One test to detect multiple cancers – where are we now?

By Jessica Lloyd, PhD , Alice Brookes

Most cancers diagnosed at an earlier stage have a better prognosis. Detecting and diagnosing cancer earlier can mean there are more treatment options for patients and, ultimately, can help people survive cancer and lead longer, better lives.  

Developments in technologies looking for markers of cancer in blood, urine, saliva or stool – often termed ‘liquid biopsy’ technologies – provide an exciting paradigm shift in early detection, diagnosis and treatment, especially as they have the potential to be used to detect multiple cancer types from a single sample. Different terms are used to refer to tests developed on the basis of these technologies. We use the term multi-cancer tests (MCT) to incorporate all potential uses of liquid biopsies in the multiple-cancer context, including diagnosing cancer and use cases that could help inform therapy selection and disease prognosis. You may see other terms like multi-cancer detection test (MCD) and multi-cancer early detection tests, but we feel MCT is the most accurate, all-encompassing terminology. They aren’t the only technology that could help us detect and treat more cancers earlier, but with multiple trials taking place across the world, MCTs are attracting a lot of interest.   

This article, the first in our MCT explainer series, will discuss the technology behind MCTs, explain how MCTs could be used for asymptomatic screening and explore the potential opportunities, substantial challenges, and uncertainties they bring. Future articles will consider some of the most important issues and innovations introduced here in more detail. 

What are multi-cancer tests?

As described above, an MCT is a tool that searches for multiple cancers in one sample, typically blood, urine, breath, or stool. All MCTs share some similarities, but the ways they identify cancers can be slightly different.  

For example, one clue that a person might have cancer is the presence of cancer DNA in their blood. So, many MCTs – like the Galleri, Shield and Delfi cancer blood tests – look for small fractions of DNA that have been released from cancer cells. Other MCTs look for the cancer cells themselves, or other metabolites or proteins they might release into the body. Typically, MCTs also use an algorithm to make sense of the data they detect. Often, the algorithm uses the signs found in the sample to predict the risk that someone has cancer and indicate the part of the body where the cancer can be found.  

Multi-cancer tests and multi-cancer earlier detection tests: what’s the difference?

There are several different ways MCTs could be used in our health system, all of which are still several years away from being ready. As well as helping detect cancers earlier, MCTs could guide GPs to make decisions about investigating symptoms, help oncologists choose the most suitable treatment for a cancer that has already been diagnosed, or detect the return of a cancer sooner. Some of these uses could have huge benefits for patient outcomes and some could help relieve some of the burden on the health system – or both.  

in this article, we’re focusing on using MCTs to screen people who seem to be healthy, or asymptomatic, and detect cancers earlier as part of population screening programmes. To be as clear as possible, and to differentiate from other uses of MCTs, we’ve decided to refer to MCTs used in this way as multi-cancer earlier detection tests (MCEDs).  

MCEDs: MCTs in screening programmes

Currently, cancer screening programmes are available for one type of cancer at a time and utilise tests specific for that cancer. For example, mammogram images can reveal signs of breast cancer and FIT tests can detect blood in stool samples, which could point to bowel cancer. These programmes aim to reduce cancer deaths in two ways: by detecting cancer precursors that doctors can stop before they become dangerous, and by detecting clinically significant cancers at earlier stages, when treatment is more likely to be successful. 

MCEDs could completely change what cancer screening looks like by making it possible to screen for multiple cancers with one test. MCEDs could also allow us to screen for cancers that aren’t covered by individual screening programmes, including less common cancers. That means MCEDs could be a more efficient way to find more types of cancer earlier, when they are more curable, helping people live longer, better lives. 

But delivering a population screening programme is complicated and requires consideration of a variety of factors, some of which relate to the way the test used for screening works. Ideally, any test used as part of a screening programme needs to pick up signs of cancer in those who have it – without missing cases or telling healthy people that they have cancer by mistake.

No test is perfect, which means they may incorrectly classify a proportion of people either as having cancer (false positive) or not having cancer (false negative). In addition, it is important to ensure that tests do not pick up cancer that would not go on to cause harm (termed overdiagnosis). Efforts are made as part of screening programmes to minimise these harms and maximise the benefits. Screening trials help us get some of this information, but because they are trying to understand risks, as well as benefits, these trials are large, long, and complicated. 

Testing multi-cancer earlier detection tests

There’s been a lot of research since we last looked at the potential of MCEDs and cancer blood tests in 2022. Many companies are developing and optimising new MCEDs, but we still need more evidence to know whether any of them are suitable for a national screening programme. Ideally, MCEDs should show good diagnostic accuracy for all early-stage, clinically significant, cancers, as well as a low number of false negatives and minimal false positives. 

But finding signs of cancer in blood (or other) samples is like looking for a needle in a haystack. This is because blood also contains material released from healthy cells, with only a minute amount from cancer cells. What’s more, early-stage cancers are often small and release a lower volume of markers into the blood, meaning they’re even harder to find. 

To date, MCEDs have mainly been trialled in people who have already been diagnosed with cancer. That helps us understand how well these technologies can pick out signs of cancer, because we can compare test results to what we already know. Generally, these trials have shown that MCEDs are able to identify people who do not have cancer (i.e., have good specificity), which is important to reduce anxiety and unnecessary testing in healthy people. The data also shows that while these tests can identify cancer, they struggle to spot certain types of cancer, including early-stage ones (i.e., they have poor sensitivity).  

We also need to investigate how MCEDs work for people who are apparently healthy, with no indication of cancer, as they’re the people who would be offered a test as part of a screening programme. One example is the NHS-Galleri trial, which began in 2021 and will finish in 2026. It’s the largest trial of this kind, with 140,000 participants across England. 

The NHS-Galleri trial is the first step towards understanding how multi-cancer screening programmes could fit into the UK health system and how good this particular test is for this purpose. It will provide evidence and give us insight into what a multi-cancer screening programme might look like, how it could impact other cancer screening programmes and diagnostic pathways, and how it could affect tests, workforce, and equipment use. A recent update from NHSE outlined that, based on the results from the first year of the trial, they were going to wait to see more data before deciding on whether or not to roll the test out more widely.

Most importantly, before the UK launches any new national multi-cancer screening programme, trials will need to show that such a programme, and the use of MCEDs, can actually help stop people dying of cancer. The gold-standard way of assessing this usually takes 10 years or more, as researchers need to follow people for a long time to understand whether diagnosing someone through MCED screening helped them live longer. It is also important to ensure that the use of these tests doesn’t lead to overdiagnosis, unnecessary treatment and worry for people who are otherwise well. 

This is the main reason that tests are still many years from being ready for use. To expedite evidence generation for new screening tests, efforts are ongoing to see if there are different types of data, or surrogate endpoints, we can collect to more quickly assess whether a test is suitable for a screening programme. These developments might help us advance innovative new technologies in much shorter timeframes while ensuring that their use improves patient outcomes. 

What’s next?

There is a lot of exciting MCED research going on around the world, so it’s important that we remain open to the different tests in development. But we need to balance that with the fact that we’re still years away from using any of these tests to screen people for cancer.  

While we gather and monitor the evidence, it’s important we set up robust frameworks to decide which tests should be introduced, and when. That’s what we’ll be thinking about next. Keep an eye out for part two of our MCT series, where we’ll look closer at the range of potential benefits, risks, and unintended consequences of using MCEDs, and consider how we need to evaluate them.