Prognostic value of interim FDG-PET in Hodgkin lymphoma: systematic review and meta-analysis

Abstract

This study aimed to systematically review and meta-analyse the value of interim (18) F-fluoro-2-deoxy-d-glucose positron emission tomography (FDG-PET) in predicting treatment failure in Hodgkin lymphoma. MEDLINE was systematically searched for original studies that used standardized international criteria for interim FDG-PET interpretation. Included studies were methodologically assessed. Summary receiver operating characteristic (sROC) analysis was performed, and pooled sensitivity and specificity were calculated using a random effects model. Heterogeneity in diagnostic odds ratios (DORs) across studies was assessed and potential sources for inter-study heterogeneity were explored using subgroup analyses. Ten studies, comprising a total of 1389 Hodgkin lymphoma patients, were included. Sensitivity, specificity, positive predictive value and negative predictive value of interim FDG-PET for predicting treatment failure ranged between 0.0-81.5%, 72.2-96.6%, 0.0-86.0% and 84.4-98.6%, respectively. The area under the sROC curve was 0.877. Pooled sensitivity and specificity were 70.8% [95% confidence interval (CI): 64.7-76.4%] and 89.9% (95% CI: 88.0-91.6%). There was heterogeneity in DORs across individual studies (I2 = 72.7). The overall prognostic value of interim FDG-PET appears to be moderate for excluding and relatively high for identifying treatment failure in Hodgkin lymphoma. However, interim FDG-PET cannot yet be implemented in routine clinical practice due to moderate-quality evidence and inter-study heterogeneity that cannot be fully explained yet.