Avneesh Chhabra1,2, Oganes Ashikyan3, Chenelle Slepicka3, Nathan Dettori3, Helena Hwang4, Alexandra Callan5, Rohit R Sharma6, Yin Xi3. 1. Radiology, UT Southwestern Medical Center, Dallas, TX, 75022, USA. avneesh.chhabra@utsouthwestern.edu. 2. Orthopedics, UT Southwestern Medical Center, Dallas, TX, USA. avneesh.chhabra@utsouthwestern.edu. 3. Radiology, UT Southwestern Medical Center, Dallas, TX, 75022, USA. 4. Pathology, UT Southwestern Medical Center, Dallas, TX, USA. 5. Orthopedics, UT Southwestern Medical Center, Dallas, TX, USA. 6. Surgical Oncology, UT Southwestern Medical Center, Dallas, TX, USA.
Abstract
AIM: To evaluate proven soft tissue musculoskeletal malignancies blinded to their Fédération Nationale des Centres de Lutte Contre le Cancer histologic grades to identify the predictive values of conventional MR findings and best fit region of interest (ROI) apparent diffusion coefficient (ADC) measurements. MATERIALS AND METHODS: Fifty-one consecutive patients with different histologic grades were evaluated by four readers (R1-4) of different experience levels. Quantitatively, the maximum longitudinal size, tumor to muscle signal intensity ratios, and ADC measurements and, qualitatively, the spatial location of the tumor, its signal alterations, heterogeneity, intralesional hemorrhage or fat, and types of enhancement were assessed. Intraclass correlation, weighted kappa, ANOVA, and Fisher exact tests were used. RESULTS: There were 22/51 (43%) men (mean age ± SD = 52 ± 16 years) and 29/51 (57%) women (mean age ± SD = 54± 17 years), with the majority of tumors 38/51 (75%) in the lower extremities. Histologic grades were I in 8/51 (16%), II in 17/51 (33%), and III in 26/51 (51%), respectively. The longitudinal dimensions were different among three grades (p = 0.0015), largest with grade I. More central enhancements and deep locations were seen in grade III tumors (p = 0.0191, 0.0246). The ADC mean was significantly lower in grade III than in grade I or II (p < 0.0001 and p = 0.04). The ADC min was significantly lower in grade III than in grade I (p = 0.02). Good to excellent agreements were seen for T1/T2 tumor/muscle ratios, longitudinal dimension, and ADC (ICC = 0.60-0.98). CONCLUSION: Longitudinal tumor dimension, central enhancement, and ADC values differentiate histology grades in musculoskeletal soft tissue malignancy with good to excellent inter-reader reliability. KEY POINTS: • The longitudinal tumor dimension of grade III malignancy is smaller than that of grade I (p < 0.0001), and higher-grade tumors are located deeper (p = 0.0246). • The ADC mean is significantly lower in grade III than in grade I or grade II (p < 0.0001 and p = 0.04). • The ADC minimum is significantly lower in grade III than in grade I (p = 0.02).
AIM: To evaluate proven soft tissue musculoskeletal malignancies blinded to their Fédération Nationale des Centres de Lutte Contre le Cancer histologic grades to identify the predictive values of conventional MR findings and best fit region of interest (ROI) apparent diffusion coefficient (ADC) measurements. MATERIALS AND METHODS: Fifty-one consecutive patients with different histologic grades were evaluated by four readers (R1-4) of different experience levels. Quantitatively, the maximum longitudinal size, tumor to muscle signal intensity ratios, and ADC measurements and, qualitatively, the spatial location of the tumor, its signal alterations, heterogeneity, intralesional hemorrhage or fat, and types of enhancement were assessed. Intraclass correlation, weighted kappa, ANOVA, and Fisher exact tests were used. RESULTS: There were 22/51 (43%) men (mean age ± SD = 52 ± 16 years) and 29/51 (57%) women (mean age ± SD = 54± 17 years), with the majority of tumors 38/51 (75%) in the lower extremities. Histologic grades were I in 8/51 (16%), II in 17/51 (33%), and III in 26/51 (51%), respectively. The longitudinal dimensions were different among three grades (p = 0.0015), largest with grade I. More central enhancements and deep locations were seen in grade III tumors (p = 0.0191, 0.0246). The ADC mean was significantly lower in grade III than in grade I or II (p < 0.0001 and p = 0.04). The ADC min was significantly lower in grade III than in grade I (p = 0.02). Good to excellent agreements were seen for T1/T2tumor/muscle ratios, longitudinal dimension, and ADC (ICC = 0.60-0.98). CONCLUSION:Longitudinal tumor dimension, central enhancement, and ADC values differentiate histology grades in musculoskeletal soft tissue malignancy with good to excellent inter-reader reliability. KEY POINTS: • The longitudinal tumor dimension of grade III malignancy is smaller than that of grade I (p < 0.0001), and higher-grade tumors are located deeper (p = 0.0246). • The ADC mean is significantly lower in grade III than in grade I or grade II (p < 0.0001 and p = 0.04). • The ADC minimum is significantly lower in grade III than in grade I (p = 0.02).
Authors: Catherina S P van Rijswijk; Patrik Kunz; Pancras C W Hogendoorn; Antonïe H M Taminiau; Joost Doornbos; Johan L Bloem Journal: J Magn Reson Imaging Date: 2002-03 Impact factor: 4.813
Authors: Jan L M A Gielen; Arthur M De Schepper; Filip Vanhoenacker; Paul M Parizel; Xiaoling L Wang; Raf Sciot; Joost Weyler Journal: Eur Radiol Date: 2004-07-29 Impact factor: 5.315
Authors: Mark D Murphey; Michael S Gibson; Bryan T Jennings; Ana M Crespo-Rodríguez; Julie Fanburg-Smith; Donald A Gajewski Journal: Radiographics Date: 2006 Sep-Oct Impact factor: 5.333
Authors: Mark J Kransdorf; Laura W Bancroft; Jeffrey J Peterson; Mark D Murphey; William C Foster; H Thomas Temple Journal: Radiology Date: 2002-07 Impact factor: 11.105
Authors: Amani Arthur; Edward W Johnston; Jessica M Winfield; Matthew D Blackledge; Robin L Jones; Paul H Huang; Christina Messiou Journal: Front Oncol Date: 2022-07-01 Impact factor: 5.738
Authors: Madelaine Hettler; Julia Kitz; Ali Seif Amir Hosseini; Manuel Guhlich; Babak Panahi; Jennifer Ernst; Lena-Christin Conradi; Michael Ghadimi; Philipp Ströbel; Jens Jakob Journal: Cancers (Basel) Date: 2022-09-05 Impact factor: 6.575
Authors: Sana Boudabbous; Marion Hamard; Essia Saiji; Karel Gorican; Pierre-Alexandre Poletti; Minerva Becker; Angeliki Neroladaki Journal: BJR Open Date: 2022-06-22