Investigating the contribution of pre- and per-treatment F-FDG PET-CT segmentation methodologies for post-treatment tumor recurrence prediction in cervical cancer

摘要

Cervical cancer is one of the most common cancers to affect women worldwide. Despite the efficiency of radiotherapy treatment, some patients present post-treatment tumor recurrence, which increases the risk of death. Several studies suggest that tumor characteristics visible with PET imaging before and during the treatment could be used to predict post-treatment recurrence. We evaluate the contribution of pre- and per-treatment 18F-FDG PET images by exploring the predictive value of features extracted through several segmentation methods. Forty-one patients with locally advanced cervix cancer treated by chemoradiotherapy were considered. For each patient, two coregistered PET/CT scan were acquired before and during the treatment. A non-rigid registration was used to match the two PET acquisitions and evaluate the tumor metabolism inside the same area. Maximum and peak standardized uptake value (SUVmax and SUVpeak), the metabolic tumor volume (MTV) and the total lesion glycolysis (TLG) were evaluated. The predictive value of the extracted features was assessed through the Harrel's C-index. Results suggest that accurate segmentation can compute early meaningful features that are related with tumor recurrence. TLG seems to be strongly informative in prediction of tumor recurrence in cervical cancer.