Preventing cancer

How Oxford's experts are investigating the causes of cancer and how changes to our lifestyles can help prevent disease

There are over 200 types of cancer, each with differing causes, means of detection, methods of treatment and, ultimately, chances of patient survival. 

Understanding what causes cancer is essential to treating existing cases and preventing avoidable ones in the future. 

Half of us in the UK will get cancer in our lifetime, but around four in ten UK cancer cases could be prevented.  

Oxford’s researchers – together with colleagues from across the globe – are uncovering the causes of cancer, identifying how changes to our lifestyle can reduce cancer risk, and discovering new precision drugs and novel therapies for treatment. 

Genetic and bacterial causes of cancer

Cancer is an uncontrolled proliferation of cells due to faults in DNA. 

DNA mutations may be acquired over an individual’s lifetime, often due to cell damage from age or carcinogens such as sunlight, smoking and some infections. Some variation is inherited, occurring in every cell in the body from birth. 

Carcinogen: 

A physical, chemical or biological agent that may induce or otherwise participate in the causation of cancer 

Many research studies investigate the separate effects of different types of genetic variation, which can generate an incomplete picture. 

Funded by the Cancer Research UK Oxford Centre Development Fund, a team of researchers from Ludwig Cancer Research at the Nuffield Department of Medicine are investigating how these genetic mutations cooperate

Ludwig Oxford researchers investigated the interplay between the variants affecting the activity of the p53 tumour suppressor protein. Mutations in p53’s gene coding are found in over half of all human cancers. 

The team found evidence that the different types of mutations – inherited or acquired – can cooperate, altering the risk of cancer, its progression and its response to therapy. This discovery can be used to identify more effective novel therapies and has the potential to guide the development of future anti-cancer drug targets and combination therapies for precision medicine.  

Genetic inheritance and breast cancer 

One example of an inherited variation is the BRCA1 gene, associated with increased risk of breast cancer. An estimated 70% of women who have faults in a BRCA gene will develop breast cancer by age 80, and mutations in the gene are one of the most common associated with cancer.

Healthy cells and tissues are normally reliant on BRCA1 to coordinate the repair of DNA breaks. If breaks aren’t repaired, which can happen when the BRCA1 gene mutates, harmful mutations will be passed down the line. This may trigger the formation of cancerous tumours, and these tumours may develop resistance to anti-cancer therapies. 

Research from the lab of Dr Ross Chapman has uncovered the precise mechanism behind how the BRCA1 protein detects and engages with DNA breaks in the genome. This discovery will help prevent the development of breast and ovarian cancers.  

Funded by the Cancer Research UK Career Development Fellowship and the CRUK Oxford Centre, the research identified that the activity of BARD1, a lesser-known partner protein of BRCA1, is responsible for the assembly of the proteins needed for accurate DNA repair. 

DNA repair: 

A variety of mechanisms that help to ensure that the genetic sequence, as expressed in the DNA, is maintained and that errors that occur during DNA replication, by mutation, are not allowed to accumulate 

The research revealed that this tumour suppressor protein evolved to counteract the threat that specialised, yet error-prone, DNA repair pathways pose to dividing cells in healthy tissues. 

Mapping tumours

Based at the Weatherall Institute of Molecular Medicine, the Chapman Group are investigating which mutations in cells give rise to which areas of cancer. They seek to better understand how faults in the regulation of recombination lead to cancer, and devise innovative strategies to exploit these faults 

This has big implications for diagnosis and also potentially for deciding which bits of cancer need treating 
Alastair Lamb, Nuffield Department of Surgical Sciences 

Knowing which cells give rise to which areas of cancer will improve our understanding of tumour growth and development. A technique called spatial transcriptomics allows scientists to see the changes that have taken place without breaking up the tissue that is being analysed.  

Researchers from Oxford – working with colleagues from the KTH Royal Institute of Technology, Science for Life Laboratory, and the Karolinska Institutet – have been able to use this technique to discover a previously unknown range of genetic variation in individual prostate tumours 

Current techniques for analysing cell genetics within tumours involve analysing the DNA of cells taken from a sample of the cancerous area. However, many cancers, such as prostate cancer, are three-dimensional, so a single sample would only give a small snapshot of a tumour. 

