Greybox: A hybrid algorithm for direct estimation of tracer kinetic parameters from undersampled DCE-MRI data.

BACKGROUND: A variety of deep learning-based and iterative approaches are available to predict Tracer Kinetic (TK) parameters from fully sampled or undersampled dynamic contrast-enhanced (DCE) MRI data. However, both the methods offer distinct benefits and drawbacks. PURPOSE: To propose a hybrid algorithm (named as 'Greybox'), using both model- as well as DL-based, for solving a multi-parametric non-linear inverse problem of directly estimating TK parameters from undersampled DCE MRI data, which is invariant to undersampling rate. METHODS: The proposed algorithm was inspired by plug-and-play algorithms used for solving linear inverse imaging problems. This technique was tested for its effectiveness in solving the nonlinear ill-posed inverse problem of generating 3D TK parameter maps from four-dimensional (4D; Spatial + Temporal) retrospectively undersampled k-space data. The algorithm learns a deep learning-based prior using UNET to estimate the K trans $\mathbf {K_{trans}}$ and V p $\mathbf {V_{p}}$ parameters based on the Patlak pharmacokinetic model, and this trained prior was utilized to estimate the TK parameter maps using an iterative gradient-based optimization scheme. Unlike the existing DL models, this network is invariant to the undersampling rate of the input data. The proposed method was compared with the total variation-based direct reconstruction technique on brain, breast, and prostate DCE-MRI datasets for various undersampling rates using the Radial Golden Angle (RGA) scheme. For the breast dataset, an indirect estimation using the Fast Composite Splitting algorithm was utilized for comparison. Undersampling rates of 8 × $\times$ , 12 × $\times$ and 20 × $\times$ were used for the experiments, and the results were compared using the PSNR and SSIM as metrics. For the breast dataset of 10 patients, data from four patients were utilized for training (1032 samples), two for validation (752 samples), and the entire volume of four patients for testing. Similarly, for the prostate dataset of 18 patients, 10 patients were utilized for training (720 samples), five for validation (216 samples), and the whole volume of three patients for testing. For the brain dataset of nineteen patients, ten patients were used for training (3152 samples), five for validation (1168 samples), and the whole volume of four patients for testing. Statistical tests were also conducted to assess the significance of the improvement in performance. RESULTS: The experiments showed that the proposed Greybox performs significantly better than other direct reconstruction methods. The proposed algorithm improved the estimated K trans $\mathbf {K_{trans}}$ and V p $\mathbf {V_{p}}$ in terms of the peak signal-to-noise ratio by up to 3 dB compared to other standard reconstruction methods. CONCLUSION: The proposed hybrid reconstruction algorithm, Greybox, can provide state-of-the-art performance in solving the nonlinear inverse problem of DCE-MRI. This is also the first of its kind to utilize convolutional neural network-based encodings as part of the plug-and-play priors to improve the performance of the reconstruction algorithm.

Higher Blood-brain barrier permeability in patients with major depressive disorder identified by DCE-MRI imaging.

BACKGROUND: Studies from animal models and clinical trials of blood and cerebrospinal fluid have proposed that blood-brain barrier (BBB) dysfunction in depression (MDD). But there are no In vivo proves focused on BBB dysfunction in MDD patients. The present study aimed to identify whether there was abnormal BBB permeability, as well as the association with clinical status in MDD patients using dynamic contrast-enhanced magnetic resonance (DCE-MRI) imaging. METHODS: Patients with MDD and healthy adults were recruited and underwent DCE-MRI and structural MRI scans. The mean volume transfer constant (Ktrans) values were calculated for a quantitative assessment of BBB leakage. For each subject, the mean Ktrans values were calculated for the whole gray matter, white matter, and 90 brain regions of the anatomical automatic labeling template (AAL). The differences in Ktrans values between patients and controls and between treated and untreated patients were compared. RESULTS: 23 MDD patients (12 males and 11 females, mean age 28.09 years) and 18 healthy controls (HC, 8 males and 10 females, mean age 30.67 years) were recruited in the study. We found that the Ktrans values in the olfactory, caudate, and thalamus were higher in MDD patients compared to healthy controls (p<0.05). The Ktrans values in the orbital lobe, anterior cingulate gyrus, putamen, and thalamus in treated patients were lower than the patients never treated. There were positive correlations between HAMD total score with Ktrans values in whole brain WM, hippocampus and thalamus. The total HAMA score was positively correlated with the Ktrans of hippocampus. CONCLUSION: These findings supported a link between blood-brain barrier leakage and depression and symptom severity. The results also suggested a role for non-invasive DCE-MRI in detecting blood-brain barrier dysfunction in depression patients.

