Strengthening the foundations for early detection

Spotting and treating cancer in its early stages remains the most powerful way to give patients the best possible chance of surviving the disease. According to England’s Independent Cancer Taskforce, diagnosing more cases early could save up to 11,000 lives a year in England alone.

On top of this, the treatments for cancers diagnosed at an early stage can have less of an impact on patients’ quality of life, and improve their experience of care.

Making sure more patients are diagnosed early is vital to making progress against cancer – it’s a theme we set out in in our strategy to beat cancer sooner.

There are various ways that better information, awareness and healthcare could improve early diagnosis, but what role will scientific research into cancer biology play?

To find ways to spot cancers in their earliest stages, we need a renewed effort to understand and track the early stages of cancer biology – a relatively new and diverse field of research called ‘early detection’. Finding ways to detect cancers early could contribute to diagnosing more cancers early, and hopefully lead to more people surviving the disease.

To try to jump-start progress in this field, and to build a bigger international community to focus on this problem, last year we started working with experts the Knight Cancer Institute at Oregon Health & Science University in the US.

And in June, at the Knight Institute in Portland, we helped run the very first in a series of joint international conferences, bringing together many of the world’s leading scientists in the field of early detection, (including many of our own leading experts). Here are a few highlights of what was discussed.

Understanding early tumour biology

It’s relatively easy to study the early biology of some cancers – for example bowel cancer. These tumours develop from benign growths called polyps, and while not all polyps will develop into cancer, bowel polyps are fairly common in older people. If doctors spot them during an endoscopy, they will surgically remove them – and so researchers can study these precancerous cells and work out what changes happen for cancer to develop.

But other types of cancer – such as lung and pancreatic tumours – are usually diagnosed at a late stage, so scientists aren’t frequently able to analyse samples of early tumours to find out what makes them tick. So other approaches are needed – and several talks at the conference focused on innovative ways to carry out research in mice to trace cancer back in time to its roots. In doing so, researchers are now beginning to identify the cells where the disease begins, and the genetic faults that trigger it. Most spectacularly, research presented by Stanford University’s Professor Mark Krasnow is homing in on the very first cell division that marks the beginning of a lung tumour.

As well as such cellular sleuthing, other talks focused on understanding the early molecular changes that help cloak cancers from the immune system, or help them spread, and how these could also provide crucial insights into cancer’s earliest signs.

Looking for cancer’s ‘fingerprints’

Another key theme was around work on so-called ‘liquid’ biopsies – tests to look in blood samples for things like cancer cells that have broken free from tumours, DNA released from cancer cells, or mysterious little packages released by cells called exosomes.

There’s already lots of focus on how liquid biopsies could improve treatment for patients after they’ve been diagnosed, including looking for warning signs that cancer is relapsing after treatment. But do liquid biopsies have more potential?

Attention is now moving towards whether liquid biopsies can pick up early warning signs of cancer in people who feel fit and well and don’t have any symptoms.

Scientists, including world-leader Professor Dennis Lo from the Chinese University of Hong Kong, talked about their research into potential ‘red flags’ that could indicate something is amiss, including faults in DNA, signs of viruses linked to cancer (such as the Epstein Barr virus (EBV)), and how to spot rare tumour cells in the bloodstream.

A blood test for cancer is an exciting prospect, but are there other bodily samples worth searching in? Some researchers think so, and discussed evidence that cancer can leave its ‘fingerprints’ in poo, sweat, urine, and exhaled breath too.

Interesting (and sometimes heated) talking points included whether tests could ever be sensitive enough to spot very early cancers or precancerous cells, and whether a single test could ever detect many different types of cancer.

Where there’s smoke…

While much of the focus is on cancer cells, some scientists are taking a different approach and looking for the smoke instead of the fire.

There were exciting ideas about monitoring the immune system as an early warning sign of cancer. Professor Lisa Coussens, from the Knight Cancer Institute, described her work studying how immune cells react to tumours, in the hope of uncovering tell-tale signs of early disease.

