Parametric net influx rate imaging of 68 Ga-DOTATATE in patients with neuroendocrine tumors: assessment of lesion detectability.

OBJECTIVES: There has been developed a clinical dynamic total-body 68 Ga-DOTATATE PET/CT imaging protocol that allows quantitative imaging of net influx rate (Ki ). Using qualitative and quantitative analyses of clinical studies, this retrospective study aims to assess whether parametric Ki images improve lesion detectability.

METHODS: Using a 194-cm axial field-of-view PET/CT scanner, 52 patients with neuroendocrine tumors underwent a 60-min dynamic total-body 68 Ga-DOTATATE scan. Parametric Ki images and static standardized uptake value (SUV) images were generated. In addition to visual inspection of both sets of images, a quantitative analysis of 249 individual lesions was conducted using the target-to-background (TBR) metric.

RESULTS: There were 52 patients who underwent dynamic total-body 68 Ga-DOTATATE PET/CT scans. A total of 249 lesions were evaluated, of which 66 lesions were biopsy-proven and 183 lesions were unproven. Ki images produced two fewer false positives than the SUV images. Overall, our results from 66 proven NET lesions suggested similar sensitivity (98.5%) but improved accuracy (from 95.6 to 97.1%) and potentially enhanced specificity with Ki over SUV imaging. Besides, there was no difference in the number of pathological lesions identified visually in both images. However, Ki TBR was significantly higher than SUV TBR quantitatively (P < 0.001).

CONCLUSIONS: Patlak Ki imaging provides nuclear physicians with a PET image with higher tumor contrast which may enhance confidence in diagnosis with possibly reduced false positive results, albeit an equivalent detectability, compared to static SUV image.