Diffusion-Weighted Imaging of Brain Metastasis: Correlation of MRI Parameters with Histologic Type

Abstract

AIM: To compare the diffusion properties of brain metastases as imaging biomarkers in various types of tumours, to determine their histology and origin. MATERIAL and METHODS: Magnetic Resonance Imaging (MRI) and diffusion-weighted imaging (DWI) were used to retrospectively study the data of 143 patients suffering from brain metastases. Four categories of primary tumours with metastases to the brain were included: lung carcinoma (n=102, 71.3%); breast carcinoma (n=27, 18.8%); colon carcinoma (n=8, 5.6%); and others (n=6, 4.2%). The Apparent Diffusion Coefficient (ADCmin ) values, as well as the normalised ADC ratio (nADC), were determined. The lesions on the DWI were categorised as follows: type 1, with negative findings on DWI; type 2, which were isointense with the normal cortical grey matter; type 3, which were hyperintense compared to the normal cortical grey matter. RESULTS: The diffusion type, mean ADCmin, and mean nADC showed statistically significant differences in different types of metastases. In the subgroup analysis, it was found that type 3 was the diffusion type found most extensively in the brain metastases of small cell carcinoma (SCLC) (n=52, 65.8%, p<0.000). Furthermore, the mean ADCmin and nADC values were the least for the brain metastases of the SCLC (552.0 +/- 134.2 and nADC = 0.8 +/- 0.1, p<0.000, respectively). The value of the mean ADCmin was low in the human epidermal growth factor receptor 2 (HER-2) negative groups than in the HER-2 positive groups at 786.8 +/- 299.1 vs 844.8 +/- 141.3 (p<0.006). CONCLUSION: Our findings indicated that there is a correlation between diffusion parameters as imaging biomarkers of the solid component of brain metastases of primary tumours and the tumour histology.