The researchers made some surprising findings

Funded by Cancer Research UK (CRUK), the study demonstrated that apparently healthy cells can have the same DNA damage as cancerous ones. Dr Henry Stennett from CRUK highlights that working out what stops them from being cancerous could allow us to help detect the disease earlier. 

The risk with predicting risk 

Genetic information on thousands of different locations in the genome can be combined to predict the likelihood of someone getting a particular disease. This is known as a polygenic risk score. 

Genome:

The complete set of genes in a cell or living thing

However, findings by Dr Lei Clifton and colleagues from Oxford Population Health have shown that there is substantial disagreement in individual risk predications between different scores for breast cancer using data from the UK Biobank

golden DNA helix against a dark background
A generated image of the BRCA1 breast cancer gene
Scan of a tumour
3D landscape of genome

Prostate cancer cells
Melanoma on white skin
Epithelial cells
H. pylori bacteria

Testosterone levels and the risk of cancer

In the UK, 1 in 8 men will be diagnosed with prostate cancer in their lifetime.

Researchers from Oxford Population Health’s Cancer Epidemiology Unit have conducted a large scale study confirming the role of two hormones in aggressive prostate cancer risk.

Previous studies by the Cancer Epidemiology Unit had shown a connection between increased risk of prostate cancer and high levels of two hormones: free testosterone and insulin-like growth factor I (IGF-1).

Using data from the Endogenous Hormones, Nutritional Biomarkers and Prostate Cancer Collaborative Group, the team found that  IGF-1 is positively associated with a greater risk of aggressive prostate cancer – and confirmed through further genetic analysis that genetically predicted levels of IGF-1 were associated with a greater risk of overall, aggressive and early-onset prostate cancer.

For the first time we show evidence that both IGF-1 and the free testosterone are important for aggressive, clinically relevant disease
Dr Eleanor Watts, formerly of the Cancer Epidemiology Unit

These results suggest that the reduction of blood levels of both IGF-1 and free testosterone, either through lifestyle or drug intervention, may be a strategy to decrease prostate cancer risk

Testosterone levels in men are also linked to a greater risk of melanoma, a potentially deadly form of skin cancer.

1 in 47 UK females, and 1 in 36 UK males, will receive a melanoma skin cancer diagnosis in their lifetime. Caused in many cases by UV exposure, 86% of melanoma cases are preventable, but other factors, such as age and genetics, can also play a role in who is most at risk.

Another study led by Dr Eleanor Watts found that testosterone is one of these risk factors.

Although we have seen associations of prostate, breast and endometrial cancer with testosterone before, this is the first time we have seen an association with risk of melanoma in men 
Dr Eleanor Watts, formerly of the Cancer Epidemiology Unit

Researchers studied blood samples collected by the UK Biobank from 182,600 men and 122,100 postmenopausal women aged 40 to 69. Looking at the total level of testosterone in the blood samples, as well as free testosterone (ie not bound to proteins), they then used NHS records and national registries to see whether participants went on to develop or die from cancer.

From this data, the researchers were able to confirm that, for men, higher levels of testosterone were associated with a greater risk of developing a malignant melanoma.

The study also highlighted that higher levels of testosterone are associated with a greater chance of endometrial and breast cancer in postmenopausal women.

By better understanding the role of growth factors underlying the epithelial structures in pre-cancerous conditions, we can detect when cancers appear and thus treat them earlier
Dr Francesco Boccellato, Leadership Fellow at Ludwig Cancer Research

In pre-cancerous conditions, the cell composition of tissues and the way that cells are organised is often different, or severely disrupted, when they progress to tumours.

If we can understand the molecular signals that cause this cell differentiation, and prompt cells to find their location within the tissue, we may be able to explain the changes observed in patients with pre-cancerous conditions.

Differentiation:

The process of a cell changing during growth and development

Leadership Fellow at Ludwig Cancer Research Dr Francesco Boccellato is investigating the mechanisms of tissue regulation in healthy stomachs, and those with the pre-cancerous condition atrophic gastritis. People with atrophic gastritis lack digestive enzymes and gastric acid-producing cells, resulting in digestive problems.

