Supplementary Fig. S1. Schematic of study design.

Supplementary tables S1-3 and figure legends. Supplementary Table S1. Antibodies and antibody dilution used for immunohistochemical analyses. Supplementary Table S2. Characterization of human HCC Supplementary Table S3. Number of tumors per analyzed animal of DEN and McA model.

Supplementary Figure 1. Summary and survival of murine cohort used in the study. Supplementary Figure 2 Consort diagram of the patient distribution of surgical subcohort 1. Supplementary Figure 3. Consort diagram of the patient distribution of surgical subcohort 2. Supplementary Figure 4. Consort diagram of the patient distribution of surgical subcohort 3. Supplementary Figure 5. Schematic of relative tumor cell and stroma contribution in PDAChigh Supplementary Table 1. Statistics of the patients used in surgical subcohort 1. Supplementary Table 2. Statistics of the patients used in surgical subcohort 2. Supplementary Table 3. Statistics of the patients used in surgical subcohort 3. Supplementary Table 4. Results of a multivariable Cox regression model fit to the data of subcohort 2 that assess whether PDAC is associated with overall survival after adjustment for common predictors. Supplementary Table 5. The permutation test analysis revealing association between overall survival and lowest ADC stratified by each of the relevant categorical variables. Data used for analysis are derived from surgical subcohort 3.

Supplementary Fig. S2. Macrophage and T-lymphocyte infiltration in DEN and McA animals.

Supplementary Fig. S4. Tumor vascularization and necrosis in human HCC.

Abstract Purpose: Tumor heterogeneity is a hallmark of pancreatic ductal adenocarcinoma (PDAC). It determines tumor biology including tumor cellularity (i.e., amount of neoplastic cells and arrangement into clusters), which is related to the proliferative capacity and differentiation and the degree of desmoplasia among others. Given the close relation of tumor differentiation with differences in progression and therapy response or, e.g., the recently reported protective role of tumor stroma, we aimed at the noninvasive detection of PDAC groups, relevant for future personalized approaches. We hypothesized that histologic differences in PDAC tissue composition are detectable by the noninvasive diffusion weighted- (DW-) MRI-derived apparent diffusion coefficient (ADC) parameter. Experimental design: PDAC cellularity was quantified histologically and correlated with the ADC parameter and survival in genetically engineered mouse models and human patients. Results: Histologic analysis showed an inverse relationship of tumor cellularity and stroma content. Low tumor cellularity correlated with a significantly prolonged mean survival time (PDAClow = 21.93 months vs. PDACmed = 12.7 months; log-rank P < 0.001; HR = 2.23; CI, 1.41–3.53). Multivariate analysis using the Cox regression method confirmed tumor cellularity as an independent prognostic marker (P = 0.034; HR = 1.73; CI, 1.04–2.89). Tumor cellularity showed a strong negative correlation with the ADC parameter in murine (r = −0.84; CI, −0.90– −0.75) and human (r = −0.79; CI, −0.90 to −0.56) PDAC and high preoperative ADC values correlated with prolonged survival (ADChigh = 41.7 months; ADClow = 14.77 months; log rank, P = 0.040) in PDAC patients. Conclusions: This study identifies high tumor cellularity as a negative prognostic factor in PDAC and supports the ADC parameter for the noninvasive identification of PDAC groups. Clin Cancer Res; 23(6); 1461–70. ©2016 AACR.

Supplementary Fig. S3. aCGH analyses.

OBJECTIVES:To evaluate the diagnostic value of MR-derived CT-like images and simulated radiographs compared with conventional radiographs in patients with benign and malignant bone tumors.METHODS:In 32 patients with a benign or malignant bone lesion (mean age 33.9 ± 18.5 years, 17 females), 3-T MR imaging was performed including a 3D T1-weighted gradient echo sequence as the basis for the CT-like images. From these, intensity-inverted MR image volumes were converted into 2D images via a forward projection to obtain simulated radiographs. Two radiologists assessed these images as well as conventional radiographs for the type of periosteal reaction, matrix mineralization and destruction pattern. Agreement between the modalities was calculated using Cohen's κ.RESULTS:The agreement between conventional radiographs and MR-derived CT-like images in combination with simulated radiographs was substantial (periosteal reaction, κ = 0.67; destruction pattern, κ = 0.75), and the sensitivity of both modalities for the final diagnosis of the lesion (aggressive vs. nonaggressive) was high (MR-derived CT-like images, 86.2% vs. conventional radiographs, 90.0%). Additional information on soft tissue extension (MR-derived CT-like images, 21.9% vs. conventional radiographs, 12.5%; p = 0.009) and lobulation (9.4% vs. 0%; p < 0.001) was significantly more often found on MR-derived CT-like images compared with conventional radiographs.CONCLUSIONS:The assessment of the destruction patterns, periosteal reaction and distinction between aggressive and nonaggressive tumors was feasible using MR-derived CT-like images and simulated radiographs and is comparable to that of conventional radiographs. Moreover, MR-derived CT-like images provided additional information on soft tissue extension and tumor architecture.KEY POINTS:• CT-like images and simulated radiographs can be generated from 3D MRI. • Evaluation of bone tumors is feasible with MR-derived images. • CT-like images and simulated radiographs provide additional information on bone tumors.

