Key Takeaways From 2024 ASCO Breakthrough

Dr. Lillian Siu and Dr. Melvin Chua discuss the new technologies and novel therapeutics that were featured at the 2024 ASCO Breakthrough meeting.
TRANSCRIPT
Dr. Lillian Siu: Hello and welcome to the ASCO Daily News Podcast. I'm Dr. Lillian Siu, a medical oncologist and director of the Phase 1 Trials Program at the Princess Margaret Cancer Center in Toronto, Canada, and a professor of medicine at the University of Toronto. On today's episode, we'll be discussing key takeaways from the 2024 ASCO Breakthrough meeting in Yokohama, Japan. Joining me for this discussion is Dr. Melvin Chua, who served as the chair of Breakthrough’s Program Committee. Dr. Chua is the head of the Department for Head, Neck and Thoracic Cancers in the Division of Radiation Oncology at the National Cancer Center in Singapore.
Our full disclosures are available in the transcript of this episode.
Dr. Chua, it's great to be speaking with you today and congratulations on a very successful Breakthrough meeting.
Dr. Melvin Chua: Thanks Dr. Siu. It was really inspiring to come together again to showcase the innovative work of world-renowned experts, clinicians, researchers, med-tech pioneers, and drug developers from around the globe. Our theme this year was inclusivity and thus it was important to bring people together again in the Asia Pacific region and to foster international collaborations that are so important in advancing cancer care. This year, we invited 65 international faculty, of which 55% were from Asia. Also, importantly, we achieved approximately a 50-50 split for male to female representation. These are remarkable statistics for the meeting, and we really hope to retain this for future Breakthrough [meetings].
Dr. Lillian Siu: The meeting featured renowned keynote speakers who shared great insights on new technologies and therapies that are shaping the future of drug development and care delivery. Let's first talk about artificial intelligence and the keynote address by Dr. Andrew Trister. He gave a very interesting talk titled, “Plaiting the Golden Braid: How Artificial Intelligence Informs the Learning Health System (meetings.asco....) .” What are the key messages from his talk?
Dr. Melvin Chua: Couldn’t agree with you more, Dr. Siu. Dr. Trister is the chief medical and scientific officer of Verily, a precision health company. He previously worked in digital health and AI at The Bill and Melinda Gates Foundation, and worked at Apple where he led clinical research and machine learning with Apple partners. But perhaps it was really his background and training as a radiation oncologist that was most pertinent as he was able to weave both the components of new AI models and the applications and pitfalls in the clinic to the audience.
Dr. Trister provided a very high-level view through the history of AI and showcased the progression of the different AI models and he basically explained between deep and shallow methods as well as deductive logic versus inductive probabilistic methods. He then provided several clinical examples where these models have shown their utility in the clinic, for example, pathology and so forth. At the same time, he illustrated several pitfalls with these models. So overall, I think Dr. Trister's talk was very well received by the audience with several key messages, including the importance of [using] high-quality data as the basis of a good AI model.
AI was also addressed in an Education Session that looked at Artificial Intelligence in the Cancer Clinic (meetings.asco....) . And we had a panel of experts that highlighted current progress and successes with AI in the clinic, advances with AI assisted pathology for clinical research and precision medicine, large language models (LLMs) for applications in the clinic, and how we could leverage AI in precision oncology. And from this session, I had several key takeaways. Dr. Alexander Pearson [of the University of Chicago] gave a very illustrative talk on how multimodal information across clinical omics, radiological information and multi omics could be used to improve diagnostic tasks and clinical prediction across different cancers. And Dr. Joe Yeong [of Singapore General Hospital] gave a very good talk on how AI can be applied in digital pathology to accelerate research in immunology and help in the development of immunotherapies. Dr. Danielle Bitterman [of Brigham and Women’s Hospital] shared very good examples of how LLMs could be used in a clinic. And I think the example that really stood out for me was how LLMs could be deployed to create responses to patient queries. And of course, the big question in the room was: How could AI eventually encapsulate compassion in their response? I think this again showcased how LLMs could really help to accelerate our clinical work going forward. And ultimately circling back to data, Dr. Caroline Chu...