Active not recruitingOBSERVATIONAL

Enhanced Outcome Prediction in Cutaneous Squamous Cell Carcinoma Using Deep-learning and Computational Histopathology

This research is investigating a new way to predict how someone with a common type of skin cancer, called squamous cell carcinoma, might get on. Doctors are using special computer technology, known as 'deep learning', which looks closely at pictures of cancer cells taken under a microscope. By analysing these images, the computer program is being trained to recognise patterns that could help predict the future health of a patient after their diagnosis. The goal is to see if this computer tool can improve how doctors understand and explain a patient's likely journey with skin cancer. This study is important because better predictions can help doctors make more informed decisions about a patient's care and can help patients understand what to expect.

At a glance

Status

Active not recruiting

Sponsor

Queen Mary University of London

Enrolment target

300

Start

20 Oct 2023

Estimated completion

01 Oct 2028

Squamous Cell Carcinoma of the Skin

What is this study about?

This study is all about trying to understand and predict how a common type of skin cancer, called squamous cell carcinoma (SCC), might affect people. Squamous cell carcinoma starts in the outer layer of the skin. While it's generally treatable, knowing more about how each person's cancer might behave can help doctors plan the best care.

Researchers are using a new computer tool that uses 'deep learning'. Think of deep learning as teaching a computer to recognise very complex patterns, much like how we learn to recognise faces or objects. In this case, the computer is being taught to look at very detailed pictures of skin cancer cells, taken from tiny samples under a microscope. These pictures contain a lot of information that the computer can analyse.

The main aim is to see if this computer program, by studying these cell pictures, can become better at predicting what might happen to someone with SCC in the future. For example, it might help predict if the cancer is likely to come back or spread. If successful, this new tool could give doctors more accurate information, helping them to explain what to expect more clearly to patients and make more personal treatment plans.

Key takeaways

This study uses computers to analyse detailed images of skin cancer.
The goal is to better predict how squamous cell carcinoma might affect patients.
It examines existing medical records and tissue samples, not new procedures.
Participation likely won't involve extra visits or tests for you.
Findings could help doctors make more informed decisions for future patients.
Your personal information will be kept confidential.

Who may be eligible?

To be able to take part in this study, you would need to have received a diagnosis from a doctor confirming that you have a type of skin cancer called primary cutaneous squamous cell carcinoma. This means the cancer started in your skin.

You also need to be at least 18 years old to participate. There isn't an upper age limit mentioned, so adults of any age from 18 upwards can be considered.

The study is open to both men and women.

Could this study suit you?

Answer these quick questions to see if you may be eligible. This is a guide only — the research team makes the final call.

Have you been diagnosed by a doctor with primary cutaneous squamous cell carcinoma?
Are you 18 years old or older?
Were your tissue samples taken as part of your standard medical care?
Are you comfortable with researchers using your existing medical information and samples securely?

Answer every question to see your result.

What does participation involve?

As this study aims to evaluate a computer model using existing tissue samples (computational histopathology), it's highly likely that your participation would primarily involve researchers using information and tissue samples that have already been collected as part of your standard medical care for squamous cell carcinoma. This usually means your participation wouldn't require any extra hospital visits, tests, or taking new medications specifically for this study. The study focuses on analysing past information and images. There is no mention of follow-up visits or a specific duration for your participation, suggesting it's about analysing existing data rather than direct patient interaction.

Potential risks and benefits

Since this study largely involves analysing existing medical information and tissue samples, there are no direct physical risks to you. Your personal information will be kept confidential and anonymised wherever possible. The potential benefit is that the findings from this research could lead to better ways for doctors to predict how squamous cell carcinoma might behave in the future, which could ultimately help many patients. You would have the right to withdraw your consent for your data or samples to be used at any point, without affecting your medical care.

Locations (1)

Some site locations are approximate. We're improving this — please verify with the trial team before travelling.

Barts Health NHS Trust
Verified postcode
London, United Kingdom

Common questions

What is squamous cell carcinoma?

It's a common type of skin cancer that starts in the outer layer of your skin. It's usually treatable.

What does 'deep learning' mean?

It's a type of computer program that can learn to recognise complex patterns in things like detailed pictures of cells, helping it to make predictions.

Will I need to do anything extra if I take part?

It's very unlikely. This study mainly uses information and tissue samples that doctors have already collected from you as part of your normal care.

How could this study help me?

While it might not directly affect your current care, the study aims to improve how doctors predict your recovery, which could lead to better decisions for future patients.

Is my personal information safe?

Yes, your personal details will be kept private and handled very carefully, often making them anonymous so you can't be identified.