How does Astrophysics help eliminate the trial-and-error approach to cancer care? Hear from Concr's newly appointed CEO Irina Babina, who has shared her experience and Concr's exciting progress in this 5 question interview.
· Congratulations on taking over as CEO of Concr. Could you give us a bit about your background and how you came to the business?
Of course. I’ve always worked in life sciences and started with building my own research in oncogenic addiction of aggressive breast and gastric cancers. After 10 years in academia, I moved into funding management, helping other scientists to develop their ideas. During that time, I completed an MBA and then moved fully into commercial side of science, as a VC and a life sciences consultant.
That’s when I got involved with Concr, initially as a strategy consultant and later became a part-time Head of Strategy. What really appealed to me was their adoption of existing and proven engineering methods to address challenges in medicine. I’m a huge advocate for convergence science and leveraging knowledge from other fields. My involvement with the company kept growing, and I eventually joined the management team and succeeded our founder-CEO in January.
· What is the problem that Concr is tackling and how does it do it?
Concr is on a mission to eliminate trial-and-error approach to cancer care. Despite all the tremendous collective research effort in cancer, patients’ benefit remains disproportionately marginal. We need better tools to select patients for existing and emerging treatments, for which we need to better understand the biology of response and resistance to these therapies. This is where Concr technology can really help move the needle.
To capture the complexity of biology and factor-in our limited knowledge of it, Concr adopted computational modelling methods from astrophysics, which allows cosmologists to simultaneously integrate enormous amounts of complex and fragmented data to model the universe. Concr adapted these Bayesian methods to integrate the diverse and disparate data gathered over decades of cancer research and clinical treatment to model human disease biology.
We use it to examine complex – typically, molecular – data to define characteristics of patients and their cancers that are associated with response and resistance to existing or emerging therapies. Importantly, we can do it at an individual- and cohort-level, which hasn’t been done before.
· Where are you in the development of the platform and what commercial traction do you have?
Most of our work currently is with biotech and pharma looking to reduce uncertainty in drug development and increase probability of success of their drug assets, although we’re beginning to get some interest from healthcare providers, given that we can make predictions for an individual.
Having delivered our first commercial projects last year, we now refined and productised our offers and built the infrastructure for rapid project set up and delivery. Furthermore, we completed beta version of FarrSight® - our cloud native software that allows interaction with the results of our predictive analytics, as well as contains self-service bioinformatics tools.
At this point, we’ve negotiated additional commercial contracts and have a rich pipeline of promising leads across the drug development supply chain. Full launch of FarrSight® is planned for June this year.
· This sounds like a competitive space so how does your approach differ from others?
You’re right, it could be difficult to differentiate all the AI/ML offers currently on the market, and I see it as a positive: this is a new and rapidly growing sector with much promise. What sets Concr apart is our positioning, our technological differentiation and validation studies across biological and clinical datasets.
For starters, Concr is not a drug discovery platform, which so many current offers focus on. We focus on predicting therapeutic response for an individual patient through in-depth understanding of the underlying biology.
What allows us to do it so effectively is the astrophysics technological backbone. It solves the challenges of integrating and learning from fragmented and diverse biological datasets, from preclinical research to patient clinical data, yielding most sophisticated scalable digital models of human disease biology. Consequently, it minimises reliance on “big data” – we can make accurate predictions with a fraction of data needed by conventional ML methods. This is particularly important in cases where research data is limited, for example if a drug asset is early in development or if it’s a rare cancer.
Of course, the only way to know if something is working is through results. We have case studies demonstrating technical, biological and clinical validation of our predictions, generated through collaborations and commercial work. We’ve just released a paper describing our approach to in silico trial predictions, which was validated in 8 historical clinical trials. Last month we launched our first prospective clinical validation study in early triple-negative breast cancer, which will test clinical utility of our platform.
· What has your funding being to date and what are your funding plans for the future?
We are fortunate to be backed by exceptional financial supporters, a testament to the groundwork laid by our former CEO, Matt Foster. Early on, our core technology was developed through our collaboration with Roche and supported by an Innovate UK grant. Securing this grant required us to demonstrate financial stability, and thanks to a swift £200K investment from Cambridge Angels, we achieved this in just two weeks.
Last year we closed a £1.94m oversubscribed seed round, including follow-on investments as well as new institutional investors, which allowed us to complete initial validation studies, secure additional contracts, and enabled us to secure another non-dilutive grant for clinical validation.
These achievements strategically position us for our imminent Series A round, a significant milestone we are about to officially embark on. This pivotal fundraise will support our commercial growth, scale business operations and allow for further improvement and refinement of our technology.
