Researchers Use Computer Model to Predict Prostate Cancer Progression

An international team of cancer researchers from Denmark and Germany have used cancer patient data to develop a computer model that can predict the course of disease for prostate cancer. The model is currently being implemented at a prostate cancer clinic in Germany. The researchers have also found the enzyme that appears to trigger some of the first mutations in prostate cancer.

Prostate cancer is one of the most frequently diagnosed cancer in men in the western world with more than a million new cases each year. One of the key challenges is to differentiate between aggressive and non-aggressive disease. An international team involving researchers from BRIC at the Faculty of Health and Medical Sciences, the University of Copenhagen, and Rigshospitalet, the German Cancer Research Center (DKFZ) and the European Molecular Biology Laboratory (EMBL) in Heidelberg will solve this challenge with a computer model.

The researchers have studied the earliest mutational events in prostate cancer to develop a computer model, described in a new study that has just been published in the scientific journal Cancer Cell. The researchers collected patient data from close to 300 men who have had their entire cancer genome sequenced to characterise all mutations present in the tumour. Based on the data set, the researchers have developed the computer model which can be used to predict how prostate cancer will develop for a given patient.

"If we have a patient with a particular set of mutations, we can use the model to predict the most likely next mutation that the patient will experience at some point - and how it will affect the patient's clinical situation. As an illustration, we can predict with some probability that if you have mutation A, you are likely to get mutation B before you get C. We can also predict if the next mutation is likely to change the clinical outcome of the disease."

"So far, our data sets comprise around 300 patients, but we expect to collect data from several thousand patients in the coming years. The model will be better the more data it can learn from," says group leader and co-author of the study Joachim Weischenfeldt, BRIC, UCPH.

The computer model is currently being implemented at a clinic in Germany. The researchers expect it to take two to three years to have the model fully implemented as an integral part of the clinic's processes. Later, in a not too distant future the researchers hope that the model can also be introduced at hospitals in other countries, for example Denmark.

Mechanism Contributing to the First Mutations in Prostate Cancer Have Been Found

The approximately 300 patients from the study all had their entire genome sequenced. With genome sequencing, it becomes possible to tailor the treatment of the individual - also referred to as personalised medicine. The patients whose data the researchers have used have primarily been so-called early onset patients. This group is defined as men who are diagnosed with prostate cancer before reaching the age of 55 years.

"Prostate cancer develops over many years. We have therefore been particularly interested in the group of patients where the cancer is detected at young age as this allows us to analyse the tumour at an early stage. This is an important element because in this way we get a cleaner picture of the first mutations and alterations that occur in the tumour, to find out what is the initiating factor," says co-author Doctor Clarissa Gerhauser, DKFZ.

So far, it has not been known precisely what initiates prostate cancer. However due to the focus on the earliest detected tumours, the researchers uncovered a mutational mechanism involving an enzyme called APOBEC. This enzyme may help trigger the disease - i.e. trigger some of the very first mutations in prostate cancer.

"We hypothesise that this enzyme mutates the prostate cells at a low but constant rate. Each time the cell divides, APOBEC is likely to cause mutations. If you have early-onset prostate cancer, you may have a couple of mutations caused by APOBEC. Twenty years later, you may have 10-20 mutations," says co-author Doctor Jan O. Korbel, group leader at EMBL.

"The most common oncogene in prostate cancer involves a certain fusion gene. The APOBEC enzyme may contribute to the formation of this fusion gene. We cannot say that there is causality, but there is a strong correlation between mutations caused by APOBEC and other alterations such as this fusion gene," says Joachim Weischenfeldt.

The researchers have also found a putative novel oncogene in prostate cancer - ESRP1 - which is associated with very fast-dividing and highly aggressive prostate cancer. It is located closely to an already known oncogene, and the researchers believe this is the reason that it has not been discovered until now. In the study, the researchers show that the oncogene ESRP1 may be used as a possible new biomarker to detect whether a patient will have aggressive prostate cancer, which they validated on a cohort consisting of 12,000 other patients with the same type of cancer.

Gerhauser C, Favero F, Risch T, Simon R, Feuerbach L, Assenov Y, Heckmann D, Sidiropoulos N, Waszak SM, Hübschmann D, Urbanucci A, Girma EG, Kuryshev V, Klimczak LJ, Saini N, Stütz AM, Weichenhan D, Böttcher LM, Toth R, Hendriksen JD, Koop C, Lutsik P, Matzk S, Warnatz HJ, Amstislavskiy V, Feuerstein C, Raeder B, Bogatyrova O, Schmitz EM, Hube-Magg C, Kluth M, Huland H, Graefen M, Lawerenz C, Henry GH, Yamaguchi TN, Malewska A, Meiners J, Schilling D, Reisinger E, Eils R, Schlesner M, Strand DW, Bristow RG, Boutros PC, von Kalle C, Gordenin D, Sültmann H, Brors B, Sauter G, Plass C, Yaspo ML, Korbel JO, Schlomm T, Weischenfeldt J.
Molecular Evolution of Early-Onset Prostate Cancer Identifies Molecular Risk Markers and Clinical Trajectories.
Cancer Cell. 2018 Dec 10;34(6):996-1011.e8. doi: 10.1016/j.ccell.2018.10.016.

Most Popular Now

What the Radiologist should Know about A…

This paper aims to provide a review of the basis for application of AI in radiology, to discuss the immediate ethical and professional impact in radiology, and to consider possible...

The Mobile Game that can Detect Alzheime…

A specially designed mobile phone game can detect people at risk of Alzheimer's - according to new research from the University of East Anglia. Researchers studied gaming data from an...

Hospital Diagnoses Critically Ill Childr…

The Rady Children’s Hospital (San Diego, USA) used Moon, software developed by the Leuven-based (Belgium) company Diploid. Moon is the first software worldwide to use Artificial Intelligence (AI) for the...

AstraZeneca Starts Artificial Intelligen…

AstraZeneca and BenevolentAI began a long-term collaboration to use artificial intelligence (AI) and machine learning for the discovery and development of new treatments for chronic kidney disease (CKD) and idiopathic...

Novartis Presents First-of-its-Kind Algo…

Novartis today announced results from a validation study of the innovative, algorithm-based digital solution MS Progression Discussion Tool, or MSProDiscussTM. The tool aims to support and facilitate a discussion between...

Bayer Joins Sensyne Health Consortium Wo…

Sensyne Health plc (LSE: SENS), the British clinical AI technology company, and Bayer, the life sciences company, announce that Bayer has become Sensyne Health's preferred pharmaceutical partner to work together...

Red Hat Helps Public Health England Use …

Red Hat, Inc. (NYSE: RHT), the world's leading provider of open source solutions, announced that Public Health England (PHE), an executive agency of the Department of Health and Social Care...

Chipmunk Health Chooses Philips HealthSu…

Royal Philips (NYSE: PHG, AEX: PHIA), a global leader in health technology, announced that telehealth service provider Chipmunk Health, with locations in the Netherlands and Canada, is leveraging the capabilities...

Siemens Healthineers and European Societ…

With their first course under the heading "Intelligence. Innovation. Imaging - The perfect vision of AI," held in Barcelona on April 5-6, Siemens Healthineers and the European Society of Radiology...

Mobile Prenatal App Shown to Reduce in-P…

Using the mobile app Babyscripts reduced in-person prenatal care visits while maintaining patient and provider satisfaction, according to research published in JMIR mHealth and uHealth by physician researchers from the...