The microbiome: a biomarker for colorectal cancer?

Colorectal cancer is the third most common cancer in the world. With more than 1.8 million people diagnosed each year and cases on the rise, simple but effective screening methods could hold the key to both reducing incidence and ensuring more people are diagnosed as early as possible, when treatment is more likely to be successful.

In two recent papers, the Cancer Grand Challenges OPTIMISTICC team begins to address this challenge, identifying a microbiome signature – a unique pattern of bacterial taxa – that appears consistent in a pilot study of stool samples from 4 nations. The signature can be reliably detected in stool collected using a cheap, robust card-based method (guaiac faecal occult blood test; gFOBT) and analysed using 16S amplicon-based sequencing.

Collectively, the findings not only hint at a global microbiome biomarker for colorectal cancer, but they could also translate to a simple, effective and scalable screening programme.

“There are two arms to these findings – one technical and one biological,” explains bioinformatician Curtis Huttenhower, co-investigator on the OPTIMISTICC team based at the Harvard TH Chan School of Public Health, Boston, US.

“From a technical perspective, this very broad, gFOBT card-based screening approach lets us detect microbial signals that are similar to much more detailed, deep shotgun metagenomic studies of people with late-stage colorectal cancer. I’m actually kind of amazed that these signals align,” Curtis laughs. He’s quick to point out the approach’s limitations. “While the signals are similar, they’re not as rich – you're not measuring the whole microbial community, or the individual strains, or all the different chemical and functional activities that you would with a higher quality sample type.

“But it’s a good first pass screen to test whether someone is at an elevated risk of colorectal cancer.”

Biologically, the findings provide evidence of a shift in the microbiome early in the development of colorectal cancer. This shift is independent of the microbiome’s starting point and seen in samples collected from a range of countries: the UK and US (countries with high incidence of colorectal cancer), Argentina and Chile (intermediate incidence) and Southern India and Vietnam (low incidence but on the rise).

“What’s interesting is that ‘normal’ microbiomes vary quite dramatically between these countries,” explains OPTIMISTICC co-investigator Phil Quirke. Phil is a pathologist based at the University of Leeds School of Medicine, UK. “But a consistent story is emerging, highlighting a group of bacteria that are associated with colorectal cancer.

“Understanding why they’re there is important – we don’t know yet if there’s a causative link between the bacteria and colorectal cancer. But just being associated with it is good enough when we’re looking to improve screening and trying to identify people at risk,” Phil explains.

To Anita Mitchell-Isler, patient advocate on the OPTIMISTICC team, the potential for improved screening is very important. “Screening for colorectal cancer is crucial for saving lives. For many people, it’s a silent disease, without clear symptoms until a later stage when the cancer is less treatable,” she says. “But if caught in the early stages, colorectal cancer is often beatable.”

It’s interesting discussing the findings with Phil and Curtis, who approach them from different perspectives. In terms of discipline, the quantitative and clinical sides of the team “fit together very well,” Curtis describes. “I tend to take a very basic, quantitative approach, using computational techniques to answer questions about microbial ecology. Phil and the team bring a more translational perspective – I wouldn’t necessarily have considered the route to screening without their input.”

The team members also find themselves operating in contrasting national healthcare contexts with different screening policies – again influencing their view of the findings. In the UK, the population-based National Health Service Bowel Cancer Screening programme looks for traces of blood in the stool to decide who should be referred for colonoscopy; in the US, while stool tests are available, screening through the private system relies heavily on colonoscopies.

As Phil explains, “microbiome analysis appears more accurate than looking for blood. The approach could provide a more effective screening method, reducing the number of unnecessary colonoscopies while ensuring those who need one are referred. That’s less of a concern in the US, but colonoscopy is one of the biggest bottlenecks in our screening programme in the UK.”

But what about countries where colorectal cancer incidence is increasing, but screening programmes are limited?

“Freezing stool samples takes huge amounts of storage space and shipping on dry ice is really expensive,” explains Caroline Young, first author of both studies and histopathologist based in Leeds, UK. “For these studies, we kept everything at room temperature – from sample collection in the pilot nations (Vietnam, Chile, Argentina and India) to shipment to and storage in the UK – and the microbiome on the card remains stable enough for screening.”

That’s important: “What the microbiome analysis of gFOBT cards begins to offer is a resource-light protocol, even in very hot countries with limited infrastructure,” Caroline elaborates. “It’ll be interesting to see if the microbiome remains similarly stable across additional populations.” The next step is to develop a simple PCR test, like that being used to test for COVID-19, that could be performed quickly and potentially translated to a wider screening programme.

Early in the programme, Caroline and colleagues visited Curtis’s team in Boston to learn and share techniques – something Caroline believes was an important part of the process. “We entered the visit with an outline of our ideas and left with all these new techniques and a strong collaborative relationship across the team,” she describes. “It really elevated our work. The underlying data was the same but working with our bioinformatician colleagues really strengthened the studies’ conclusions.”

Looking forward, the team has clear plans to scale up the project. The microbial biomarker pilot will be expanded – is detection of the signature reproducible across wider populations? The UK screening programme recently moved from gFOBT to the faecal immunochemical test (FIT), a similar test that has higher levels of uptake. Are the findings consistent in this different collection method? And could probing the microbiome signature reveal more information about the causes and early development of colorectal cancer?

“The question now is how we tie it all together, and tie it back in with the rest of the OPTIMISTICC programme,” Phil muses. “It’s a fascinating story that may lead us to the causes and prevention of colorectal cancer.”

OPTIMISTICC is generously supported by Nick and Annette Razey.

Find out more about the OPTIMISTICC team. Read the papers in Genome Medicine and Clinical Cancer Research.

Written by Emily Farthing. Image: Cross-section of an ulcerated cancer growing through the wall of the large bowel and invading a large vein. Credit: Leeds University OPTIMISTICC team