Pathology test uses AI to predict prostate cancer progression following surgery

October 2, 2018, The Mount Sinai Hospital

A pathology test that applies artificial intelligence (AI) to characterize tissue samples can accurately predict clinically significant prostate cancer disease progression following surgery, according to a study conducted at the Icahn School of Medicine at Mount Sinai and published in Nature Prostate Cancer and Prostatic Diseases. The Precise MD post-op test automates the Gleason score (a grading system that has been used since the 1960s to establish the prognosis for a prostate cancer) through the application of an algorithm that integrates image analysis with protein biomarkers. Precise MD is a pathology platform that uses artificial intelligence and complex algorithms to translate data into clinical knowledge.

According to the American Cancer Society, is the second leading cause of cancer death in American men, behind lung cancer. Surgery is a treatment option that generally yields a good prognosis, but 25 to 30 percent of men who have surgery will have a recurrence. Accurate risk stratification post-surgery is essential to identifying at intermediate to high risk for clinically significant disease progression. Those patients could benefit from more attentive monitoring and possibly additional therapy, such as radiation or chemotherapy.

Researchers at the Center for Computational and Systems Pathology at Mount Sinai used AI-guided machine learning techniques to analyze cancer tissue samples from 590 patients who underwent a radical prostatectomy, an operation to remove the prostate gland and the tissues surrounding it. The Precise MD platform relies on cutting-edge microscopy with multispectral immunofluorescence to analyze cancer tissue architecture and biomarkers, enabling pathologists to see what the human eye cannot. Its analysis uses mathematical features to define tumor aggressiveness.

Researchers at the Center for Computational and Systems Pathology discuss how computer vision and artificial intelligence are more accurately categorizing disease. Credit: Mount Sinai Health System

The Precise MD post-op test predicted significant disease progression with a greater degree of accuracy as compared with models that incorporated only clinical features such as the traditional Gleason score, or a Prostate-Specific Antigen (PSA) test; PSA is a substance produced by the prostate gland and elevated levels may indicate prostate cancer. Importantly, the Precise Post-op test reclassified 58 percent of intermediate risk patients as low risk and 42% as high risk for significant disease progression. Men identified as high risk by the Precise Post-op test may be appropriate candidates for additional monitoring and treatments, including chemotherapy and radiation.

"By refining diagnoses, we can guide patients toward the best treatment option and optimize care," said senior author Carlos Cordon-Cardo, MD, Ph.D., Chair of the Department of Pathology at the Mount Sinai Health System and Professor of Pathology, Genetics and Genomic Sciences, and Oncological Sciences at the Icahn School of Medicine.

"The ability to generate more quantitative, less subjective, and enhanced cancer-grading systems will bring precision medicine to the practicing pathologist and provide treating physicians and their patients important information for guiding management decisions," said lead author Michael Donovan, MD, Ph.D., Research Professor of Pathology at the Icahn School of Medicine at Mount Sinai.

"The opportunity to provide a more objective and comprehensive understanding of prostate cancer Gleason grading is especially important as our patients come from all over the world where practice patterns and expertise are often quite variable," said Ash Tewari, MBBS, MCh, Chair of the Department of Urology at the Mount Sinai Health System and the Kyung Hyun Kim, MD Professor of Urology at the Icahn School of Medicine. "A test that improves risk discrimination, specifically for durable response to adjuvant androgen therapy, represents a significant step towards providing a comprehensive integrated patient-focused care path. We are currently using the Precise Post-Op test for our intermediate and high risk patients to better understand the available parameters of risk assignment in the postsurgical setting where appropriate management planning is oftentimes challenging."

"Precision medicine is an innovative model of health care, and Mount Sinai is well positioned to provide our patients with more accurate diagnosis and tailored treatments," said Dennis S. Charney, MD, Anne and Joel Ehrenkranz Dean, Icahn School of Medicine at Mount Sinai, and President for Academic Affairs, Mount Sinai Health System. "Machine learning systems in prostate cancer grading provide a more objective measure of risk assessment."

The Henry Ford Hospital in Detroit and the Roswell Park Comprehensive Cancer Center in Buffalo, New York, provided patient samples for the study. Researchers say the Precise MD platform could be used to characterize any number of disease states, including but not limited to breast, melanoma, lung, and colon cancers as well as chronic inflammatory conditions such as inflammatory bowel disease.

"The introduction of machine learning systems in traditional grading represents an important step towards a more objective and biological reflection of personalized risk assignment," said Howard Soule, Ph.D., Executive Vice President and Chief Scientific Officer of the Prostate Cancer Foundation.

Explore further: New RNA and DNA-sequencing platform matches thousands of drugs to late-stage cancer patients

Related Stories

New RNA and DNA-sequencing platform matches thousands of drugs to late-stage cancer patients

August 8, 2018
A comprehensive RNA and DNA sequencing platform benefits late-stage and drug-resistant multiple myeloma patients by determining which drugs would work best for them, according to results from a clinical trial published in ...

PSA level three months after radiation tx may predict outcome

May 24, 2018
(HealthDay)—Prostate-specific antigen (PSA) levels three months after radiotherapy (RT) are strong markers of prostate cancer outcomes for patients with intermediate- and high-risk disease, according to a study published ...

Variations in key gene predict cancer patients' risk for radiation-induced toxicity

July 8, 2014
Key genetic variants may affect how cancer patients respond to radiation treatments, according to a study published this week in Nature Genetics. The research team, which included researchers at the Icahn School of Medicine ...

Common genetic fusion event may be associated with low-risk prostate cancer

November 10, 2017
Establishing the way in which a genetic alteration called a TMPRSS2-ERG gene fusion forms in a prostate cancer, rather than the presence of the gene fusion itself, could help identify patients with prostate cancer with a ...

Implications of very low risk prostate cancer assessed

September 20, 2013
(HealthDay)—The risk of adverse findings at surgery for men with very low risk (VLR) prostate cancer is significantly lower than for those with low risk (LR), according to research published in the October issue of the ...

Recommended for you

Potential seen for tailoring treatment for acute myeloid leukemia

December 8, 2018
Advances in rapid screening of leukemia cells for drug susceptibility and resistance are bringing scientists closer to patient-tailored treatment for acute myeloid leukemia (AML).

Study may offer doctors a more effective way to treat neuroblastoma

December 7, 2018
A very large team of researchers, mostly from multiple institutions across Germany, has found what might be a better way to treat patients with neuroblastoma, a type of cancer. In their paper published in the journal Science, ...

Major breakthrough in quest for cancer vaccine

December 6, 2018
The idea of a cancer vaccine is something researchers have been working on for over 50 years, but until recently they were never able to prove exactly how such a vaccine would work.

'Chemo brain' caused by malfunction in three types of brain cells, study finds

December 6, 2018
More than half of cancer survivors suffer from cognitive impairment from chemotherapy that lingers for months or years after the cancer is gone. In a new study explaining the cellular mechanisms behind this condition, scientists ...

Scientists develop new technology for profiling unique genetic makeup of myeloma tumor cells

December 6, 2018
Cancer arises when cells lose control. Deciphering the "blueprint" of cancer cells—outlining how cancer cells hijack specific pathways for uncontrolled proliferation—will lead to more efficient ways to fight it. Joint ...

Putting the brakes on tumor stealth

December 6, 2018
New research undertaken at Monash University has shed new light on how some cancers are able to escape our immune system.

0 comments

Please sign in to add a comment. Registration is free, and takes less than a minute. Read more

Click here to reset your password.
Sign in to get notified via email when new comments are made.