Immune cell identity behind the Ktrans mapping of mouse glioblastoma.

Dynamic contrast-enhanced MR imaging (DCE-MRI) can assess the integrity of the blood brain barrier (BBB) and has been used in GBM patients to determine glioma grade, predict prognosis, evaluate treatment response, and differentiate treatment-induced effect from recurrence. The volume transfer constant Ktrans is the most frequently used metric in tumor assessment. Based on previous studies that a higher WHO grade of brain tumor was associated with greater impairments of immunity and that Ktrans value was associated with the pathological grading, the relationship between differential composition of immune cells in GBM tissue and dynamic changes in Ktrans mapping was anticipated in this study. The present study utilized an orthotopic allograft model of GBM in which mouse GL26 cells are implanted into Ccr2RFP/wtCx3cr1GFP/wt mice on a C57 background. The brain tumors exhibited heterogenous Ktrans values with the coefficients of variation (CV) above 75%, or relatively homogeneous Ktrans maps with CV values below 50%. The Ktrans values of homogeneous tumors ranged between 0.02/min-0.32/min with a median value of 0.10/min. The immune cell composition defined by quantitative immunohistochemistry and cell sorting was compared between the tumors with Ktrans values above 0.10/min (higher Ktrans) or below 0.10/min (lower Ktrans). Histological analysis showed that tumors with higher Ktrans values exhibited greater numbers of CCR2pos cells (257.60 ± 16.42/mm2 vs 203.23 ± 12.20/mm2, p = 0.04) and an increased ratio of CCR2pos cells to CX3CR1pos cells (1.20 ± 0.02 vs 0.38 ± 0.04, p = 0.001), the numbers of CX3CR1pos cells did not differ significantly based on Ktrans values (219.70 ± 16.20/mm2 vs 250.38 ± 21.20/mm2, p = 0.19). Flowcytometry analysis showed that tumors with higher Ktrans values (above 0.1/min) were associated with greater numbers of both overall monocytes (54.93 ± 6.81% vs 29.75 ± 3.54%, p = 0.01) and inflammatory monocytes (72.38 ± 1.49% vs 59.52 ± 2.44%, p = 0.001). In contrast, tumors with lower Ktrans values (below 0.1/min) exhibited greater numbers of patrolling monocytes (75.65 ± 4.14% vs 63 ± 6.94%, p = 0.05). In the tumors with lower Ktrans values, all three types of tumor associated cells, including patrolling monocytes, inflammatory monocytes, and microglia cells possessed a higher proportion of cells at pro-inflammatory status (41.77 ± 6.13% vs 25.06 ± 6.72%, p = 0.05; 27.50 ± 2.11% vs 20.62 ± 1.87%, p = 0.03; and 55.80 ± 9.88% vs 31.12 ± 7.31%, p = 0.05), inflammatory monocytes showed fewer anti-inflammatory cells (1.25 ± 0.62% vs 3.16 ± 3.56%, p = 0.04). Taken together, differences in Ktrans values were associated with differential immune cell phenotypes and polarizations. Ktrans mapping may therefore represent a novel approach for defining the immune status of GBM.

Comparison of DCE-MRI parametric mapping using MP2RAGE and variable flip angle T1 mapping.

Quantitative dynamic contrast enhanced magnetic resonance imaging (DCE-MRI) measures the rate of transfer of contrast agent from the vascular space to the tissue space by fitting signal-time data to pharmacokinetic models. However, these models are very sensitive to errors in T1 mapping. Accurate T1 mapping is necessary for high quality quantitative DCE-MRI studies. This study compares magnetization prepared rapid (two) gradient echo sequence (MP2RAGE) T1-mapping accuracy to the conventional variable flip angle (VFA) approach, and also determines the effect of the new T1-mapping method on the Ktrans parameter. VFA and MP2RAGE T1 values were compared to the gold standard inverse recovery (IR) method in phantom over manually drawn ROIs. In vivo, ROIs were manually drawn over prostate and prostatic lesions. Average T1 values over ROIs were compared and Ktrans maps for each method were calculated via the extended Tofts model. VFA-T1 maps overestimated T1 values by up to 50% compared to gold standard IR T1 values in phantom. MP2RAGE differed by up to 9%. MP2RAGE-T1 and Ktrans values were significantly different from VFA values over prostatic lesions (p < 0.05). Ktrans was consistently underestimated using VFA compared to MP2RAGE (p < 0.05). MP2RAGE T1 maps are shown to be more accurate, leading to more reliable pharmacokinetic modeling. This can potentially lead to better lesion characterization and improve clinical outcomes.

