A new genomic urine test can help doctors accurately predict bladder cancer as many as 12 years before clinical signs and symptoms emerge and long before a diagnosis can be made with cystoscopy, the most common method of detection, according to a new study.
Developed by Convergent Genomics, the UroAmp test uses next-generation DNA sequencing and Machine Learning to analyse urine for mutations across 60 genes linked to bladder cancer while also measuring the entire genome. In the study, a team of American, French and Iranian researchers focused on a subset of 10 genes that hold the greatest predictive power for a future bladder cancer diagnosis. Funding for the study was provided by the National Cancer Institute through a Small Business Innovation Research (SBIR) grant to Convergent Genomics.
“This is the first study to show comprehensive genomic profiling of somatic mutations can detect preclinical urothelial cancer more than a decade ahead of a natural diagnosis,” said Yair Lotan, MD, Professor of Urology and Chief of Urologic Oncology at University of Texas (UT) Southwestern Medical Center in Dallas. “With further studies, this discovery could improve how we identify, risk stratify and monitor patients at increased risk of developing bladder cancer.”
Researchers studied UroAmp in two ways. First, they conducted a case-control study with urine samples from 96 control subjects and 70 bladder cancer patients (22 de novo, 48 surveillance). This found UroAmp had an 86% sensitivity in predicting new tumours and 71% sensitivity overall (new and recurrent). Specificity, or the percentage of people who tested negative for bladder cancer and do not have the disease, was 94%.