease genomics and mouse phenotype data.^[17] The authors demonstrated that this combined approach of computational drug discovery could be used to design precision medicine for patients with different disease subtypes and genetics.

The majority of precision medicine programs (such as AoURP) today involve the use of genetics and genomes, mostly due to DNA data being more available and interpretable. To advance the use of precision medicine, datasets for proteome, epigenome, phenotypes, neuroimaging and neurochemical biomarkers are also necessary in the future. They can give researchers specific targets to develop antagonizing drugs and therapies.

Changes are also needed in the fundamental animal models so that they can better mimic individual patients. For example, genetically engineered animal models based on an individual patient’s genetics are first steps that some researchers are already undertaking. Here, we have discussed how animal models have been used to model variations in dementia, Rett syndrome, and cancer.

Researchers still need to take into consideration individual epigenomes, biomarkers and environment interactions in the future. Although the remaining tasks are still vast and difficult, enabling precision medicine is an inevitable trend. Bringing animal models closer to human patients is an important step in accelerating precision medicine that will pave the way for progress and medicinal advancements.

1. Tibbetts, J. Precision Medicine Brings Mouse Models Closer to Human Patients BioScience 68, 828 (2018).
2. Jones, D. S. How personalized medicine became genetic, and racial: Werner Kalow and the formations of pharmacogenetics. Journal of the History of Medicine and Allied Sciences, 68(1), 1–48. (2013).
3. Roberts, J., & Ostergren, S. Direct-to-Consumer Genetic Testing and Personal Genomics Services: A Review of Recent Empirical Studies. Current Genetic Medicine Reports, 1(3), 182–200. https://doi.org/10.1007/s40142-013-0018-2 (2013).
4. Dunn, J., Runge, R., Snyder, M. Wearables and the medical revolution. Personalized Medicine 15(5), pp. 429-449. (2018)
5. Denny, J. C., Rutter, J. L., Goldstein, D. B., Philippakis, A., Smoller, J. W., Jenkins, G., … Brilliant, Murray. The “All of Us” Research Program. The New England Journal of Medicine, 381(7), 668–676. https://doi.org/10.1056/NEJMsr1809937 (2019).
6. Lida Anestidou, Erin Hammers Forstag, National Academies of Sciences, Engineering, Medicine, Division on Earth Life Studies, & Institute for Laboratory Animal Research. Advancing Disease Modeling in Animal-Based Research in Support of Precision Medicine: Proceedings of a Workshop. National Academies Press. https://doi.org/10.17226/25002 (2018).
7. Lloyd, K. C. K., Robinson, P. N., & Macrae, C. A. Animal-based studies will be essential for precision medicine. Science Translational Medicine, 8(352), 352ed12. https://doi.org/10.1126/scitranslmed.aaf5474(2016).
8. Gatz, Margret, Pedersen, Nancy, Berg, Stig, Johansson, Boo, Johansson, K, Mortimer, J A, … Ahlbom, A.Heritability for Alzheimer’s disease: The study of dementia in Swedish twins. Journals Of Gerontology. Series A, Biological Sciences And Medical Sciences, 52(2), M117–125. https://doi.org/10.1093/gerona/52A.2.M117 (1997).
9. Neuner, S. M., Heuer, S. E., Huentelman, M. J., O’connell, K. M. ., & Kaczorowski, C. C. Harnessing Genetic Complexity to Enhance Translatability of Alzheimer’s Disease Mouse Models: A Path toward Precision Medicine. Neuron, 101(3), 399–411.e5. https://doi.org/10.1016/j.neuron.2018.11.040 (2019).
10. Liu, M. N., Lau, C. I. & Lin, C. P. Precision Medicine for Frontotemporal Dementia. Front Psychiatry 10, 75, doi:10.3389/fpsyt.2019.00075 (2019).
11. Che, X. Q., Song, N., Gao, Y., Ren, R. J. & Wang, G. Precision medicine of frontotemporal dementia: from genotype to phenotype. Front Biosci (Landmark Ed) 23, 1144-1165, doi:10.2741/4637 (2018).
12. Fanny M. Elahi, & Bruce L. Miller. A clinicopathological approach to the diagnosis of dementia. Nature Reviews Neurology, 13(8), 457–476. https://doi.org/10.1038/nrneurol.2017.96 (2017).
13. Matsuoka, Y. et al. A neuronal microtubule-interacting agent, NAPVSIPQ, reduces tau pathology and enhances cognitive function in a mouse model of Alzheimer’s disease. J Pharmacol Exp Ther 325, 146-153, doi:10.1124/jpet.107.130526 (2008).
14. Gandaglia, Brivio, Carli, Palmieri, Bedogni, Stefanelli, … Landsberger. A Novel Mecp2 Y120D Knock-in Model Displays Similar Behavioral Traits But Distinct Molecular Features Compared to the Mecp2 -Null Mouse Implying Precision Medicine for the Treatment of Rett Syndrome. Molecular Neurobiology, 56(7), 4838–4854. https://doi.org/10.1007/s12035-018-1412-2(2019).
15. Lopes, M. H. et al. Disruption of prion protein-HOP engagement impairs glioblastoma growth and cognitive decline and improves overall survival. Oncogene 34, 3305-3314, doi:10.1038/onc.2014.261 (2015).
16. Weller, M. et al. Rindopepimut with temozolomide for patients with newly diagnosed, EGFRvIII-expressing glioblastoma (ACT IV): a randomised, double-blind, international phase 3 trial. The Lancet Oncology, 18(10), pp.1373–1385. (2017)
17. Chen, Y., Gao, Z., Wang, B. & Xu, R. Towards precision medicine-based therapies for glioblastoma: interrogating human disease genomics and mouse phenotypes. BMC Genomics 17 Suppl 7, 516, doi:10.1186/s12864-016-2908-7 (2016).