Artificial Intelligence in Oncology

Podcast Transcript

Dale Shepard, MD, PhD: Cancer Advances, a Cleveland Clinic podcast for medical professionals, exploring the latest innovative research and clinical advances in the field of oncology. Thank you for joining us for another episode of Cancer Advances. I'm your host, Dr. Dale Shepard, a Medical Oncologist, Director of International Programs for the Cancer Institute and Co-Director of the Sarcoma Program at Cleveland Clinic. Today, I'm happy to be joined by Dr. Jame Abraham, Chair of the Department of Hematology & Medical Oncology at the Cleveland Clinic Cancer Institute. He was previously a guest on this podcast to discuss developments in breast cancer clinical trials, and that episode is still available for you to listen to. He's here today to talk about the use of artificial intelligence in oncology. So, welcome back.

Jame Abraham, MD: Thank you, Dale. I really appreciate. Thank you for having me again.

Dale Shepard, MD, PhD: Sure. So, we're going to talk about artificial intelligence today. Everyone hears all kinds of things about artificial intelligence, and sometimes it's not even clear what exactly that means these days. We have a number of different ways this may be coming into play with oncology. You actually helped put together a conference here at Cleveland Clinic on artificial intelligence. Tell us a little bit about why you decided to put together a conference, and we'll talk a little bit more later about some of the things that may have been discussed.

Jame Abraham, MD: Sure. Absolutely. And thank you for this opportunity. So, AI or artificial intelligence, in my mind, this is going to change the way we live, and it's going to redefine even our whole existence. So, when I say it's going to redefine how we live, how we work, how we relate, it's going to redefine how we work in the healthcare space. It's going to change how we interact with our patients. It's going to change how we interact with our electronic records. It's going to change how we deal with data on a day-to-day basis.

So, the reason I organized the AI healthcare conference with Dr. Steele and with the support of our entire AI leadership at Cleveland Clinic, it's mainly for people like you and me who know that it's going to change our life, but we don't really understand what that means. So, again, we're just clinicians. So, for me, I think this is an opportunity for us to push ourselves to this unknown space so that we can start understanding how this AI is going to impact us. That's number one reason.

And number two, it's for clinicians. It's for clinicians. It's not a technology conference. It's for clinicians, how this is going to impact the clinical space.

And the third, and probably most important, is, when the EMR came to the healthcare space, the EMR was actually electronic medical record. Electronic medical record was defined or created by people other than clinicians. So, now AI is coming to our world. It's really important for people like you and me, clinicians, to get to the table and start having the conversation so that we, the clinicians, who really take care of patients, can define how it's going to change our day-to-day work in a patient care space.

Dale Shepard, MD, PhD: So, instead of trying to take something that's out there and make it work for us, try to actually start at the beginning and have it work from the get-go.

Jame Abraham, MD: Absolutely. If we don't do that, Dale, someone else is going to tell us how it's going to do it for us. But we want to set the tone so that it's going to be meaningful for our patients and for us.

Dale Shepard, MD, PhD: What were some of the areas that were covered in the conference in terms of how AI might be used in oncology?

Jame Abraham, MD: In the meeting, it was actually a really successful meeting, we had about 550 people, 350 at site and 200 online, 30 different states, 10 different countries. So, it went really well. So, why do I say those numbers? It actually shows the appetite for clinicians. When I say clinicians, it's doctors, pharmacists, nurses, students, administrators, all showed up. So, there's a strong appetite for people to understand the space. So, we covered our inpatient space, our outpatient space, ICUs and critical care, medical education, patient safety and quality, radiology and imaging, pathology. So, we kind of looked at the full spectrum of what we do on a daily basis.

Dale Shepard, MD, PhD: What were some of the surprises? I mean, we have our own interactions in terms of how we have encountered AI, how we perceive AI. What were the biggest things from people presenting like, "Wow, I never really thought that it was going to be entering that space"?

Jame Abraham, MD: People were so excited. They were wide-eyed and they were really excited. They really want to hear all these news. And there were really interesting questions. And surprising thing were we have number of experts really in the space, and they all, the limited number of doctors or clinicians and pharmacists and nursing leaders in this space keep telling the same thing. It's really important for us to set the tone, how it's going to shape our world. So, I think that's the big takeaway I have. And Dr. Ridgeway was really kind enough to give the opening statement, and she was really excited about how this is shaping up.

Dale Shepard, MD, PhD: Do you get a sense that... certainly, tech companies are the ones driving development. Are they being receptive to clinician input, clinician sort of co-development of products?

Jame Abraham, MD: They are, but-

Dale Shepard, MD, PhD: That seems like a hesitant they are.

Jame Abraham, MD: I'll tell you, Dale. And you know that. I mean, again, I don't interact with Silicon Valley people all the time, but most of the Silicon Valley people are in their 20s and 30s. They just came out of some fancy college and fancy degree. They never lived or experienced the complexities of healthcare. So, it's really important for people like us, who lived in this space for 20-30 years, to make them understand how complex. This is not a tech solution. It's a human issue. So, people like us really need to educate them to be aware of what we're doing on a daily basis.

