68Ga-FAPI and 18F-FAPI PET/CT for detection of nodal metastases prior radical cystectomy in high-risk urothelial carcinoma patients.
Background: To determine the best therapeutic strategy for muscle-invasive bladder cancer (BC), the accuracy of lymph node staging is of paramount importance. However, diagnostic performance of conventional computed tomography in BC prior to radical cystectomy (RC) remains unsatisfactory. There is an increased interest in evaluating 18F-FAPI PET/CT for hybrid imaging due to their logistical advantages compared to [68Ga]Ga-based FAPI tracers in clinical routine. Recently, the potential diagnostic value of [68Ga]Ga-FAPI- 46 PET/CT was demonstrated in BC. Thus, we aimed to examine the diagnostic performance of [18F]F-FAPI- 74 and [68Ga]Ga-FAPI- 46 PET/CT for preoperative evaluation of locoregional lymph node metastases.
Methods: Fifty-one patients underwent FAPI PET/CT with either [68Ga]Ga-FAPI- 46 (n = 23) or [18F]F-FAPI- 74 (n = 28) prior to RC and PLND. SUVmax, SUVmean and the ratio between the SUVmax of lymph nodes and the SUVmean of the background (SUVmax_lymph node/SUVmean_background) were assessed. Additionally, short axis diameter (SAD) for a representative lymph node were documented in each lymph node region (n = 123) and compared to histopathological findings. Each scan was interpreted visually and quantitatively. ROC-analyses were performed to determine cut-off values with highest diagnostic accuracy.
Results: 20/123 (16.3%) lymph node regions showed UC lymph node metastases. Histopathologically positive lymph nodes were associated with a significantly higher FAPI uptake compared to negative lymph nodes regarding SUVmax, SUVmean values and SUVmax_lymph node/SUVmean_background ratios. Visual analysis based on FAPI uptake showed a sensitivity and specificity, PPV and NPV of 63.6%, 95.8%, 77.7%, and 92.0% for [68Ga]Ga-FAPI- 46 and 55.5%, 98.1%, 83.3%, and 93.1% for [18F]F-FAPI- 74, respectively. ROC analysis revealed an optimal cut-off for SUVmax, SUVmean and SUVmax_lymph node/SUVmean_background of 1.35, 1.20 and 5.95 for [68Ga]Ga-FAPI- 46 and 1.55, 1.25 and 4.15 for [18F]F-FAPI- 74 to discriminate between histopathologically proven lymph node metastases and non-malignant lymph nodes resulting for example using SUVmax in a sensitivity and specificity, PPV and NPV of 81.8%, 89.5%, 64.2%, 95.5% for [68Ga]Ga-FAPI- 46 and 100%, 81.8%, 47.3%, 100% for [18F]F-FAPI- 74, respectively. CT visual analysis of locoregional lymph nodes showed a sensitivity, specificity, PPV and NPV of 30.0%, 97.0%, 66.6% and 87.7%, respectively. ROC analysis regarding SAD revealed a cutoff at 0.8 cm with a sensitivity, specificity, PPV and NPV of 75.0%, 84.4%, 48.3%, 94.5%, respectively.
Conclusions: Overall, FAPI PET imaging shows a significantly higher sensitivity than CT analysis for detection of locoregional lymph node metastases in UC. [18F]F-FAPI- 74 demonstrates a comparable diagnostic performance compared to [68Ga]Ga-FAPI- 46. Of note, the quantitative analysis with a pre-defined SUVmax as well as SUVmean values, and SUVmax_lymph node/SUVmean_background ratio-based cut-offs provided a higher sensitivity compared to visual assessment.