How AI is advancing and transforming future cancer medical treatments and how a cure for cancer may be closer than we think.
AI is transforming medical treatment in many areas and across many diseases, not just cancer. Within oncology however, which is a particularly complex and challenging disease, AI is making an even more significant difference.
Technology is accelerating the way molecules are designed and developed in new approaches against cancer, allowing us to invent innovative molecules that could work better and safer in a shorter amount of time. This stands to bring improved treatments faster to patients.
Findings published by the WHO last year revealed that in 2022 there were an estimated 20 million new cancer cases and 9.7 million deaths. Globally, about 1 in 5 people develop cancer in their lifetime, and approximately 1 in 9 men and 1 in 12 women die from the disease. This is the scale of the challenge.
Pharmaceutical and biomedical companies are now using increasingly powerful computational methods to fight this battle, strengthening their discovery efforts with AI to help find new compounds for killing cancer. In this early-stage, pre-clinical setting, AI is being used to search for, analyse and develop compounds aiming to eliminate the disease. As such, these new computational methods are gradually increasing accuracy, speed and reliability of treatments.
How does AI drug discovery work?
Our computational platform, called Synth AI, achieves acceleration by computationally producing molecules that meet three key objectives simultaneously.
Firstly, the platform ensures that prospective molecules can be synthesised using known chemical methods. Many computational approaches suggest potential drugs candidates that can’t actually be made or have significant challenges in their synthesis. Synth AI avoids this, ensuring the molecules can be not only made but also scaled using known technologies. This significantly boosts the chances of these treatments being successful when they reach clinical or commercial use.
The third key criterion is that Synth AI optimises the chances that these molecules have the desired biological effect. The platform delivers molecules that are not only synthesisable and scalable but also stand to be biologically effective, critically increasing the overall accuracy, speed and reliability of the drug discovery process.
These three metrics – ensuring the molecule can be active biologically, that it can be made in the laboratory, and that it may be scaled cost-effectively – are making these treatments more likely to come to market. This potentially more cost-effective drug development is a positive prospect for both patients as well as for investors in the space.
Balancing efficacy with side effects in oncology
AI drug discovery is playing a crucial role in improving efficacy, dose response and toxicity in cancer treatment as well.
Considering the unmet need and the lethal consequences of cancer, efficacy is typically prioritised, namely how effective the treatment is at killing cancerous cells. However, the market is being populated with an arsenal of anti-cancer agents which have side effects that remain a major burden for patients.
Our efforts, through the use of AI, are focused on improving efficacy while overcoming the side effect disadvantages that plague many existing treatments.
This comes back to the precision of AI.
As a concrete demonstration of how computationally sourced molecules are being validated, our initial tests against cancer in mice have produced clear evidence of tumour regression and a good safety profile. Remarkably, this is being achieved with a drug candidate that has not even been optimised in the lab.
This points to the chemical scaffolds identified using the proprietary AI computational component already producing molecules with significant advantages that would normally be far behind in the optimisation curve. The benefit for patients, physicians and also investors, is that because treatments are starting their journey to the clinic at a more advanced point, the time and capital at risk during the optimisation phase is reduced overall, speeding up the process and prospectively providing faster economic returns.
The fact that AI drug discovery can achieve such superior results with an unoptimised drug candidate means that we’ve jumped ahead in time compared to what would have normally been required to reach this point.
What AI means for patients and cancer treatment
AI is therefore getting better, more effective and safer treatment to cancer patients more quickly with the associated reduced risk for investors.
When such a validated, strong technology is devised and applied by a leading scientific team, the result can be potential treatments which begin their life cycle as if they’ve been optimised over many months or years. The saved time and valuable resources for drug discovery and development companies means investment then goes further and money can be used more efficiently within the industry.
Importantly, this new approach can positively impact disease areas beyond oncology. Within our own pipeline we have additionally demonstrated the versatility of AI in creating new potential treatments also against resistant infections.
And it isn’t just us. Drug discovery efforts of other pharmaceutical and biotech companies in numerous other conditions are demonstrating the broad applicability of AI computational technology in drug discovery.
Many novel AI methods being utilised within the industry today have already been found to work for a diversity of targets, independently of the therapeutic area or the nature of the target itself.
By Dr. Alan D Roth, CEO of Oxford Drug Design
