Premium
Whole‐tumor apparent diffusion coefficient measurements in nephroblastoma: Can it identify blastemal predominance?
Author(s) -
Littooij Annemieke S.,
Sebire Neil J.,
Olsen Øystein E.
Publication year - 2017
Publication title -
journal of magnetic resonance imaging
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.563
H-Index - 160
eISSN - 1522-2586
pISSN - 1053-1807
DOI - 10.1002/jmri.25506
Subject(s) - effective diffusion coefficient , medicine , histopathology , receiver operating characteristic , wilms' tumor , magnetic resonance imaging , diffusion mri , spearman's rank correlation coefficient , nuclear medicine , confidence interval , radiology , pathology , mathematics , statistics
Purpose To explore the potential relation between whole‐tumor apparent diffusion coefficient (ADC) parameters in viable parts of tumor and histopathological findings in nephroblastoma. Materials and Methods Children ( n = 52) with histopathologically proven nephroblastoma underwent diffusion‐weighted magnetic resonance imaging (MRI) (1.5T) before preoperative chemotherapy. Of these, 25 underwent an additional MRI after preoperative chemotherapy, shortly before resection. An experienced reader performed the whole‐tumor ADC measurements of all lesions, excluding nonenhancing areas. An experienced pathologist reviewed the postoperative specimens according to standard SIOP guidelines. Potential associations between ADC parameters and proportions of histological subtypes were assessed with Pearson's or Spearman's rank correlation coefficient depending on whether the parameters tested were normally distributed. In case the Mann–Whitney U ‐test revealed significantly different ADC values in a subtype tumor, this ADC parameter was used to derive a receiver operating characteristic (ROC) curve. Results The 25 th percentile ADC at presentation was the best ADC metric correlated with proportion of blastema (Pearson's r = –0.303, P = 0.026). ADC after preoperative treatment showed moderate correlation with proportion stromal subtype at histopathology ( r = 0.579, P = 0.002). By ROC analysis, the optimal threshold of median ADC for detecting stromal subtype was 1.362 × 10 −3 mm 2 /s with sensitivity and specificity of 100% (95% confidence interval [CI] 0.65–1.00) and 78.9% (95% CI 0.57–0.92), respectively. Conclusion ADC markers in nephroblastoma are related to stromal subtype histopathology; however, identification of blastemal predominant tumors using whole‐tumor ADC measurements is probably not feasible. Level of Evidence : 3 J. MAGN. RESON. IMAGING 2017;45:1316–1324