Loading...
To assess and compare the diagnostic accuracy of 68Ga-pentixafor PET/CT and 18F-FDG PET/CT in patients with newly diagnosed, relapsed, or highly suspected hematological malignancies. Accuracy will be evaluated using a composite reference standard, including histopathological confirmation, clinical follow-up, and additional imaging findings.
To evaluate the concordance between [⁶⁸Ga]Ga-pentixafor PET/CT-based staging and conventional clinical staging systems (e.g., Ann Arbor, Durie-Salmon PLUS, R-ISS/R2-ISS) in patients with hematological malignancies. Concordance will be measured using Cohen's kappa statistic (κ).
To assess whether imaging biomarkers from [⁶⁸Ga]Ga-pentixafor PET/CT, such as SUVmax or total lesion CXCR4 uptake, predict progression-free survival (PFS) in patients with hematological malignancies. Comparison will be made with [¹⁸F]FDG PET/CT-derived parameters.
To evaluate whether [⁶⁸Ga]Ga-pentixafor PET/CT-based biomarkers can predict overall survival in patients with hematological malignancies, compared with [¹⁸F]FDG PET/CT.
Evaluation of the overlap between artificial intelligence-generated lesion segmentations and expert manual annotations on PET/CT images using the Dice similarity coefficient (DSC).
To assess the correlation between CXCR4 expression levels in biopsy samples and semiquantitative parameters from [⁶⁸Ga]Ga-pentixafor PET/CT (e.g., SUVmax).
To assess the correlation between the presence or absence of del(17p), as identified by fluorescence in situ hybridization (FISH), and total lesion uptake (TLU) on [68Ga]Ga-pentixafor PET/CT at baseline. The degree of correlation will be measured using the Spearman correlation coefficient (rho), a unitless value ranging from -1 to +1.
To develop and validate an AI-based radiogenomic model integrating PET imaging features and genomic data to predict disease aggressiveness in patients with hematological malignancies.