Taking on cancer with AI: Niven Narain, BERG
This week, New Orleans has played host to an enormous international gathering of cancer researchers at the American Association for Cancer Research Annual Meeting. At the meeting, FLG’s own Katie Draper caught up with Niven Narain of Boston-Based biopharma company BERG to learn about cancer drug development, and explore how artificial intelligence can contribute to precision medicine.
FLG: Tell us a bit about yourself and your role at BERG.
NN: Sure, I’m Niven Narain, Founder and CEO of BERG, and as Founder much of my role involves bringing forth the vision of the technologies, and to put forward these products that are now in clinical trials for cancer, namely BPM31510. The vision is really to get these drugs to patients as quickly as possible, but importantly using a precision medicine approach so that we’re not falling into a one-size-fits all.
FLG: So what particular work are you here at AACR to present?
NN: We had a total of five presentations, most notably on pancreatic cancer. In one, we were able to demonstrate pre-clinically that when our drug BPM31510 combines with gemcitabine, a traditionally-used chemotherapy for pancreatic cancer, we see that there’s greater efficacy of the drug. We’re pretty excited to take that forward into further clinical trials based on that pre-clinical evidence that we have.
In addition we have started a really unique project called Project Survival, with Harvard Medical School, where we are using the samples and tissues from individuals with pancreatic cancer and others who are otherwise healthy, and really using the technology that we have at BERG to figure out what’s so unique in patients with pancreatic cancer as far as their molecular markers, proteins, their lipids, and then using that as a backdrop to define and validate the first ever biomarker for pancreatic cancer.
The other big news for us is in triple-negative breast cancer. Our ability to really see BPM31510 as a promising agent, because there are really no drugs that are approved for triple-negative breast cancer, it’s a huge unmet need, and so we’re excited that there is another indication for us to look at. And then there’s some other mechanistic work that we show with fatty acid metabolism and how it correlates to the onset of cancer.
FLG: So could you tell me a little bit more about your drug, BPM31510, because I’m aware that it’s the first drug developed through artificial intelligence.
NN: Sure, so this drug works by switching the metabolism of the tumour. The tumour likes to operate on lactic acid, that’s a fuel that cancer has a preferential utility for because it’s able to evade cell death. So this drug comes in an switches the fuel of the cancer micro-environment to make it behave much like a normal cell. As for the AI component, the use of this drug was guided by artificial intelligence. During our phase one trials and in our discovery models we were able to look at over 40 different types of cancer and find the trigger points that govern most prominently how cancer works. The AI helps us to figure out which folks have the specific biology that correlates to this treatment, and helps us to be much more efficient in the trials that we do. It’s helping us develop the drug in a way that’s geared towards a specific population, and most importantly it’s helping us to figure out which tumours this drug works on, because it’s a very broad spectrum mechanism. So instead of us having to do a tremendous trial and error, it’s helping us to develop and guide a drug in ways that have never been done before.
FLG: So in terms of those time scales and costs that you mentioned, do you have a sense of how using AI can/has improved those?
NN: Obviously things are still ongoing, so we don’t have the hardcore data points, but I would say that our goal is to decrease the cost by at least 50% and reduce the time to under ten years. Because it is just not sustainable to wait that long and spend huge amounts of money on therapies. So I think the cost and efficiency that we’re going to see, in addition to the precision is really going to be significantly more beneficial to the patient.
FLG: Pancreatic cancer is infamously one of the hardest cancers to treat. Could you go into a little more detail about the data that you have seen with your drug?
NN: What we’ve reported here at AACR is that when we looked at the survival times in a preclinical study, what we noticed is that when we gave our drug first and then added on gemcitabine the survival times of the animals increase by over 50% in some cases. At the end of the day, as a cancer drug we have a responsibility to look at survival. You can look at all the tumour effects, but our goal should be to keep folks alive, right, and so so safely and effectively. Our drug alone has an effect, but when you combine with gemcitabine there seemed to be a significant augmented effect, so we’re very excited as we now put forth protocols into clinical development to translate this to patients.
FLG: Do you have plans to explore using your drug in other therapeutic combinations?
NN: Absolutely, we have an ongoing phase I trial to look at our drug in combination with 5FU, and we’re also examining our drug with other chemotherapies and immunotherapies.
FLG: You’ve also got a series of posters at AACR that are covering some different work. Could you tell us a bit about that?
NN: Sure, we did some studies, the first of their kind, to look at how fatty acid metabolism correlates to the state of cancer. We were able to show that, under certain states in the micro-environment and the metabolic phenotype, certain alterations in fatty acid metabolism has a direct correlation to the oncogenicity or severity of the breast cancer. But when we looked at how fatty acid metabolism correlated to the aggressiveness of those breast cancers, and when the fatty acid metabolism was changed in certain ways by using certain drugs or certain models, we did see that these cancers were also very altered. This shows pretty clearly that the control of fatty acid metabolism has an effect on these types of cancers.
FLG: As well as your progress into next-stage clinical trials, what else does 2016 hold for BERG?
NN: In cancer we’re really excited to expand out into mid and late stage trials in solid tumours. And we also have a few drugs for diabetes that are now going to go from pre-clinical, and we have a goal to complete those studies so that we can begin first-in-man trials in 2017. We have some follow-on cancer drugs that are in pre-clinical development, and in addition we have a programme in Parkinson’s and Alzheimer’s disease. We’ve built these molecular models that have novel drug targets and biomarkers. So there’s a lot going on that we’re excited about.
FLG: Thank you so much for your time. Is there anything else that you would like to share with our readers?
NN: I think what allows the breadth and the diversity of this conversation is really this platform that we’ve built called the Interrogative Biology Platform. That platform takes a patient-centric approach, beginning with patient tissue samples and their biofluids, and of course it ends, as we go through their biology and the use of artificial intelligence, coming back to the patient. By making the right decisions for the patient we allow better tools and arbitration for the physician, but we also allow the patient to be a voice. I think globally in medicine that’s been a big issue. We still dictate to diseases and patients, and I think because of the technologies that we have built at BERG we are bringing that patient voice into the process, and that is what’s going to drive the evolution of drugs that really make a difference.
I use this term “biology-based drugs”, and I think that’s absolutely the future and our company is going to drive that. The other component of that is that precision medicine is getting a lot of attention, but it needs to be done the right way. It needs to look beyond genomics to all the types of data, it needs to look at diversity of patients, not just one sector of patients. I think we’re sitting at a really exciting time in healthcare, where awareness and technology and investment and patient involvement are the nexus of this fusion. If we make hay of what’s ongoing we can make enormous advances in the next five years.