We read with interest the letter to the editor by Huang et al.1 The authors raised several concerns regarding the methodology and conclusions of our recent meta-analysis that aimed to compare the performance of 18F-FET-PET with 18F-FDG-PET for diagnosing and grading new brain tumors.2
As noted by the authors, only a few articles have directly compared both tracers. The respective performance thus needs more exploration to help in choosing the best examination for daily practice. Higher performances were demonstrated with 18F-FET-PET compared with 18F-FDG-PET for the diagnosis of brain tumors (our Fig. 2 in2) and glioma. We also reported performance heterogeneity and inconsistency among the 5 selected studies with 18F-FDG-PET but not with 18F-FET-PET. Huang et al1 noted that inclusion criteria and population characteristics were different between studies, which seems to introduce an analytic bias from their point of view. First, while it may sustain performance heterogeneity with 18F-FDG-PET, these differences did not result in significant heterogeneity for 18F-FET-PET performance, as reported in the Results section and measured by Cochrane Q. Second, the per-patient analysis allowed pooling all patients regardless of the pre-test probability (ie, whether or not there was contrast enhancement on MRI). Indeed, this avoids a selection bias based on MRI findings and reflects the spectra of patients’ disease in daily practice. Pooling patients from studies of brain lesions with different histological characteristics can thus, to our point of view, not be considered a bias of this meta-analysis.
However, we agree with Huang et al, who noted that there were differences in imaging protocols and analyses among the 5 studies. These may have influenced reported tumor-to-background ratio and cutoff values. We have reported and discussed this limitation for 18F-FET-PET in a previous meta-analysis3 and have supported the need to standardize PET imaging and analysis protocols using amino acid analogues, as recommended by the European Association of Nuclear Medicine.4 Moreover, as noticed by Rapp et al5 and discussed in the present meta-analysis, several of these 18F-FET-PET studies came from a small number of centers due to the low availability of 18F-FET. This highlights the need for larger multicenter standardized studies to avoid both selection bias and overlapping populations.
With the aim of exploring the performance of both 18F-FDG and 18F-FET-PET in patients with new brain tumor regardless of their MRI characteristics, we concluded that 18F-FET-PET should be preferred in daily practice when it is available. However, while keeping in mind the limitations of 18F-FDG-PET, the use of 18F-FDG metrics could be an acceptable alternative to help with grading glioma, for example, when biopsy would come with a high complication risk.
Conflict of interest statement. None declared.
References
- 1.Huang X, Bai H, Zhou H, Tang H, Yang L. Performance of 18F-FET-PET versus 18F-FDG-PET for the diagnosis and grading of brain tumors: inherent bias in meta-analysis not revealed by quality metrics. Neuro Oncol. 2016;18(7):1028–1029. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Dunet V, Pomoni A, Hottinger A, Nicod-Lalonde M, Prior JO. Performance of 18F-FET versus 18F-FDG-PET for the diagnosis and grading of brain tumors: systematic review and meta-analysis. Neuro Oncol. 2016;18(3):426–434. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Dunet V, Rossier C, Buck A, Stupp R, Prior JO. Performance of 18F-fluoro-ethyl-tyrosine (18F-FET) PET for the differential diagnosis of primary brain tumor: a systematic review and Metaanalysis. J Nucl Med. 2012;53(2):207–214. [DOI] [PubMed] [Google Scholar]
- 4.Vander Borght T, Asenbaum S, Bartenstein P et al. EANM procedure guidelines for brain tumour imaging using labelled amino acid analogues. Eur J Nucl Med Mol Imaging. 2006;33(11):1374–1380. [DOI] [PubMed] [Google Scholar]
- 5.Rapp M, Heinzel A, Galldiks N et al. Diagnostic performance of 18F-FET PET in newly diagnosed cerebral lesions suggestive of glioma. J Nucl Med. 2013;54(2):229–235. [DOI] [PubMed] [Google Scholar]