To help identify those who might have cancer, and to find new ways of preventing its progression, the team exposed healthy human gut tissue to a variety of growth factors, including EGF (epidermal growth factor), BMP (bone morphogenetic protein 4) and a protein called Noggin

The combination of these factors helps determine the cell differentiation that forms the stomach-lining gastric glands. These glands contain cells that produce digestive enzymes, gastric acid, and mucus for the protection of the stomach lining.

Growth factors:

Polypeptides (chains of amino acids) whose primary role is the maturing of cells

The team found elevated levels of EGF in the biopsies from the tissues affected by atrophic gastritis. The specific localisation of growth factors in the tissue could be responsible for the differentiation process. Changes in the growth factors’ relative quantities could trigger a change in cellular composition and the epithelium structure over time.

This could then lead to cancer. The next step is to build a high-resolution dynamic map of growth factors during the progression of cancer.

In September 2022, Dr Francesco Boccellato won a CRUK Early Detection and Diagnosis primer award to study how tissue shape changes in atrophic gastritis and another pre-cancerous stomach condition, intestinal metaplasia.

Understanding infection’s role in cancer

The investigation of how infection alters the growth factor microenvironment is also essential in understanding the potential development of cancer. For instance, long-term infection with Heliobacter pylori bacteria has been associated with an increased risk of stomach cancer.

Heliobacteria pylori (H. pylori):

A bacterium that infects the stomach and is a major factor in gastric (stomach) and probably duodenal (small intestine) ulcers

Stomach cancer is the second leading cause of cancer deaths globally, with around 770,000 deaths in 2020. Approximately 478,000 were from China.

The World Health Organization estimates that around 800,000 cases of stomach cancer were caused by H. pylori in 2018, though this is based on mainly on evidence from Western populations.

There had been previous studies linking H. pylori with cancer in the main area of the stomach, known as non-cardia gastric cancer (NCGC).

However, there was conflicting evidence as to whether it also causes cancer at the top of the stomach, called cardia gastric cancer (CGC). CGC accounts about one third of global stomach cancer cases.

To generate conclusive evidence, researchers from Oxford Population Health used data from the China Kadoorie Biobank to investigate the link between H. pylori and stomach cancer in China.

Participants were tracked through electronic death and cancer registries, and national health insurance databases.

Our study showed that H. pylori infection was associated with a six-fold higher risk of non-cardia gastric cancer and a three-fold higher risk of cardia gastric cancer
Dr Ling Yang, senior epidemiologist at Oxford Population Health

Participants who had a H. pylori bacterial infection were six times more likely to develop NCGC, and three times more likely to develop CGC, than uninfected participants. It was estimated that 78.5% of NCGC cases and 62.1% of the CGC cases in the study were caused by H. pylori.

Primarily funded by Cancer Research UK, the study estimated that H. pylori infection – spread through direct contact with saliva, vomit or stool, or by contaminated food or water – is responsible for 340,000 new cases of stomach cancer in China per year.

Diet and lifestyle

Around 40% of cancer cases in the UK, over 135,000, could be prevented by lifestyle changes. It’s important to know how we might change our behaviour to limit risks.

Investigating cancer risk in new technologies

It’s also important to assess new and emerging factors. The launch of 5G technology reignited fears that mobile phones and their networks might increase the risk of brain tumours.

Oxford Population Health and the International Agency for Research on Cancer demonstrated that there was no significant difference in brain tumours among mobile phone users and non-users.

776,000 participants were surveyed in 2001 and again in 2011, and were followed up for an average of 14 years through their NHS records.

Assessing how alcohol impacts cancer risk for different populations

One modifiable risk factor that there is much more evidence about is alcohol.

An estimated 3 million deaths per year are caused by alcohol, including over 400,000 from cancer.

Evidence from Europe already suggests that alcohol is a direct cause of cancer in the head, neck, oesophagus, liver, colon and breast. It has, however, been difficult to establish whether alcohol directly causes cancer, or whether related factors such as smoking and diet could generate biased results.

Alcohol consumption is rising in rapidly developing countries, such as China, so there is an urgent need to understand how alcohol affects disease risks in different populations. To do so, researchers from Oxford Population Health, Peking University and the Chinese Academy of Medical Sciences have investigated gene variants linked to lower alcohol consumption in Asian populations.

