Dear Sir,In the letter to the editor regarding the EANM guidelinefor FDG PET and PET/CT tumour imaging [1]theauthors raise two issues, i.e.partial volume effects (PVE)and use of total lesion glycolysis (TLG) as a quantitativeparameter, which might not have been addressed in theguideline. PVE does indeed affect quantification oftumour uptake causing an increasing negative bias insmaller tumours. PVE has, however, been addressed in theguideline. Rather than promoting the use of PVC methods,the guideline aims at matching SUV contrast recoveriesbetween different PET/CT systems and institutes, therebymatching partial volume effects, as a combined effect ofscanner resolution, image reconstruction algorithms andsettings (zoom, voxel size and filters). To this end aspecific multi-centre image quality QC is suggested. Thisstrategy is proposed as, unlike the authors of the letterstate, there are not yet validated, accepted and widelyavailable PVC methods for use in oncology FDG PETstudies. Although there is much experience in using PVCin oncology and for PET brain studies, there is still largevariability regarding methodology being used or explored[2]. For example, most frequently used methods for PVCof brain studies depend e.g. on a coregistered MRI imageand grey and white matter segmentations, for which againmany different algorithms exist. The quality of theacquired MRI data as well as the algorithms being usedfor coregistration of PETand MRI images, grey and whitematter segmentations and PVC can have a large effect onthe quantitative results of these studies, as was alsoobserved by Zaidi et al. [3]. Likewise, use of PVC inoncology FDG PETstudies is still in an exploratory phase.Yet, it should be noted that further development of PVCmethods both for brain as well as for oncology studies ishighly supported by the authors of the EANM guidelineand, in fact, many are involved in the development andevaluation of these methods [2]. The EANM guidelinedoes not discourage improvement of PET technology or(correction) algorithms, but aims at defining minimalstandards to allow for multi-centre quantitative PETstudies that should be feasible to be carried out in aclinical setting. Future guidelines will probably recom-mend use of PVC provided that these methods arevalidated, accepted and are widely available, i.e. can beroutinely applied in a multi-centre setting.As indicated in the letter to the editor, apart fromSUV, other parameters such as TLG and/or metabolic
Read full abstract
Jun 1, 2018
Mai Do + 3
Open Access
