A Study of the Correlation of Perfusion Parameters in High‐Resolution GRASP MRI With Microvascular Density in Lung Cancer. Issue 4 (3rd November 2018)

• Record Type: Journal Article
• Authors: Chen, Lihua; Zeng, Xianchun; Wu, Youli; Yan, Xiaochu; Huang, Xuequan; Chen, Hui; Zhang, Jiuquan; Wang, Jian; Feng, Li
• Is Part Of: Journal of magnetic resonance imaging. Volume 49:Issue 4(2019)
• Page Start: 1186
• Page End: 1194
• Publication Date: 2018-11-03
• Subjects: lung cancer -- microvascular density -- GRASP -- free‐breathing -- golden‐angle radial -- compressed sensing; Magnetic resonance imaging -- Periodicals; 616
• DOI: 10.1002/jmri.26340
• Languages: English
• ISSNs: 1053-1807

Abstract

Abstract : Background: The histological count of microvascular density (MVD) is the current clinical standard for assessing tumor angiogenesis. Although it is hypothesized that perfusion MRI can be a noninvasive alternative to MVD, there have been few studies to validate their correlations, particularly in lung cancer. Purpose: To investigate the correlation between MVD and perfusion parameters obtained from high‐resolution GRASP (Golden‐angle RAdial Sparse Parallel) dynamic contrast‐enhanced (DCE)‐MRI in a cohort of lung cancer patients, and to validate that GRASP MRI can serve as a free‐breathing, noninvasive imaging approach for studying tumor angiogenesis. Study Type: Prospective. Population: Twenty‐five lung cancer patients (16 male, 9 female, mean age = 57.3 ± 11.7 years). Field Strength/Sequence: 3T MRI; a prototype golden‐angle stack‐of‐stars sequence. Assessment: Contrast‐enhanced MR data were acquired during free breathing and were reconstructed using GRASP with a temporal resolution of ∼3 sec/phase. For all data, perfusion analysis was performed using a standard Tofts model to generate the volume transfer coefficient (K trans ) and the interstitial volume (Ve ). The MVD of corresponding tumor specimens, obtained from Computed Tomography‐guided biopsies, were counted with CD34 staining. Statistical Tests: Pearson correlation analysis; one‐way analysis of variance analysis; least significant difference‐t method of multiple comparisons. Results: The correlation coefficient was 0.983 and 0.972 for the measurement and remeasurement of K trans and Ve . The mean values of K trans, Ve, and MVD were 0.33 ± 0.22 min ‐1, 0.25 ± 0.12, and 49.68 ± 27.08 vessels/0.723 mm 2, respectively, in all patients ( n = 25); 0.36 ± 0.26 min ‐1, 0.27 ± 0.13, and 49.09 ± 29.84 vessels/0.723 mm 2, respectively, in adenocarcinoma ( n = 15); 0.34 ± 0.17 min ‐1, 0.26 ± 0.12, and 53.85 ± 23.53 vessels/0.723 mm 2, respectively, in squamous cell carcinoma ( n = 8); and 0.13 ± 0.15 min ‐1, 0.14 ± 0.06, and 37.20 ± 28.28 vessels/0.723 mm 2, respectively, in small‐cell carcinoma ( n = 2). There was a positive relationship between the K trans and MVD in all patients ( r = 0.738, P < 0.001). Data Conclusion: High spatiotemporal resolution DCE‐MRI using GRASP is a promising noninvasive alternative to the histological count of MVD for assessing tumor angiogenesis in lung cancer. Level of Evidence: 1 Technical Efficacy: Stage 2 J. Magn. Reson. Imaging 2019;49:1186–1194.