Dale Shepard, MD, PhD: What have you encountered along the way with our colleagues, with other clinicians in terms of sort of apprehension and pushback? You mentioned the electronic medical record and good and bad. Ultimately, there's probably a lot more good than bad. But tremendous pushback when it first was introduced. What kind of apprehensions do you face and people sort of being worried about?

Jame Abraham, MD: Let's just say any change is hard. Anything new is tough. And as you know, there's always the bell curve, and the so-called early adopters and then the other end. "Well, I'm going to do it," "I'll never do it" kind of people on the other side. And they all have their own reasons, and I respect that. But I think it's important for us to really first make sure it's user-friendly, it's not adding one more layer like the EMR did to us, right?

Dale Shepard, MD, PhD: One more layer or 10 more clicks.

Jame Abraham, MD: 10 more clicks. And the electronic medical record really made us highly-paid typists, right? That's what we do. And a lot of nurses and pharmacists and doctors spend more time in front of computers. That's the initial skepticism. "Oh, is this going to force us to do something more?" But I think we really need to educate and make sure the tech industry understand this too. Our purpose is to spend time with our patients and find solutions for the patients' problem. The technology should not be a hindrance. It should help us to do that. So, I think that's the major concern from our providers.

Dale Shepard, MD, PhD: Talking about some maybe specific applications. So, you've mentioned a couple times about interacting with patients and importance there. There are too many times busy clinicians really don't even look at the patients. They're typing the note, looking at the computer as the patient talks. Tell us a little bit about how we've incorporated AI into developing clinical notes and really just being able to talk to the patient and not even take notes ourselves.

Jame Abraham, MD: Right, right. Yeah, absolutely right. Like you, I've been part of the transition from the paper record to the electronic medical record, and that really took away. And actually, the computer came in between the doctor and the patient. That happened in 2007 or something, and it kind of kept getting worse in the past 5-10 years. The AI-enabled technologies for capturing the meaningful interaction between the patient and doctor, and that's translating into an electronic health record. And as you know, we've been working with the ambience AI technology.

It's a lifesaver for me. Listening to the conversation between the doctor and the patient, we can really talk to the patient. Then the ambience, this AI, generative AI is smart enough to understand the conversation, filter out the noises, focus on the actual problem. And at the end... We should, of course... There's always a need for human editing, and we can actually save lot of time in capturing all those things. And then at the end, we can actually summarize what we told the patient in a separate document, what we call as the after-patient-visit summary. So, it helps the patients. They'll get a nice summary at the end. And helps the doctors. We don't have to sit and type. Save a lot of time for us.

Dale Shepard, MD, PhD: I've had a great experience with it. You start the recording on your phone. You just sit and talk and it captures everything. And it doesn't just transcribe, it actually puts things together in a meaningful way and captures details that you might not have actually put in your note otherwise. So, it's been good. Patient responses. Have you had positive or negative patient interaction with doing this sort of AI?

Jame Abraham, MD: I don't think even the patients are bothered or they don't even... And I say that "I'm using this technology to capture our conversation," they don't care because the technology is so pervasive. People are so used to using different, different technology. So, I think that's one phase. But I think the AI in a oncology space, there'll be so many different applications. So, if I look at patient care is one, even within patient care, I strongly feel the safety and quality can be helped by the AI technology. Just imagine, Dale, let's just say if you have a patient with a low white cell count and has a fever, let's just say if the AI can enable us and say alert, "Oh, this patient is high risk for sepsis." So, kind of safety and quality. That's one side.

Then research, as you know, it's going to play a major role, major role. Even within the research, just talk about our space, the clinical research space. Now what you do on a daily basis, most of the research is driven by the complex molecular picture of the patient's tumor genomics or germline, which is so complex for people like all of us to understand, especially someone is doing so many different types of tumor. Eventually, the AI is going to be the one which is going to help us to filter through all this information, find out what's the right potential standard care or clinical trial option for that patient.

Dale Shepard, MD, PhD: Yeah. And I guess, Jame, you mentioned clinical trials a couple times. Maybe you can sort of highlight the things we're doing clinical trial basis. So, one of the hard parts about clinical trial, you see a patient in front of you and you say, "Well, I've got all these trials. What might my patient qualify for?" Maybe highlight a little bit about how we're trying to do trial matching sort of activities.

Jame Abraham, MD: As all of us clearly understand, clinical trial is the gold standard of cancer treatment. And here, in the adults in the US, less than 10% of patients go on clinical trial. And that's due to many reasons. One reason is, as you said, it's so complex to understand the inclusion criteria, who meets the criteria, who doesn't meet the criteria. So, the AI-based technology, Cleveland Clinic is partnering with a company to help us to identify patients who can potentially eligible for certain trials. And they'll kind of help the clinicians to cut through the noise and select the patients for the trial.

