Read the Full Video Transcript

Alicia Morgans: Hi. I'm so excited to be here with Andre Esteva, who is the CEO and co-founder of Artera. Thank you so much for being here with me today.

Andre Esteva: Thank you for having me, Alicia.

Alicia Morgans: Wonderful. Can you tell us, who is Artera and why should we be paying attention?

Andre Esteva: Artera is a company that develops medical AI tests that will help personalize therapy for cancer patients. At the moment, we've developed a suite of tests for prostate cancer patients and in the future our long-term mission is to become a global company that helps to personalize decision-making across diseases.

Alicia Morgans: You mentioned AI, artificial intelligence, this has many meanings, I think, in our culture and in science, of course. Why is AI or artificial intelligence important to clinical decision-making? How can it actually help us do what we do on a daily basis that we think we do pretty well?

Andre Esteva: The promise of AI is absolutely remarkable. What AI can do that conventional techniques and computers cannot do is it can learn from huge amounts of very heterogeneous data in ways that people can't possibly hope to learn from. By doing so, you can develop specialized tools to support clinical care.

Alicia Morgans: Can you tell me a little bit about how you actually make the AI algorithm? What are the components and how does it all come together?

Andre Esteva: Our AI is a very unique multi-modal algorithm composed of two pieces. One piece learns from imagery, and the other piece learns from a patient's clinical data. It's been trained on enormous amounts of data, tens of thousands of patients, hundreds of thousands of pathology slides, and millions of images, all combined to give it the algorithm. Very, very strong prognostic and predictive abilities.

Alicia Morgans: Do these tools replace the doctor or how do they work to contribute to that decision-making?

Andre Esteva: Absolutely not. We do not replace physician care in any way, shape or form. We are a complement to the way that physician care is performed today. Suppose that you have a patient in front of you and they've been diagnosed with cancer and you're wondering as a physician how to treat them. You're wondering how aggressive their disease is. You're wondering how they would respond to one therapy versus another. That's where AI can enter and help guide therapy for that patient.

Alicia Morgans: Tell me a little bit about Artera's AI as it relates to pathology. When we make decisions in prostate cancer, for example, we think about clinical or pathologic features that are defined by our pathologist or measured in labs, and we try to use that information to understand, to your point earlier, how aggressive the cancer is. How does AI in pathology help us get a better sense of how aggressive the cancer is or, perhaps, how aggressive we need to be in terms of applying certain treatments?

Andre Esteva: The way AI works is it's composed of two pieces. It's a multi-modal AI, as we call it, that looks at the patient's digital pathology and it also looks at the patient's clinical data. It fuses those two streams of information to be able to predict two things about the patient; one, their likeliest prognosis, and two, their response to a particular therapy such as, for instance, hormone therapy.

Alicia Morgans: And that's so important, I think, as we try to really match a patient with a treatment that's just aggressive enough but not overdoing things in terms of extra side effects that are not really going to be associated with a meaningful change in the patient's disease control outcome, which is so important. But when I think about this, any additional testing that I do on a patient might use up tissue or may make it more difficult for me to do more testing in the future. How does AI in pathology, the Artera way I guess, how does that consume tissue? How does that use things up or maybe make it harder in the future?

Andre Esteva: That's actually one of the key benefits of leveraging artificial intelligence, is that we don't consume any tissue. The way that our tests work is we will take the patient's pathology slide, we will digitize it, and we will run an AI classifier over it. It takes seconds of cloud compute, and then we will return a report to the referring physician. No tissue is consumed and the pathology slide is untouched.

Alicia Morgans: Well, what's so interesting and important too, from what I hear you saying, is that you're providing a standardized report that is going to give information to the clinician and to the patient that doesn't really rely on any local interpretation or any doctor education or patient education that might be necessary with some advances in technology that we have, for example with imaging, where we really rely on the education and expertise of the specialist who reads the imaging. In this case, we all get the standardized report. We all get the best information that you can offer. Is that true?

Andre Esteva: That's right. No training is required to understand or interpret or work with an Artera report test.

Alicia Morgans: Can you tell me, what have you developed specifically for prostate cancer when it comes to AI?

Andre Esteva: We've developed, at this point, a suite of tests for prostate cancer patients that will help prognosticate patient outcomes and predict response to therapy. For instance, let's say you have an intermediate risk patient in front of you. You are considering radiation for their primary local therapy and you're wondering if they would benefit from hormone therapy as an adjuvant. Our test will tell you.

Alicia Morgans: Wonderful. Are these tests available? Are they in guidelines or available for clinicians to use through standard of care pathways at this point?

Andre Esteva: We are commercially available. The test can be ordered today, and we are recommended by NCCN guidelines as a standard of care test.

Alicia Morgans: Wonderful. Well, I'd love to hear just a closing thought. We're learning so much about who Artera is, what the overall mission is, and how we could think about AI provided by Artera in our clinical practice even today in terms of prostate cancer, so what is your message?

Andre Esteva: Our number one value is really patient success. We've developed a suite of tests to support patients that can prognosticate patient outcome and predict response to therapy. Those tests are commercially available today in prostate cancer and in the future they will be commercially available across a variety of cancers.

Alicia Morgans: Wonderful. Well, thank you so much for getting us up to speed, it was really a pleasure to talk to you.

Andre Esteva: Thank you, Alicia, great to be here.