Some cancers can be linked to things we do and aspects of our environment, for example smoking and lung cancer. An unsolved problem in lung screening in the US is that not all suspicious signs on a chest x-ray turn out to be cancer. So Professor Avrum Spira, based at Boston University, is looking at the collateral DNA damage inflicted upon the cells in smokers’ airways, and how this can help predict how likely a suspicious shadow on a scan is to be cancer. Crucially, said Spira, this could save people needing unnecessary surgery.

Pioneering new technology

But studying early tumour biology is only one half of the equation. There also needs to be more investment into new technology to make early detection a reality.

Professor Sam Gambhir, from Stanford University, illustrated the scale of the problem with an analogy – the difficulty of spotting early signs that an aeroplane’s engine is malfunctioning. Monitoring a jet engine requires several hundred sensors, he said, taking readings every 30 seconds and generates 10 terabytes of data (that’s about the same as 1400 CD-ROMs) every 30 minutes during flight.

With so much information coming in, the biggest challenge that engineers face is finding which readings show a real problem, and which are red herrings.

And we face the same challenge in developing technology to detect early signs of disease.

Ghambir talked about the emergence of ‘smart’ technology that could monitor us all the time – from contact lenses and intelligent toilets to imaging sensors in our clothing. And many researchers believe that such sophisticated devices could one day collect enough real-time information to sound an alarm about our health.

Of course there are big challenges facing this type of technology too. It has to be cheap, robust, non-invasive, accurate and sensitive – teasing apart the data to find that crucial bit of information among the reams of information pouring in is no easy task. And it has to be a level of monitoring that people want, and not cause too much inconvenience or unnecessary worry.

But there’s encouraging progress. For example, Professor Rebecca Fitzgerald, from the University of Cambridge, gave an update on the development of a device called the Cytosponge – a small sponge on a string that collects cells from the oesophagus of people with persistent acid reflux. This can be analysed to see if they have a disease called Barrett’s Oesophagus, which increases the risk of oesophageal cancer.

But only a small number of people with Barret’s actually develop cancer. So Fitzgerald’s team is now looking at the genetic changes in these cells, to find out if people with Barrett’s can be split into groups at low and high risk of developing oesophageal cancer, to whittle down the number of people who needs further tests, such as endoscopy.

Lessons from screening programmes

While national screening programmes have saved thousands of lives, they’ve not come without downsides. Some people have been diagnosed with cancers that would never have caused harm (a phenomenon called overdiagnosis). And this has led to some people being given unnecessary treatments.

To oversee our renewed focus on early detection, we recently appointed a Director of Population and Early Detection Research – Dr Zahid Latif. And he thinks that the true challenge of early detection is finding the most reliable ‘red flags’ that need further investigation in a health care setting.

“One of the greatest challenges is not only finding early warning signs that cancer has developed, but identifying which of those alarms need following up with more tests and, potentially, treatment – and those which we can safely ignore,” he says.

And that the ultimate goal comes in “combining biology with technology and coming up with an inexpensive, non-invasive and convenient test that can be used to monitor healthy people for warning signs that something is not right and needs further investigation”.

Dr Sadik Esener, Director of the Center for Early Detection at OHSU adds, “to make screening affordable, we also need to do better at identifying people who would benefit the most, because they’re at high risk of developing certain types of cancer.”

According to Esener, any new technologies will have to be “reliable, accurate, and not create too many false alarms, while at the same time not missing cases”. And this challenge is even greater given this would be a test for healthy people with no symptoms.

The journey into early detection research is still in its beginnings – understanding what early detection means, bringing together experts from cancer biology, cancer care, and detection technology, and turning the idea into a clinical reality. But if we’re going to achieve a shift towards diagnosing cancer early and reducing late stage disease – potentially saving thousands of lives – it’s a vital path to set out on.

Emma

• Read more about the conference on the OHSU blog