Purpose: We investigated the prognostic value of the tumour heterogeneity index determined on preoperative [18F]FDG PET/CT in patients with uterine leiomyosarcoma (LMS).
Methods: We retrospectively reviewed patients with uterine LMS who underwent preoperative [18F]FDG PET/CT scans at three tertiary referral hospitals. The PET/CT parameters maximum standardized uptake value of the primary tumour (SUVmax), metabolic tumour volume (MTV) and total lesion glycolysis were assessed. The negative values of the MTV linear regression slope (nMLRS) according to the SUV thresholds of 2.5 and 3.0 were determined as the tumour heterogeneity index. The value of PET/CT-derived parameters in predicting progression-free survival (PFS) and overall survival (OS) were determined in regression analyses.
Results: Clinicopathological and PET/CT data from 16 patients were reviewed. The median postsurgical follow-up was 21 months (range 4-82 months), and 12 patients (75.0%) experienced recurrence. Tumour size (P = 0.017), SUVmax (P = 0.019), MTV (P = 0.016) and nMLRS (P = 0.008) were significant prognostic factors for recurrence. MTV (P = 0.048) and nMLRS (P = 0.045) were significant prognostic factors for patient survival. nMLRS was correlated with clinicopathological parameters including tumour size (Pearson's correlation coefficient γ = 0.825, P < 0.001) and lymph node metastasis (γ = 0.721, P = 0.004). Patient groups categorized according to the nMLRS cut-off value showed significant differences in PFS (P = 0.033) and OS (P = 0.044).
Conclusion: The preoperative tumour heterogeneity index obtained using the MTV linear regression slope may be a novel and useful prognostic marker in uterine LMS.
Keywords: Linear regression; Tumour heterogeneity; Uterine leiomyosarcoma; [18F]FDG PET/CT.