To prospectively compare image quality and myocardial T1 relaxation times of modified Look-Locker inversion recovery (MOLLI) imaging at 3.0 T (T) acquired with patient-adaptive dual-source (DS) and conventional single-source (SS) radiofrequency (RF) transmission. Pre- and post-contrast MOLLI T1 mapping using SS and DS was acquired in 27 patients. Patient wise and segment wise analysis of T1 times was performed. The correlation of DS MOLLI measurements with a reference spin echo sequence was analysed in phantom experiments. DS MOLLI imaging reduced T1 standard deviation in 14 out of 16 myocardial segments (87.5%). Significant reduction of T1 variance could be obtained in 7 segments (43.8%). DS significantly reduced myocardial T1 variance in 16 out of 25 patients (64.0%). With conventional RF transmission, dielectric shading artefacts occurred in six patients causing diagnostic uncertainty. No according artefacts were found on DS images. DS image findings were in accordance with conventional T1 mapping and late gadolinium enhancement (LGE) imaging. Phantom experiments demonstrated good correlation of myocardial T1 time between DS MOLLI and spin echo imaging. Dual-source RF transmission enhances myocardial T1 homogeneity in MOLLI imaging at 3.0 T. The reduction of signal inhomogeneities and artefacts due to dielectric shading is likely to enhance diagnostic confidence.

New therapeutic regimens for otherwise lethal pancreatic ductal adenocarcinoma (PDAC) are rapidly emerging. Thus, imaging-based biomarkers that would allow early assessment of therapy response would be of great need.

PurposeTo study the effects of refocusing angle modulation with 3D turbo spin echo (TSE) on signal and sharpness of small oblique nerves embedded in muscle and suppressed fat in the lumbar plexus.Materials and MethodsFlip angle trains were generated with extended phase graphs (EPG) for a sequence parameter subspace. Signal loss and width broadening were simulated for a single-pixel nerve embedded in muscle and suppressed fat to prescribe a flip angle modulation that gives the best compromise between signal and sharpness of small nerves. Two flip angle trains were defined based on the simulations of small embedded nerves: design denoted A, predicting maximum global signal, and design denoted B, predicting maximum signal for minimum width broadening. In vivo data of the lumbar plexus in 10 healthy volunteers was acquired at 3.0T with 3D TSE employing flip angle trains A and B. Quantitative and qualitative analyses of the acquired data were made to assess changes in width and signal intensity.ResultsChanging flip angle modulation from A to B resulted in: 1) average signal losses of 23% in (larger) L5 nerves and 9% in (smaller) L3 nerves; 2) average width reductions of 4% in L5 nerves and of 16% in L3 nerves; and 3) statistically significant sharpness improvement (P=0.005) in L3 nerves.ConclusionAn optimized flip angle train in 3D TSE imaging of the lumbar plexus considering geometry-specific blurring effects from both the nerve and the surrounding tissue can improve the delineation of small nerves. J. Magn. Reson. Imaging 2016;43:789-799

