The ingenuity of medical science has created effective treatments for a multitude of severe diseases that have plagued humanity. Infectious diseases such as the bubonic plague, influenza, and cholera are largely controlled with the development of antibiotics, antivirals, and vaccines and science will eventually prevail with COVID-19. Certain chronic diseases such as cardiovascular and metabolic disorders are now preventable, manageable, or even reversible to some extent with the establishment of the causal link to diet and exercise, and availability of effective therapeutics. Cancer is no longer a death sentence, with a better understanding of genetics and breakthroughs in immuno-oncology.

 

The cure for Alzheimer’s disease may be the “Holy Grail” of medicine today

Mental illness remains the next big challenge for medicine; we are adopting the definition of mental illness that collectively includes neurological, psychiatric, and substance-abuse disorders. The neurological disorder, Alzheimer’s disease (AD) in particular has received much attention because of its high unmet medical need (high disease severity and a dearth of effective treatments), growing prevalence, and high societal costs. AD is a debilitating chronic neurodegenerative disorder characterized by dementia and premature death and accounts for up to 75 percent of all dementia cases. This is a particularly nefarious affliction that robs its victims of the memories and self-knowledge that define them as a sentient being. Today in the US alone, almost six million people suffer from AD, afflicting ten percent of people over 65 years of age. This number will grow due to the rapid aging population, with projections of almost 14 million by 2050. Current US costs from dementia are estimated at USD 470 billion US annually in direct healthcare and unpaid caregiver services.

 

Made with Visme Infographic Maker

 

Currently, there are no validated disease-modifying treatments to effectively treat, slow progression, or reverse AD. The available approved drugs are marginally effective symptomatic treatments. The most recent novel AD compound approval was in 2003; there have been two recent (late 2019) drug developments, but we await real-world evidence of their effectiveness. Oligomannate (Green Valley) was conditionally approved by China’s regulatory agency, with requirements for a global Phase III trial and post-marketing surveillance. Aducanumab (Biogen/Eisai), initially deemed a failure based on interim analysis, was later reported to be effective at the highest dose tested; FDA filing has been completed and the regulatory decision awaits.

Notwithstanding oligomannate and aducanumab, 2003 was the year of last novel AD drug approval by the FDA and EMA. Evaluation of AD clinical trials (from ClinicalTrials.gov) showed that from 2004 onwards, close to 500 interventional clinical trials of Phase II and III compounds were completed, with close to 100 unique compound failures. In fact, since the beginning of 2017, there have been 14 Phase III compound failures. This begs the question of why there has been such lack of success in developing AD treatments: Is there something wrong with our current method to developing treatments for AD?

 

With so many clinical trial failures over the past two decades, Alzheimer’s disease has been the “graveyard” of biotechnology companies.

Perhaps a systems approach, using Artificial Intelligence (AI) and Quantitative Systems Pharmacology (QSP) that considers the organism holistically, and incorporates a multitude of data sources will be more successful. Rather than the classic reductionistic approach to AD drug development, a systems approach can look at the whole organism with its multitude of interactions amongst the many physiological systems. We now know that AD is a complex multifaceted disease linked to, for example, dysregulation of cholesterol homeostasis, changes in energy metabolism and mitochondrial dysfunction, activation of inflammatory pathways, and the function of neurotrophic factors. The AI-driven modeling approach, can accommodate huge and varying data types of the latest scientific advances, real-world clinical evidence, failed clinical trials, and AD patient databases.

We are advocating for an approach that incorporates the strength of explainable AI models to work with many different data types, with the mechanistic framework of QSP models trained to the patient data from AD databases (e.g., ADNI, RADC*) to develop an AD evidence platform. Such a platform may aid in explaining the failure of previous targets, and enable finding new targets or a combination of targets for specific cohorts of AD patients, to maximize the chances of demonstrating clinical efficacy.

Given the general lack of success to date in developing effective treatments for AD over the past two decades, perhaps it is time to adopt different and potentially paradigm-shifting approaches.

 

* Longitudinal databases set up for the express purpose of following aging and AD patients:

ADNI = Alzheimer’s Disease Neuroimaging Initiative; RADC = Rush Alzheimer’s Disease Center

 

References

  1. 2019 Alzheimer’s disease facts and figures. Alzheimer’s Dement. 15, 321–387 (2019).
  2. Masters, C. L. et al. Alzheimer’s disease. Nat. Rev. Dis. Prim. 1, 15056 (2015).
  3. Lane, C. A., Hardy, J. & Schott, J. M. Alzheimer’s disease. Eur. J. Neurol. 25, 59–70 (2018).
  4. Green Valley Announces NMPA Approval Of Oligomannate For Mild To Moderate Alzheimer’s Disease. Shanghai Green Valley Pharmaceuticals (2019). Available at: https://www.greenvalleypharma.com/En/Index/pageView/catid/48/id/28.html.
  5. BIOGEN PLANS REGULATORY FILING FOR ADUCANUMAB IN ALZHEIMER’S DISEASE BASED ON NEW ANALYSIS OF LARGER DATASET FROM PHASE 3 STUDIES. Biogen (2019). Available at: http://investors.biogen.com/news-releases/news-release-details/biogen-plans-regulatory-filing-aducanumab-alzheimers-disease.
  6. Wang, C. et al. The relationship between cholesterol level and Alzheimer’s disease-associated APP proteolysis/Aβ metabolism. Nutr. Neurosci. 22, 453–463 (2019).
  7. Tzioras, M., Davies, C., Newman, A., Jackson, R. & Spires-Jones, T. Invited Review: APOE at the interface of inflammation, neurodegeneration and pathological protein spread in Alzheimer’s disease. Neuropathol. Appl. Neurobiol. 45, 327–346 (2019).
  8. Calsolaro, V. & Edison, P. Alterations in Glucose Metabolism in Alzheimer’s Disease. Recent Pat. Endocr. Metab. Immune Drug Discov. 10, 31–39 (2016).
  9. Cenini, G. & Voos, W. Mitochondria as Potential Targets in Alzheimer Disease Therapy: An Update. Front. Pharmacol. 10, 902 (2019).
  10. Bradshaw, E. L. et al. Applications of Quantitative Systems Pharmacology in Model-Informed Drug Discovery: Perspective on Impact and Opportunities. CPT pharmacometrics Syst. Pharmacol. 8, 777–791 (2019).
  11. Wang, H. et al. Conducting a Virtual Clinical Trial in HER2-Negative Breast Cancer Using a Quantitative Systems Pharmacology Model With an Epigenetic Modulator and Immune Checkpoint Inhibitors. Front. Bioeng. Biotechnol. 8, 141 (2020).
  12. Qiu, S. et al. Development and validation of an interpretable deep learning framework for Alzheimer’s disease classification. Brain (2020).

 

 

Nawal Roy is founder and CEO of Holmusk. Holmusk is building the world’s largest real-world evidence platform for neurosciences and aims to leverage its technology to work with real-world data across the whole spectrum of chronic diseases.