18F-FDG positron emission tomography for the assessment of histological response to neoadjuvant chemotherapy in osteosarcomas: A meta-analysis

Purpose: The purpose of this meta-analysis was to evaluate the predicting value of fluorine-18-fluorodeoxyglucose positron emission tomography with computed tomography (18F-FDG PET-CT) in the assessment of histological response to neoadjuvant chemotherapy in patients with osteosarcomas. Methods: A detailed search was made in MEDLINE, EMBASE and the Web of Knowledge for relevant original articles published in English; methodological quality of the included studies were also assessed. Two reviewers extracted data independently. Sufficient data was presented to construct a 2 x 2 contingency table. Pooled sensitivity and specificity, positive and negative likelihood ratios were estimated. A summary receiver operating characteristic curve (SROC) was constructed with the Moses' constant of linear model. A chi(2) test was performed to test for heterogeneity. Results: Eight studies comprising 178 patients met the inclusion criteria. The pooled sensitivity and specificity for standardized uptake values (SUV) after chemotherapy (SUV2) <= 2.5 were 0.734 (95% CI, 0.537-0.867) and 0.864 (95% CI, 0.510-0.975), for the ratio of standardized uptake values after (SUV2) to before (SUV1) chemotherapy SUV 2: 1 <= 0.5 were 0.690 (95% CI, 0.497-0.833) and 0.653 (95% CI, 0.492 -0.786), the positive and negative likelihood ratio (LR+/LR-) for SUV2 <= 2.5 were 5.397 (95% CI, 1.169 -24.920) and 0.308 (95% CI, 0.165-0.577), for SUV 2:1 <= 0.5 were 1.989 (95% CI, 1.145-3.457) and 0.475 (95% CI, 0.247-0.915). There was no significant difference between-study heterogeneity for either LR+ or LR- in any of these analyses. The area under the SROC curve for SUV2 <= 2.5 and SUV 2:1 <= 0.5 were 0.81 and 0.72, respectively. Conclusions: The present meta-analysis showed that 18F-FDG PET-CT scan, as measured by the SUV before and after treatment, SUV2 <= 2.5 and SUV 2:1 <= 0.5 are valuable for predicting the histological response to chemotherapy. SUV2 <= 2.5 have better predicting performance than SUV 2:1 <= 0.5. (C) 2012 Elsevier Ltd. All rights reserved.