Purpose: Cervical cancer is the second most frequently diagnosed cancer in women worldwide. About one-third of patients experience recurrent disease. A better chance of survival might be achieved by the early detection of recurrent cervical cancer. [(18)F]fluoro-2-deoxy-D-glucose (FDG) PET could be a promising imaging modality for this purpose, given that FDG PET has high diagnostic efficacy. Ideally, pre-selection of patients should be performed before considering FDG PET. The purpose of this study was to investigate parameters of primary cervical cancer associated with recurrence as a basis for pre-selection of patients in whom FDG PET should be performed.
Methods: Thirty-eight cervical cancer patients, clinically suspected of having recurrent disease, underwent FDG PET. Tissue from primary tumours and nine histologically confirmed metastases was analysed for biomarkers possibly related to glucose metabolism and prognosis (vascular endothelial growth factor, CD31 for microvessel density, glucose transporter-1, hexokinases I, II and III, Ki67, p53, hypoxia-inducible factor 1alpha, and degree of infiltration by lymphocytes and macrophages).
Results: Based on clinical outcome, sensitivity and specificity of FDG PET were 96% and 100%, respectively. Cox regression revealed microvessel density and p53 (tumour suppressor protein) to be the two most important biomarkers for prediction of recurrence (hazard ratios 2.54 and 2.28, respectively). By combining these two biomarkers in a parallel test, sensitivity and specificity in predicting recurrence were 87% and 71%, respectively. Leave-one-out cross-validation demonstrated predictive validity of a model based on microvessel density and p53.
Conclusion: In this first study of its kind, we have demonstrated that microvessel density and p53 profiles could be important in pre-selecting cervical cancer patients for detection of recurrence by FDG PET.