Prediction of prognostic factors in breast cancer: A noninvasive method utilizing histogram parameters derived from Adc maps

OBJECTIVE: The aim of this study is to investigate the relationship between histogram parameters and prognostic factors of breast cancer and to reveal the diagnostic performance of histogram parameters in predicting prognostic factors status. MATERIALS AND METHODS: Ninety-two patients with a confirmed histopathological diagnosis of breast cancer were included in the study. Magnetic resonance imaging (MRI) was performed using a 1.5T scanner and two different b values were used for diffusion-weighted imaging (DWI) (b values: 0 s/mm(2), b: 800 s/mm(2)). For 3D histogram analysis, regions of interest (ROI) were drawn each slice of the lesion on apparent diffusion coefficient (ADC) maps. The following data were derived from the histogram analysis data: percentiles, skewness, kurtosis, and entropy. The relationship between prognostic factors and histogram analysis data was investigated using the Kolmogorov−Smirnov test, Shapiro−Wilk test, skewness-kurtosis test, independent t-test, Mann−Whitney U test, and Kruskal−Wallis test. Receiver operator characteristic (ROC) curve analysis was performed to evaluate the diagnostic performance of the histogram parameters. RESULTS: ADC(max), kurtosis, and entropy parameters were statistically significantly correlated with tumor diameter (p = 0.002, p = 0.008, and p = 0.001, respectively). There was a significant difference in ADC(90%) and ADC(max) values, depending on estrogen receptor (ER) and progesterone receptor (PR) status. These values were lower in ER- and PR-positive than ER- and PR-negative patients (p = 0.02 and p = 0.001 vs. p = 0.018, p = 0.008). All ADC percentage values were lower in patients with a positive Ki-67 proliferation index as compared with those with a negative Ki-67 proliferation index (all p = 0.001). The entropy value was high in high-grade lesions and lesions with axillary involvement (p = 0.039 and p = 0.048, respectively). The highest area under the curve (AUC) for ER and PR status was calculated for the ADC(90%) value with ROC curve analysis. The highest AUC for Ki-67 proliferation index was found for the ADC(50%). CONCLUSION: Histogram analysis parameters derived from of ADC maps of whole lesions can reflect histopathological features of the tumors. Based on our study, it was concluded that histogram analysis parameters were related to the prognostic factors of the tumor.