By Conor Hale

Using its artificial intelligence programs, Insilico Medicine said it has identified more than a handful of previously untapped genomic targets that could provide avenues to new treatments for amyotrophic lateral sclerosis and potentially rescue patients from the neurodegenerative effects of the disease better known as ALS.

Through a collaboration with Answer ALS—a global research project working to assemble a massive repository of patient and sample data—Insilico employed its PandaOmics engine to analyze the underlying mechanisms of ALS and parse the sequences of molecular reactions that take place deep within cells but eventually add up to the debilitating and ultimately fatal condition.

Also known as Lou Gehrig’s disease, people diagnosed with ALS rapidly lose voluntary muscle movement, including the ability to walk, talk, eat and breathe. About 700,000 people have ALS worldwide, according to Insilico.

Pharmaceutical drug developers continue to realize the strong therapeutic potential of highly potent active pharmaceutical ingredients (HPAPIs). This is particularly true for therapeutic areas such as oncology, where highly potent drugs are becoming more commonplace. Download this Executive Summary to learn more about Catalent’s approach on safe handling procedures and effective formulation strategy.

Alongside scientists from Johns Hopkins University School of Medicine, Massachusetts General Hospital, Harvard Medical School, the Mayo Clinic, the University of Zurich, 4B Technologies, Tsinghua University and the Buck Center for Aging Research, researchers were able to uncover 28 targets from samples of central nervous system tissue and motor neurons collected by Answer ALS.

Out of those, the act of blocking 18 of those targets showed moderate to strong effects on neurodegeneration, measured through early-stage, preclinical experiments with fruit fly models that help mimic the course of the disease. The researchers’ findings were published in the journal Frontiers in Aging Neuroscience.

 “The results of this collaborative research effort show what is possible when we bring together human expertise with AI tools to discover new targets for diseases where there is a high unmet need,” Insilico founder and co-CEO Alex Zhavoronkov, Ph.D., said in a statement. “This is only the beginning.”

Answer ALS has collected and publicly released 2.6 trillion data points to date as it works toward an ultimate goal of 20 trillion. The study includes health information from more than 800 people with ALS, plus more than 100 volunteers acting as a control group, and spans clinical, genomic and protein expression data.

“We are truly excited to see the Answer ALS data being used to identify possible ALS disease-causing pathways and candidate drugs,” said Jeffrey Rothstein, director of the Robert Packard Center for ALS Research at Johns Hopkins and Answer ALS, Jeffrey Rothstein, M.D., Ph.D.,. “The work by Insilico is exactly how this unprecedented program was envisioned to help change the course of ALS.”

In the study, 17 therapeutic targets were listed as high-confidence, while 11 were described as novel approaches—each representing a potentially different method of interrupting the progression of the disease. Eight genes were verified as previously unreported.

They have been linked to cellular pathways governing protein aggregation and degradation, oxidative stress, mitochondrial dysfunction and inflammation as well as the formation of neurons within the brain and ultimately cell death.

One gene target, dubbed ADRA2B, has been linked to receptors that regulate the release of neurotransmitters within the central nervous system. By interfering with the expression of that gene, the eyes of fruit flies in the ALS model showed markedly less deterioration. According to the study, ADRA2B has been evaluated as a potential target for conditions such as bipolar disease, brain injury and Parkinson’s disease, but not ALS—meaning certain agonist and antagonist drugs could potentially be repurposed for the condition.

A separate pair of targets described as potentially novel for ALS drugs, known as KCNB2 and KCNS3, also performed well in fly models. They would tap into the potassium channels found on the surface of cells in skeletal muscle as well as in the brain and pituitary gland and help regulate the flow of electric current across the cell membranes.

They have also been linked to cell cycle progression, proliferation and apoptosis, while KCNS3 has previously been studied in the early stages of Alzheimer’s disease and as a risk gene in Parkinson’s.

“It is exciting to see the power of AI to help understand ALS biology,” said the paper’s corresponding author Merit Cudkowicz, M.D., chief of neurology and director of the Healey & AMG Center for ALS at Mass General Hospital and Harvard Medical School. “We immediately saw the potential to connect the Insilico team with the multidisciplinary Answer ALS team. We look forward to the next steps to turn this knowledge into new targets for treatments for people living with ALS.”

Insilico previously announced an ALS-focused partnership with the Suzhou, China-based 4B Technologies late last year, with the goal of designing drugs that can enhance neuroprotection and tamp down inflammation.

“From AI-powered target discovery based on massive datasets, to biological validation by multiple model systems (fly, mouse, human iPS cells), to rapid clinical testing through investigator-initiated trials, this represents a new trend that may dramatically reduce the costs and duration and more importantly [improve] the success rate of developing medicines, especially for neurodegenerative diseases” said 4B founder Bai Lu, Ph.D., a professor at Tsinghua University.

Going forward, Insilico plans to work with collaborators to advance some targets toward human clinical trials while also expanding the use of PandaOmics to hunt for targets in cancer, immunology and fibrosis, according to the former Fierce 15 winner’s chief scientific officer and co-CEO, Feng Ren, Ph.D.