Emerging technologies like AI and spatial transcriptomics take center stage

Insights from Owkin's Director of Biomarkers at AACR 2024

The AACR Annual Meeting is the first key meeting of the year for the cancer research community. AACR draws scientists, clinicians, healthcare professionals, survivors, patients and advocates. It serves as a platform for sharing the latest advances in cancer science and medicine; boasting more than 7,200 abstracts — a new record for the AACR annual meeting —  with 6 plenary sessions, over 50 major symposia, and nearly 300 poster sessions. 

This year’s AACR Annual Meeting was truly inspiring

During the opening session, Christina Curtis, PhD, the RZ Cao Professor of Medicine, Genetics, and Biomedical Data Science at Stanford University said that AACR made,

“a concerted effort to weave technology into the scientific program by pairing presentations on discoveries in basic science, translation, and clinical science with presentations on the tools that support and amplify those discoveries.”

This is precisely what Owkin is doing: using technologies such as AI and spatial omic discoveries - along with human insight - to discover new targets, identify relevant patient subgroups, and deliver better drugs and diagnostics at scale. We are excited about tackling this challenging paradigm. At Owkin, we believe that achieving this requires breaking down research silos through great academic collaborations (e.g. MOSAIC) and heavy investment in novel AI technologies such as foundation models (e.g. Bioptimus). I am proud to be part of this new paradigm! 

We were delighted to have strong representation this year, with 3 presentations and 4 posters. Our Owkin team manned our booth, had many excellent face-to-face meetings and shared in the excitement and energy of the event. 

My highlights from the opening plenary session: Inspiring Science, Fueling Progress, Revolutionizing Care

1. From cell atlases to medicines - Aviv Regev, PhD, FAACR, head and executive vice president of research and early development at Genentech presented the Human Cell Atlas project, which aims to map all of the cells in the body to better understand human development and disease. Comically, she likened the data generated by bulk RNA sequencing and genomics methods to a fruit smoothie. However, with the advancement of new technologies such as single-cell genomics and spatial transcriptomics, researchers are now able to look at tissue in a more granular manner, as though one could examine each piece of fruit in great detail - like within a fruit salad or a fruit tart. She concludes that the current challenges involve analyzing this vast library of data in ways that can benefit patients. And that’s where ML/AI comes into the field. However, she also presented the view that AI/ML is not enough on its own; we will need to go back to the lab to perform experiments and start developing new drugs

2. Artificial intelligence-based biomarkers in cancer histopathology - Jakob Nikolas Kather, MD, a professor of clinical artificial intelligence at Technical University Dresden in Germany, presented the role of histopathology images and the use of AI as the next step for biomarker research in cancer. He and his team have shown that they can feed histopathology slides into a deep learning model and train the model to recognize which tumors exhibit high microsatellite instability, a marker of eligibility for certain therapies. Jakob Nikolas Kather was part of the validation of this model development by Owkin, MSIntuit® CRC. “We can use these biomarkers on very inexpensive, routinely available H&E data,” Kather said,

“The technology is here, ready to move to the clinic in the near future."

Our MOSAIC initiative aims to achieve precisely this: creating a multi-omic spatial atlas in cancer to enable precision oncology. During the session 'Advances in Technology - Unlocking Spatial Complexity: Harnessing Machine Learning for Spatial Biology Insights,' our Chief R&D Officer, Jean-Philippe Vert, PhD, and Professor Raphael Gottardo, PhD, Founding Director, Biomedical Data Sciences Center, CHUV/UNIL, presented how new technologies such as spatial transcriptomics and AI can revolutionize the characterization of the tumor microenvironment (TME) for thousands of patients. This advancement promises to unlock new opportunities, including the discovery of novel targets, a deep understanding of patient subgroups, and accurate predictions of responses to perturbations and treatments. 

My insights from AACR - after having the chance to see the partial solar eclipse on Monday, April 8th from directly outside the convention center!

• Cancer vaccines: ready for prime time? This session showcased significant clinical results across various cancer types, definitively highlighting that cancer vaccines represent a new era for patients.
• Designing the next generation of Antibody Drug Conjugates (ADC) - The presentation titled “Biology meets design: Lessons from 30 years of ADC development” by Elaine Hurt from AstraZeneca R&D, Gaithersburg, MD, provided valuable insights into the intricate process of ADC design and underscored the importance of discovering new targets in this field. The significance of ADCs is evident as they represent a crucial avenue for developing new therapies across numerous cancer indications. 

With a record number of more than 23,000 registrants, the conference provided an abundance of research and cutting-edge technologies to enhance our understanding of cancer. The conference signifies a positive outlook for the future of cancer treatment and the ongoing dedication of the field to bringing precision medicine to each step of the patient journey.

Now that AACR is behind us, we look forward to reconnecting at ASCO 2024 in Chicago in June. 

‍