Breslow thickness and 18 F-FDG PET/CT results in initial staging of cutaneous melanoma: Can a cut-off point be established?

Autor: J.L. González Larriba, J.L. Carreras-Delgado, R. Cano-Carrizal, C. Rodríguez-Rey, L. Lapeña-Gutiérrez, M. García García-Esquinas, M.E. Fuentes-Ferrer, E. Cala-Zuluaga, A. Ortega-Candil, María Nieves Cabrera-Martín, A. Jiménez-Ballvé, M.J. Pérez-Castejón
Rok vydání: 2016
Předmět:
Zdroj: Revista Española de Medicina Nuclear e Imagen Molecular (English Edition). 35:96-101
ISSN: 2253-8089
DOI: 10.1016/j.remnie.2016.01.002
Popis: Aim To establish a Breslow thickness (BT) cut-off point for indication of PET/CT of cutaneous melanoma in early stages and evaluate its prognostic value. Material and methods Retrospective analysis of 347 PET/CT studies with diagnosis of melanoma, of which 108 were performed for initial staging. Thirty-one patients were excluded, and a final sample of 77 patients remained. A ROC curve analysis was performed to establish an optimal cut-off point. A survival analysis was performed, considering death assignable to melanoma as the main event, for the evaluation of its prognostic value. Results Forty-seven (61.04%) of all 77 patients selected were men, and 11 (14.29%) had a positive PET/CT result. Mean age was 65.17 ± 15.00 years. The median BT in patients with a negative PET/CT result was 2.75 mm (IQR 1.83–4.50) and in the positive group 6.25 mm (IQR 5.40–7.50) ( p = 0.0013). In the ROC curve analysis (AUC 0.804, SE 0.054), an optimal value of 5 mm BT with the following values was obtained: sensitivity 90.91%, specificity 78.79%, negative predictive value (NPV) 98.1%, positive predictive value (PPV) 41.7%, diagnostic OR 37.1, and accuracy 80.52%. Mean follow-up was 18.66 ± 14.35 months, detecting 2/53 (3.77%) deaths in the BT p = 0.0013). Conclusions A 5 mm cut-off point correctly distinguishes those patients with positive PET/CT from those with negative results in the early stages of cutaneous melanoma; therefore it could be included in initial staging of this subgroup of patients.
Databáze: OpenAIRE