In Chinese and other East Asian populations, two genetic variants reduce alcohol tolerance. Both are common in East Asians populations, but rare in European ones. As these two variants are present at birth and are independent of other lifestyle factors, they allow researchers to assess how alcohol consumption affects the risk of disease.

Researchers used approximately 150,000 DNA samples from the China Kadoorie Biobank, tracking participants for a number of years.

They found that men with a variant had a 13–25% lower risk of both overall cancer and alcohol-related cancer. Men who carried two variants drank very little alcohol, and had a 14% lower risk of overall cancer, and a 31% lower risk of cancers previously linked to alcohol.

However, men who drank regularly despite carrying one of the variants had significantly higher risks of head, neck and oesophageal cancer. For non- and occasional drinkers, there was no increased cancer risk.

These findings indicate that alcohol directly causes several types of cancer, and that these risks may be increased further in people with inherited low alcohol tolerability who cannot properly metabolise alcohol
Dr Pek Kei (Becky) Im, Oxford Population Health

A previous study, also led by Dr Im, demonstrated that reducing alcohol consumption in China could be an important cancer prevention strategy.

Regular male drinkers in the study had a 26% higher risk for cancers previously associated with alcohol – with a clear link between the amount of alcohol consumed and the risk of these cancers.

Does dairy in the diet increase chances of cancer?

Another large-scale study by researchers from Oxford Population Health, Peking University and the Chinese Academy of Medical Sciences set out to investigate dairy consumption and cancer risk in Chinese adults.

In China, consumption of dairy is far lower than in Western populations.

Supported by Wellcome as part of Oxford’s Livestock, Environment and People (LEAP) programme, the study looked at over 510,000 individuals. It showed that people who consumed dairy products regularly had a higher risk of liver and breast cancer. The study doesn’tprove causation; there are several plausible biological mechanisms that may explain the associations, from increased levels of growth factors to higher hormone levels from cow’s milk.

While the study shows that there may be a link between regular dairy consumption and cancer, it’s important to be aware that dairy products are also a source of protein, vitamins and minerals for many.

Red meat eaters at higher risk

Another possible source of protein, vitamins and minerals in your diet is red meat.

However, this also comes with an associated cancer risk. Even moderate amounts of red and processed meats could potentially increase the risk of bowel cancer, according to a study co-authored by Professor Tim Key, Deputy Director of Oxford’s Cancer Epidemiology Unit.

Prof Key’s work highlights that even eating in line with UK Government recommendations increases the risk of developing bowel cancer.

The risk of cancer rose 19% for every 25g of processed meat per day, roughly equivalent to a rasher of bacon or a slice of ham. For red meat, the increase was 18% for every 50g, approximately a thick slice of roast beef or a bit of lamb chop.

Our results strongly suggest that people who eat red and processed meat four or more times a week have a higher risk of developing bowel cancer than those who eat red and processed meat less than twice a week
Professor Tim Key, Professor of Epidemiology and Deputy Director of the Cancer Epidemiology Unit

Is vegetarianism the solution?

Results from a large-scale analysis show that following a vegetarian or pescatarian (fish-eating) diet could reduce the risk of developing cancer.

The study found that low meat-eaters had a 2% lower chance of developing any type of cancer than regular meat-eaters. This is further decreased for pescatarians (10% lower) and vegetarians (14% lower).

An Apple iPhone
Alcohol bottles stacked on a bar
A bottle and a glass of milk
Various packages of red meat

A gloved pair of hands helping give a blood test via someone's finger
A coiled measuring tape

Diet is not only a modifiable risk factor in preventing cancer, but also in preventing chronic diseases such as diabetes – which may themselves contribute to cancer. A global review, led by researchers from The George Institute for Global Health, has shown that diabetes raises the risk of cancer, especially in women.

Diabetes was a risk factor for most types of cancer in both men and women; however, women with diabetes were at a higher risk of developing kidney (11%), oral (13%) and stomach cancer (14%), and leukaemia (15%) than men with the disease.

In order to improve detection and prognosis of pancreatic cancer, researchers are looking to use computer modelling to predict which patients who have been diagnosed with diabetes may go on to develop pancreatic cancer.

