Markers of sarcopenia quantified by computed tomography predict adverse long-term outcome in patients with resected oesophageal or gastro-oesophageal junction cancer

Objectives

To assess the impact of sarcopenia and alterations in body composition parameters (BCPs) on survival after surgery for oesophageal and gastro-oesophageal junction cancer (OC).

Methods

200 consecutive patients who underwent resection for OC between 2006 and 2013 were selected. Preoperative CTs were used to assess markers of sarcopenia and body composition (total muscle area [TMA], fat-free mass index [FFMi], fat mass index [FMi], subcutaneous, visceral and retrorenal fat [RRF], muscle attenuation). Cox regression was used to assess the primary outcome parameter of overall survival (OS) after surgery.

Results

130 patients (65 %) had sarcopenia based on preoperative CT examinations. Sarcopenic patients showed impaired survival compared to non-sarcopenic individuals (hazard ratio [HR] 1.87, 95 % confidence interval [CI] 1.15–3.03, p?=?0.011). Furthermore, low skeletal muscle attenuation (HR 1.91, 95 % CI 1.12–3.28, p?=?0.019) and increased FMi (HR 3.47, 95 % CI 1.27–9.50, p?=?0.016) were associated with impaired outcome. In the multivariate analysis, including a composite score (CSS) of those three parameters and clinical variables, only CSS, T-stage and surgical resection margin remained significant predictors of OS.

Conclusion

Patients who show signs of sarcopenia and alterations in BCPs on preoperative CT images have impaired long-term outcome after surgery for OC.

Key Points

? Sarcopenia is associated with impaired OS after surgery for oesophageal cancer.? Other body composition parameters are also associated with impaired survival.? This influence on survival is independent of established clinical parameters.? Sarcopenia provides a better estimation of cachexia than BMI.? Sarcopenia assessment could be considered in risk/benefit stratification before oesophagectomy.

     

Human phase I metabolism of the novel synthetic cannabinoid 5F-CUMYL-PEGACLONE

Purpose

5F-CUMYL-PEGACLONE is a recently emerged γ-carbolinone derived synthetic cannabinoid. The present study aimed to identify phase I metabolites to reliably prove consumption of the substance by urine analysis and to differentiate from the uptake of the non-fluorinated analog CUMYL-PEGACLONE.

Methods

For metabolite characterization, phase I metabolites were analyzed by liquid chromatography–high resolution mass spectrometry after incubation with pooled human liver microsomes. Reliability of the biomarkers was evaluated by analysis of human urine samples (n?=?20) by liquid chromatography–triple quadrupole tandem mass spectrometry. Sample preparation included β-glucuronidase treatment followed by liquid-liquid extraction.

Results

In total, 15 metabolites were detected in vivo and characterized. Metabolic reactions were primarily observed at the γ-carbolinone core and the 5-fluoropentyl chain, and included N-dealkylation, hydroxylation, hydrolytic defluorination, formation of a dihydrodiol, oxidation to the pentanoic acid metabolite and formation of the propionic acid metabolite. Six of these metabolites were identical with phase I metabolites of CUMYL-PEGACLONE, which must be considered for interpretation of analytical findings in urine samples.

Conclusions

5F-CUMYL-PEGACLONE was subject to extensive metabolism in humans. The propionic acid metabolite was the most abundant metabolite in all urine samples and should be targeted when maximum sensitivity is needed (e.g., drug abstinence control). However, this metabolite also occurs in the biotransformation of the non-fluorinated analog and is, therefore, not a compound-specific marker. For differentiation, a metabolite hydroxylated at the γ-carbolinone core showed to be the most reliable marker and should be used as an additional target analyte.

     

Automated measurement of lymph nodes: a phantom study

The purpose of this study was to assess the accuracy of automated nodal quantification in a phantom. MDCT of a phantom with 17 synthetic lymph nodes of different sizes (diameter 6.0–30.0 mm) was performed at varying tube currents, reconstruction kernels and slice thicknesses. RECIST diameter and volume were measured using an automated software tool. Results were compared with the reference diameter and volume by calculating the absolute percentage error (APE). Degree of agreement between software and reference measurements was evaluated by computing corresponding concordance correlation coefficients (CCC). Under varying tube currents the mean APE (CCC) varied between 5.18% and 10.12% (0.95–0.99) for RECIST diameter and between 7.22% and 16.21% (0.94–1.00) for the volume. At different reconstruction kernels the mean APE values ranged between 7.20% and 7.55% (0.99) (RECIST) and between 8.96% and 14.42% (1.00) (volume). With different slice thicknesses the mean APE values differed from 5.81% to 9.20% (0.97–0.99) (RECIST) and from 8.16% to 22.66% (0.99–1.00) (volume). Regarding RECIST criteria and volume, automated evaluation of lymph nodes in a phantom demonstrated a high accuracy under varying MDCT parameters.     

