Oncotarget · 2018 · 40 citations · 22 references
MedicineOncologyPathologyHistologyHistopathologyNeoplasiaSquamous Cell CarcinomaTumor CharacterizationTumor GradingTumor HistopathologyTumor ImagingTumor MorphologyApparent Diffusion Coefficient AnalysisHistogram AnalysisHistogram MetricsHistopathological GradingTumor Activity GradingTumor Diffusion ImagingTumor Morphology CorrelationTumor Proliferation Assessment
// Alexey Surov 1, * , Hans Jonas Meyer 1, * , Karsten Winter 2 , Cindy Richter 2 and Anna-Kathrin Hoehn 3 1 Department of Diagnostic and Interventional Radiology, University Hospital of Leipzig, Leipzig 04103, Germany 2 Institute of Anatomy, University Hospital of Leipzig, Leipzig 04103, Germany 3 Department of Pathology, University Hospital of Leipzig, Leipzig 04103, Germany * These authors contributed equally to this work Correspondence to: Alexey Surov, email: Alexey.Surov@medizin.uni-leipzig.de Keywords: MRI; ADC; histogram analysis; KI 67; HNSCC Abbreviations: MRI: magnetic resonance imaging; DWI: diffusion weighted imaging; ADC: apparent diffusion coefficient; HNSCC: head and neck squamous cell carcinoma; ROI: region of interest Received: January 24, 2018 Accepted: April 06, 2018 Published: May 04, 2018 ABSTRACT Our purpose was to analyze associations between apparent diffusion coefficient (ADC) histogram analysis parameters and histopathologicalfeatures in head and neck squamous cell carcinoma (HNSCC). The study involved 32 patients with primary HNSCC. For every tumor, the following histogram analysis parameters were calculated: ADCmean, ADCmax, ADC min , ADC median , ADC mode , P10, P25, P75, P90, kurtosis, skewness, and entropy. Furthermore, proliferation index KI 67, cell count, total and average nucleic areas were estimated. Spearman's correlation coefficient (p) was used to analyze associations between investigated parameters. In overall sample, all ADC values showed moderate inverse correlations with KI 67. All ADC values except ADCmax correlated inversely with tumor cellularity. Slightly correlations were identified between total/average nucleic area and ADC mean , ADC min , ADC median , and P25. In G1/2 tumors, only ADCmode correlated well with Ki67. No statistically significant correlations between ADC parameters and cellularity were found. In G3 tumors, Ki 67 correlated with all ADC parameters except ADCmode. Cell count correlated well with all ADC parameters except ADCmax. Total nucleic area correlated inversely with ADC mean , ADC min , ADC median , P25, and P90. ADC histogram parameters reflect proliferation potential and cellularity in HNSCC. The associations between histopathology and imaging depend on tumor grading.
22