Finding Career Fulfillment Working in Immune Diagnostics

Can you tell us what work you’re leading in immune diagnostics?

My team works at GE Healthcare Molecular Imaging for oncology, specifically for nuclear medicine diagnostics. That is whole-body imaging of patients, as opposed to in vitro diagnostics. My role is to speak to scientists, and clinicians in academia, pharmaceutical and biotech companies to understand what they are doing and relay that back to my team to make the right decisions about developing the right diagnostics in the right way to address the unmet needs.

Whole-body PET imaging has the advantages of being noninvasive, with no biopsies needed and being able to image over time. You get longitudinal analysis of patients, and very importantly for oncology, it is whole body. You overcome the limitations of getting information from small biopsies as opposed to a big picture for a patient, which is very important. We are interested in imaging various biomarkers to understand both the likelihood that a patient will benefit from various immunotherapies, and then also to understand early on whether those patients are responding to their treatment. 

What are you looking for to determine if a patient is a good fit for a therapy? 

To use an example, we are just about to start a clinical trial in patients and are looking at a biomarker for CD8+ T cells. That will inform whether, for example, they are likely to respond to immune checkpoint inhibitors because they have got CD8+ T cells in their tumors already, or whether they will benefit from treatments to generate CD8+ T cells or whether they will require other therapies such as CAR T cell therapies, etc.

What is the current group of technologies that folks are using for diagnostics? 

For certain cancer types, patients are evaluated for expression of PD-L1, using immunohistochemistry staining of a tumor biopsy to understand if they are more likely to respond to immune checkpoint inhibitor therapy. It is the best available, however there are opportunities to improve. And we think that whole body PET imaging of CD8+ T cells will add value. 

"We are developing whole body PET scanning with CD8 to look at T-cells very early on after patients receive their therapy to understand in perhaps one or two weeks, as opposed to months."

And how is treatment response currently assessed? 

This is currently being evaluated using different types of RECIST criteria. Patients have a CT scan, and the size and volume of their tumors are measured at baseline and tracked on treatment. Response evaluation with RECIST often takes months. And some responding patients have an initial increase in the size of their tumors - immune cells infiltrate the tumors and proliferate, so on a CT scan it looks like the tumor has grown, when in fact the patient is responding to the immunotherapy

We are developing whole body PET scanning with CD8 to look at T-cells very early on after patients receive their therapy to understand in perhaps one or two weeks, as opposed to months, whether a patient is responding. This could prevent patients receiving drugs that they are not benefiting from, which are both costly and can lead to immune-related toxicities.

You’re also using AI in your work. Can you tell us how it’s enabling better whole-body scanning? 

We have a program at GE Healthcare that we have conducted and developed in partnership with Vanderbilt University in the US. It is a predictive tool using real world, baseline clinical data from patients before they start on treatment. It predicts both the likelihood of patients responding to their treatment, and the likelihood of them developing immune-related toxicities. 

What data is this predictive tool pulling to make these predictions? 

It uses their electronic health record data, before patients start any IO treatment, to be able to predict their likelihood of efficacy and toxicity. This dataset was trained purely using routinely acquired clinical data from over 3,300 patients receiving immune checkpoint inhibitors across all cancer types. We believe that this makes our models very scalable.

What are the next steps for your AI tools?

We are now planning to deploy them using our GE Healthcare Edison platform, and also working with pharmaceutical companies to be able to incorporate our predictive tool to enrich their trials. This increases the likelihood of success for their trials by increasing efficacy of their drugs and decreasing toxicities. We also think that there is utility in clinical practice by providing decision support to clinical oncologists and by, e.g., predicting which patients may require more surveillance as they are prone to developing toxicities whilst they also have a good likelihood of responding to the therapy.

What do you think is a misconception around the use/usefulness of whole-body PET scanning? 

I don't think it is widely understood. When I was working in academia, immune PET would have been something I would have been very grateful to have been able to use for many of my studies over the years. It has value in its utility, and the longitudinal amount of information you can get. 

It's in the early days. As it becomes more understood, the benefits and the utility of whole-body immune system imaging will generate the momentum for itself. 

How did you transition from a career in research to one at GE? 

My research career was going well at Oxford University; I was publishing, and I was mentoring and teaching there. I made knock-in and knockout mice, which involved working seven days a week for many months. You have to when you are growing and culturing embryonic stem cells to make animals. My experiments meant that I worked late nights and many weekends. 

When I had young children, I chose to step away from the bench. I actually had a different career for a few years. I went into project management. I also learnt computer programming and did scientific consultancy work, so that I could manage my work around a young family. But my bedtime reading was immuno-oncology. When my children were a little bit older, I found I wanted to go back because my passion was immunology and oncology. I saw a job opportunity at GE Healthcare, and I thought it sounded perfect for my skill set and my experience. 

Why did you choose to pursue work in diagnostics, rather than, for example, drug development? 

What I love about it is that you can provide diagnostics across all the different cancer types, and across all of the different treatment types. To be able to have utility and provide value for potentially all cancers, and all treatments expands the impact. Having spent so many years understanding the immune system and understanding the molecular mechanisms of cancer progression, it was a huge tick in the box for me to be able to develop tools to cover all types of cancers and all types of treatments. 

What made you so passionate about your work? 

I worked at Great Ormond Street Hospital for about 10 years. I worked on childhood diseases, first in immunodeficiency disease where children eventually got cancer and didn't survive beyond their teens, and then I worked on childhood leukemia. Working in a hospital where children would come in with cancer and seeing the impact it had on them and their families, it makes me want to do better every day. The team I worked with at GOSH are developing game-changing cell and gene therapies that are now curing these previously incurable diseases.

"For me, it is important to try and do what you are good at and what the world needs."

Do you have a piece of career advice that’s always stuck with you? 

I was in a meeting recently with someone. We ended up talking about science, and he said to me, “You’ve found your ikigai.” I didn’t know what that meant. But it’s a Japanese word that means the intersection of four things. What you're good at; what you love; what you're paid for; and what the world needs. And if you meet all of those in the center, that's your ikigai. And I think as long as you’re doing that, it doesn’t feel like work. 

I didn't have any specific career goals. When I started, I just kept doing what I enjoyed. And that is what has brought me to where I am. For me, it is important to try and do what you are good at and what the world needs.

What would you characterize the overall usage of whole-body scanning in IO as? 

It's emerging. When you talk about PET scans for oncology, people automatically think of FDG scans. Using immune PET in IO is starting to emerge and be accepted, certainly by pharmaceutical companies who are developing IO drugs. Other companies have also been developing imaging agents for IO that have been successful in their early development. 

I think that it will take some more time. However, it is really encouraging that leading experts in immuno-oncology understand the need for PET imaging of the immune system with this whole-body technology. If the products are good, then it is a matter of time before we see it being adopted more widely for use. We are first developing it to enable pharmaceutical companies to enrich their clinical trials. But ultimately, we want to incorporate it into clinical patient management and to benefit patients routinely in the clinic.

For more information, visit gehealthcare.com.