Predictive decision trees optimized to decide if a cancer mutation will become...
Predictive decision trees optimized to decide if a cancer mutation will become visible to the immune system.
Source: Rik G.H. Lindeboom, Radboud University.

AI system predicts which cancer patients benefit from immunotherapy

Scientists at the Institute for Research in Biomedicine (IRB Barcelona), in collaboration with the Centre for Genomic Regulation (CRG) and Radboud University, have developed an algorithm that can predict which cancer patients are more likely to benefit from immunotherapy.

Mutations in our DNA can disrupt protein synthesis, sometimes causing truncated proteins which don’t work as intended. Known as nonsense mutations,these types of alterations can give rise to hereditary diseases and different types of cancer. To keep the number of truncated proteins to a minimum, human cells recognise and remove RNAs with nonsense mutations through a quality control process known as nonsense-mediated mRNA decay (NMD).

To better understand the effect of NMD on human disease, researchers built NMDetective, a tool describing every possible nonsense mutation that can occur in the human genome. Developed by large-scale statistical analyses based on machine learning, the algorithm identifies which mutations in the genome are susceptible to NMD.

The scientists used NMDetective to analyse thousands of genetic variants that give rise to hereditary diseases in humans. “We were surprised to observe that, in many cases, NMD activity was predicted to lead to a greater severity of the disease,” says Fran Supek, ICREA researcher, head of the Genome Data Science laboratory at IRB Barcelona and leader of the team that built the tool.

The results of the study suggests that pharmacological NMD inhibition could slow the progression of many different genetic diseases. To distinguish which patients would benefit from this therapy, it is necessary to apply a precision medicine approach to determine the mutation responsible for the disease and the effect of NMD on this mutation, and this is precisely where NMDetective comes into play.

Researchers also studied the role of NMD in cancer and the interaction between the tumour and the immune system. “We discovered that NMD activity is important for the prediction of successful outcome of immunotherapy in cancer,” explains Supek. Researchers found NMD hides mutations that would otherwise trigger the immune system. Therefore, NMDetective can be used to analyse the mutations present in the tumour, in order to better distinguish between cancer patients that respond to immunotherapy from those who do not respond to immunotherapy.

“Tumours are riddled with genetic mutations that should make all sorts of weird proteins. The immune system should pick these up, identifying the cells gone wrong and destroying them,” says Ben Lehner, a researcher at the Centre for Genomic Regulation who also took party in the study. “But many of these weird proteins never actually get made because of NMD.”

“This algorithm can distinguish which mutations will and won’t trigger this error-checking system. What’s exciting is that drugs that block NMD already exist, which could be used in conjunction with other treatments to help the immune system better recognise tumour cells,” explains Lehner.

“Here we see an example where publicly available cancer genomics data can be ‘reused’ through machine learning approaches to better understand biological processes such as NMD,” says lead author Rik Lindeboom, a researcher at Radboud University in the Netherlands. “What makes this study especially exciting, is that we could directly translate this fundamental research into insights that are relevant for clinicians and patients.”

Subscribe to our newsletter

Related articles

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.

Using AI to predict risk of thyroid cancer on ultrasound

Using AI to predict risk of thyroid cancer on ultrasound

Researchers from Thomas Jefferson University use machine learning on ultrasound images of thyroid nodules to predict risk of malignancy.

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 classifies lung cancer slides at pathologist level

AI classifies lung cancer slides at pathologist level

Researchers have utilized machine learning capabilities to assist with the challenging task of grading tumor patterns and subtypes of lung adenocarcinoma.

Machine learning model predicts cancer patients’ symptoms

Machine learning model predicts cancer patients’ symptoms

Doctors could get a head start treating cancer thanks to new AI developed at the University of Surrey that is able to predict symptoms and their severity throughout the course of a patient’s treatment.

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.

Popular articles