PET/MRI in Cervical Cancer: Associations Between Imaging Biomarkers and Tumor Stage, Disease Progression, and Overall Survival.

Abstract

Positron emission tomography (PET)/MRI biomarkers have been shown to have prognostic significance in patients with cervical cancer. Their associations with progression-free survival (PFS) and overall survival (OS) merit further investigation. To evaluate the association between PET/MRI biomarkers and tumor stage, PFS, and OS in patients with cervical cancer. Prospective cohort study. In all, 54 patients with newly diagnosed cervical cancer and measurable tumors (>1 cm) were included in the image analysis. 3.0T integrated PET/MRI including diffusion-weighted echo-planar imaging (b = 50 and 1000 s/mm2 ) and [18F]fluorodeoxyglucose PET. Two radiologists measured the minimum and mean apparent diffusion coefficient (ADCmin and ADCmean ), maximum standardized uptake value (SUVmax ), metabolic tumor volume (MTV), and total lesion glycolysis (TLG) of the primary tumors. A Mann-Whitney U-test was used to evaluate the association between the imaging biomarkers and tumor stage. A Cox proportional hazards model was used to assess the relationships between the imaging biomarkers and survival. In advanced tumors (T ≥ 1b2, M1, stage ≥ IB3), ADCmin was significantly lower and MTV, TLG, MTV/ADCmin , and TLG/ADCmin were significantly higher (P values between <0.001 and 0.036). In N1 tumors, ADCmin was significantly lower and MTV and MTV/ADCmin were significantly higher (P values between 0.005 and 0.016). In survival analysis, SUVmax was an independent predictor of PFS (hazard ratio [HR] = 4.57, P < 0.05), and ADCmin was an independent predictor of OS (HR = 0.02, P < 0.05). In subgroup analysis of patients with different stages, MTV/ADCmin was a predictor of PFS in stage I disease (P = 0.003), ADCmin (P = 0.038), and MTV (P = 0.020) in stage II, SUVmax (P = 0.006), and TLG (P = 0.006) in stage IV; and ADCmin was a predictor of OS in stage III disease (P = 0.008). PET/MRI biomarkers of cervical cancer are associated with tumor stage and survival. SUVmax and ADCmin are independent predictors of PFS and OS, respectively. 1 TECHNICAL EFFICACY: 3.