Dale Shepard, MD, PhD: Yeah. And I guess the good part about that is some of these products, I guess, if someone might not have qualified because their white blood cell count was too low, it will go back and then tell you, if those numbers change, whether they're now eligible.

Jame Abraham, MD: Right. Right, right.

Dale Shepard, MD, PhD: So, which is good.

Jame Abraham, MD: Which is huge.

Dale Shepard, MD, PhD: Yeah.

Jame Abraham, MD: Right.

Dale Shepard, MD, PhD: So, when you think about AI, do you think this is going to be something that... I mean, there's always capital costs to these things. Is this going to be something that you think is going to be readily accessible to most institutions, community practices? I mean, you think these are things that, I guess, we'll be able to actually provide to everyone in the short term or no?

Jame Abraham, MD: I think so, like any technology. But the fascinating thing with this technology, unlike many other technology, just think about how the internet came or the cell phones came, the pace of how this technology is being adopted is much faster than any other technology when we look at the history of mankind. So, I think the exponential growth of the computing power, exponential growth of how AI is being used by many purposes on a daily basis potentially will bring down the cost. But of course, there is so many other complex things which I don't understand, including conception of energy. The AI technologies use so much energy. And how do we pay that for as a society? Again, you and me, we're just clinicians. We don't understand that.

But coming back to this thing, you're absolutely right. I think this will transform not just select few of the institutions. It will get to everybody. And then this, potentially, it can help us to level the playing field. Like, the cell phones. Of course, I'm from India. Now, my mom can send a message sitting in the kitchen to me, a video message. Just imagine, she's 79 and I never thought that she even knows how to use a phone, but she can send a video message to me. So, I'm thinking this is going to level the playing field, like other technologists did in the past.

Dale Shepard, MD, PhD: Yeah. I mean, you mentioned about even selection of therapies. And the ability to just look and review literature in a very, very rapid way, things like open evidence and things like that are really powerful in terms of instead of trying to remember there might've been a paper about something, you can very quickly just summarize a massive amount of information, which is good.

Jame Abraham, MD: Which is huge. As you know, Google just partnered with ASCO to help us exactly that with ASCO guidelines. And so, people like you and me who sees handful of diseases, we can try to be really up-to-date. But our colleagues who are seeing vast majority of patients in the community, who sees 10 or 15 different diagnosis on a given day, it's really tough to keep up because the field is moving so fast. I always admire our colleagues who can do that. But I think the technology should, will help them to do that, including NCCN guidelines or ASCO guidelines, AI-based technologies.

But, Dale, the thing which I'm really fascinated is how the intersection of genomics and the drug discovery will play in the future. As you know really well, today, drug discovery or development of a drug will take anywhere between 5 to 10 years, or at least 5 years. And historically, 10-15 years. Now it cut down to a little less. And it's an expensive thing, right? 1 to 2 billion dollars per drug. And I'm really hoping the AI-based technologies will allow us to really shorten that time span, that's number one, and also reduce the cost. That's where I think it's going to make the major impact, especially in cancer field and oncology space.

Dale Shepard, MD, PhD: As you sat through the conference talking about AI applications in oncology, certainly lots of fascinating things you see are happening. What was missing? Is there something where you're like, "This is where we really need more progress"?

Jame Abraham, MD: I mean, I think patient safety and quality is something I think we need to continue to focus. How do we get meaningful input from patients? I felt like when I finished the conference, I was thinking, "Okay, probably I should bring a patient voice, how the patients perceive this. What's their anxiety?" like you were asking me some questions, "What's their anxiety? What do they need? What do they think which is going to help them?" I think that's important. And then, medical education is one area which is going to be completely changed because our kids are going to learn this technology in a completely different way, completely different way. I think that's a huge gap. And they're going to shape the future. So, those are the few things I thought we can do better for the next year's conference.

Dale Shepard, MD, PhD: Yeah. The patient education piece, it's interesting. I've been using the AI-driven note process for four or five months. I've only had one patient, though, who was leery of me using the AI. He was like, "I don't know if I trust it." And so, for the most part, it seemed to be pretty well received. So, that's good.

Jame Abraham, MD: Right. It's an exciting time. Really exciting time.

Dale Shepard, MD, PhD: It is an exciting time.

Jame Abraham, MD: Yeah.

Dale Shepard, MD, PhD: So, exciting time. Lots of things we know. And there's going to be, by the time we talk about the same topic after the conference next year, 10 times more things or more. So, yes, that's great. So, well, appreciate you being with us to give us some insights.

Jame Abraham, MD: Thank you, Dr. Shepard.

Dale Shepard, MD, PhD: To make a direct online referral to our Cancer Institute, complete our online cancer patient referral form by visiting clevelandclinic.org/cancerpatientreferrals. You will receive confirmation once the appointment is scheduled.

This concludes this episode of Cancer Advances. For more podcast episodes, visit our website, clevelandclinic.org/canceradvancespodcast. Subscribe on Apple Podcasts, Spotify, or wherever you listen to podcasts.

Thank you for listening. Please join us again soon.