Detecting, diagnosing, and treating prostate cancer may become simpler following the development of a novel computer model that uses medical images to reproduce prostate cancer growth on the anatomy of a patient’s prostate.
The model, described in the paper “Tissue-scale, personalized modeling and simulation of prostate cancer growth,” and published in the Proceeding of the National Academy of Sciences, may lead to better personalized treatment and prediction of each patient’s clinical outcomes.
“There is a lot of room for improvement in both the diagnosis and management of prostate cancer,” study co-author Michael Scott, Brigham Young University professor of civil and environmental engineering, said in a press release.
“We’re using computer modeling to capture the behavior of prostate tumor growth, which will hopefully lead to minimally invasive predictive procedures which can be used in clinical practice,” he said.
Current medical practice requires men to undergo periodical screening for prostate cancer through PSA tests and digital rectal exams. If either test is positive, patients are asked to undergo a biopsy to detect cancerous cells and the aggressiveness of the tumor. But this often lead to patients being over- or under-treated.
Although medical images obtained with MRI and computed tomographies are already used to complement prostate cancer screening techniques, the researchers believe the new model may help replace the invasive biopsies with imaging.
Their attention was turned to prostate cancer because the prostate is a small organ; its growth can be estimated using the serum PSA concentration; and since some patients aren’t treated but are monitored for their PSA levels, the model could be easily validated in prostate cancer patients.
Using patient-specific information, including imaging data and PSA levels combined with factors like age, life expectancy, overall health status, and patient preferences, the model can not only construct a precise virtual anatomic model of the prostate and prostate tumors for any given patient, it can also simulate and compare alternative treatments, helping urologists make more informed diagnosis and treatment plans.
Although the mathematical model still needs extensive validation in prostate cancer patients, the researchers believe that “it’s likely that these types of models will eventually turn up in medical practice,” Scott said.
“We are entering an age where we will see the emergence of tools which leverage computation to improve diagnosis of disease,” Scott said. “And we’re not the only people working in this area — it’s rapidly growing.”