Purpose: To evaluate the feasibility of early metabolic change assessed by PET in predicting clinical response to chemotherapy and investigate its prognostic value in patients with advanced gastric cancer.
Experimental design: A total of 64 patients with advanced gastric cancer were prospectively enrolled and examined by PET with (18)F-fluorodeoxyglucose (FDG) and (18)F-fluoro-3'-deoxy-3'-L-fluorothymidine (FLT) at baseline and 14 days after treatment initiation. PET findings were analyzed for the correlation with best clinical response of patients, disease control status, and survival after identifying the threshold of metabolic change percentage by ROC analysis.
Results: For FDG-PET, the total uptake value reduction percentage (δ-SUV) of 40% was the cut-off point with the maximum of sensitivity (70%) and specificity (83%) to predict clinical responding and that of prediction for disease control status was 30%, with the highest sensitivity (58%) and specificity (100%). The δ-SUV of FLT-PET played no predictive role for clinical response (AUC = 0.62; P= 0.134) and disease control (AUC = 0.66; P= 0.157). The univariate Cox regression analysis revealed no significant prognostic impact. FDG uptake reduction in liver metastases could predict both clinical response (P= 0.010) and disease control status (P= 0.002) at thresholds of 35% and 15%, respectively. Those with greater FDG uptake reduction in liver lesions had a longer overall survival (P= 0.004).
Conclusions: Early metabolic change in FDG-PET might be a predictive marker for response and disease control in advanced gastric cancer. Early FDG uptake change in liver metastases might be a useful prognostic factor and needs further exploration.
©2015 American Association for Cancer Research.