Computational analysis clarifies cancer risk for families with genetic variants microbiologystudy

QIMR Berghofer-led research has shown that new advanced computational prediction tools can improve the accuracy of genetic testing for families affected by an inherited condition that significantly increases their risk of developing cancer, paving the way to better targeted care.

The findings have been published in the American Journal of Human Genetics alongside complementary studies by international collaborators, which together show how incorporating the new computational biology tools with existing modeling methods improved the predictive power of genetic test results.

Computational tools are used to predict if and how a genetic variant is likely to impact the function of the protein encoded by the gene.

Genetic testing is recommended for people suspected to have a fault in a gene, to provide a genetic diagnosis and determine the most appropriate clinical management, such as regular screening, preventative measures, and targeted treatments.

However, not all genetic variants will cause harm. Determining which are benign and which are likely to cause disease is complex. Rapid advances in genetic sequencing have also led to increasing numbers of genetic variants being discovered that are yet to be classified.

Professor Amanda Spurdle heads an internationally recognized team at QIMR Berghofer that is developing and improving methods to fill this gap in knowledge, by better understanding which genetic variants are disease-causing (pathogenic), and which are benign.

One area of focus is Li Fraumeni Syndrome, a rare but devastating condition which occurs when people have a fault in the TP53 gene. This gene is known as the “guardian of the genome” because of its key tumor-suppressing role. A fault in the way this gene works can increase the risk of developing multiple cancers by the age of 60 to as high as 95%.

The study by QIMR Berghofer in collaboration with The Health Research Institute of the Hospital Clinico San Carlos in Spain, and clinical genomic testing company Ambry Genetics in the U.S., aimed to reduce the number of TP53 gene variants with unclear impact by using new computer-based methods to predict which variants impact function.

These included several tools trained to predict possible structural changes in proteins that could result from a genetic variant, and their effect on protein stability. The researchers showed using these types of tools in combination with existing prediction methods could improve the accuracy of variant classification, providing clearer answers for individuals undergoing testing. This could mean the difference between an individual having access to intensive screening or not.

Study first author Nitsan Rotenberg from QIMR Berghofer said that this will ultimately support more informed clinical decision-making and better patient care.

“Li-Fraumeni Syndrome is rare, but for the people affected it’s extremely important to give them certainty. We want to reduce the worry that comes with finding out they have a variant in the TP53 gene, but no one knows if it’s pathogenic or not.

“Improving genetic variant classification means those with a pathogenic variant can receive targeted screening which increases the chances of early diagnosis and more effective treatment,” Ms. Rotenberg said.

Senior author Professor Amanda Spurdle said the studies by QIMR Berghofer and their international collaborators can help inform international guidelines and clinical practice.

“The use of new computational biology tools focused on predicting protein structure represents a paradigm shift in genetic research, offering the potential to transform genomics research and its clinical application.

“We hope to continue to use these tools to improve the accuracy of our assessment of newly discovered and poorly understood genetic variants in hereditary cancer and other areas,” Professor Spurdle said.

This latest research adds to another important study by QIMR Berghofer’s Molecular Cancer Epidemiology Group, published in the journal Human Genomics.

That work, led by Dr. Cristina Fortuno revealed another way in which changes in the TP53 gene can potentially lead to abnormal protein production, by disrupting the crucial “splicing” process, where the genetic instructions for forming the tumor-suppressing proteins are edited together.

Dr. Fortuno said, “Together these studies have significant implications for personalized cancer risk prediction and prevention strategies. The findings highlight the rapid evolution of genetic knowledge, and the critical need for regular reassessment of genetic test results.”

Provided by
QIMR Berghofer