Reducing prostate cancer risk with a healthy weight

There is nothing men can do about their height, but at least it is now more evident that they may reduce their risk of aggressive prostate cancer by having a healthy weight
Dr Aurora Pérez-Cornago, Associate Professor and Senior Nutritional Epidemiologist

Prostate cancer is the most common cancer in men in the UK, with more than 47,500 cases and over 11,500 deaths per year.

A number of factors that increase the risk of prostate cancer are not modifiable, such as ethnicity and age; however, researchers have identified modifiable factors that could increase the risk of high grade disease.

A study led by the Cancer Epidemiology Unit showed that there was an 18% greater risk of death from prostate cancer and a 13% greater risk of high-grade cancer with every 10cm increase in waist circumference.

A further study, also led by Dr Pérez-Cornago, showed that greater body fat may increase the risk of dying from prostate cancer by 10%.

Body Mass Index (BMI):

The measure of body mass relative to a person’s size, calculated by dividing a person’s weight in kilograms by the square of their height in metres

If the average BMI of men aged 55–64 was reduced to the ideal range (18.5 to 24.9), the study suggests that there would be 1,300 fewer prostate cancer deaths per year.

Using data from the Million Women Study, additional research has suggested that women with ovarian cancer have a 6% higher risk of death for each five point increase in their BMI.

Menopause and menopausal treatment

A woman post mastectomy and partial reconstruction
mammogram image
A packet of pink pills
A newspaper photo of three women: Valerie Beral, Emily Banks and Julietta Patnick

Valerie Beral, Emily Banks and Julietta Patnick featured in The Times newspaper on the launch of the Million Women Study in 1996.

Valerie Beral, Emily Banks and Julietta Patnick featured in The Times newspaper on the launch of the Million Women Study in 1996.

Body weight at different ages can tip the scale of cancer risk

Being overweight early in life, but not adulthood, may protect women against breast cancer, according to data from the Million Women Study.

A study led by Oxford Population Health’s Cancer Epidemiology Unit combined information about participants’ body size at age 10, clothes size at age 20 and BMI at age 60 with data from national cancer and death registries.

To our knowledge, this is the largest study of early-life body size and the risk of postmenopausal breast cancer
Dr Owen Yang, lead researcher, Cancer Epidemiology Unit

Researchers found that greater body size at ages 10 and 20 were associated with a decreased risk of postmenopausal breast cancer. However, those who had a greater BMI at around age 60 were associated with an increased risk of postmenopausal breast cancer, with the risk increasing 20% for each five-point increase in BMI.

Can menopausal treatments cause cancer?

An international collaboration, the Collaborative Group on Hormonal Factors in Breast Cancer, investigated data from more than 100,000 women with breast cancer from 58 studies worldwide. They found that using menopausal hormone therapy (MHT) is associated with an increased risk of breast cancer, and that some risk persists for more than a decade after therapy stops. Risks are greater for users of oestrogen–progestogen therapy than for oestrogen-only therapy.

A co-author of the study, Professor Valerie Beral, was also the Principal Investigator of the Million Women Study.

The Million Women Study recruited around 1.3 million women over five years through 66 NHS breast screening centres in England and Scotland. It has increased our understanding of cancer risk in key areas, including the demonstration that women using hormone replacement therapy were at an increased risk of breast cancer, and the discovery that using the contraceptive pill for ten years cuts the risk of womb cancer by half.

Cancer recurrence

Menopausal treatment and recurring risks of cancer

The Early Breast Cancer Trialists’ Collaborative Group (EBCTCG) looked at whether aromatase inhibitors or tamoxifen is better at reducing the risk of breast cancer recurrence in both premenopausal and postmenopausal women.

Aromatase inhibitors:

Any one of a class of drugs used in the treatment of early and advanced oestrogen-dependent breast cancer in postmenopausal women. These drugs act by inhibiting the action of aromatase, which can reduce oestrogen levels

The study, primarily funded by Cancer Research UK, found that for women with hormone-sensitive, operable breast cancer, tamoxifen treatment after surgery reduces the risk of death from breast cancer within 15 years by about a third. Aromatase inhibitors are even more effective for postmenopausal women, reducing this risk by a further 30%.

Aromatase inhibitors are ineffective for premenopausal women, however, since their ovaries respond by increasing oestrogen production.

