Quantitative MR-perfusion and -diffusion imaging in a case study of CNS-lymphomas under therapy

Introduction: With improved treatment of lymphoma more specialized imaging techniques for monitoring the follow-up are sought for. Dynamic contrast enhanced (DCE) MRI provides detailed information of the integrity of the blood brain barrier, which is severely disrupted in untreated CNS lymphoma. CNS lymphoma are known to show a restricted diffusion behaviour caused by a high cellularity of the tumors. The aim of this case study was to get multiparametric information of primary CNS lymphoma (PCNSL) during treatment by extracting perfusion and diffusion characteristics amongst others. Three patients (A, B, C) with different location of CNS-lymphoma (A: splenium of the corpus callosum, B: frontal white matter, C: frontal grey/white matter junction) have been under investigation for 10 (A), 4 (B) and 14 (C) months.

Material and Methods: Three patients with different locations of a primary CNS lymphoma have been examined before and during therapy. Using a 1.5T MR-scanner (Sonata, Siemens Medical Solutions) a protocol was used including T1 weighted DCE-MRI, apparent diffusion constant (ADC) maps and chemical shift imaging (CSI) spectroscopy (not shown here). For the DCE-MRI an Inversion-Recovery (IR) TrueFISP sequence was used to calculate T1-maps every 5 seconds during 6 minutes [1]. Hence contrast agent concentration values in tissue were calculated [2] and the variation in time could be analysed for each pixel. DCE-MRI data analysis was performed using a custom-built software package developed under Matlab. To evaluate the perfusion of the lesions, maps were calculated concerning the initial area under concentration curve for the first 60 seconds after the bolus reached the tissue (iAUC60). A region of interest (ROI) analysis was chosen to evaluate the follow-up of the particular lesions.

Results: Maps of iAUC60, of the T1-rate before contrast agent administration and ADC, displaying the heterogenous diffusion, are exemplarily shown for patient C pre-treatment. All tumors presented high local iAUC60-values before treatment. Under therapy a decrease of iAUC60 was observed in all cases. Two tumors showed a rapid and one a slow decrease. The characteristic of the diffusion constant was particularly dissimilar. One lesion showed a slow increasing diffusion over the whole period of observation. In contrast another lesion showed a peak at the second examination.

Discussion: The different behaviour of iAUC60 and ADC of the lymphoma seems to reflect a different pathophysiological origin. One could imagine that the high iAUC60 before therapy might be attributed to a pathological capillary constitution. The increases of ADC over the time are thought to reflect an increase of extracellular space.

References: [1] Scheffler K., MRM 49:781–783 (2003), [2] Rohrer M., Invest. Radiol. 40:715–724 (2005)