Correlation of immunohistochemistry markers with textural characteristics derived from dual point F-18 fluorodeoxyglucose PET-CT in treatment naive locally advanced breast cancer patients

Aim Radiomics is an emerging field in medical imaging that leverages advanced computational algorithms to extract textural characteristics from medical images. It promises to transform breast cancer management by providing more accurate, reproducible, and noninvasive tools for clinicians, ultimately improving patient outcomes. We assessed the correlation of immunohistochemistry markers with textural characteristics derived from dual point F-18 fluorodeoxyglucose (FDG) positron emission tomography-computed tomograhy (PET-CT) in treatment naive locally advanced breast cancer patients. Materials and methods A total of 83 histopathologically proven, locally advanced breast cancer patients underwent baseline FDG PET-CT, 1-h postinjection. A total of 58 of 83 patients underwent PET-CT scan at 2-h postinjection. The primary lesion was segmented semiautomatically with 40% maximum standard uptake value (SUV max ) threshold, and textural characteristics were extracted using LIFEx software. Pearson correlation test was used to analyse relationships between textural metrics and immunohistochemistry markers (estrogen receptor, progesterone receptor, Her2neu, and Mib-1 index) of breast cancer. Results In 1-h FDG-PET, we observed no significant correlation between immunohistochemistry markers and textural parameters; however, in 2-h FDG-PET, there was moderate positive correlation of estrogen receptors with variance, moderate negative correlation with small-zone emphasis, zone-size nonuniformity, and grey-level variance (GLV). A moderate negative correlation of progesterone receptor with GLV and SUV max , a moderate positive correlation of Mib-1 index with volume, voxel counting, kurtosis, large zone emphasis, and a moderate negative correlation with skewness was observed. None of the parameters showed significant correlation with Her2neu. Conclusion This study indicates an association between textural parameters on delayed FDG PET-CT and immunohistochemistry markers. These textural parameters can be potentially useful in understanding tumor biology noninvasively.