Estimation of Contrast Agent Concentration in DCE-MRI Using 2 Flip Angles

Abstract

Purpose: The aim of this study was to investigate the feasibility of using 2 flip angles (FAs) with an ultrashort echo time during dynamic contrast-enhanced (DCE)-magnetic resonance imaging (MRI) for estimation of plasma gadolinium (Gd) concentration without using a precontrast longitudinal relaxation time T1 (T10) measurement.

Methods: T1-weighted DCE-MRI experiments were carried out with C57BL/6J mice using the scan protocol with 2 FAs over 3 sequential segments during 1 scan. The data with 2 FAs were used to estimate T10 (T1T) during conversion of a time-intensity curve to the time-concentration curve. Three dosages of gadolinium-based contrast agent were used to achieve a wide range of variability in Gd concentrations when measured at 10 minutes postinjection: 0.05 mmol/kg (n = 6), 0.1 mmol/kg (n = 11), and 0.15 mmol/kg (n = 7). For comparison, the signal-to-concentration conversion was also conducted using the T10 measured from the precontrast scan (T1M) as well as a constant T10 (2.1 seconds) from the literature (T1C). The Gd concentrations ([Gd]) estimated using DCE-MRI data for the time of retro-orbital blood collection ([Gd]T1T, [Gd]T1M, and [Gd]T1C, respectively) were compared against the [Gd] of the blood samples measured by inductively coupled plasma mass spectrometry ([Gd]MS). In addition, contrast kinetic model analysis was conducted on mice with GL261 brain tumors (n = 5) using the 3 different methods for T10.

Results: T1T strongly correlated with T1M (r = 0.81). [Gd]T1M and [Gd]T1T were significantly different from [Gd]T1C. [Gd]T1M and [Gd]T1T were in good agreement with [Gd]MS with strong correlations (mean percentage error ± standard deviation) of r = 0.70 (16% ± 56%) and r = 0.85 (15% ± 44%), respectively. In contrast, [Gd]T1C had a weak correlation of r = 0.52 with larger errors of 33% ± 24%. The contrast kinetic model parameters of GL261 brain tumors using T1T were not significantly different from those using T1M.

Conclusions: This study substantiates the feasibility of using the 2-FA approach during DCE-MRI scan to estimate [Gd] in the plasma without using an extra scan to perform precontrast T1 measurements.

Figure 1.

Workflow of the experiment. The schematic describes the experimental workflow for pre-contrast T1 mapping, DCE-MRI, blood sampling and ICP-MS. It also shows how the three different methods are used to determine pre-contrast T₁ values of vascular voxels; T_1T from the DCE-MRI data with two flip angles, T_1M from the 3D T₁ mapping prior to contrast injection, and T_1C as a constant value fixed to the average of T_1T and T_1M values of all mice. These three pre-contrast T₁ values were used to convert the signal intensity curves of DCE-MRI to three Gd concentration curves, [Gd]_T1T, [Gd]_T1M, and [Gd]_T1C, respectively. The [Gd] data, extrapolated to 90s after the end of the scan, were compared with the [Gd] measured from the blood sampling using ICP-MS, [Gd]_MS.

Figure 2.

Representative pre-contrast T₁ maps of one C57BL/6J mouse measured using the 3D-UTE-GRASP sequence. (A) 3D T₁-weighted images of a mouse brain with flip angle of 8° in axial/coronal/sagittal center slices. The T₁ maps (B) and 3D rendering of the whole mouse head T₁ map (C) with the isotropic resolution same as the DCE-MRI data.

Figure 3.

Selection of vascular voxels for the arterial input function was based on the initial area under the curve (IAUC). (A) Representative 3D maximum intensity projection (MIP) rendering of DCE-MR images at baseline (30 s), the peak enhancement (75 s), high flip angle (300 s), and washout end (610 s), with the red spots indicating the location of the selected 10 vessels. (B) The 10 selected vessel pixels (red) used for AIF shown on axial slices of the 3D image at the peak enhancement of AIF.

Figure 4.

Conversion of signal intensity to [Gd] using the top 10 most enhanced voxels (shown in Figure 2) from the IAUC-based selection. (A) Signal enhancement curves of the selected voxels. (B) [Gd] estimated from the signal enhancement curves (A) using T_1C. Since the T_1C value assumed in this study is close to the actual value, the [Gd]_T1C curves do not show distinct discontinuities between different flip angle segments, unlike the signal enhancement curves in (A). (C) [Gd] curve estimated from the signal enhancement curves (A) using T_1M. The smooth curves without sudden changes between the segments with different FAs support the validity of T_1M. (D) [Gd] curves estimated from the signal enhancement curves (A) using T_1T. (E) Comparison of three [Gd] curves using the three different methods to determine the pre-contrast T₁ values. Median concentrations obtained using the median values from (B)-(D). All signal enhancement curves overlaid with a dashed box over the tail end. (F) A zoomed-in image for the tail end of the signal enhancement curves. The extrapolated concentrations (700 s) were used to cross validation with ICP-MS method.

Figure 5.

Comparison of the T₁₀ values estimated for the vascular voxels by IAUC selection using T_1T (based on the two-flip-angle DCE-MRI data) and T_1M (based on the pre-contrast T₁ mapping), for the 24 mice. The red solid line is the regression line with correlation coefficient of 0.81. The gray dash line is the unity line. The filled red star indicates the fixed T₁, T_1C=2.1 s.

Figure 6.

Comparison of the MRI-based [Gd] estimates and ICP-MS measured [Gd] values. (A-C) The top row shows the scatter plots of [Gd]_T1C, (B) [Gd]_T1M, and [Gd]_T1T against the reference standard [Gd] _MS for all 24 mice. The red lines are regression lines. The gray dashed lines are unity lines. (D-F) The bottom row shows the Bland-Altman plots of the MRI-based [Gd] values vs [Gd]_MS. The dotted lines are for the lines of agreement. The red lines show the bias of the plot.

Figure 7.

Application of the two-flip angle method in analysis of tumor DCE-MRI data. (A) Representative contrast kinetic parameter maps for one mouse bearing a GL261 tumor, based on [Gd] estimates of the tumor voxels using the three T₁₀ estimation methods (T_1C, T_1M, and T_1T). (B-G) Comparisons of the median contrast kinetic parameters of the tumors estimated using the [Gd] estimates based on the three T₁₀ estimation methods. “*” Indicates a significant difference (p < 0.05 using the paired t-test) between the parameters and the ones using T_1T.