Cancer

Lead Researcher: Dr Cathrin Gudd

Supported by the Digestive Diseases Theme

Immune checkpoint inhibitors (CPIs) are a new class of cancer therapy. They have proved to be highly effective in treating a range of tumour types, including skin, kidney and lung cancer. Immune checkpoints function as the brake of the immune system. Cancer cells can exploit these to hide from immune cells. CPI therapy blocks this brake and encourages immune cells to become more active and to identify and kill cancer cells.

While CPI therapies can be very effective in treating cancer, around two-thirds of patients experience side effects because the immune system becomes too active. One serious side effect is CPI-induced hepatitis (CPI-hepatitis), where the liver becomes inflamed due to the cancer treatment. This condition occurs in up to 30% of patients receiving CPI cancer therapy and can range from mild to severe illness or even death. If a patient develops CPI-hepatitis, they must stop their cancer treatment and take strong medications to suppress their immune system. This unfortunately can reduce the chances of cancers responding to treatment. Up to 19% of CPI-hepatitis patients must permanently stop CPI therapy because of the severity of liver inflammation. CPI-hepatitis is therefore a pressing concern for patients and oncologists.

Our research to date found two types of immune cells (monocytes and cytotoxic T cells) to play a key role in causing liver inflammation during CPI treatment. We discovered that these cells enter the liver and interact in ways that lead to damage. This pilot project focused on studying the genes of these cells to better understand how they cause liver damage. With this we hope to find ways to block this harmful interaction, allowing patients to continue their cancer treatments with fewer risks.

In this pilot project, we have identified a cohort of patient and control liver biopsy samples accessible for genetic analysis. All samples were successfully retrieved, prepared and assessed for their genetic makeup. We identified a number of changes in inflammatory genes specific to patients with CPI-hepatitis. We are now investigating these changes in detail in the lab to test whether they could lead to treatments for liver inflammation in CPI-hepatitis.

Lead Researcher: Dr Zoe Hall

Supported by the Digestive Diseases Theme

Liver cancer is one of the leading causes of cancer-related deaths worldwide, with more than a million cases expected by 2025. Hepatocellular carcinoma (HCC) is the most common form of liver cancer and is a very complex form of cancer which makes the development of new therapies is challenging. Drugs which target the immune system and get it to attack the cancer cells in HCC are often successful. At the same time patients with HCC can also have fatty liver disease, where fats accumulate in the liver and affect the liver’s ability to function properly. However, the immune targeting drugs have limited success in patients with HCC and who also have fatty liver disease.

The liver is made of many different cell types, including immune cells. The immune cells may be protective and “search” for and removing tumour cells, but they may also cause inflammation and increase cancer risk. When the liver starts to have problems metabolising fats it can contribute to the development and progression of HCC. Fats can interfere with the liver’s immune cells, and this can affect their ability to find and kill the HCC cells.

In the past decade, new methods and technologies have been developed that can quantify DNA, RNA, proteins and chemicals in biological samples. In this project, we will use a combination of these new methods to measure levels of chemicals across different parts of tissues and cells. We will explore the how different fats are found in the tumour and how they are connected to different tumour cells. Using these technologies, we can build a “chemical signature” across the tumour, and between tumours from different patients, to understand better the link between the immune system and metabolism. From this signature we will decide the best treatment for patients and lead to design of novel drugs for HCC.

Lead Researcher : Dr Laura Tookman

Supported by the Surgery & Cancer Theme

Ovarian cancer is a complex disease. To ensure that each individual patient receives the best treatment, doctors need more information about why the cancer developed. Cancer can be caused by genetic changes inherited from parents or due to changes within the cancer that drives the cancer to grow. The most commonly inherited gene abnormalities in ovarian cancer are in the BRCA1 and BRCA2 genes. Identifying harmful changes in genes is important for treatment decisions and managing future cancer risk for patients and their relatives.

