Quantitative MR model may shed light on prostate cancer aggressiveness

Twitter icon
Facebook icon
LinkedIn icon
e-mail icon
Google icon
69-year-old man with Gleason 7 (3 + 4) prostate cancer in the right midgland PZ at transrectal biopsy. A, Axial T2-weighted turbo spin-echo image shows area of decreased T2 signal within this sextant (solid arrow). B, ADC map, constructed by using a standard Gaussian model, and, C, corrected diffusion coefficient, or D, map, constructed by using a non-Gaussian kurtosis model, both show decreased signal intensity in this region (solid arrows).
Source: Radiology, published online May 1, 2012

A diffusional kurtosis (DK) model may help differentiate benign regions from malignant prostate cancer and distinguish low- from high-grade regions, according to a study published online May 1 in Radiology.

Prostate cancer presents multiple clinical challenges. Heightened awareness of the indolent nature of the disease in many cases and multiplying treatment options, coupled with the limitations of PSA testing and ultrasound-guided biopsy, have created the need for improved detection and evaluation.

Although diffusion-weighted (DW) MR imaging has shown promise as a marker of tumor aggressiveness and apparent diffusion coefficients (ADCs) may predict prostate cancer progression, there is room for improvement in the evaluation of prostate cancer. Specifically, DW imaging assumes Gaussian behavior of water diffusion, which may not provide full details of the disease.

DK imaging, which leverages advanced postprocessing, may overcome some limitations of DW MR by accounting for non-Gaussian water diffusion. DK is expected to increase with tumor aggressiveness, wrote Andrew B. Rosenkrantz, MD, from the department of radiology at New York University Langone Medical Center in New York City, and colleagues.

The researchers sought to determine the feasibility of the DK model and examine its potential role. They devised a retrospective analysis and reviewed imaging data of 41 patients who underwent 3T DW MR between December 2010 and February 2011.

The researchers post-processed DW imaging data, reviewed biopsy findings and generated parametric maps for kurtosis metrics for all patients.

Analysis of the receiver operating characteristics revealed statistically significant greater sensitivity for kurtosis at 93.3 percent than for ADC or diffusion, at 78.5 percent and 83.5 percent, respectively.  

The researchers also reported that D, a diffusion coefficient corrected for non-Gaussian diffusion behavior, was nearly double the value of standard ADC, and DK was greater in cancerous sextants than benign areas and further elevated in sextants with a higher Gleason score.

Although ADCs and Gleason scores have been shown to correlate with each other, Rosenkrantz et al emphasized the importance of metrics beyond ADCs and Gleason scores, as previous studies have shown a large degree of overlap in ADCs between tumors of various grades.

“The greater relative contrast, as well as increased sensitivity for cancerous sextants, by using DK imaging metrics suggests a potential clinical advantage for use of the DK imaging model when incorporating diffusion techniques into prostate imaging protocols,” wrote Rosenkrantz and colleagues.