Scientists at the University of East Anglia (UEA) say they have discovered a new analytical method that helps explain why some prostate cancers are more aggressive, spread to different parts of the body, and ultimately cause death. The team, which published its study, “A novel stratification framework for predicting outcome in patients with prostate cancer” in the British Journal of Cancer, hopes its findings will help transform patient treatment.
“Unsupervised learning methods, such as Hierarchical Cluster Analysis, are commonly used for the analysis of genomic platform data. Unfortunately, such approaches ignore the well-documented heterogeneous composition of prostate cancer samples. Our aim is to use more sophisticated analytical approaches to deconvolute the structure of prostate cancer transcriptome data, providing novel clinically actionable information for this disease. We apply an unsupervised model called Latent Process Decomposition (LPD), which can handle heterogeneity within individual cancer samples, to genome-wide expression data from eight prostate cancer clinical series, including 1,785 malignant samples with the clinical endpoints of PSA failure and metastasis,” write the investigators.
“We show that PSA failure is correlated with the level of an expression signature called DESNT (HR = 1.52, 95% CI = [1.36, 1.7], P = 9.0 × 10−14, Cox model), and that patients with a majority DESNT signature have an increased metastatic risk (X2 test, P = 0.0017, and P = 0.0019). In addition, we develop a stratification framework that incorporates DESNT and identifies three novel molecular subtypes of prostate cancer. These results highlight the importance of using more complex approaches for the analysis of genomic data, may assist drug targeting, and have allowed the construction of a nomogram combining DESNT with other clinical factors for use in clinical management.”
The findings come after the same team developed a test that distinguishes between aggressive and less harmful forms of prostate cancer, helping to avoid sometimes-damaging unnecessary treatment. The new study shows how the number of aggressive cells in a tumor sample defines how quickly the disease will progress and spread. The study also reveals three new subtypes of prostate cancer that could be used to stratify patients for different treatments.
Lead researcher Colin Cooper, PhD, from UEA’s Norwich Medical School, said: “Prostate cancer is the most common cancer in men in the U.K. It usually develops slowly, and the majority of cancers will not require treatment in a man’s lifetime. However, doctors struggle to predict which tumors will become aggressive, making it hard to decide on treatment for many men. This means that many thousands of men are treated unnecessarily, increasing the risk of damaging side effects, including impotence from surgery.
The team developed a test to distinguish aggressive prostate cancers from less threatening forms of the disease, by applying complex math known as Latent Process Decomposition.
A collaborator, Vincent Moulton, PhD, from UEA’s School of Computing Sciences, added “By applying the Latent Process Decomposition process and analyzing global prostate cancer datasets, we discovered an aggressive form of prostate cancer known as DESNT, which has the worst clinical outcomes for patients.”
In the latest study, published today, the team studied gene expression levels in 1,785 tumor samples. They found that the amount of DESNT subtype cells in a sample is linked with the likelihood of disease progression, the more DESNT cells, the quicker the patient is likely to progress.
Co-lead researcher Daniel Brewer, PhD, noted: “If you have a tumor that is majority DESNT you are more likely to get metastatic disease, in other words it is more likely to spread to other parts of your body. This is a much better indication of aggressive disease. We also identified three more molecular subtypes of prostate cancer that could help doctors decide on different treatment options for patients. This research highlights the importance of using more complex approaches for the analysis of genomic data.”
