Cancer is the single-most feared medical diagnosis imaginable, even though it’s really an umbrella term for hundreds of different destructive illnesses. Uncontrolled growth of cells disrupts the body’s core functions, and the way these cells can be altered not only differs between locations in the body, but also from person to person.
“We’re going to be able to help people to live with cancer, as people are living with HIV today”
“Cancer is a living monster and constantly evolving,” says Jurgi Camblong, CEO of Sophia Genetics – a firm that hopes one day to ensure a cancer diagnosis becomes more benign than it is today. “There is a very good hope that cancer will somehow be mastered, a bit like HIV – we’re going to be able to help people to live with cancer, as people are living with HIV today,” he explains.
While few would have confidently predicted the advances in HIV care we’ve made in the last few decades, cancer is a different beast, and Sophia Genetics aren’t the only company trying to tackle the problem. Because of the variables involved, and the constant evolution of the illness, there’s no single right way to treat it, and even the world’s most prominent oncologists ultimately have a limited experience of the many aggressive forms it can take and the right cocktail of drugs to utilise. That’s where big data comes in: by connecting 160 hospitals around Europe, Sophia Genetics has been able to examine 45,000 patients’ genetics, providing the quantity of case studies that most doctors could only dream of. In the coming year they hope that DNA sequencing will help 80,000 patients.
A revolution in data driven medicine
“There are many different types of cancers, because you can have many different types of rearrangements in your genome. What’s really important is being exposed to the maximum number of cases – the more cases your algorithm is exposed to, the more they will learn and the more it will be able to solve,” says Camblong.
“It really does feel like we’re at the beginning of a revolution in data driven medicine”
“If you understand the cards the tumour has in its hand, you not only know what it does now, but what it’ll do next,” adds Doctor Mike Lynch, from Invoke Capital, who invested in the company back in 2014.
It’s not exclusively cancer related – the company also deals in cardiology, pediatrics and other hereditary diseases, but cancer’s uniquely frustrating traits and elusive nature make it a particularly appealing target. While our progress in tackling cancer has undoubtedly been slow, we’ve never had this kind of technology so freely available before. DNA sequencing used to be prohibitively expensive, but now prices are if not cheap, then at least accessible. And given the cost of giving the wrong medicines – both in lives and in pounds and pence – increasingly, it feels like a price worth paying. “It really does feel like we’re at the beginning of a revolution in data driven medicine”, comments Camblong.
This is helped by the government’s move towards a paperless NHS, which will not just be useful for Sophia Genetics, but other companies too – something Nuance were keen to highlight when I spoke to them last month. The hope is that Sophia Genetics and similar may prove to be a driving force in data uptake in healthcare.
Use of available data, Lynch explains, is not something the health sector has always excelled at. “Sadly, medicine has been a little slower than some might hope to use all the data we have available, but there’s obviously complexities around security and privacy.”
Privacy matters
Which brings me to the two elephants in the room: privacy and security. These elephants are particularly prominent coming just off the back of the reveal that Google Deepmind has 16 million NHS patient records. Still, these concerns have been preempted by Sophia Genetics from the very beginning. The company has its own servers to ensure data privacy, and has ISO 27001:2012 certification. “Data privacy is a must,” Camblong argues. “It is a must so that you can build trust and motivate people sharing knowledge.” Without trust, the amount of data shrinks and progress simply cannot be made.
“We still have got to overcome public perception about data privacy,” adds Lynch. “Some of the technical solutions to that don’t work. Anonymising is a lot more difficult than you might think as there’s so much information you can use to de-anonymise things. But I think as the public starts to see the benefits from these things, the debate will become a bit more nuanced.”
Assuming public unease can be overcome, it may look to an outsider that Sophia Genetics is moving towards a position where the human doctor is a thing of the past. That’s not happening any time soon, according to Camblong. “We don’t replace the human, because ultimately a human is taking the decision about a patient. Some companies talk about the future of healthcare and machines that will be so powerful that doctors eventually won’t be needed – we don’t believe that’s going to happen. The machine is the tool of diagnostics, but you still need the human.”
“The machine is the tool of diagnostics, but you still need the human”
Lynch agrees with this. “It’s not about replacing the human, it’s about giving the human the right information to start with,” he explains. “We’re talking about a snapshot – the clinician will have had longitudinal information of the patient for the last ten years, but for the moment the infrastructure of bringing that kind of information in just doesn’t exist.”
If we can collect even more patient data and overcome public squeamishness about the two P words (that’s “privacy” and “personal data”), then this kind of development should be genuinely transformational. Early signs suggest both of these barriers are eminently surmountable, and it genuinely gives me hope that the way we view cancer could be transformed within my lifetime – something I’d almost given up hope on.
READ NEXT: How connected devices could revolutionise healthcare
Image: Shaury Nash used under Creative Commons
Disclaimer: Some pages on this site may include an affiliate link. This does not effect our editorial in any way.