Evaluation and comparison of cartilage repair tissue of the patella and medial femoral condyle by using morphological MRI and biochemical zonal T2 mapping

The objective of this study was to use advanced MR techniques to evaluate and compare cartilage repair tissue after matrix-associated autologous chondrocyte transplantation (MACT) in the patella and medial femoral condyle (MFC). Thirty-four patients treated with MACT underwent 3-T MRI of the knee. Patients were treated on either patella (n = 17) or MFC (n = 17) cartilage and were matched by age and postoperative interval. For morphological evaluation, the MR observation of cartilage repair tissue (MOCART) score was used, with a 3D-True-FISP sequence. For biochemical assessment, T2 mapping was prepared by using a multiecho spin-echo approach with particular attention to the cartilage zonal structure. Statistical evaluation was done by analyses of variance. The MOCART score showed no significant differences between the patella and MFC (p ≥ 0.05). With regard to biochemical T2 relaxation, higher T2 values were found throughout the MFC (p < 0.05). The zonal increase in T2 values from deep to superficial was significant for control cartilage (p < 0.001) and cartilage repair tissue (p < 0.05), with an earlier onset in the repair tissue of the patella. The assessment of cartilage repair tissue of the patella and MFC afforded comparable morphological results, whereas biochemical T2 values showed differences, possibly due to dissimilar biomechanical loading conditions.     

Whole heart magnetization‐prepared steady‐state free precession coronary vein MRI

Purpose

To compare two coronary vein imaging techniques using whole‐heart balanced steady‐state free precession (SSFP) and a targeted double‐oblique spoiled gradient‐echo (GRE) sequences in combination with magnetization transfer (MT) preparation sequence for tissue contrast improvement.

Materials and Methods

Nine healthy subjects were imaged with the proposed technique. The results are compared with optimized targeted MT prepared GRE acquisitions. Both quantitative and qualitative analyses were performed to evaluate each imaging method.

Results

Whole‐heart images were successfully acquired with no visible image artifact in the vicinity of the coronary veins. The anatomical features and visual grading of both techniques were comparable. However, the targeted small slab acquisition of the left ventricular lateral wall was superior to whole‐heart acquisition for visualization of relevant information for cardiac resynchronization therapy (CRT) lead implantation.

Conclusion

We demonstrated the feasibility of whole‐heart coronary vein MRI using a 3D MT‐SSFP imaging sequence. A targeted acquisition along the lateral left ventricular wall is preferred for visualization of branches commonly used in CRT lead implantation. J. Magn. Reson. Imaging 2009;29:1293–1299. © 2009 Wiley‐Liss, Inc.     

Impact of postoperative radiotherapy and HER2/new overexpression in salivary duct carcinoma

Aim

Retrospective Investigation of the prognostic relevance of clinicopathologic parameters in patients with salivary duct carcinoma (SDC).

Methods

An experienced pathologist reviewed 67 patients with de novo SDC or SDC ex pleomorphic adenoma. Paraffin-embedded tumor samples were examined by immunohistochemistry for expression of HER2/neu, androgen (AR), progesterone (PR), estrogen (ER), epidermal growth factor (EGFR) and programmed death ligand 1 (PD-L1-R) receptor. In 45 patients who had cM0 and follow-up data available, survival rates were calculated (Kaplan–Meier method) and prognostic variables were analyzed (univariate analysis: log-rank test; multivariate analysis: Cox-regression analysis).

Results

Overexpression of HER2/neu, AR, ER, PR, EGFR, PD-L1-R was found in 25.4%, 84%, 0%, 0%, 17.9%, 16.4% of patients. Overall (OS), disease-free (DFS), distant-metastases-free survival (DMFS) and locoregional control (LRC) were 92.3/72.4/56.9%, 78.2/58.1/58.1%, 85.4/65.2/65.2% and 89.7/81.9/81.9% after 1/3/5 years (medial follow-up 26 months). In univariate analysis a positive resection margin (p = 0.008) and no postoperative radiotherapy (p = 0.001) predict an increased locoregional recurrence rate. In multivariate analysis only postoperative radiotherapy is statistically significant (p = 0.004). Presence of lymph node metastases, a lymph node density >4 and HER2/neu overexpression predict decreased DFS and DMFS. In multivariate HER2/neu overexpression was the only significant predictor for reduced DFS (p = 0.04) and DMFS (p = 0.02).

Conclusion

Postoperative radiotherapy is the only significant predictor for LRC. HER2/neu receptor expression is an independent prognostic factor for decreased DFS and DMFS in patients with SDC. In addition to radio(chemo)therapy, intensified first-line treatment regimens should also be evaluated in the future.