Blood-brain barrier dysfunction significantly correlates with serum matrix metalloproteinase-7 (MMP-7) following traumatic brain injury.

OBJECTIVES: To determine if radiological evidence of blood brain barrier (BBB) dysfunction, measured using Dynamic Contrast Enhanced MRI (DCE-MRI), correlates with serum matrix metalloproteinase (MMP) levels in traumatic brain injury (TBI) patients, and thereby, identify a potential biomarker for BBB dysfunction. PATIENTS AND METHODS: 20 patients with a mild, moderate, or severe TBI underwent a DCE-MRI scan and BBB dysfunction was interpreted from KTrans. KTrans is a measure of capillary permeability that reflects the efflux of gadolinium contrast into the extra-cellar space. The serum samples were concurrently collected and later analysed for MMP-1, -2, -7, -9, and -10 levels using an ELISA assay. Statistical correlations between MMP levels and the KTrans value were calculated. Multiple testing was corrected using the Benjamin-Hochberg method to control the false-discovery rate (FDR). RESULTS: Serum MMP-1 values ranged from 1.5 to 49.6 ng/ml (12 ± 12.7), MMP-2 values from 58.3 to 174.1 ng/ml (109.5 ± 26.7), MMP-7 from 1.5 to 31.5 ng/mL (10 ± 7.4), MMP-9 from 128.6 to 1917.5 ng/ml (647.7 ± 749.6) and MMP-10 from 0.1 to 0.6 ng/mL (0.3 ± 0.2). Non-parametric Spearman correlation analysis on the data showed significant positive relationship between KTrans and MMP-7 (r = 0.55, p < 0.01). Correlations were also found between KTrans and MMP-1 (r = 0.74, p < 0.0002) and MMP-2 (r = 0.5, p < 0.025) but the actual MMP values were not above reference ranges, limiting the interpretation of results. Statistically significant correlations between KTrans and either MMP-9 or -10 were not found. CONCLUSION: This is the first study to show a correlation between DCE measures and MMP values in patients with a TBI. Our results support the suggestion that serum MMP-7 may be considered as a peripheral biomarker quantifying BBB dysfunction in TBI patients.

Response to neoadjuvant chemoradiotherapy for locally advanced rectum cancer: Texture analysis of dynamic contrast-enhanced MRI.

BACKGROUND: Tumor heterogeneity can be assessed by texture analysis (TA). TA has been applied using diffusion-weighted imaging and apparent diffusion coefficient maps to predict pathological responses to preoperative chemoradiation therapy (CRT) in patients with locally advanced rectal cancer (LARC). PURPOSE: To evaluate the texture parameters obtained from K trans maps derived from dynamic contrast-enhanced (DCE)-MRI for predicting pathological responses to preoperative CRT for LARCs. STUDY TYPE: Retrospective. POPULATION: Altogether, 83 patients (26 women, 57 men) with rectal cancer met the inclusion criteria. FIELD STRENGTH/SEQUENCE: 3.0T/T1 -weighted DCE-MRI sequence. ASSESSMENT: After CRT, each tumor was assessed by a pathologist who assigned a tumor regression grade (TRG), thereby identifying pathologically complete responders (pCR; TRG 1) and good responders (GR; TRG1 + TRG2). TA was then applied to the DCE-MRI K trans maps. The K trans value, several TA parameters, and tumor volumes were calculated. STATISTICAL TESTS: The Shapiro-Wilk test was used to verify that the data had normal distribution. Results of parameters measured before and after CRT were compared using paired-sample t-tests. Value changes of each parameter in the combined pCR/GR group were compared using independent sample t-tests. Receiver operating characteristic curves and areas under the curve (AUC) were calculated to assess the diagnostic performance of each parameter related to CRT effectiveness. RESULTS: There were 15 pCR (16.9%) and 21 GR (25.3%) patients. Tumor volume, mean K trans , entropy, and correlation decreased and energy values increased significantly in these groups compared with those of the non-PCR and non-GR groups. ΔCorrelation (Δcorrelation = postcorrelation - precorrelation) was found to be a valuable parameter for identifying pCR/GR patients (AUC 0.895, sensitivity 86.7%, specificity 81.8%). DATA CONCLUSION: TA parameters from the DCE-MRI K trans map can predict the efficacy of CRT for treating LARCs. Also, Δcorrelation may be useful for identifying patients who will be responsive to CRT. LEVEL OF EVIDENCE: 4 Technical Efficacy: Stage 3 J. Magn. Reson. Imaging 2019;49:885-893.

