Shooting for the stars: The healthtech using cosmology and AI to try to cure cancer

Dubbed the “emperor of all maladies” by doctor and Pulitzer Prize winner Siddhartha Mukherjee, most cancers is the reason for one in six deaths worldwide. The illness blights the lives of hundreds of thousands due to its complexity and shape-shifting capability to elude the human immune system and adapt to remedies. 

However there are shimmers of sunshine on the horizon. The diagnostic and therapeutic arsenal for oncology is turning into extra focused due to advances in synthetic intelligence, DNA sequencing and precision oncology, which might analyse the genetic and molecular make-up of cancers and establish the cell modifications that drive the unfold of the illness. 

These approaches promise to sort out most cancers based mostly on its distinctive fingerprints and biomarkers, moderately than blanket approaches like chemotherapy, which kill all fast-growing cells each wholesome and cancerous. 

Breakthroughs in non-invasive detection strategies, like liquid biopsies, are enhancing the percentages of catching most cancers earlier when it may be eradicated extra simply. However to design really efficient and dependable therapies, biotech and pharma corporations face an information downside. 

A UK and Australia-based startup, Concr, is popping to the sphere of astrophysics for concepts. Right here’s the huge knowledge downside confronted by most cancers remedy — and the way startups like Concr may assist. 

Information energy

The drug growth journey is expensive and failure-prone. The common value of oncology drug growth stands at $4.4bn per drug and new therapies not often attain sufferers, with virtually 97% failing to move part I scientific testing.

Drug builders typically shelve their analysis, dropping data gained in regards to the illness and the huge knowledge collected throughout most cancers analysis research, together with genomics, proteomics, transcriptomics, imaging knowledge and affected person remedy response and outcomes. 

 Past the lab, loads of knowledge can also be gathered as sufferers criss-cross the well being system. 

“Sufferers are handed from one self-discipline to a different. They go to their GP, then to their oncologist and perhaps the radiologist, after which to the medical oncologist for chemotherapy or different remedies. Plenty of completely different applied sciences are used and every will generate their very own completely different knowledge stream,” says Dr Tom Gibbs, senior funding director at Debiopharm Innovation Fund, which not too long ago dedicated a £200k seed funding in Concr, as a part of an oversubscribed £1.94m spherical.

Every dataset presents a slice of the general image, however they’re siloed, held by completely different stakeholders, creating an innovation barrier for analysis. “It’s not clear that every one that knowledge ever actually will get assembled in a harmonised method,” says Gibbs. “Sharing knowledge would assist to beat this, however there must be a solution to join the completely different knowledge, equivalent to by means of machine studying.”

Capturing for the celebrities 

Concr makes use of established statistical studying and modelling strategies utilized in astrophysics to deal with knowledge sparsity in pre-clinical knowledge units.

“We will combine knowledge from the early phases of preclinical analysis, all the way in which to scientific outcomes, and combine and share learnings from these knowledge in a single, unified mannequin,” says Irina Babina, Concr’s chief enterprise officer and a most cancers biologist by background. 

“The mannequin can then be used to foretell remedy response in sufferers — pan-cancer and pan-treatment,” she continues.

What oncologists and astrophysicists have in frequent is a reliance on giant quantities of sparse knowledge moderately than battling with excessive quantity, or “large” knowledge. 

“Most cancers researchers and astrophysics researchers are comparable in working with huge quantities of disparate and fragmented knowledge, and desirous to study from and share info from these knowledge units; and mannequin one thing complicated that you may’t instantly observe,” says Dr Babina. “Within the case of astrophysics that’s darkish matter, in most cancers biology, it’s affected person outcomes.” 

Strike early

 Debiopharm Innovation Fund’s seed funding in Concr is a part of a strategic evolution to develop from Collection A investments, of round €2-5m, to incorporate smaller pre-Collection A offers.

“The concept is to place in smaller cheques, as a foot within the door. We aren’t anticipating to be concerned on boards of administrators, we take a really small minority place and we watch the corporate whereas we’re serving to them develop,” says Gibbs.

He provides that Debiopharm was particularly intrigued by Concr’s thought of taking the maths from astrophysics and making use of it to most cancers biology: “If you concentrate on the historical past of biology, it’s not with out precedent. A number of physicists got here on the start of molecular biology and introduced a tough science method to what was earlier than a bit like stamp amassing.”

Concr is Debiopharm’s fourth seed funding, with curiosity sparked at an opportunity assembly at a biotech occasion. The fund is anticipating to make 10-15 seed investments yearly, with a fifth already accepted, and to grow to be main traders on the Collection A stage the place they just like the course of the corporate. 

“These earlier stage corporations enable us to see additional into the way forward for life science applied sciences,” says Gibbs. 

Due to the help from Debiopharm alongside present and new traders, together with R42 Group, Oncology Ventures, Cambridge Enterprise and SyndicateRoom, Concr has superior from proof-of-concept to technical, organic and scientific validation of their predictive know-how throughout numerous datasets. Moreover, the corporate has cast a number of partnerships, together with with Roche and Step Pharma.

Pharmaceutical curiosity in Concr’s know-how is fuelled by the price and excessive failure charge drug growth — higher high quality knowledge may also help corporations make extra knowledgeable choices about which therapeutic avenues are most promising, based mostly on historic knowledge and insights from previous trials. 

The startup benefit

The prevalence of most cancers means Huge Pharma will possible proceed to speculate closely in quest of a remedy, however the analysis and growth course of is so pricey, inefficient and failure-prone, there’s a want for contemporary pondering and experimentation that startups supply.

“We’re removed from routine use of precision medication in oncology and though we and all different corporations are claiming precision medication, there’s nonetheless work to do,” says Cedric Odje, funding affiliate at Debiopharm Innovation Fund.

As AI fashions grow to be extra accessible, the know-how life cycles are getting shorter, Odje provides. “One of many causes we launched our seed funding exercise is to remain nearer to, and elevate up, the newest innovation, rising the probabilities to seize essentially the most impactful innovation on the market.

“Yearly, each month, there’s a new firm with a brand new method and applied sciences,” he continues. “By way of scouting and screening, there are entrepreneurs taking over the difficulty, though there are large gamers on this house, and that’s very thrilling.”