Knowledge about genetic changes within the cancer itself can also help guide treatment.  Some ovarian cancers grow due to an inability to repair defective cells properly.  This process is known as homologous recombination deficiency or HRD. Identifying this defect is crucial because studies have shown that a new therapy called PARP inhibitors improves outcomes for patients with advanced ovarian cancer. The benefit is greatest in those with these harmful changes in the BRCA genes and those cancers that show HRD.

Genetic testing is vital for patients with ovarian cancer to help determine the best treatment.  However, a recent national survey of health record use by gynaecological cancer professionals found that finding these genetic records is challenging and time consuming. To improve this process, we wish to investigate new ways of evaluating results.

This project will set up a secure database so we can examine genetic test results from over 200 ovarian cancer patients. We will follow their progress to assess how accurately the test results predicted response to treatment. We will work with data scientists to develop new automated processes to interpret genetic results from patient notes to create accurate datasets. This will allow us and other researchers to perform high quality research, enhance efficiency and improve care for patients with ovarian cancer.

Lead Researcher: Dr Naomi Laskar

Supported by the Surgery & Cancer Theme

Contrast dyes can be used to target and light up cancer cells. The contrast dye we are using, 5-Aminolevulinic acid (5-ALA), is not visible to the naked eye. This is beneficial as it does not interfere with the operation however, it means we must use specialised cameras, microscopes and devices in order to detect the presence of the dye.

The initial step to use a contrast agent to highlight a tumour boundary is aiming to prove that the contrast only accumulates within the tumour cells and not the healthy, normal cells. This is why we have begun our study with taking small samples of tumour and healthy tissue and comparing the levels of uptake of dye within them. We have compared samples of breast tumour and normal breast tissue with the dye in and also some without as a control sample so we can more be sure that our results are accurate. This is particularly because there are agents within normal healthy tissue that naturally light up when you look at them with specialised lenses. It can then be difficult to distinguish between the light from the normal healthy tissue and the light from the contrast dye.

So far, we have recruited 12 patients to drink the dye before their surgery. We have cut the samples of tumour and healthy tissue of 10 of these patients into small sections and analysed them with our specialised microscope. We have measured the levels of dye accumulating and lighting up tumour cells and found that there is a considerable difference between the amount of dye within the breast tumour compared to the healthy breast tissue. Given that this is a small sample of patients, we still need to recruit many more and prove this on a larger scale before we can start to safely use this equipment in theatre on patients to guide an operation.

We are also using mass spectrometry (a weighing scale system) to identify the presence of 5-ALA within the tumour cells. However, before this can be used in theatre on patients in real-time, first we need to demonstrate it works within a laboratory setting. The initial step we have performed is to find the specific weight of the contrast dye that we can see on a graph so we have something to look for when we perform the analysis on the samples of tissue we have taken

Lead Researcher: Maira Salman

Supported by the Surgery & Cancer Theme

Lung cancer causes the most cancer death in the UK, because it is usually diagnosed when the cancer is more advanced, and treatment is less effective. Screening can improve chances of survival by finding cancer earlier. The Targeted Lung Health Check (TLHC) is lung cancer screening being offered in some areas of the UK for people who have a higher risk of getting lung cancer. However, uptake is low nationally, at around 40%.

We conducted a survey as part of a larger NHS Behavioural Science Unit project to understand why people would choose to attend or not attend a lung cancer screening appointment. The survey ran in October 2023 and only those individuals who were eligible to take part in lung cancer screening participated in the survey (current or previous smoking history, ages 55-74 and based in England). A total of 3274 participants took part in the survey and were recruited through Bilendi, the NHS App Panel, and the NHS Vaccines panel run by NHS England.