Correlation of Preoperative Von Willebrand Factor with Magnetic Resonance Imaging Perfusion and Permeability Parameters as Predictors of Prognosis in Glioblastoma.

BACKGROUND: Angiogenesis has been shown to be strictly related to tumor malignancy. Glioblastoma (GBM) is highly vascularized and von Willebrand Factor (VWF) plays a potent proangiogenic role. Dynamic contrast-enhanced and dynamic susceptibility contrast magnetic resonance imaging (MRI) represent a widely accepted method to assess GBM microvasculature. Our aim was to investigate the correlation between plasma VWF:Ag, permeability, and perfusion MRI parameters and examine their potential in predicting GBM patient prognosis. METHODS: We retrospectively analyzed preoperative dynamic contrast-enhanced, dynamic susceptibility contrast MRI, and VWF:Ag level of 26 patients with GBM. We assessed the maximum values of relative cerebral blood flow and volume, volume transfer constant Ktrans, plasma volume (Vp) and reflux rate constant between fractional volume of the extravascular space and blood plasma (Kep). Nonparametric Mann-Whitney test and Kaplan-Meier survival analyses were conducted and a P value < 0.05 was considered statistically significant. RESULTS: The median VWF:Ag value was 248 IU/dL and the median follow-up duration was about 13 months. We divided patients according to low-VWF:Ag and high-VWF:Ag and we found significant differences in the median follow-up duration (19 months vs. 10 months; P = 0.04) and in Ktrans (0.31/minute vs. 0.53/minute; P = 0.02), and Kep (1.79/minute vs. 3.89/minute; P = 0.005) values. The cumulative 1-year survival was significantly shorter in patients with high-VWF:Ag and high-Kep compared with patients with low-VWF:Ag and low-Kep (37.5% vs. 68%; P = 0.05). CONCLUSIONS: These findings, in a small group of patients, suggest a role for VWF:Ag, similar to Ktrans, and Kep as a prognostic indicator of postoperative survival of patients with GBM.

Dynamic Contrast-Enhanced MRI.

Dynamic contrast-enhanced MRI in pre-clinical imaging allows the in-vivo monitoring of vascular, physiological properties in normal and diseased tissue. There is considerable variation in the methods employed owing to the different questions that can be asked and answered about the physiologic alterations as well as morphologic changes in tissue. Here we review the typical decisions in the design and execution of a dynamic contrast-enhanced MRI study in mice although the findings can easily be transferred to other species. Emphasis is placed on highlighting the many pitfalls that wait for the unaware pre-clinical MRI practitioner and that go often unmentioned in the abundant literature dealing with dynamic contrast-enhanced MRI in animal models.

Dynamic Contrast-Enhanced MR Imaging in Head and Neck Cancer.

Dynamic contrast-enhanced (DCE) MR imaging uses rapid sequential MR image acquisition before, during, and after intravenous contrast administration to elucidate information on the microvascular biologic function of tissues. The derived pharmacokinetic parameters provide useful information on tissue perfusion and permeability that may help to evaluate entities that otherwise appear similar by conventional imaging. When specifically applied to the evaluation of head and neck cancer, DCE-MR imaging may provide valuable information to help predict treatment response, discriminate between posttreatment changes and residual tumor, and discriminate between various head and neck neoplasms.

Early Prediction of Response to Neoadjuvant Chemotherapy Using Dynamic Contrast-Enhanced MRI and Ultrasound in Breast Cancer

OBJECTIVE: To determine the diagnostic performance of dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI) and DCE ultrasound (DCE-US) for predicting response to neoadjuvant chemotherapy (NAC) in breast cancer patients. MATERIALS AND METHODS: This Institutional Review Board-approved prospective study was performed between 2014 and 2016. Thirty-nine women with breast cancer underwent DCE-US and DCE-MRI before the NAC, follow-up DCE-US after the first cycle of NAC, and follow-up DCE-MRI after the second cycle of NAC. DCE-MRI parameters (transfer constant [Ktrans], reverse constant [kep], and leakage space [Ve]) were assessed with histograms. From DCE-US, peak-enhancement, the area under the curve, wash-in rate, wash-out rate, time to peak, and rise time (RT) were obtained. After surgery, all the imaging parameters and their changes were compared with histopathologic response using the Miller-Payne Grading (MPG) system. Data from minor and good responders were compared using Wilcoxon rank sum test, chi-square test, or Fisher's exact test. Receiver operating characteristic curve analysis was used for assessing diagnostic performance to predict good response. RESULTS: Twelve patients (30.8%) showed a good response (MPG 4 or 5) and 27 (69.2%) showed a minor response (MPG 1–3). The mean, 25th, 50th, and 75th percentiles of Ktrans and Kep of post-NAC DCE-MRI differed between the two groups. These parameters showed fair to good diagnostic performance for the prediction of response to NAC (AUC 0.76–0.81, p ≤ 0.007). Among DCE-US parameters, the percentage change in RT showed fair prediction (AUC 0.71, p = 0.023). CONCLUSION: Quantitative analysis of DCE-MRI and DCE-US was helpful for early prediction of response to NAC.

