Objective: The aim of this study was to assess the performance of fluorine-18-fluorodeoxyglucose (18F-FDG) positron emission tomography (PET) in predicting pathological response to preoperative chemoradiotherapy (CRT) in patients with primary rectal cancer.
Methods: Potentially relevant articles were searched in the databases of PubMed and Embase from January 1990 to September 2013. The Quality Assessment for Diagnostic Accuracy Studies criteria was employed to assess the quality of all of the included studies. The pooled sensitivity and specificity were calculated, and the area under the curve of the summary receiver operating characteristic curve was obtained. Subgroup analysis was conducted to explore the sources of heterogeneity.
Results: Thirty-one eligible studies involving 1527 patients were ultimately included in the meta-analysis. Four main quantitative or qualitative parameters [response index (RI), post-treatment maximum standardized uptake value (SUVmax-post), visual response (VR) and the percentage change in total lesion glycolysis (TLG) before and after CRT (deltaTLG%)] related to PET or positron emission tomography/computed tomography (PET/CT) were assessed for the prediction of histopathological response. The pooled sensitivities of these four parameters were comparable and were 74, 74, 75 and 78%, respectively (P>0.05). The pooled specificity of deltaTLG% was higher than that of the other three parameters (RI, SUVmax-post and VR) and was 81, 66, 64 and 67%, respectively (P<0.05). The results from subgroup analysis showed that the RI and SUVmax-post had higher specificity in predicting tumor regression grade (TRG) than complete pathological response (pCR) [RI, 71 vs. 59% (P=0.0275); SUVmax-post, 72 vs. 61% (P=0.0178)].The diagnostic sensitivity and specificity of the RI and SUVmax-post when the post-treatment PET or PET/CT scan was performed at two different time points (during CRT and after the completion of CRT) were 82 vs. 72% (P=0.0630) and 78 vs. 63% (P=0.0059), respectively.
Conclusions: 18F-FDG PET could be a potentially powerful non-invasive tool for predicting pathological response; the related parameters RI and SUVmax-post may be more suitable for the prediction of TRG than pCR. The current data also suggested that the optimum post-treatment 18F-FDG PET scan could be carried out during CRT.