Lantern Pharma is harnessing the power of AI and Machine Learning to Fight Cancer

This year, the United States expects to tally more than two million new cancer diagnoses— a first-ever prediction— and more than 611,000 cancer-related deaths. While the death rate from cancer is going down, largely due to improvements in early diagnosis and healthy lifestyle changes, the problem is that fewer people aren’t dying of cancer; that’s because the more people who are diagnosed, statistically the more who will die. For those developing treatments for cancer, this stark reality means that best efforts must get better and breakthroughs must come at a faster pace. That’s where artificial intelligence (AI) and machine learning (ML) can make a difference, and already are changing the biopharma landscape.

AIML is curating data sets in the billions to help scientists find treatments for the incurable

Research data from large biopharma companies is typically locked away in silos, inaccessible and unusable, and often in formats that would have little meaning outside of the lab where it was collected and processed. The global shift to AI biopharma development, however, is pushing companies to make their data machine-ready. This in turn makes it easier for AI to utilize that data to the advantage of researchers and, ultimately, patients who are waiting for life-saving medicines. Machine-ready data reduces the amount of time AI spends on cleaning, processing, sorting, and formatting data, and increases the amount of time it has to make critical connections. As a result, AI is being successfully deployed in discovery and development platforms to seek out, collect, and curate massive, relevant datasets from research that is currently underway, or that has already occurred, all over the world.

Why does access to other scientist’s datasets matter in drug development?

Developing and bringing to market effective therapeutics for difficult-to-treat diseases is a challenge that biopharma has rallied against for decades. Typically, it takes between 10 and 12 years for a novel therapeutic to successfully complete clinical trials and achieve FDA approval. This long and arduous— and deeply necessary, for the safety of patients— process can sometimes drag on even longer if clinical trials have difficulty enrolling enough qualified patients. What’s more, errors in data acquisition or processing can end research long before results and data have been published. This leaves billions of potentially useful datasets in bottomless pits, inaccessible to new research projects that could make use of the data to strengthen their own science. It is expected that global data creation will reach 147 zettabytes by the end of 2024, and more than 30% of that— or 44 zettabytes— will be health data. Until AI allowed scientists to start collecting this data, or really just a fraction of it, into one place, it wasn’t understood just how much data was out there not being used or shared. Access to this data, particularly in machine-ready form, could be the key scientists need to unlocking treatments and cures for diseases of all types.

Lantern Pharma is harnessing AIML to speed up the development of novel drugs for cancer

Lantern Pharma believes that AIML can be used to speed up the drug development process while simultaneously enhancing the effectiveness of potential therapeutics. Lantern’s proprietary RADR® AI and ML platform is leading the way in breaking down information silos and taking on the time-consuming task of cleaning and formatting data so that scientists can focus on what they do best: develop life-saving medicines. RADR® has been collecting and curating health data— including from public, proprietary, research, and anonymized sources— since 2020, all the while getting smarter at its tasks, and giving scientists at Lantern more to work with as they continue the development of therapeutics like LP-300, a treatment for non-small cell lung cancer (NSCLC) in never-smokers and LP-184, a treatment for pancreatic cancer. The company began additive measures in 2021, when RADR® had just 10 million datapoints, and the addition of antigen, immune-response, and protein data from a wide variety of sources set the AI platform on the path to explosive growth. In March 2024, RADR® exceeded 60 billion data points, giving the AI everything needed to contribute further by creating its own models and testing combinations of molecules at speeds scientists can’t achieve on their own. For scientists, that has also meant quicker determinations about whether to spend time and resources on certain compounds.

Importantly, the swift advancement of AI platforms like RADR® is rapidly changing the biopharma landscape. It’s not only speeding up the initial stages of drug discovery, but the entire pipeline, from discovery to market. That’s more than just a breakthrough in science and technology. It’s a message to the two million Americans who will be diagnosed with cancer before the year is out that someone is out there working to ensure they get access to effective therapies and cures when it counts.