• Sustainability
  • DE&I
  • Pandemic
  • Finance
  • Legal
  • Technology
  • Regulatory
  • Global
  • Pricing
  • Strategy
  • R&D/Clinical Trials
  • Opinion
  • Executive Roundtable
  • Sales & Marketing
  • Executive Profiles
  • Leadership
  • Market Access
  • Patient Engagement
  • Supply Chain
  • Industry Trends

Using AI to Improve Precision Medicine: Q&A With Dr. Kate Sasser, Chief Scientific Officer at Tempus


Sasser discusses how AI is directly helping with problems specific to precision and personalized medicine research.

Kate Sasser

Dr. Kate Sasser
Chief scientific officer

Dr. Kate Sasser, chief scientific officer at Tempus, recently spoke with Pharmaceutical Executive about her work with AI. Tempus is focused on precision medicine, mainly in oncology but also other disease areas like cardiology and neurology. According to Sasser, AI is utilized across the entire precision medicine platform.

It’s used for unlocking different patient subgroups and understanding which medicines go best with those populations, developing next generation genomics test, trials, and other applications.

Pharmaceutical Executive: How is AI being used in the drug development process?
Kate Sasser: AI is really interesting in the drug development space. In the last few years, we’ve all seen how it’s entered the entire drug development process. The two big areas we’ve heard a lot about is in the target discovery and validation space and also the biomarker space, where its used to understand patient populations and matching them with therapies.

In the target discovery and validation space, you see AI and machine learning applications moving very fast in this area. This includes entire companies that are popping up just in this space. What we start to see is a convergence of a few things, such as really large data sets that are now available for AI and machine learning to come in and understand new targets that weren’t able to be unlocked with smaller data sets that couldn’t be combined together.

We’ve heard about Google using something like DeepMind to uncover and unlock how proteins fold and how chemistry behaves in certain applications. Those AI tools are moving very rapidly. Since we now have large data sets in different areas, those types of tools and applications can be applied to large data and it allows for things to be unlocked.

Tempus is also playing a big role in that area. Although we are not drug developers ourselves, we partner with drug development companies in both bio tech and large pharma. We utilize our large, multi-modal data that we now have, which is a combination of clinical, molecular, imaging data that is combined with AI and machine learning capabilities so drug developing companies can unlock new targets and pathways.

There’s also the way AI is being used for biomarkers and patient populations. It can be used to understand which patients will respond best to therapeutic interventions. The field at large is working on this, and Tempus is also playing a big role here as well.

PE: AI’s use for biomarkers goes together with precision medicine, correct?
Stasser: The whole idea of precision medicine is really matching the right therapeutic or intervention with the right patient population. This has been the mantra of a lot of drug companies for several years, but it’s only been in the last few years that we see this becoming a reality. A large part of the promise of AI will be to unlock precision or personalized medicine.

We can think about that in multiple ways. One big way is matching patients to the right trial, which is a really big problem in today’s world. The more precise the medicine or therapeutic becomes, the more difficult it is to find those patients to find those studies to get the medicine approved. That becomes a bottleneck that AI with access to large data sets can really help unlock.

The other big promise for precision medicine is to move away from single biomarkers and move into multi-parametric biomarkers or precision medicine using all types of data. What we start to think about and start to work on at Tempus is being able to combine molecular data with images and using all the different modalities available to really unlock or personalize a particular approach for a patient.

Instead of single genes, we’re going to start to see algorithmic types of approaches being the signature biomarkers at large.

PE: How will these breakthroughs in AI benefit the oncology field specifically?
Stasser: We’ve talked about some of the ways that AI can and already is helping oncology. It can uncover new targets and modalities. It can also find broader multi-parametric biomarkers that can be applied to pull patients into the right therapy and run trials at a faster timeline so drugs can get approved and onto the market.

Outside of the drug development, it will also help physicians as new therapies enter the real world. It will be able to help physicians understand a rapidly evolving space, such as what medicines are available and how to get them to their patients. We have two large customer bases at Tempus: one is on the drug developer, pharmacy side and the other is physicians. AI can help bring those two worlds together.

Recent Videos
Related Content