Whether [18F]FDG PET/CT radiomic features provide information on tumor proliferation beyond conventional PET parameters remains uncertain. We investigated their associations with Ki-67 in untreated primary breast carcinoma using a predefined panel of technically reproducible features. This retrospective, single-center study included 42 women and 19 radiomic features previously selected based on interobserver reproducibility and redundancy, independently of Ki-67. Associations with continuous Ki-67 were evaluated using Spearman correlations, permutation tests, bootstrap confidence intervals, and false discovery rate (FDR) correction, alongside four conventional PET parameters. Five radiomic features showed significant associations with Ki-67 after FDR correction. The strongest association was observed for firstorder_90Percentile (ρ = 0.556; 95% CI, 0.279–0.745; q = 0.002), comparable to that of SUVmax (ρ = 0.565). However, none of these five associations remained significant after adjustment for metabolic tumor volume (MTV), SUVmax, or both, followed by FDR correction. Excluding six small lesions attenuated the correlations for both SUVmax (ρ = 0.440) and firstorder_90Percentile (ρ = 0.427). Although several radiomic features were associated with Ki-67, an independent contribution beyond conventional PET parameters was not demonstrated. These findings require confirmation in independent cohorts.
A multinodular and vacuolating neuronal tumor (MVNT) is a rare non-malign central nervous system tumor. The cerebral cortex of young to middle-aged adults is affected and seizures and focal neurological deficits could reveal it. Despite being considered a low-grade tumor, MVNT can be challenging to diagnose and manage due to its histopathological similarities to other neuronal tumors. The limited number of reported cases makes it difficult to establish definitive guidelines for treatment and follow-up. Given the relatively recent discovery of MVNT, there is a need for further research to better understand its pathogenesis, optimal treatment and long-term strategies. Furthermore, studying MVNT can contribute to the broader understanding of brain tumor biology and the intricacies of tumor heterogeneity. This paper explores the challenges and perspectives surrounding MVNT to improve clinical decision-making and patient outcomes. Based on the present narrative review, advancements in diagnostic technologies have shown promise in addressing the challenges associated with diagnosing MVNTs. Liquid biopsy is an innovative and non-invasive diagnostic tool that analyzes biomarkers to provide information about tumors, and radiomics is a method that extracts quantitative data from medical images to provide detailed insights into tumor characteristics. Such innovative diagnostic technologies are important because they have the potential to significantly improve the current diagnostic landscape for MVNTs, allowing for earlier detection, accurate classification, and personalized treatment strategies.