Reproducibility and relative stability in magnetic resonance imaging indices of tumor vascular physiology over a period of 24h in a rat 9L gliosarcoma model.

PURPOSE: The objective was to study temporal changes in tumor vascular physiological indices in a period of 24h in a 9L gliosarcoma rat model. METHODS: Fischer-344 rats (N=14) were orthotopically implanted with 9L cells. At 2weeks post-implantation, they were imaged twice in a 24h interval using dynamic contrast enhanced magnetic resonance imaging (DCE-MRI). Data-driven model-selection-based analysis was used to segment tumor regions with varying vascular permeability characteristics. The region with the maximum number of estimable parameters of vascular kinetics was chosen for comparison across the two time points. It provided estimates of three parameters for an MR contrast agent (MRCA): i) plasma volume (vp), ii) forward volumetric transfer constant (Ktrans) and interstitial volume fraction (ve, ratio of Ktrans to reverse transfer constant, kep). In addition, MRCA extracellular distribution volume (VD) was estimated in the tumor and its borders, along with tumor blood flow (TBF) and peritumoral MRCA flux. Descriptors of parametric distributions were compared between the two times. Tumor extent was examined by hematoxylin and eosin (H&E) staining. Picrosirus red staining of secreted collagen was performed as an additional index for 9L cells. RESULTS: Test-retest differences between population summaries for any parameter were not significant (paired t and Wilcoxon signed rank tests). Bland-Altman plots showed no apparent trends between the differences and averages of the test-retest measures for all indices. The intraclass correlation coefficients showed moderate to almost perfect reproducibility for all of the parameters, except vp. H&E staining showed tumor infiltration in parenchyma, perivascular space and white matter tracts. Collagen staining was observed along the outer edges of main tumor mass. CONCLUSION: The data suggest the relative stability of these MR indices of tumor microenvironment over a 24h duration in this gliosarcoma model.

Quantitative in vivo imaging of tissue factor expression in glioma using dynamic contrast-enhanced MRI derived parameters.

OBJECTIVE: Tissue Factor (TF) has been well established in angiogenesis, invasion, metastasis, and prognosis in glioma. A noninvasive assessment of TF expression status in glioma is therefore of obvious clinical relevance. Dynamic contrast-enhanced (DCE) MRI parameters have been used to evaluate microvascular characteristics and predict molecular expression status in tumors. Our aim is to investigate whether quantitative DCE-MRI parameters could assess TF expression in glioma. MATERIALS AND METHODS: Thirty-two patients with histopathologically diagnosed supratentorial glioma who underwent DCE-MRI were retrospectively recruited. Extended Tofts linear model was used for DCE-MRI post-processing. Hot-spot, whole tumor cross-sectional approaches, and histogram were used for analysis of model based parameters. Four serial paraffin sections of each case were stained with TF, CD105, CD34 and α-Sooth Muscle Actin, respectively for evaluating the association of TF and microvascular properties. Pearson correlation was performed between percentage of TF expression area and DCE-MRI parameters, multiple microvascular indexes. RESULTS: Volume transfer constant (Ktrans) hot-spot value best correlated with TF (r=0.886, p<0.001), followed by 90th percentile Ktrans value (r=0.801, p<0.001). Moreover, histogram analysis of Ktrans value demonstrated that weak TF expression was associated with less heterogeneous and positively skewed distribution. Finally, pathology analysis revealed TF was associated with glioma grade and significantly correlated with these two dynamic angiogenic indexes which could be used to explain the strong correlation between Ktrans and TF expression. CONCLUSION: Our results indicate that Ktrans may serve as a potential clinical imaging biomarker to predict TF expression status preoperatively in gliomas.

Differential diagnosis of oligodendroglial and astrocytic tumors using imaging results: the added value of perfusion MR imaging.

