Artificial intelligence has developed a treatment for cancer in only 30 days and might predict a patient’s survival rate.
In a latest study published within the journal Chemical Science, researchers on the University of Toronto together with Insilico Medicine developed a possible treatment for hepatocellular carcinoma (HCC) with an AI drug discovery platform called Pharma.AI.
HCC is essentially the most common kind of liver cancer and occurs when a tumor grows on the liver, in response to Cleveland Clinic.
Researchers applied AlphaFold, an AI-powered protein structure database, to Pharma.AI to uncover a novel goal — a previously unknown treatment pathway — for cancer and developed a “novel hit molecule” that might bind to that focus on without aid.
The creation of the potential drug was achieved in only 30 days from the choice of the goal and after synthesizing just seven compounds.
After a second round of generating compounds, they found a stronger hit molecule — but any potential drug would want to undergo clinical trials before widespread use.
“While the world was fascinated with advances in generative AI in art and language, our generative AI algorithms managed to design potent inhibitors of a goal with an AlphaFold-derived structure,” Alex Zhavoronkov, founder and CEO of Insilico medicine, said in a press release.
AI is rapidly changing the way in which drugs and medicine are discovered and developed, as the standard approach to trial and error is slow, expensive and limits the scope of exploration.
“This paper is further evidence of the capability for AI to rework the drug discovery process with enhanced speed, efficiency, and accuracy,” Michael Levitt, a Nobel Prize winner in chemistry, said. “Bringing together the predictive power of AlphaFold and the goal and drug-design power of Insilico Medicine’s Pharma.AI platform, it’s possible to assume that we’re on the cusp of a latest era of AI-powered drug discovery.”
In 2022, AlphaFold made an enormous breakthrough in each AI and structural biology by predicting protein structure for the entire human genome.
“AlphaFold broke latest scientific ground in predicting the structure of all proteins within the human body,” co-author Feng Ren, chief scientific officer and co-CEO of Insilico Medicine, said.
“At Insilico Medicine, we saw that as an incredible opportunity to take these structures and apply them to our end-to-end AI platform with a view to generate novel therapeutics to tackle diseases with high unmet need. This paper is a crucial first step in that direction.”
Researchers also explained how different AI information can revolutionize health care.
“What this paper demonstrates is that for health care, AI developments are greater than the sum of their parts,” Alan Aspuru-Guzik, a professor of chemistry and computer science at U of T’s Faculty of Arts & Science, said. “If one uses a generative model targeting an AI-derived protein, one can substantially expand the range of diseases that we are able to goal. If one adds self-driving labs to the combo, we will probably be in uncharted territory. Stay tuned!”
A separate study published within the journal JAMA Network Open showed an AI system invented by scientists on the University of British Columbia and BC Cancer was in a position to predict cancer patient survival rates using doctors’ notes.
The model uses natural language processing (NLP), which is a component of AI that may understand complex human language.
The NLP can analyze doctors’ notes after an initial consultation visit and discover individual characteristics specifically for every patient.
It was in a position to predict six-month, 36-month and 60-month survival with an accuracy rate of over 80%. This model also can determine rates for all cancers, while previous models were only in a position to apply to certain cancer types.
“The AI essentially reads the consultation document much like how a human would read it,” lead writer Dr. John-Jose Nunez, a psychiatrist and clinical research fellow with the UBC Mood Disorders Centre and BC Cancer, said in a press release. “These documents have many details just like the age of the patient, the kind of cancer, underlying health conditions, past substance use, and family histories. The AI brings all of this together to color a more complete picture of patient outcomes.”
Cancer survival rates are traditionally calculated retrospectively and only categorized by just a few generic aspects akin to tissue type and cancer site.
This model was tested using data from 47,625 patients across six BC cancer sites positioned in British Columbia.
“Since the model is trained on B.C. data, that makes it a potentially powerful tool for predicting cancer survival here within the province,” Nunez said.
“The wonderful thing about neural NLP models is that they’re highly scalable, portable and don’t require structured data sets,” he added. “We will quickly train these models using local data to enhance performance in a latest region. I’d suspect that these models provide a superb foundation anywhere on the earth where patients are in a position to see an oncologist.”
AI could possibly be a cutting-edge technology for future cancer care that could possibly be applied in cancer clinics all over the world.
“Predicting cancer survival is a crucial factor that may be used to enhance cancer care,” Nunez said. “It’d suggest health providers make an earlier referral to support services or offer a more aggressive treatment option upfront. Our hope is that a tool like this could possibly be used to personalize and optimize the care a patient receives immediately, giving them the perfect consequence possible.”
