Different18F-FDG PET parameters for the prediction of histological response to neoadjuvant chemotherapy in pediatric Ewing sarcoma family of tumors

Background The histological response to neoadjuvant chemotherapy (NAC) in pediatric patients with Ewing sarcoma family of tumors (ESFT) can predict the disease-free survival. Therefore, a noninvasive method for response assessment is needed. Using the currently established imaging modalities, mass reduction does not always correlate with the percentage of necrosis. Objective To determine the potential role of(18)fluorine-labeled fluoro-2-deoxyglucose positron emission tomography (F-18-FDG PET) metabolic parameters in the prediction of poor histological response to NAC in pediatric patients with ESFT. Methods Thirty-six patients who were treated with NAC and surgery at the Children's Cancer Hospital, Egypt, were prospectively included in this study. All patients underwent two studies; a PET/CT study before NAC and another one after NAC completion. Metabolic PET parameters were measured in each study. The ability of each of these parameters, their pretreatment and pre-local control values, as well as the percentage reduction between their pretreatment and pre-local control values, were evaluated to differentiate between good and poor responders using the histological response as a standard reference. Results Neither the pretreatment value nor the percentage reduction of any of the measured PET parameters predicted poor histological response. After NACcompletion, metabolic tumor volume (MTV) at the threshold of an SUV of 2.5 isocontour (MTV(2.5)(post)), MTV at the threshold of hepatic reference SUVmean (MTV(HR)(post)), and total lesion glycolysis at the threshold of hepatic reference SUVmean (TLG(HR)(post)) predicted poor histological response (P = 0.006, 0.018, and 0.003, respectively). The cutoff values of 90% reduction of TLG(HR) and maximum standardized uptake value (SUVmax)(post)<= 2.5 could differentiate between good and poor responders. Conclusion FDG PET parameters can predict poor histological response to NAC in ESFT patients. MTV and TLG at the thresholds of an SUV of 2.5 isocontour and hepatic reference SUVmean are the two most promising thresholds in predicting the response of patients. The cutoff value of SUVmax(post)<= 2.5 predicts poor histological response.