PURPOSE: The purposes of the present study are to assess whether different characteristics of oligodendrogliomas and astrocytic tumors are visible on MR imaging and to determine the added value of perfusion imaging in conventional MR imaging when differentiating oligodendrogliomas from astrocytic tumors. METHODS: We retrospectively studied 22 oligodendroglioma and 54 astrocytic tumor patients, including glioblastoma multiforme (GBM). The morphological tumor characteristics were evaluated using MR imaging. The rCBV, K trans, and V e values were recorded. All imaging and clinical values were compared. The ability to discriminate between the two entities was evaluated using receiver operating characteristic curve analyses. Separate comparison analysis between oligodendroglioma and astrocytic tumors excluding GBM was also performed. RESULTS: The presence of calcification, higher cortex involvement ratio, and lower V e value were more representative of oligodendrogliomas than astrocytic tumors (P = <0.001, 0.038, and <0.001, respectively). The area under the curve (AUC) value of a combination of calcification and cortex involvement ratio was 0.796. The combination of all three parameters, including V e, further increased the diagnostic performance (AUC = 0.881). Comparison test of the two AUC areas revealed significant difference (P = 0.0474). The presence of calcification and higher cortex involvement ratio were the only findings suggestive of oligodendrogliomas than astrocytic tumors with exclusion of GBMs (P = 0.014 and <0.001, respectively). CONCLUSION: Cortex involvement ratio and the presence of calcification with V e values were diagnostically accurate in identifying oligodendrogliomas. The V e value calculated from dynamic contrast-enhanced MR imaging could be a supportive tool for differentiating between oligodendrogliomas and astrocytic tumors including GBMs.

Effects of arterial input function selection on kinetic parameters in brain dynamic contrast-enhanced MRI.

PURPOSE: Kinetic parameters derived from dynamic contrast-enhanced MRI (DCE-MRI) were suggested as a possible instrument for multi-parametric lesion characterization, but have not found their way into clinical practice yet due to inconsistent results. The quantification is heavily influenced by the definition of an appropriate arterial input functions (AIF). Regarding brain tumor DCE-MRI, there are currently several co-existing methods to determine the AIF frequently including different brain vessels as sources. This study quantitatively and qualitatively analyzes the impact of AIF source selection on kinetic parameters derived from commonly selected AIF source vessels compared to a population-based AIF model. MATERIAL AND METHODS: 74 patients with brain lesions underwent 3D DCE-MRI. Kinetic parameters [transfer constants of contrast agent efflux and reflux Ktrans and kep and, their ratio, ve, that is used to measure extravascular-extracellular volume fraction and plasma volume fraction vp] were determined using extended Tofts model in 821 ROI from 4 AIF sources [the internal carotid artery (ICA), the closest artery to the lesion, the superior sagittal sinus (SSS), the population-based Parker model]. The effect of AIF source alteration on kinetic parameters was evaluated by tissue type selective intra-class correlation (ICC) and capacity to differentiate gliomas by WHO grade [area under the curve analysis (AUC)]. RESULTS: Arterial AIF more often led to implausible ve >100% values (p<0.0001). AIF source alteration rendered different absolute kinetic parameters (p<0.0001), except for kep. ICC between kinetic parameters of different AIF sources and tissues were variable (0.08-0.87) and only consistent >0.5 between arterial AIF derived kinetic parameters. Differentiation between WHO III and II glioma was exclusively possible with vp derived from an AIF in the SSS (p=0.03; AUC 0.74). CONCLUSION: The AIF source has a significant impact on absolute kinetic parameters in DCE-MRI, which limits the comparability of kinetic parameters derived from different AIF sources. The effect is also tissue-dependent. The SSS appears to be the best choice for AIF source vessel selection in brain tumor DCE-MRI as it exclusively allowed for WHO grades II/III and III/IV glioma distinction (by vp) and showed the least number of implausible ve values.

Response-Derived Input Function Estimation for Dynamic Contrast-Enhanced MRI Demonstrated by Anti-DLL4 Treatment in a Murine U87 Xenograft Model.

