TY - JOUR
T1 - 68Ga-Labeled Fibroblast Activation Protein Inhibitor (68Ga-FAPI) PET for Pancreatic Adenocarcinoma
T2 - Data from the 68Ga-FAPI PET Observational Trial
AU - Kessler, Lukas
AU - Hirmas, Nader
AU - Pabst, Kim M.
AU - Hamacher, Rainer
AU - Ferdinandus, Justin
AU - Schaarschmidt, Benedikt M.
AU - Milosevic, Aleksandar
AU - Nader, Michael
AU - Umutlu, Lale
AU - Uhl, Waldemar
AU - Reinacher-Schick, Anke
AU - Lugnier, Celine
AU - Witte, David
AU - Niedergethmann, Marco
AU - Herrmann, Ken
AU - Fendler, Wolfgang P.
AU - Siveke, Jens T.
N1 - Publisher Copyright:
COPYRIGHT ß 2023 by the Society of Nuclear Medicine and Molecular Imaging.
PY - 2023
Y1 - 2023
N2 - The fibroblast activation protein (FAP) is highly expressed on carcinoma-associated fibroblasts in the stroma of pancreatic cancer and thus is a promising target for imaging and therapy. Preliminary data on PET imaging with radiolabeled FAP inhibitors (FAPIs) demonstrate superior tumor detection. Here we assess the accuracy of FAP-directed PET in patients with pancreatic cancer. Methods: Of 64 patients with suspected or proven pancreatic cancer, 62 (97%) were included in the data analysis of the 68Ga-FAPI PET observational trial (NCT04571086). All of these patients underwent contrast-enhanced CT, and 38 patients additionally underwent 18F-FDG PET. The primary study endpoint was the association of 68Ga-FAPI PET uptake intensity and histopathologic FAP expression. Secondary endpoints were detection rate, diagnostic performance, interreader reproducibility, and change in management. Datasets were interpreted by 2 masked readers. Results: The primary endpoint was met: The association between 68Ga-FAPI SUVmax and histopathologic FAP expression was significant (Spearman r, 0.48; P 5 0.04). For histopathology-validated lesions, 68Ga-FAPI PET showed high sensitivity and positive predictive values (PPVs) on per-patient (sensitivity, 100%; PPV, 96.3%) and per-region (sensitivity, 100%; PPV, 97.0%) bases. In a head-to-head comparison versus 18F-FDG or contrast-enhanced CT, 68Ga-FAPI detected more tumor on a per-lesion (84.7% vs. 46.5% vs. 52.9%), per-patient (97.4% vs. 73.7% vs. 92.1%), or per-region (32.6% vs. 18.8% vs. 23.7%) basis, respectively. 68Ga-FAPI PET readers showed substantial overall agreement on the basis of the Fleiss k: primary k, 0.77 (range, 0.66–0.88). Minor and major changes in clinical management occurred in 5 patients (8.4%) after 68Ga-FAPI PET. Conclusion: We confirmed an association of 68Ga-FAPI PET SUVmax and histopathologic FAP expression in pancreatic cancer patients.
AB - The fibroblast activation protein (FAP) is highly expressed on carcinoma-associated fibroblasts in the stroma of pancreatic cancer and thus is a promising target for imaging and therapy. Preliminary data on PET imaging with radiolabeled FAP inhibitors (FAPIs) demonstrate superior tumor detection. Here we assess the accuracy of FAP-directed PET in patients with pancreatic cancer. Methods: Of 64 patients with suspected or proven pancreatic cancer, 62 (97%) were included in the data analysis of the 68Ga-FAPI PET observational trial (NCT04571086). All of these patients underwent contrast-enhanced CT, and 38 patients additionally underwent 18F-FDG PET. The primary study endpoint was the association of 68Ga-FAPI PET uptake intensity and histopathologic FAP expression. Secondary endpoints were detection rate, diagnostic performance, interreader reproducibility, and change in management. Datasets were interpreted by 2 masked readers. Results: The primary endpoint was met: The association between 68Ga-FAPI SUVmax and histopathologic FAP expression was significant (Spearman r, 0.48; P 5 0.04). For histopathology-validated lesions, 68Ga-FAPI PET showed high sensitivity and positive predictive values (PPVs) on per-patient (sensitivity, 100%; PPV, 96.3%) and per-region (sensitivity, 100%; PPV, 97.0%) bases. In a head-to-head comparison versus 18F-FDG or contrast-enhanced CT, 68Ga-FAPI detected more tumor on a per-lesion (84.7% vs. 46.5% vs. 52.9%), per-patient (97.4% vs. 73.7% vs. 92.1%), or per-region (32.6% vs. 18.8% vs. 23.7%) basis, respectively. 68Ga-FAPI PET readers showed substantial overall agreement on the basis of the Fleiss k: primary k, 0.77 (range, 0.66–0.88). Minor and major changes in clinical management occurred in 5 patients (8.4%) after 68Ga-FAPI PET. Conclusion: We confirmed an association of 68Ga-FAPI PET SUVmax and histopathologic FAP expression in pancreatic cancer patients.
KW - Ga-FAPI
KW - PDAC
KW - PET
KW - cancer imaging
KW - fibroblast activation protein
KW - pancreatic ductal adenocarcinoma
UR - http://www.scopus.com/inward/record.url?scp=85178651747&partnerID=8YFLogxK
U2 - 10.2967/jnumed.122.264827
DO - 10.2967/jnumed.122.264827
M3 - Article
C2 - 37973185
AN - SCOPUS:85178651747
SN - 0161-5505
VL - 64
SP - 1910
EP - 1917
JO - Journal of Nuclear Medicine
JF - Journal of Nuclear Medicine
IS - 12
ER -
