Can this AI platform change the future of cancer treatments?

Panna Sharma, CEO of clinical stage oncology biotech Lantern Pharma, speaks to EPM about how his company’s artificial intelligence platform could help revolutionise drug development and ultimately change the way we treat cancer.

As a disease, cancer is perhaps both a patient’s and a clinician’s biggest fear.

For patients, a diagnosis of cancer often means undergoing invasive surgery to cut out whatever tumour resides inside them, alongside months or even years of body-wrecking chemotherapy. Often it means a death sentence.

For clinicians, cancer acts as almost a final frontier, an unbeatable and mythologised disease at which surgery and chemotherapy are used in the hope of beating it. And while recent years have seen the emergence of immunotherapies target cancers, the truth is that treatment for the disease has largely remained the same for decades.

Lantern Pharma wants to change that.

Founded in 2014, Lantern Pharma wanted to discover new ways to utilise precision medicine and biomarkers to rescue cancer drugs that had failed in late stage trials. Oncology continues to remain pharma’s biggest market but its success in clinical trials can be devastatingly low.

Lantern Pharma decided on a different approach to clinical trials; one that would involve using artificial intelligence (AI) and biomarker signatures to understand how patients respond to specific drug classes.

The AI platform the company eventually developed - ‘RADR’ - is one of the big reasons CEO Panna Sharma was attracted to the company.

“I really thought that this was a company that could be a game changer because they weren’t just a technology and service provider, they were also biotech,” Sharma mentions.

RADR - which stands for Response Algorithm for Drug Positioning & Rescue – works by leveraging big data and genomics to analyse millions of data points from clinical trials to determine how drugs will work with tumour samples. What this means for clinical trials is that RADR is able to identify patients that will respond better to certain therapies, ultimately de-risking trials for drug development companies. 

“The platform today can very exquisitely tease out who’s going to respond and who’s not going to respond and generate a fairly, smart, practical, genetic signature. And we get that genetic signature from all the way down from the 20,000 genes, to 2,000, to 20-50,” Sharma explains.

Whilst RADR sounds impressive, the platform is only a fragment of what the company is wanting it to be. The platform currently holds around 100 million data points of specific drug tumour interactions, but the company wants it to be bigger - much bigger.

“Our goal is to be at about 300+ million [data points] next year, a billion the year after. At that point it’ll be a fairly unique platform. We’ll be able to not only understand which patients respond to the majority of drug classes, but we’ll also be able to start identifying which combinations of drugs can work best in different tumour dynamics,” Sharma says.

More than this, Sharma hopes that by using RADR, drug development companies will be able to start designing clinical trials in a more efficient way.

“If you can generate the assay or the lab test that can look for a certain biomarker, a certain genomic sequence, or a certain protein like PD-L1, and you can test for that or more importantly you can enroll based on that, well that totally changes the way you should think about the trial,” Sharma explains.

Of course there’s another use-case for RADR and that is for helping reduce the excessive costs pharma companies currently face when conducting clinical trials. Ultimately, if successful, this could mean cheaper drugs becoming more accessible to patients; because despite all of the advances in technology and emergences of the likes of AI, pharma is yet to see any cost returns. Drugs are still overly expensive, so much so that patient advocacy groups regularly criticise the industry and, in the UK, NICE and the NHS are constantly battling to approve drugs that are cost effective.

“We live in an era where drugs have gotten obnoxiously more expensive and everything else in multiple other industries has gotten cheaper. Computers, computing power, clothes. The only thing that’s gotten more expensive is real estate and drugs - pharma and healthcare in the United States. Ridiculously expensive,” Sharma says seriously.

“The opportunity here is to leverage all the advances from the computing and data industry and fundamentally change the cost curve – so that $2 billion cancer drug can hopefully be brought down to $200 million or $100 million and hopefully enable greater lifesaving care.”

When questioned on whether pharma companies will continue to charge extortionate prices for their products - despite his company’s aim of drug development costs being reduced - Sharma is perhaps more optimistic than some.

“I think the future of leadership in pharma and the success of those organisations is going to be based on incorporating patient data more directly and providing transparency into pricing and therapy effectiveness,” he says going on to mention how this kind of activity is already occurring in the rare disease community.

It makes sense. Diseases such as spinal muscular atrophy and cystic fibrosis now have treatments that can vastly improve the lives of patients but at massive costs. So while progress in pharma never comes cheap, the industry must take a step forward towards more responsible pricing, and more transparent conversations surrounding those prices.

But how is all of this going to change the way we treat cancer?

Imagine that if in 10 or 15 years, a patient diagnosed with cancer could be ‘genomically profiled’ as Sharma puts it, so that instead of having to undergo surgery or needless chemotherapy, the oncologist would know which drug or drug combinations would work best against their tumour and prescribe them accordingly.

That’s exactly what Lantern Pharma is targeting with a clinical trial investigating the efficacy of a drug to treat prostate cancer. In collaboration with The Clinical Translational Research and Innovation Centre (C-TRIC), Lantern Pharma is investigating how its drug LP-184 can be used to reduce tumour growth in patients with advanced prostate cancer – with an ultimate aim of bringing the therapy to market.

The unique quality of the trial however comes from the way it is being undertaken. By taking fresh biopsies from the patients’ tumours, Lantern Pharma is hoping to ‘gain real insight into the biology of the disease,’ Sharma says.

It’s fairly uncharted work, as Sharma explains, that enables the company to follow patients to see how they respond to the drug. This way, if a patient isn’t responding to their current standard of care, the data gathered will enable Lantern Pharma to guide clinicians towards other treatments that may be best suited to their cancer.

“In many ways this is the way to start pioneering and thinking about very importantly personalised medicine. I think this is how a lot of future cancer treatments will be done,” Sharma says.