Algorithm, MRI images eliminate need for kidney cancer biopsies

Register now

Researchers at UT Southwestern Harold C. Simmons Comprehensive Cancer Center have developed an algorithm used with current magnetic resonance imaging (MRI) technology that can help physicians determine whether kidney tumors are benign or malignant without having to perform tissue biopsies.

Typically, a diagnosis of kidney cancer cannot be positively made without looking at some of the tumor under a microscope. To confirm that it is cancer, a sample of tissue is removed by an interventional radiologist through a biopsy, an invasive procedure that is painful and not without medical risks.

However, investigators at UT Southwestern’s Kidney Cancer Program have successfully used a standardized diagnostic algorithm to evaluate targeted renal masses on specific MRI images, enabling doctors to recognize clear cell carcinoma (ccRCC)—the most common and aggressive form of kidney cancer—with 80 percent accuracy.

“What we have tested in this study is a predefined algorithm on how to interpret the images,” says Ivan Pedrosa, MD, professor of radiology and chief of MRI at UT Southwestern. “We actually tested the performance of seven different radiologists using that algorithm to predict the histology.”

Also See: University of Texas supercomputer speeds real-time MRI analysis

Pedrosa and Jeffrey Cadeddu, MD, professor of urology and radiology at UT Southwestern, co-authored a study published in The Journal of Urology on the diagnostic accuracy of multiparametric MRI (mpMRI) protocols allowing investigators to assess the chemical composition of tumors without a biopsy.

“A clear cell likelihood score used with magnetic resonance imaging can reasonably identify clear cell histology in small renal masses and may decrease the number of diagnostic renal mass biopsies,” concludes the article.

“Using mpMRI, multiple types of images can be obtained from the renal mass and each one tells us something about the tissue,” adds Pedrosa, who notes that T2-weighted images and those immediately after intravenous dye reaches the kidney are leveraged, as well as other images that indicate whether fat is present in the tumor.

According to Pedrosa, the research team at UT Southwestern is currently working on further improving the performance of the algorithm by introducing artificial intelligence in the process. So far, he concludes that they have achieved a “very crude form of artificial intelligence.”

Pedrosa and Cadeddu are supported by a Specialized Program of Research Excellence (SPORE) award from the National Cancer Institute, one of two such awards for kidney cancer in the country.

For reprint and licensing requests for this article, click here.