Progression-free survival prediction and comparative metastasis detection accuracy of ¹⁸F-FES PET/CT for ER-Positive breast cancer: Systematic review, meta-analyses and meta-regressions.

Abstract

The role of ¹⁸F-fluoroestradiol (¹⁸F-FES) PET/CT has been investigated in breast cancer recurrence identification and noninvasive hormone receptor profiling. The detection of distant metastasis, however, lacks high quality evidence comparing FES to standard imaging protocols. Additionally, if ¹⁸F-FES PET/CT is to become one of the first-line exams in ER positive breast cancer patients, obtaining its maximum utility will be beneficial and this includes prognosis prediction. Here, we address these issues by analyzing literature on diagnostic accuracy comparisons and reconstructing progression-free survival data to investigate the ideal PET parameter for predicting survival. A systematic review was conducted on PubMed, Scopus and Embase to identify studies assessing survival or diagnostic outcomes on breast cancer patients undergoing ¹⁸F-FES PET/CT for metastasis detection. Random-effects inverse variance meta-analyses were conducted for summarizing hazard ratios, sensitivity, specificity and likelihood ratios. Meta-regressions were conducted to compare ¹⁸F-FES PET/CT and ¹⁸F-FDG PET/CT; and to evaluate the impact of clinical covariates on the survival prediction potential of ¹⁸F-FES PET/CT. Risk of bias was assessed using the QUADAS-2 and QUADAS-C tools, for diagnostic studies; and the Evidence Project risk of bias tool for the survival trials. 19 papers were included. 10 studies had survival outcomes, and 9 had diagnostic endpoints. Overall, patients with higher ¹⁸F-FES SUV_max, complete lesion concordance or no complete lesion discordance had significantly longer PFS (HR:0.29 [0.16-0.54]). Stratification by complete lesion concordance had numerically longer PFS than stratification by ¹⁸F-FES SUV_max. Out of the 9 metastasis detection studies, 3 were pooled in the bivariate model, resulting in overall estimates for ¹⁸F-FES PET/CT and ¹⁸F-FDG PET/CT or CT, respectively, of 87.9% [82.6-91.7%] and 83.4% [80-87%] for sensitivity; and 97.1% [85.4-99.5%] and 45.5% [4.8-93.2%] for specificity. For confirmation, 6 studies had their sensitivities meta-analyzed in a univariate approach, with similar results. A bivariate meta-regression showed a significant difference in sensitivity between modalities. Patients with higher ¹⁸F-FES SUV_max or full lesion concordance on ¹⁸F-FES/FDG PET/CT have significantly longer PFS. Lesion concordance approaches were associated with longer PFS. ¹⁸F-FES PET/CT has slightly higher sensitivity than ¹⁸F-FDG PET/CT for metastasis detection in the ER+/HER2- metastatic breast cancer setting. Cost-benefit analysis will be essential in the development of guidelines to integrate its clinical applicability, on a case-by-case basis.