Associations between histogram analysis parameters derived from morphological sequences and histopathological tissue alterations in myositis and other myopathies: a preliminary study.

ABSTRACT

OBJECTIVES: Magnetic resonance imaging (MRI) is a cornerstone in diagnosis of myopathies. Recently, imaging techniques, such as histogram analysis are used to obtain novel imaging biomarkers. The present study sought to elucidate possible associations between histopathology derived from muscle biopsies and histogram parameters derived from clinical MRI in myositis and other myopathies. METHODS: 20 patients with myopathies were included in this retrospective study. MRI was performed using a 1.5T MRI scanner including T2- and T1- weighted images. The histogram parameters of the MRI sequences were obtained of the biopsied muscle. The histopathology analysis included the scoring systems proposed by Tateyama et al., Fanin et al., Allenbach et al. and immunohistochemical stainings for MHC-I, CD68, CD8 and CD4. RESULTS: Entropy derived from T2-weighted images showed strong positive associations with the inflammation scores (r=0.71, p=0.0005 with Allenbach et al score and r=0.68, p=0.001 with Tateyama score). Furthermore, there were strong associations between entropy derived from T2-weighted images with MHC-I staining (r=0.67, p=0.022), with the amount of CD20 cells (r=0.70, p=0.022), with the amount of CD4 positive cells (r=0.78, p=0.0075) and with the amount of CD8 positive cells (r=0.79, p=0.004). Other parameters showed no associations with the investigated histopathology features. CONCLUSIONS: Entropy derived from T2-weighted images showed strong associaitions with inflammation scores and with the sole amount of immune cells in myopathies. These results need to be confirmed by clinical studies, whether it is also related to clinical performance or can predict treatment response.