     

Accuracy of Right and Left Ventricular Functional Assessment by Short-Axis vs Axial Cine Steady-State Free-Precession Magnetic Resonance Imaging: Intrapatient Correlation With Main Pulmonary Artery and Ascending Aorta Phase-Contrast Flow Measurements

Objective

The left ventricle (LV) is routinely assessed with cardiac magnetic resonance imaging (MRI) by using short-axis orientation; it remains unclear whether the right ventricle (RV) can also be adequately assessed in this orientation or whether dedicated axial orientation is required. We used phase-contrast (PC) flow measurements in the main pulmonary artery (MPA) and the ascending aorta (Aorta) as nonvolumetric standard of reference and compared RV and LV volumes in short-axis and axial orientations.

Methods

A retrospective analysis identified 30 patients with cardiac MRI data sets. Patients underwent MRI (1.5 T or 3 T), with retrospectively gated cine steady-state free-precession in axial and short-axis orientations. PC flow analyses of MPA and Aorta were used as the reference measure of RV and LV output.

Results

There was a high linear correlation between MPA-PC flow and RV–stroke volume (SV) short axis (r = 0.9) and RV-SV axial (r = 0.9). Bland-Altman analysis revealed a mean offset of 1.4 mL for RV axial and –2.3 mL for RV–short-axis vs MPA-PC flow. There was a high linear correlation between Aorta-PC flow and LV-SV short-axis (r = 0.9) and LV-SV axial (r = 0.9). Bland-Altman analysis revealed a mean offset of 4.8 m for LV short axis and 7.0 mL for LV axial vs Aorta-PC flow. There was no significant difference (P = .6) between short-axis–LV SV and short-axis–RV SV.

Conclusion

No significant impact of the slice acquisition orientation for determination of RV and LV stroke volumes was found. Therefore, cardiac magnetic resonance workflow does not need to be extended by an axial data set for patients without complex cardiac disease for assessment of biventricular function and volumes.     

Improved fat water separation with water selective inversion pulse for inversion recovery imaging in cardiac MRI

Purpose:

To develop an improved chemical shift‐based water‐fat separation sequence using a water‐selective inversion pulse for inversion recovery 3D contrast‐enhanced cardiac magnetic resonance imaging (MRI).

Materials and Methods:

In inversion recovery sequences the fat signal is substantially reduced due to the application of a nonselective inversion pulse. Therefore, for simultaneous visualization of water, fat, and myocardial enhancement in inversion recovery‐based sequences such as late gadolinium enhancement imaging, two separate scans are used. To overcome this, the nonselective inversion pulse is replaced with a water‐selective inversion pulse. Imaging was performed in phantoms, nine healthy subjects, and nine patients with suspected arrhythmogenic right ventricular cardiomyopathy plus one patient for tumor/mass imaging. In patients, images with conventional turbo‐spin echo (TSE) with and without fat saturation were acquired prior to contrast injection for fat assessment. Subjective image scores (1 = poor, 4 = excellent) were used for image assessment.

Results:

Phantom experiments showed a fat signal‐to‐noise ratio (SNR) increase between 1.7 to 5.9 times for inversion times of 150 and 300 msec, respectively. The water‐selective inversion pulse retains the fat signal in contrast‐enhanced cardiac MR, allowing improved visualization of fat in the water‐fat separated images of healthy subjects with a score of 3.7 ± 0.6. Patient images acquired with the proposed sequence were scored higher when compared with a TSE sequence (3.5 ± 0.7 vs. 2.2 ± 0.5, P < 0.05).

Conclusion:

The water‐selective inversion pulse retains the fat signal in inversion recovery‐based contrast‐enhanced cardiac MR, allowing simultaneous visualization of water and fat. J. Magn. Reson. Imaging 2013;37:484–490. © 2012 Wiley Periodicals, Inc.     

Measurement of vibration‐induced volumetric strain in the human lung

Noninvasive image‐based measurement of intrinsic tissue pressure is of great interest in the diagnosis and characterization of diseases. Therefore, we propose to exploit the capability of phase‐contrast MRI to measure three‐dimensional vector fields of tissue motion for deriving volumetric strain induced by external vibration. Volumetric strain as given by the divergence of mechanical displacement fields is related to tissue compressibility and is thus sensitive to the state of tissue pressure. This principle is demonstrated by the measurement of three‐dimensional vector fields of 50‐Hz oscillations in a compressible agarose phantom and in the lungs of nine healthy volunteers. In the phantom, the magnitude of the oscillating divergence increased by about 400% with 4.8 bar excess air pressure, corresponding to an effective‐medium compression modulus of 230 MPa. In lungs, the averaged divergence magnitude increased in all volunteers (N = 9) between 7 and 78% from expiration to inspiration. Measuring volumetric strain by MRI provides a compression‐sensitive parameter of tissue mechanics, which varies with the respiratory state in the lungs. In future clinical applications for diagnosis and characterization of lung emphysema, fibrosis, or cancer, divergence‐sensitive MRI may serve as a noninvasive marker sensitive to disease‐related alterations of regional elastic recoil pressure in the lungs. Magn Reson Med, 2013. © 2012 Wiley Periodicals, Inc.