Lung cancer: AI could predict risk of recurrence

Computer scientists working with pathologists have trained an artificial intelligence (AI) tool to determine which patients with lung cancer have a higher risk of their disease coming back after treatment.

Photo
Non-small cell lung cancer.
Source: Dr Ed. Uthman via Yale Rosen on Flickr, CC BY-SA 2.0

The AI tool was able to differentiate between immune cells and cancer cells, enabling researchers to build a detailed picture of how lung cancers evolve in response to the immune system in individual patients. Although this research is in its early stages, this ground-breaking approach could speed up how doctors can predict which patients are more likely to see their lung cancer return, so they can be closely monitored with tailored treatment plans.

The new AI tool—developed by researchers at The Institute of Cancer Research, London, in collaboration with scientists at University College London Cancer Institute and the Francis Crick Institute—was trained by pathologists to pick out immune cells from cancer cells. This allowed the tool to map out areas in tumours where the number of immune cells were high compared to the number of cancer cells, in patients with lung cancer.

Using the AI tool, the team found that while some parts of the tumour were packed with immune cells, described as 'hot' regions, other parts of the tumour appeared to be completely devoid of them, which they described as 'cold' regions.

When the researchers followed the progress of patients who had a higher number of 'cold' regions, they found patients were at a higher risk of relapse. This study is part of the TRACERx (Tracking Cancer Evolution through therapy [Rx]) lung study—a £14 million, 9-year study funded by Cancer Research UK.

Scientists led by Dr. Yinyin Yuan from The Institute of Cancer Research (ICR) combined AI pathology image-mapping technology with next-generation sequencing. They used this tool to analyse samples from 100 patients with non-small cell lung cancer who took part in the TRACERx study.

Comparing immune hot or cold areas within the same tumour, the team assessed the extent to which their genetic make-up differed, giving them clues as to how they had changed in response to evolutionary pressures.

Evolutionary pressure from the immune system

The team's work revealed that cancer cells found in immune cold regions may have evolved more recently than cancer cells found in immune hot regions that are packed with immune cells. The researchers suggest that areas of the tumour with fewer immune cells may have developed a 'cloaking' mechanism under evolutionary pressure from the immune system allowing them to hide from the body's natural defences. Their AI tool can assess how many regions with this cloaking mechanism exist within a tumour—which is critical given that these immune cold areas are associated with cancer relapses.

This early research could in future be used in the clinic to give doctors more detailed information about the cellular make-up of tumours than can be seen with the naked eye, informing the development of tailored treatment strategies for individual patients. "In our new study, we applied artificial intelligence to genetic data and pathology images, to create a new tool that could in the future help pick out those patients with lung cancer who are at highest risk of their cancer coming back," Dr. Yinyin Yuan, Team Leader in Computational Pathology at the ICR, said. "We've gained new insight into how lung cancers can cloak themselves to escape the attention of the immune system—and in doing so can continue to evolve and develop. Cancer's ability to evolve and to come back after treatment is one of the biggest challenges facing cancer researchers and doctors today. Our research has revealed fresh insights into why some lung cancers are so difficult to treat, and we wouldn't have been able to do this without the scale and scope of the TRACERx project."

Mapping the evolution of lung cancer

Professor Charles Swanton, Cancer Research UK's chief clinician who leads the TRACERx initiative, said: "Much like the finches Darwin described in the Galapagos islands, cancer cells also grow and evolve in response to their unique surroundings, weeding out the weak to let the strongest survive. This unstoppable force of nature means curing cancer after it has spread is a difficult task and helps us understand how tumours develop resistance to even the most advanced and powerful treatments we have. Focusing on the intricacies of the tumour alone isn't enough, we need to explore the environment that each tumour is growing in and to understand its influences. TRACERx has given us the platform to explore this in exceptional detail and has shown us that multiple approaches at different stages of the disease are needed to outwit cancer."

Dr. Mariam Jamal-Hanjani, senior clinical lecturer and consultant medical oncologist at UCL Cancer Institute, who has worked on TRACERx since its inception, said: "Unfortunately, lung cancer survival remains among the lowest of all cancer types despite new treatment options for patients. This can be due to lung cancer cells evolving, becoming resistant to treatment causing the disease to return or worsen. TRACERx is enabling us to map the evolution of lung cancer from diagnosis to cure after surgery or death, and is already revealing how we can offer patients better care and treatment. Not only have we found new ways to predict treatment response and outcome, but we can now detect when lung cancer has returned before it is seen on a scan. And new cutting-edge treatments that use a patient's own immune cells that have been reprogrammed to target lung cancer cells could become a reality."

Subscribe to our newsletter

Related articles

Machine learning improves diagnostics of head and neck cancers

Machine learning improves diagnostics of head and neck cancers

Researchers used artificial intelligence to develop a new classification method which identifies the primary origins of cancerous tissue based on chemical DNA changes.

AI finds patterns of mutations in tumour images

AI finds patterns of mutations in tumour images

Researchers have developed an AI algorithm that uses computer vision to analyze tissue samples from cancer patients.

AI model accurately classifies colorectal polyps

AI model accurately classifies colorectal polyps

An AI model for automated classification of colorectal polyps could benefit cancer screening programs by improving efficiency, reproducibility, and accuracy.

AI and laser-based imaging system identify brain tumors

AI and laser-based imaging system identify brain tumors

A novel method of combining advanced optical imaging with an artificial intelligence algorithm produces accurate, real-time intraoperative diagnosis of brain tumors.

AI tracks down acute myeloid leukemia

AI tracks down acute myeloid leukemia

Artificial intelligence can detect one of the most common forms of blood cancer—acute myeloid leukemia (AML)—with high reliability.

AI detects even the smallest metastases

AI detects even the smallest metastases

Researchers have developed a new algorithm that enables automated detection of metastases at the level of single disseminated cancer cells in whole mice.

ConvPath software uses AI to identify cancer cells

ConvPath software uses AI to identify cancer cells

A software tool uses artificial intelligence to recognize cancer cells from digital pathology images — giving clinicians a powerful way of predicting patient outcomes.

Deep learning identifies molecular patterns of cancer

Deep learning identifies molecular patterns of cancer

An AI platform can analyze genomic data extremely quickly, picking out key patterns to classify different types of colorectal tumors and improve the drug discovery process.

AI predicts effectiveness of immunotherapy

AI predicts effectiveness of immunotherapy

Scientists can determine which lung-cancer patients will benefit from expensive immunotherapy.

Popular articles