18F-FDG PET radiomics approaches: comparing and clustering features in cervical cancer.

OBJECTIVES: The aims of our study were to find the textural features on 18F-FDG PET/CT which reflect the different histological architectures between cervical cancer subtypes and to make a visual assessment of the association between 18F-FDG PET textural features in cervical cancer. METHODS: Eighty-three cervical cancer patients [62 squamous cell carcinomas (SCCs) and 21 non-SCCs (NSCCs)] who had undergone pretreatment 18F-FDG PET/CT were enrolled. A texture analysis was performed on PET/CT images, from which 18 PET radiomics features were extracted including first-order features such as standardized uptake value (SUV), metabolic tumor volume (MTV) and total lesion glycolysis (TLG), second- and high-order textural features using SUV histogram, normalized gray-level co-occurrence matrix (NGLCM), and neighborhood gray-tone difference matrix, respectively. These features were compared between SCC and NSCC using a Bonferroni adjusted P value threshold of 0.0028 (0.05/18). To assess the association between PET features, a heat map analysis with hierarchical clustering, one of the radiomics approaches, was performed. RESULTS: Among 18 PET features, correlation, a second-order textural feature derived from NGLCM, was a stable parameter and it was the only feature which showed a robust trend toward significant difference between SCC and NSCC. Cervical SCC showed a higher correlation (0.70 ± 0.07) than NSCC (0.64 ± 0.07, P = 0.0030). The other PET features did not show any significant differences between SCC and NSCC. A higher correlation in SCC might reflect higher structural integrity and stronger spatial/linear relationship of cancer cells compared with NSCC. A heat map with a PET feature dendrogram clearly showed 5 distinct clusters, where correlation belonged to a cluster including MTV and TLG. However, the association between correlation and MTV/TLG was not strong. Correlation was a relatively independent PET feature in cervical cancer. CONCLUSIONS: 18F-FDG PET textural features might reflect the differences in histological architecture between cervical cancer subtypes. PET radiomics approaches reveal the association between PET features and will be useful for finding a single feature or a combination of features leading to precise diagnoses, potential prognostic models, and effective therapeutic strategies.

Different Prognostic Implications of 18F-FDG PET Between Histological Subtypes in Patients With Cervical Cancer.

This study aimed to investigate whether the predictive values of intensity- and volume-based PET parameters are different between histological subtypes in patients with cervical cancer. Ninety patients, 65 with squamous cell carcinoma (SCC) and 25 with non-SCC (NSCC), who underwent pretreatment ¹8F-FDG PET/CT and pelvic MRI, were studied retrospectively. In addition to SUVmax and SUVmean, metabolic-tumor-volume (MTV) was determined by thresholding of 40% SUVmax and total-lesion-glycolysis (TLG) was calculated. Clinical factors and PET metabolic indices were compared between SCC and NSCC. Progression-free survival (PFS) and overall survival (OS) were estimated using the Kaplan-Meier method with cut-offs determined by ROC analyses to stratify SCC and NSCC patients separately. Factors associated with survival were assessed with univariate and multivariate analyses using the Cox regression model. No significant differences were observed in clinical factors other than tumor size or ¹8F-FDG PET metabolic indices between SCC and NSCC. The Kaplan-Meier estimates of 2-year PFS and OS rates were 60% and 70% for SCC and 40% and 76% for NSCC, respectively. Multivariate analyses showed that MTV and TLG were the independent prognostic factors for PFS and OS in SCC; in contrast, SUVmax was the independent prognostic factor for PFS and OS in NSCC. Metabolic burden (MTV and TLG) could be beneficial for the prognostic prediction of cervical SCC patients; in contrast, metabolic intensity (SUVmax) could be beneficial for the prognostic prediction of NSCC patients. The different prognostic implications might be based on the differences of tissue integrity and histological heterogeneity between SCC and NSCC.