PURPOSE: Dynamic contrast-enhanced magnetic resonance imaging (DCE MRI) is an accepted method to evaluate tumor perfusion and permeability and anti-vascular cancer therapies. However, there is no consensus on the vascular input function estimation method, which is critical to kinetic modeling and K trans estimation. This work proposes a response-derived input function (RDIF) estimated from the response of the tumor, modeled as a linear, time-invariant (LTI) system. PROCEDURES: In an LTI system, an unknown input can be estimated from the system response. If applied to DCE MRI, this method would eliminate need of distal image-derived inputs, model inputs, or reference regions. The RDIF method first determines each tumor pixel's best-fit input function, and then combines the individual fits into a single input function for the entire tumor. The method was tested with simulations and a xenograft study with anti-vascular drug treatment. RESULTS: Simulations showed successful estimation of input function expected values and good performance in the presence of noise. In vivo, significant reductions in K trans and AUC occurred 2 days following anti-delta-like ligand 4 treatment. The in vivo study results yielded K trans consistent with published data in xenograft models. CONCLUSION: The RDIF method for DCE analysis offers an alternative, easy-to-implement method for estimating the input function in tumors. The method assumes that during the DCE experiment, the changes observed by MRI result solely from vascular perfusion and permeability kinetics, and that information can be used to model the input function. Importantly, the method is demonstrated in a murine xenograft study to yield K trans results consistent with literature values and suitable for compound studies.

Intra-individual comparison of different gadolinium-based contrast agents in the quantitative evaluation of C6 glioma with dynamic contrast-enhanced magnetic resonance imaging.

This experiment aimed to compare the ionic (Gadodiamide, Gd-DTPA-BMA) and non-ionic (Gadopentetate dimeglumine, Gd-DTPA) gadolinium-based contrast agents (GBCA) in the quantitative evaluation of C6 glioma with dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI). A C6 glioma model was established in 12 Wistar rats, and magnetic resonance (MR) scans were performed six days after tumor implantation. Imaging was performed using a 3.0-T MR scanner with a 7-inch handmade circular coil. Pre-contrast T1 mapping and dynamic contrast-enhanced T1WI after a bolus injection (0.2 mL s-1) of GBCA at 0.4 mmol kg-1 were performed. Each rat received two DCE-MRI scans, 24 h apart. The first and second scans were performed using Gd-DTPA-BMA and Gd-DTPA, respectively. Image data were processed using the Patlak model. Both K trans and V p maps were generated. Tumors were manually segmented on all 3D K trans and V p maps. Pixel counts and mean values were recorded for use in a paired t-test. Three radiologists independently performed the tumor segmentation and value calculation. The agreements from different observers were subjective to the intra-class correlation coefficient (ICC). Readers demonstrated that the pixel counts of tumors in K trans maps were higher with Gd-DTPA-BMA than with Gd-DTPA (P<0.001, all readers). Although the K trans values were higher with Gd-DTPA-BMA than with Gd-DTPA, there was no statistical significance (P>0.05, all readers). The pixel counts of tumors in V p maps, as well as V p values, showed no obvious difference between the two agents (P>0.05, all readers). Excellent interobserver measurement reproducibility and reliability were demonstrated in the ICC tests. The Gd-DTPA-BMA contrast agent had significantly higher pixel counts of glioma in the K trans maps, and an increased tendency for average K trans values, indicating that DCE-MRI with Gd-DTPA-BMA may be more suitable and sensitive for the evaluation of glioma.

Dynamic Contrast-Enhanced Magnetic Resonance Imaging as a Pharmacodynamic Biomarker for Pazopanib in Metastatic Renal Carcinoma.

INTRODUCTION/BACKGROUND: Traditional imaging assessment criteria might not correlate well with clinical benefit from vascular endothelial growth factor pathway-directed therapy in metastatic renal cancer. Preclinical data suggest tumor growth is preceded by a rise in Ktrans level, a parameter derived from dynamic contrast-enhanced (DCE) magnetic resonance imaging (MRI) that reflects vascular permeability. We thus hypothesized that Ktrans might be a predictive biomarker for pazopanib. PATIENTS AND METHODS: Patients with metastatic renal cancer were treated with pazopanib at 800 mg oral daily until disease progression. MRI of the abdomen and pelvis with a DCE-MRI sequence was obtained at baseline and every 8 weeks. RESULTS: Seventy-three DCE-MRI scans were completed and 66 were technically assessable. Of the 17 patients with at least 1 DCE-MRI scan after the baseline scan, 16 (94%) had a decline in Ktrans level. Changes in Ktrans compared with baseline after 1, 8, 16, and 24 weeks were -49%, -65%, -63%, and -53%, respectively (P = .0052, repeated measures analysis of variance). The median Ktrans nadir occurred at 8 weeks. The median progression-free survival (PFS) was 32.1 weeks. PFS was longer in patients with higher baseline Ktrans values (P = .036, log rank). Baseline Ktrans did not reach significance in a Cox proportional hazard model including clinical prognostic index and previous treatments (P = .083). CONCLUSION: We show that Ktrans is a pharmacodynamic biomarker for pazopanib therapy in metastatic renal cancer. Because of the small sample size, the predictive capacity of Ktrans recovery could not be assessed, but baseline Ktrans correlated with PFS.