Purpose: Tumor heterogeneity is a hallmark of pancreatic ductal adenocarcinoma (PDAC). It determines tumor biology including tumor cellularity (i.e., amount of neoplastic cells and arrangement into clusters), which is related to the proliferative capacity and differentiation and the degree of desmoplasia among others. Given the close relation of tumor differentiation with differences in progression and therapy response or, e.g., the recently reported protective role of tumor stroma, we aimed at the noninvasive detection of PDAC groups, relevant for future personalized approaches. We hypothesized that histologic differences in PDAC tissue composition are detectable by the noninvasive diffusion weighted- (DW-) MRI-derived apparent diffusion coefficient (ADC) parameter.Experimental design: PDAC cellularity was quantified histologically and correlated with the ADC parameter and survival in genetically engineered mouse models and human patients.Results: Histologic analysis showed an inverse relationship of tumor cellularity and stroma content. Low tumor cellularity correlated with a significantly prolonged mean survival time (PDAClow = 21.93 months vs. PDACmed = 12.7 months; log-rank P < 0.001; HR = 2.23; CI, 1.41-3.53). Multivariate analysis using the Cox regression method confirmed tumor cellularity as an independent prognostic marker (P = 0.034; HR = 1.73; CI, 1.04-2.89). Tumor cellularity showed a strong negative correlation with the ADC parameter in murine (r = -0.84; CI, -0.90- -0.75) and human (r = -0.79; CI, -0.90 to -0.56) PDAC and high preoperative ADC values correlated with prolonged survival (ADChigh = 41.7 months; ADClow = 14.77 months; log rank, P = 0.040) in PDAC patients.Conclusions: This study identifies high tumor cellularity as a negative prognostic factor in PDAC and supports the ADC parameter for the noninvasive identification of PDAC groups. Clin Cancer Res; 23(6); 1461-70. ©2016 AACR.

To compare prospectively image quality and diagnostic confidence of flow-sensitive 3D turbo spin echo (TSE)-based non-contrast-enhanced MR angiography (NE-MRA) at 3.0 T using dual-source radiofrequency (RF) transmission with contrast-enhanced MRA (CE-MRA) in patients with peripheral arterial occlusive disease (PAOD).

PurposeTo determine changes in the bone marrow fat fraction (BMFF) in obesity after dietary intervention in comparison with changes in abdominal fat, liver fat, and serum lipids.Materials and MethodsTwenty obese (BMI 34.92 ± 3.8 kg/m2) women participated in a 4‐week dietary intervention of 800 kcal/d plus additional vegetables. They underwent anthropometric and blood value measurements before and after the intervention. Abdominal 3T MRI was performed to measure changes in subcutaneous adipose tissue (SAT) and visceral adipose tissue (VAT) volume and single‐voxel magnetic resonance spectroscopy (MRS) to measure fat content changes in the liver and L5 vertebral body.ResultsThe greatest relative change after dietary intervention was found in the liver (−40.3%), followed by VAT volume (−15.1%), serum lipids (−12.6 to −14.5%), and SAT volume (−8.5%). There were no statistically significant changes in BMFF after dietary intervention (P = 0.39), but absolute changes in the BMFF were positively associated with SAT volume (r = 0.489) and negatively associated with nonadipose tissue volume (r = –0.493) before dietary intervention.ConclusionBone marrow behaves differently compared to SAT volume, VAT volume, liver fat, and serum lipids after a 4‐week dietary intervention in obesity and BMFF changes depend on abdominal tissue volumes before intervention. J. Magn. Reson. Imaging 2015;42:1272–1280.

Poster: ECR 2015 / C-0841 / Assessment of wall shear stress in patients without aortic disease and with aortic dissection using velocity encoding 4D MRI by: P. Rudolph, M. Rasper, C. Maegerlein, B. M. Gramer, M. Settles, C. Reeps, B. Lutz, E. J. Rummeny, A. Huber; Munich/DE

Abstract Purpose: Preclinical model systems should faithfully reflect the complexity of the human pathology. In hepatocellular carcinoma (HCC), the tumor vasculature is of particular interest in diagnosis and therapy. By comparing two commonly applied preclinical model systems, diethylnitrosamine induced (DEN) and orthotopically implanted (McA) rat HCC, we aimed to measure tumor biology noninvasively and identify differences between the models. Experimental Design: DEN and McA tumor development was monitored by MRI and PET. A slice-based correlation of imaging and histopathology was performed. Array CGH analyses were applied to determine genetic heterogeneity. Therapy response to sorafenib was tested in DEN and McA tumors. Results: Histologically and biochemically confirmed liver damage resulted in increased 18F-fluorodeoxyglucose (FDG) PET uptake and perfusion in DEN animals only. DEN tumors exhibited G1–3 grading compared with uniform G3 grading of McA tumors. Array comparative genomic hybridization revealed a highly variable chromosomal aberration pattern in DEN tumors. Heterogeneity of DEN tumors was reflected in more variable imaging parameter values. DEN tumors exhibited lower mean growth rates and FDG uptake and higher diffusion and perfusion values compared with McA tumors. To test the significance of these differences, the multikinase inhibitor sorafenib was administered, resulting in reduced volume growth kinetics and perfusion in the DEN group only. Conclusions: This work depicts the feasibility and importance of in depth preclinical tumor model characterization and suggests the DEN model as a promising model system of multifocal nodular HCC in future therapy studies. Clin Cancer Res; 21(19); 4440–50. ©2015 AACR. See related commentary by Weber et al., p. 4254

