Assessment of tumor heterogeneity in treatment-naïve adrenocortical cancer patients using (18)F-FDG positron emission tomography.

Abstract

As an orphan malignancy, only limited treatment options are available in adrenocortical carcinoma (ACC). Non-invasive risk assessment has not been described but may be of value to stratify patients for treatment. We aimed to evaluate the potential value of intra-individual tumor heterogeneity as assessed by (18)F-fluorodeoxyglucose ((18)F-FDG) positron emission tomography/computed tomography (PET/CT) for outcome prediction in treatment-naïve ACC patients. Ten patients with primary diagnosis of ACC were included in this study. Prior to any treatment initiation, baseline (18)F-FDG PET scans were performed. Tumor staging was performed using the European Network for the Study of Adrenal Tumors (ENS@T). Intratumoral heterogeneity of the primary tumor was assessed by manual segmentation using conventional PET parameters (standardized uptake values and tumor-to-liver ratios) and textural features. The impact of tumoral heterogeneity based on pre-therapeutic (18)F-FDG PET to predict progression-free (PFS) and overall survival (OS) was evaluated by receiver operating characteristic analysis. On average, tumor recurrence or progression was detected after median of 561days (range 71-1434days) after the pre-therapeutic baseline PET scan. 50% of the patients died of ACC within the follow-up period (mean 983±404days). Pre-therapeutic tumor volume was associated with PFS (r=-0.67, p=0.05) and Ki67 index with OS (r=-0.66, p=0.04). ENS@T tumor stage was the only parameter to correlate with both PFS and OS (r=-0.82, p=0.001, and r=-0.72, p=0.01, respectively). In the subgroup of patients without distant metastases (ENS@T stages II and III), age and pre-therapeutic tumor volume correlated significantly with PFS (r=0.96, p=0.01 and r=-0.93, p=0.02, respectively) and OS (r=0.95, p=0.02 and r=-0.90, p=0.04, respectively). None of the investigated classic or textural PET parameters predicted PFS or OS. In this pilot study in treatment-naïve ACC patients, conventional (18)F-FDG PET-derived parameters and textural tumor heterogeneity features were not suitable to identify high-risk patients.