The quantification of blood-brain barrier disruption using dynamic contrast-enhanced magnetic resonance imaging in aging rhesus monkeys with spontaneous type 2 diabetes mellitus.

Microvascular lesions of the body are one of the most serious complications that can affect patients with type 2 diabetes mellitus. The blood-brain barrier (BBB) is a highly selective permeable barrier around the microvessels of the brain. This study investigated BBB disruption in diabetic rhesus monkeys using dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI). Multi-slice DCE-MRI was used to quantify BBB permeability. Five diabetic monkeys and six control monkeys underwent magnetic resonance brain imaging in 3 Tesla MRI system. Regions of the frontal cortex, the temporal cortex, the basal ganglia, the thalamus, and the hippocampus in the two groups were selected as regions of interest to calculate the value of the transport coefficient Ktrans using the extended Tofts model. Permeability in the diabetic monkeys was significantly increased as compared with permeability in the normal control monkeys. Histopathologically, zonula occludens protein-1 decreased, immunoglobulin G leaked out of the blood, and nuclear factor E2-related factor translocated from the cytoplasm to the nuclei. It is likely that diabetes contributed to the increased BBB permeability.

Role of pharmacokinetic parameters derived with high temporal resolution DCE MRI using simultaneous PET/MRI system in breast cancer: A feasibility study.

PURPOSE: To evaluate the reliability of pharmacokinetic parameters like Ktrans, Kep and ve derived through DCE MRI breast protocol using 3T Simultaneous PET/MRI (3Tesla Positron Emission Tomography/Magnetic Resonance Imaging) system in distinguishing benign and malignant lesions. MATERIALS AND METHODS: High temporal resolution DCE (Dynamic Contrast Enhancement) MRI performed as routine breast MRI for diagnosis or as a part of PET/MRI for cancer staging using a 3T simultaneous PET/MRI system in 98 women having 109 breast lesions were analyzed for calculation of pharmacokinetic parameters (Ktrans, ve, and Kep) at 60s time point using an in-house developed computation scheme. RESULTS: Receiver operating characteristic (ROC) curve analysis revealed a cut off value for Ktrans, Kep, ve as 0.50, 2.59, 0.15 respectively which reliably distinguished benign and malignant breast lesions. Data analysis revealed an overall accuracy of 94.50%, 79.82% and 87.16% for Ktrans, Kep, ve respectively. Introduction of native T1 normalization with an externally placed phantom showed a higher accuracy (94.50%) than without native T1 normalization (93.50%) with an increase in specificity of 87% vs 84%. CONCLUSION: Overall the results indicate that reliable measurement of pharmacokinetic parameters with reduced acquisition time is feasible in a 3TMRI embedded PET/MRI system with reasonable accuracy and application may be extended to exploit the potential of simultaneous PET/MRI in further work on breast cancer.

Permeability Imaging as a Biomarker of Leptomeningeal Collateral Flow in Patients with Intracranial Arterial Stenosis.

Different methods of angiography are of great clinical utility; however, it still remains unstandardized as which method would be suitable to determine cerebral collateral circulation. Here we compared digital subtraction angiography (DSA), computer tomography angiography (CTA) and dynamic contrast-enhanced T1-weighted imaging magnetic resonance imaging (MRI) findings in seven patients with severe intracranial arterial stenosis, and determine whether volume transfer constant (K(trans)) maps of permeability imaging could be used as the biomarkers of cerebral collateral circulation. We retrospectively reviewed seven adult patients with severe intracranial arterial stenosis or occlusion with a complete parenchymal and vascular imaging work-up. DSA, CTA source imaging (CTA-SI), arterial spin labeling (ASL), and K(trans) maps were used to assess their collateral flow. Cohen's Kappa coefficient was calculated to test the consistency of their collateral scores. A reasonable agreement was found between DSA and K(trans) maps (Kappa = 0.502, P < 0.001) when all 15 regional vascular sites were included, and a better agreement found after exclusion of perforating artery territories (N = 10 sites, Kappa = 0.766, P < 0.001). The agreement between CTA-SI and DSA was moderate on all 15 sites (Kappa = 0.413, P < 0.001) and 10 sites (Kappa = 0.329, P < 0.001). The agreement between ASL and DSA was least favorable, no matter for all 15 sites (Kappa = 0.270, P < 0.001) or 10 sites (Kappa = 0.205, P = 0.002). K(trans) maps are useful and promising for leptomeningeal collateral assessment, when compared to CTA-SI or ASL. Further studies are requited for verify its validity in a large registry of patients.