At this meeting I talked about the idea of how we can use informatics across a wide variety of different projects that are going on in the immuno-oncology space and specifically how we at the Parker Institute for Cancer Immunotherapy are doing this and thinking through the use of informatics all the way from clinical trials, consortia-wide projects, to data projects where we’re bringing data across our institutes together to answer key questions in the field.
What are informatics?
Informatics is the idea of what can we do with the information that you can collect both molecularly and clinically about patients and apply statistics and computer science methods, data science methods, to better understand how we can use that information to better understand molecular mechanisms, understand different ways of treating patients and potentially different ways of dosing patients and new drugs as well.
What methods have you developed?
At the Parker Institute one of the key things and one of the key benefits from an informatics standpoint is we’ve really thought about how we can weave that into a variety of the different components of translational clinical research. So we’re involved in the ideation phase, how can we think about the hypotheses that our scientists are looking at or that we can look at by looking through the data, both data we’ve collected and publically, to better understand the actions that we can take.
We’re also involved in execution and operationalising both the clinical and the research trials that we’re doing, including clinical reporting and biostatistics, as well as thinking through the translational information we get by translation information – I mean sequencing, imaging, cytometry, microbiome, epigenetics. And then finally the analysis of all of that. A lot of the work, as you’ve heard through the conference, has been around how can we better understand the biology of the patient, how can we bring that information earlier on and how can we make that actionable. In order to do all of that we need to get a better understanding of the immune landscape and the immune measurements of our patients and be able to then understand how to make those actionable.
How does the cost weigh up against the increased efficiency?
The cost piece of the work ongoing is small compared to the cost of the drugs themselves and, I would almost argue, very necessary to be able to understand best how to make the information you’re collecting actionable or best how to understand what drug therapy and drug regimens are going to work and increasingly important from a precision medicine standpoint.
How could this impact practice?
At the Parker Institute we’ve approached this from a couple of different ways from an endgame perspective. We think a lot about what is important for the patient today, what is going to be important for the patient five years from now and sometimes even ten years from now. In terms of what is important for the patient today, just being able to understand how best the therapies will work coming together, so we do a lot of work from a clinical trial perspective where we’re not only understanding what the primary and secondary outcomes are for the trial but also bringing in some deep translational information.
Then as we think about what is going on in the realm of patients becoming resistant to immune therapies or the toxicities that are associated downstream, how can we better get a sense of what that is going to look like two, three, four years from now and understand both the molecular mechanisms that might be causing it but also the clinical features that if we can understand earlier on will help us better drive treatment or drive toxicity management. So we’ve taken an approach of really thinking through deep integrative analysis of data for trials that are working now and collecting. We’ve taken an approach of doing prospective collections where we’re collecting and monitoring information around toxicities and resistance to better understand what is happening to a patient.
Then a third part that also we do is work across industry, across academics and across other institutes to try and get an understanding of the state of the field for a variety of areas like neoantigen prediction and then bringing that to the community so that everyone can benefit from the learnings.
If another institute was coming to me to ask about using informatics the key thing I would be really interested in conveying is informatics is a very broad term, it means a lot of different things to different people but all the different pieces it means are critical to bring together. It’s really understanding both what is the outcome of what you are defining as informatics and bringing that to the table to make sure you have that conversation.
The other piece to really understand, especially from a translational or a data sense perspective, is that it is often not a push button answer, it is not about coming to someone after you’ve run an experiment and telling them to press a button to analyse data. It is very much about involving people earlier on in the questions you’re asking and the ways you’re designing both the trials and the experiments that are going to be critical to be able to maximise the insight you’re going to get downstream from it.