[68Ga]Ga-DATA5mSA.FAPi PET detects a subgroup of neuroendocrine tumor patients with high risk of progression

Introduction: Gallium-68-labeled inhibitors of the fibroblast activation protein (FAPi) enable positron emission tomography/computed tomography (PET/CT) imaging of fibroblast activation. We performed [⁶⁸Ga]Ga-DATA^5m.SA.FAPi (FAPi) PET in a collective of patients with progressive NET with SSTR-positive metastases in [⁶⁸Ga]Ga-DOTA-TOC-PET and determined (i) the number of patients with FAPi-positive metastases, (ii) the number of patients with both FAPi-positive and -negative metastases and (iii) the number of patients with mismatch metastases only positive for FAPi and negative for DOTA-TOC, but detectable by CT. Results were correlated with tumor grading as determined by Ki-67 values as well as clinical progression of disease and survival.

Methods: We analyzed a collective of 55 patients with metastasized SSTR-positive NET. The age was in mean 68 ± 9 years. 33 patients were male and 22 were female. Of these 23 had intestinal NET, 21 pancreatic NET, 7 patients had CUP-NET and 4 lung NET. 14 patients had G1-, 30 G2- and 11 G3 tumors. The follow-up period was 25 ± 7 month. All patients underwent PET/CT with [⁶⁸Ga]Ga-DATA^5m.SA.FAPi and [⁶⁸Ga]Ga-DOTA-TOC. PET/CT examinations were performed on a Vereos PET/CT scanner (Philips Healthcare, Eindhoven, Netherlands). Mean injected activities were 184 ± 22 MBq for [⁶⁸Ga]Ga-DATA^5m.SA.FAPi, and 161 ± 18 MBq for [⁶⁸Ga]Ga-DOTA-TOC. Mean incubation times were 79 ± 34 minutes for [⁶⁸Ga]Ga-DATA^5m.SA.FAPi and 42 ± 16 minutes for [⁶⁸Ga]Ga-DOTA-TOC. PET data were reconstructed iteratively, including scatter, random and decay correction RAMLA 3D, 2 iterations, 0.05 relaxation. Reconstructed PET images were using a 5 mm Gaussian filter.

Results: In our collective only 3 patients (5%) showed FAPI-negative metastases and 52 (95%) were FAPI-positive (group i). 34 patients (62%) showed heterogeneous FAPi-expression of DOTA-TOC-positive tumor-lesions with FAPI-positive and -negative metastases (group 2). 9 patients (16%) showed FAPi-positive and CT-detectable but DOTA-TOC-negative metastases, referred to as mismatch lesions (group iii). In the whole collective 28 patients (51%) showed progressive disease within the follow-up period. In the group with mismatch lesions 100% of patients had progressive disease. In the heterogeneous group 17 patients (50%) had progressive disease. Time to progression was not significantly different in the whole collective, the heterogeneous and the mismatch group (13.6 vs. 12.2 vs. 13.6 months). Proliferation rate as determined by Ki67 immunohistochemistry was not significantly different in the whole collective, the heterogeneous- and mismatch group. In the heterogeneous group Ki67 was in mean 14.3  19.1, in the mismatch group 12.912.2. Time to death and death rates were not significantly different in the groups. The number of PRRT in each group was not significantly different, as in the whole collective received 5.4 cycles of PRRT, 5 in the discordant group and 5.3 in the heterogeneous group.

Conclusions: In this retrospective analysis we found that in 95% of patients with metastasized DOTA-TOC-positive NET FAPi-PET was positive (i). 34% of patients showed FAPi-positive and -negative metastases (ii). 9% of patients showed metastases that were only FAPi-positive and did not match with DOTA-TOC-positive metastases (iii). The mismatch group showed progression within approximately one year in all patients. In contrast, in the whole collective and the group with heterogeneous metastases only 50% of patients showed progression, indicating that metastases positive for FAPi and negative for DOTA-TOC define at patient group with aggressive disease. This is remarkable as the Ki67-values in the three groups analyzed were not significantly different. Therefore, FAPi-PET may help to detect a subgroup of patients with FAPi-positive mismatch metastases who have a high risk of progression.