FAP-Specific Signalling Is an Independent Diagnostic Approach in ACC and Not a Surrogate Marker of MRI Sequences

SIMPLE SUMMARY: Fibroblast Activation Protein (FAP) is a new target for positron emission tomography and computed tomography (PET/CT) imaging of epithelial tumours, such as adenoid cystic carcinomas (ACCs). The current gold standard for ACC imaging is contrast enhanced (ce) MRI, where intertumoural heterogeneity leads to variable appearance on T1-weighted (T1w) and T2-weighted (T2w) images. The correlation of (68)Gallium ((68)Ga)-labelled FAP-Inhibitor (FAPI) PET signals and MRI signals (contrast-enhanced T1w and T2w) of ACCs has not been described yet. Here, we analysed the correlation of (68)Ga-FAPI PET signals and MRI signals of 12 ACC patients. (68)Ga-FAPI PET signals showed a very weak positive correlation with ceT1w values (pooled correlation 0.114, 0.147 and 0.162 at 10, 60 and 180 min) and a weak negative correlation with T2w values. These results underline that (68)Ga-FAPI PET signalling is not only a surrogate marker of MRI sequences but an independent signal in ACC patients. ABSTRACT: Background: Fibroblast Activation Protein (FAP) is a new target for positron emission tomography and computed tomography (PET/CT) imaging of epithelial tumours embedded in a fibrous stroma. Adenoid cystic carcinomas (ACCs) have shown elevated tracer uptake in (68)Gallium ((68)Ga)-labelled FAPIs in previous studies. The current gold standard for ACC imaging is contrast-enhanced (ce) MRI, where intertumoural heterogeneity leads to variable appearance on T1-weighted (T1w) and T2-weighted (T2w) images. In this retrospective analysis, we correlated (68)Ga-FAPI PET signalling at three time points with ceT1w and T2w MRI signals to further characterise the significance of (68)Ga-FAPI uptake in ACCs. Methods: Clinical PET/CT scans of 12 ACC patients were performed at 10, 60 and 180 min post i.v. administration of (68)Ga-labelled-FAPI tracer molecules. (68)Ga-PET- and corresponding MRI-scans were co-registered, and 3D volumetric segmentations were performed on ceT1w and T2w lesions of co-registered MRI slides. Signal intensity values of (68)Ga-FAPI PET signalling and ceT1w/T2w MRI scans were analysed for their pixelwise correlation in each patient. Pooled estimates of the correlation coefficients were calculated using the Fisher z-transformation. Results: (68)Ga-FAPI PET signals showed a very weak positive correlation with ceT1w values (pooled correlation 0.114, 0.147 and 0.162 at 10, 60 and 180 min) and a weak negative correlation with T2w values (pooled correlation −0.148, −0.121 and −0.225 at 10, 60 and 180 min). Individual r-values at 60 min ranged from −0.130 to 0.434 in ceT1w and from −0.466 to 0.637 in T2w MRI scans. Conclusion: There are only slight correlations between the intensity of (68)Ga-FAPI PET signals and tumour appearance in ceT1w or T2w MRI scans, which underlines that (68)Ga-FAPI PET signalling is not a surrogate marker of MRI sequences but an independent signal.