Oncolytic viruses are promising new agents in cancer therapy. Success of tumor lysis is often hampered by low intra-tumoral titers due to a strong anti-viral host immune response and insufficient tumor targeting. Previous work on the co-assembly of oncolytic virus particles (VPs) with magnetic nanoparticles (MNPs) was shown to provide shielding from inactivating immune response and improve targeting by external field gradients. In addition, MNPs are detected by magnet resonance imaging (MRI) enabling non-invasive therapy monitoring. In this study two selected core-shell type iron oxide MNPs were assembled with adenovirus (Ad) or vesicular stomatitis virus (VSV). The selected MNPs were characterized by high r2 and r2(*) relaxivities and thus could be quantified non-invasively by 1.5 and 3.0 tesla MRI with a detection limit below 0.001 mM iron in tissue-mimicking phantoms. Assembly and cell internalization of MNP-VP complexes resulted in 81 - 97 % reduction of r2 and 35 - 82 % increase of r2(*) compared to free MNPs. The relaxivity changes could be attributed to the clusterization of particles and complexes shown by transmission electron microscopy (TEM). In a proof-of-principle study the non-invasive detection of MNP-VPs by MRI was shown in vivo in an orthotopic rat hepatocellular carcinoma model. In conclusion, MNP assembly and compartmentalization have a major impact on relaxivities, therefore calibration measurements are required for the correct quantification in biodistribution studies. Furthermore, our study provides first evidence of the in vivo applicability of selected MNP-VPs in cancer therapy.

PurposeTo propose and optimize diffusion‐weighted stimulated echo acquisition mode (DW‐STEAM) for measuring fat unsaturation in the presence of a strong water signal by suppressing the water signal based on a shorter T2 and higher diffusivity of water relative to fat.MethodsA parameter study for point‐resolved spectroscopy (PRESS) and STEAM using oil phantoms was performed and correlated with gas chromatography (GC). Simulations of muscle tissue signal behavior using DW‐STEAM and long–echo time (TE) PRESS and a parameter optimization for DW‐STEAM were conducted. DW‐STEAM and long‐TE PRESS were applied in the gastrocnemius muscles of nine healthy subjects.ResultsSTEAM with TE and mixing time (TM) up to 45 ms exhibited R2 correlations above 0.98 with GC and little T2‐weighting and J‐modulation for the quantified olefinic/methylene peak ratio. The optimal parameters for muscle tissue using DW‐STEAM were b‐value = 1800 s/mm2, TE = 33 ms, TM = 30 ms, and repetition time = 2300 ms. In vivo measured mean olefinic signal‐to‐noise ratios were 72 and 40, mean apparent olefinic water fractions were 0.19 and 0.11 for DW‐STEAM and long‐TE PRESS, respectively.ConclusionOptimized DW‐STEAM MR spectroscopy is superior to long‐TE PRESS for measuring fat unsaturation, if a strong water peak prevents the olefinic fat signal's quantification at shorter TEs and water's tissue specific ADC is substantially higher than fat. Magn Reson Med 75:32–41, 2016. © 2015 Wiley Periodicals, Inc.

The kinematics of the human hand is optimal with respect to force distribution during pinch as well as power grasp, reducing the tissue strain when exerting forces through opposing fingers and optimising contact faces. Quantifying this optimality is of key importance when constructing biomimetic robotic hands, but understanding the exact human finger motion is also an important asset in, e.g. tracking finger movement during manipulation. The goal of the method presented here is to determine the precise orientations and positions of the axes of rotation of the finger joints by using suitable magnetic resonance imaging (MRI) images of a hand in various postures. The bones are segmented from the images, and their poses are estimated with respect to a reference posture. The axis orientations and positions are fitted numerically to match the measured bone motions. Eight joint types with varying degrees of freedom are investigated for each joint, and the joint type is selected by setting a limit on the rotational and translational mean discrepancy. The method results in hand models with differing accuracy and complexity, of which three examples, ranging from 22 to 33 DoF, are presented. The ranges of motion of the joints show some consensus and some disagreement with data from literature. One of the models is published as an implementation for the free OpenSim simulation environment. The mean discrepancies from a hand model built from MRI data are compared against a hand model built from optical motion capture data.