The most commonly reported reasons for intention to attend included beliefs that it was better to detect cancer at an earlier state (65%) and that lung cancer is more treatable if diagnosed earlier(55%). Most common reasons for non-attendance were fear of results (28%), lack of symptoms (25%), stigma of smoking (19%), and lack of endorsement from their GP (19%). Analysis of subgroups showed those who did not identify as “White” were significantly more likely to consider whether attendance was free as a reason to attend screening. It is worth noting though, that the percentage of non-white ethnicity participants in the survey was very low (3.8%). Women were more likely than men to attend because they felt it was something they felt they could do to feel in control of their health and because they felt lung cancer was more treatable if diagnosed earlier. Women were less likely to attend compared to men, due to feeling scared of the results, feeling judged, and anxiety around the screening process. Men were more likely than women to report that their decision about whether or not to attend screening was affected by whether their GP had recommended it to them..

These results may be used by lung cancer screening services to create targeted messaging to improve the uptake of lung cancer screening

Lead Researcher: Dr Stefan Antonowicz

Supported by the Surgery & Cancer Theme

We have discovered that oesophageal cancer chemicals vary between different tumour areas (see Pilot Data). Biologically these areas behave differently despite looking identical, and respond differently to anti-cancer drugs. This might explain why oesophageal cancer usually responds poorly to chemotherapy.

To address this, we made a new collaboration between Departments of Surgery and Cancer, and the Dyson School of Design Engineering. We have created a device (iSlice) which specifically tests how this variation affects chemotherapy results. iSlice passes tiny streams of different anti-cancer drugs across thin slices of patients’ tumours grown in the lab. The anti-cancer effects are then measured using microscope techniques. In each tumour slice, many drug combinations/concentrations can be tested, in replicate, which makes the approach flexible and robust.

Our objective is a trial to improve oesophageal cancer chemotherapy by targeting this chemical variation. This project built the case for this, by:

(i) assessing whether this chemical variation can be measured in archived patient tissue, to provide a complete description and unlock a vast potential resource

(ii) listing drug predictions between patients, to develop the clinical trial plan

(iii) demonstrating the iSlice technology, so we can confirm this plan.

This study is now completed. We were able to show that archived patient tissue can be used for metabolic analysis, however more work is needed to reliably measure chemical variation across these tumours. Despite this, we were able to generate a list of three promising new drugs for tackling chemical variation in these tumours. We also successfully established tumour slice cultures of a number of patients and tested traditionally chemotherapy on these slices. This progress takes us significantly closer to the intended trial. More work is needed to finalise these plans and apply for funding, and the results of this project directly contributed to a successful 5 year research award to fund this work.

Our patients believe tailored cures with less side effects are their research priority. Our approach is designed to meet this need, and can later be developed to deliver personalised treatment plans after surgery, in oesophageal and other solid tumours.

Lead Researcher: Dr Gina Brown

Supported by the Surgery & Cancer Theme

Cancer is deadly due to its ability to spread. It is traditionally believed that tumour spreads to lymph nodes and then to other organs, forming the current basis of disease staging. This is then used to decide pre-operative treatments patients are offered as well as surgery.

Results from the MARVEL trial, completed with support of this grant, showed that out of: tumour stage, lymph node stage, circumferential margin involvement and vascular spread- only vascular spread before and after chemoradiotherapy significantly affected patient disease-free survival (DFS). 5-year DFS for patients without vascular spread was 71% compared to 54.8% with vascular spread (95% CI, p = 0.006). Patients with vascular spread before or after chemoradiotherapy had roughly twice the risk of their cancer returning, disease spread elsewhere or death ( Baseline MRI vascular spread positive- Hazard Ratio 2.02, CI 1.20-3.41, p = 0.008; Post chemoradiotherapy MRI vascular spread positive- Hazard Ratio 1.98, CI 1.12 – 3.45, p = 0.018 ). Furthermore, when patient had vascular spread that responded to chemoradiotherapy and disappeared, their survival was similar to that of patients who had never had vascular spread.

On preliminary review, there was anecdotally a reduced immune infiltrate for patients who had response to CRT or no vascular invasion, whilst those with aggressive disease had an active immune response. More detailed immune profiling is pending due a delay in our workstream, and we aim to report on this by September 2025.