Tamoxifen:

A synthetic antioestrogen used to treat oestrogen receptor-positive breast carcinoma

 The risk of breast cancer recurrence was reduced significantly (21%) for those who took aromatase inhibitors instead of tamoxifen. The first five years of treatment saw the greatest benefit. Women under the age of 35, who have a higher risk of recurrence, found that aromatase inhibitors were just as effective for them as they were for older women who have a lower risk.

The length of treatment also has an impact on the risk of breast cancer recurrence.

An earlier study from the EBCTCG suggested that hormonal treatments should continue for longer to avoid the risk of late recurrence.

The EBCTCG pooled data from over 60,000 women from 88 clinical trials between 1976 and 2011. All had been diagnosed with oestrogen receptor-positive breast cancer, and proscribed anti-oestrogen therapy for five years. Their recurrence risk remained constant in each subsequent year from year five through to year twenty.

Study co-author Professor Daniel F Hayes from the University of Michigan says that the study shows anti-oestrogen therapy can reduce the risk of recurrence beyond five years; however, he says that it’s difficult to know whether the benefits outweigh the possible side effects of continued treatment.

This improved understanding of the factors that raise the risk of a late recurrence will help women and their doctors make individual decisions on how long to continue hormonal treatment
Dr Hongchao Pan, Associate Professor, Clinical Trial Service Unit and Epidemiological Studies Unit

Identifying biomarkers to predict early recurrence of post-surgery pancreatic cancer

Recurrence has also been a focus of research by the Oxford Centre for Early Cancer Detection, which has worked on predicting early recurrence of pancreatic cancer.

In the UK, only 7% of pancreatic cancer patients survive the disease for five years. Researchers are trying to understand which patients are at high risk of early relapse so that treatments can be better tailored. For example, this could be through pre-operative chemotherapy, by improving patient selection for surgery or by altering the frequency of post-operative monitoring.

Researchers from the Department of Oncology and the Oxford Hepato-Pancreato-Biliary surgical unit at Oxford University Hospitals NHS Foundation Trust set about looking for biomarkers that are predictive of pancreatic cancer recurrence.

Biomarker:

A naturally occurring molecule, gene or characteristic by which a particular pathological or physiological process, disease, etc. can be identified

The team, including Dr Daniel Hughes and Professor Eric O’Neill, studied RNA sequencing data from the Cancer Genome Atlas database, looking at 95 patients with pancreatic ductal adenocarcinoma. They identified five genes that were expressed at different levels in two patient groups.

One was a promising biomarker candidate: patients who had tumours that recurred early were seen to have higher levels of it than those whose disease did not recur within a year of surgery.

This is a positive step towards using predictive biomarkers to understand and treat pancreatic cancer, with validation in a larger patient cohort now needed.

A computer drawing of aromatase
A computer model of tamoxifen
Hands holding two pills
Renderings of the pancreas and surrounding organs

Precision prevention 

MRI scan of a brain
A package of Metformin tablets

Early detection, or the ‘watch and wait’ strategy, is currently the only option for people with Li Fraumeni Syndrome (LFS), an inherited familial disposition to a range of often rare cancers.

LFS impacts around 600 people in the UK and is caused by gene mutations. People with LFS have a 70–90% risk of developing cancers, particularly those of the brain, breast, blood and soft tissue.

In collaboration with St. George’s Hospital London, and the Universities of Cambridge and Manchester, the University of Oxford is leading a world-first £2 million national cancer prevention trial in patients with LFS: the Metformin in Li Fraumeni (MILI) study.

It will be the world’s largest study into LFS – and the first ever cancer prevention study for this community
Professor Sarah Blagden, Chief Investigator for the MILI study and Professor of Experimental Oncology

The MILI study is a global collaboration with parallel studies in Canada, Germany and the USA that aims to recruit over 200 adults with LFS to assess whether the diabetes drug Metformin can prevent or delay cancer emergence.

For over 60 years, Metformin has been used to treat Type 2 diabetes. If successful, it could be the first cancer intervention for those with LFS. All participants in the study will undergo regular imaging scans, blood tests and other assessments to detect any emerging cancers and monitor any side effects.

The first outcomes from the trial are expected in 2028.