Objective: The purpose of this study was to assess the predictive value of (18)F-FDG PET/ CT for pathologic response to neoadjuvant chemoradiotherapy in locally advanced rectal cancer.
Materials and methods: A systematic search was performed (PubMed and Cochrane databases) for potentially relevant studies up to January 2014. Pooled sensitivity and specificity were calculated. The AUCs of the global cohort and for "major response," "complete response," "only PET after chemoradiotherapy," "ad interim PET," "major response excluding ad interim studies," "complete response excluding ad interim studies," "response index (RI)," "posttreatment maximum standardized uptake value (SUV(max) post)," "visual response analysis (VRA)," "percentage reduction of the total lesion glycolysis (ΔTLG)," and "percentage reduction of the metabolic tumor volume (ΔMTV)" were analyzed. Heterogeneity was explored for multiple variables. Pooled prevalence and 95% CI were calculated.
Results: Thirty-four of 131 (26%) initial articles met the inclusion criteria. Those articles included 1526 patients. PET/CT showed good pooled accuracy both in the global cohort (pooled sensitivity, 73%; pooled specificity, 77%; pooled AUC, 0.83) and for subgroups. Pooled accuracy was similar for early PET restaging and at 1 and 2 weeks after beginning chemoradiotherapy (pooled sensitivity, 84%; pooled specificity, 81%; pooled AUC, 0.89). The major response group showed similar sensitivity to the complete response group (74% and 71%, respectively). RI, SUV(max), and VRA were the most frequent parameters used. Pooled RI and SUV(max) postcutoff values were 63% and 4.4. Pooled time to PET during and after chemoradiotherapy was 1.5 and 6.5 weeks, respectively.
Conclusion: This meta-analysis supports the use of FDG PET for restaging locally advanced rectal cancer.
Keywords: FDG PET/CT; locally advanced rectal cancer; meta-analysis; predictive value; preoperative chemoradiotherapy.