Systematic review of the diagnostic accuracy of (18)F-fluorodeoxyglucose positron emission tomography in melanoma patients

Abstract

Background: Positron emission tomography (PET) with (18)F-fluorodeoxyglucose (FDG) is a rapidly developing new imaging technique in the diagnosis and staging of melanoma. The objective of the current study was to determine the diagnostic accuracy of FDG-PET in patients with melanoma.

Methods: A systematic review and meta-analysis of clinical studies regarding FDG-PET and cutaneous melanoma was conducted. Studies were identified by a comprehensive search of the MEDLINE, EMBASE, and Current Contents databases, without any language restrictions. Eleven studies were selected. The methodologic quality of these studies was assessed independently by two reviewers. Levels of evidence and grades of recommendation were determined for each study. Six studies could be included in the statistical pooling. Sources of heterogeneity were studied by meta-regression of the diagnostic odds ratio (DOR). A summary receiver operating characteristic curve was calculated.

Results: The pooled sensitivity and specificity of FDG-PET in the detection of melanoma metastases were 0.79 (95% confidence interval [95% CI], 0.66-0.93) and 0.86 (95% CI, 0.78-0.95), respectively. The pooled DOR of 33.1 (95% CI, 21.9-54.0) suggests a high diagnostic accuracy for PET. Subgroup analysis revealed that PET is more accurate for systemic staging (DOR of 36.4) than for regional staging (DOR of 19.5). When used for regional staging, PET performed better in patients with American Joint Committee on Cancer Stage III disease, compared with patients with Stage I and Stage II disease. However, the methodologic quality of the studies was limited. Major problems were verification, review, and selection bias.

Conclusions: Due to the poor methodologic quality of the available studies, to the authors' knowledge it is yet not possible to develop guidelines for the effective use of PET in patients with melanoma. Future accuracy studies should meet the methodologic criteria outlined in the current review.