Greg O. Cron PhD, Presenter: Nothing to Disclose

Adnan Mohammad Sheikh MD, Abstract Co-Author: Nothing to Disclose

Ian Cameron, Abstract Co-Author: Nothing to Disclose

Mark E. Schweitzer MD, Abstract Co-Author: Nothing to Disclose

Jing Zhang MD, Abstract Co-Author: Nothing to Disclose

Xiao Guang Cheng MD, PhD, Abstract Co-Author: Nothing to Disclose

Joel Werier MD, Abstract Co-Author: Nothing to Disclose

Gina Anna Di Primio MD, Abstract Co-Author: Nothing to Disclose

On MR there is significant overlap in enhancement between malignant and benign masses. Therefore, specific parameters of contrast pharmacokinetics have been used, most commonly Ktrans, a method to assess blood flow and capillary permeability. Since the data acquisition and analysis methods used to compute Ktrans are usually specific to the site, vendor, and model, we applied a modified Ktrans (using kurtosis) which can be used across institutions.

86 patients (38 benign masses, 48 malignant tumors) were studied at 1.5T with histologic confirmation of diagnosis. Dynamic contrast-enhanced MRI was performed with inversion-prepared spoiled gradient echo imaging with matrix = 288 x 160, sixteen 6-mm-thick slices, TI=21 ms, TR=6.3 ms, TE=1.56 ms, readout flip = 20 degrees, temporal resolution = 20 s, total scan time = 4 minutes. Patient motion was corrected by an automatic sub-pixel registration algorithm. Pre-contrast T1 of the tumors was estimated by referencing their signal to muscle (T1 of muscle assumed to be 1000 ms), and a population AIF was utilized. Ktrans was then calculated pixel-by-pixel for each patient. For each patient, a region of interest was drawn around the tumor for all slices by an experienced MSK radiologist, thereby creating a distribution of Ktrans values for that tumor. The kurtosis of each distribution was calculated. Kurtosis values for benign vs malignant tumors were compared using a two-tailed student t-test with unequal variances.

Benign tumors included 17 giant cell tumors, 4 chondroblastomas, 3 desmoid tumors, 3 neurilemmomas, and 6 other. Malignant tumors included 12 metastases, 11 osteosarcomas, 8 chondrosarcomas, and 4 Ewing’s family tumors, and 9 other. The kurtosis values of malignant tumors were higher than those of the benign tumors (p=0.02).

We have successfully applied Ktrans kurtosis, a method which can be applied across imaging platforms, as a mesaure of biological aggressiveness (surrogate for blood flow / capillary perbeability), showing that this variable is greater in malignant musculoskeletal tumors. This now validates this technique to be used for multi-institution studies.

Ktrans kurtosis shows potential for distinguishing malignant from benign musculoskeletal tumors and can be used for multi-institution studies.

Cron, G, Sheikh, A, Cameron, I, Schweitzer, M, Zhang, J, Cheng, X, Werier, J, Di Primio, G, Ktrans Kurtosis: A Potential Surrogate for Biological Aggressiveness Suitable for Multi-institutional Musculoskeletal Trials.  Radiological Society of North America 2010 Scientific Assembly and Annual Meeting, November 28 - December 3, 2010 ,Chicago IL. http://archive.rsna.org/2010/9001719.html