Skip to main content

There may be four distinct types of bowel cancer

Wednesday 14 October 2015

"Bowel cancer is four distinct diseases, each with a different prognosis," BBC News reports after new research suggests there are four genetic sub-types of bowel cancer. It's hoped adapting treatment to each type will lead to better outcomes.

Most bowel (colorectal) cancers are known as adenocarcinomas. The name comes from gland cells in the lining of the bowel where the cancer first develops.

The researchers pooled the genetic details of these types of cancer cells from 4,151 people with colorectal cancer to reassess previously published classifications of the disease.

They coupled this with clinical information about how the cancers behaved and came up with four new categories:

  • consensus molecular subtype (CMS) 1 – this accounts for 14% of cases
  • CMS2 – 37% of cases
  • CMS3 – 13% of cases
  • CMS4 – 23% of cases

The remaining 13% of cases did not fit into any of the categories – it could be the case they represented one category transforming into another.

Researchers say this new classification could help personalise treatments to achieve better outcomes. For example, CMS4 is thought to be particularly aggressive and may benefit from extensive treatment.

Further research is now required to validate the results and work out which treatments are most effective for each type.

Effective ways you can reduce your risk of bowel cancer include eating more fibre from cereals, beans, fruit and vegetables, limiting how much red meat you eat to no more than 70g a day, taking regular exercise, quitting smoking, and moderating your consumption of alcohol.  

Where did the story come from?

The study was carried out by an international consortium of researchers involving scientists from the UK, the Netherlands, Spain, the US and Hong Kong.

It was funded by various different organisations, including the National Cancer Institute, La Caixa International Program for Cancer Research & Education, the Dutch Cancer Society, Worldwide Cancer Research, the European Research Council, and the US National Institutes of Health.

The study was published in the peer-reviewed medical journal Nature Medicine.

It was widely reported by the UK media both accurately and responsibly. However, some of the study's limitations were not pointed out.

The BBC quoted one of the authors of the study, Dr Anguraj Sadanandam, who said the next step is to match currently available treatments with the newfound colorectal cancer types so they can begin to personalise treatment.  

What kind of research was this?

This was mainly a laboratory study where researchers from around the world came together to address the issue of different gene expression (information provided by specific genes) being reported in current classifications of colorectal cancer types.

Researchers say the differences in previously reported colorectal cancer types are the result of different ways of processing data and different methods used for different population groups. They aimed to solve this problem by forming an international consortium dedicated to large-scale data sharing and analytics.

Laboratory-based studies are good at understanding the structure of individual cancer cells. However, as they only assess single cells in a controlled environment, the results may differ from inside a living organism, where the cancer cells are exposed to different agents.  

What did the research involve?

This research involved a large gene expression data set collected by six groups of researchers who had previously developed and published a methodology for the classification of colorectal cancer using gene expression data. An additional evaluation group was set up who were responsible for unbiased analysis and reporting.

Researchers focused on secondary analysis of the already existing gene and clinical data. Since the data sets were collected using different methods in different labs at different points in time, they developed a methodology that was suitable for the aggregation of these data sets.

The researchers say all the patients had previously given informed consent for the use of their data in future colorectal cancer research.  

What were the basic results?

A total of 18 colorectal cancer data sets from 4,151 patients were used in the analysis. Based on gene expression of the cancer cells, researchers grouped colorectal cancer into four main types: CMS1, CMS2, CMS3 and CMS4.

Some of the main findings are listed below:

  • CMS1 tumours were more commonly diagnosed in females with right-sided colon lesions and had undergone a lot of mutations. These tumours caused a very poor survival rate after relapse.
  • CMS2 tumours displayed higher chromosomal instability and were mainly in the left side of the colon.
  • CMS3 tumours had lower levels of mutations.
  • CMS4 tumours were usually diagnosed at more advanced stages (stage 3 or 4) and had the worst overall survival rate.  

How did the researchers interpret the results?

The researchers concluded by saying: "We believe that the framework presented here provides a common foundation for colorectal cancer subtyping, and is to be further refined in the future as other sources of 'omics' data are integrated and clinical outcomes under specific drug interventions become available."

They added: "We hope that this model of expert collaboration and data sharing among independent groups with strong clinical and preclinical expertise will be emulated by other disease areas to accelerate our understanding of tumour biology." 


Based on genes, the findings of this study allowed researchers to broadly classify colorectal cancer into four main types: CMS1, CMS2, CMS3 and CMS4.

To address the issue of inconsistent reporting of colorectal cancer types, researchers from different parts of the world formed an international consortium dedicated to large-scale data sharing and analysis.

They pooled pre-existing data on gene expression from colorectal cancer patients to assess the previously published colorectal cancer types.

While this study extends our understanding of the different types of colorectal cancer and in the future may lead to more targeted treatments for better outcomes, the results must be interpreted with caution.

The study does have strengths in its large sample size. However, although researchers have used sophisticated methods to aggregate this large data set, the samples were collected using different methods in different labs and at different points in time.

This means there may be some factors that were not taken into account when aggregating the different data sets, which might reduce the quality and reliability of the data.

Also, the clinical implementation of these results may take a long time as further research is required to fully understand these types of colorectal cancer.

Having a healthy diet that includes plenty of fibre-packed vegetables and fruits and exercising regularly is very important for reducing your risk of developing colorectal cancer.

Identifying bowel cancer early improves outcomes, so if you have experienced any of the symptoms of bowel cancer, it is best to see your GP.

Analysis by Bazian
Edited by NHS Website

Links to the headlines

Bowel cancer is four distinct diseases

BBC News, 13 October 2015

Bowel cancer breakthrough as scientists find there are FOUR distinct types of the disease - paving the way for better treatment

Mail Online, 12 October 2015

Links to the science

Guinney J, Dienstmann R, Wang X, et al.

The consensus molecular subtypes of colorectal cancer

Nature Medicine. Published online October 12 2015