Abstract
17
Objectives: Diagnosis of immunoglobulin light-chain amyloidosis, which can occur in association with multiple myeloma (MM), is clinically important because unrecognition may result in fatal organ damage. [18F]Florbetaben is a positron emission tomography (PET) tracer that specifically binds to amyloid beta-pleated structure. We aimed to evaluate the feasibility of [18F]Florbetaben PET for detecting amyloid deposits in MM patients and to explore the most optimal method for PET analysis.
Methods: Fourteen patients with multiple myeloma were prospectively enrolled (6 patients with amyloidosis and 8 controls). After injection of 300 MBq of [18F]Florbetaben, dynamic imaging of the kidneys was acquired for 20 minutes. Time-activity curve was generated by using a region of interest placed over the kidney and retention index (RI) was obtained. At 90 minutes post-injection, PET image was acquired, with a scan field from the vertex to mid-thigh. All images were assessed both qualitatively and quantitatively. Comparison between amyloidosis and non-amyloidosis groups was performed by using SUVmax, SUVmean, and SUV ratio (SUVR = tissue of interest/reference region), which were obtained from both sphere and manually drawn volumes of interest (VOI). Amyloid deposition was confirmed according to international consensus guidelines.
Results: All amyloidosis patients showed abnormal tracer uptake while the controls did not. Detection rate by visual analysis was excellent (100%) in the heart, stomach, and tongue, moderate in the kidneys (66.7%), and poor (0%) in the esophagus, liver and colon. Both sphere and manual VOIs showed similar performances in discriminating amyloidosis patients and controls. SUVmean could distinguish between the two groups while SUVmax could not in the stomach. The SUVmean (sphere) in the heart, stomach, tongue were significantly higher in the amyloidosis patients than in the controls. On [18F]Florbetaben PET, the thoracic descending aorta had the most consistent SUV and was chosen as the reference region for SUVR. SUVRs in the heart, stomach, tongue, thyroid, and spleen were significantly different between the two groups. On dynamic analysis, renal uptake peaked within the first minute after injection and reached plateau at around 20 minutes. Two out of the 3 renal amyloidosis patients had significantly higher RI (0.70, 0.83) from the rest.
Conclusions: [18F]Florbetaben PET can accurately detect systemic amyloid deposits in MM patients, especially in the heart, stomach, and tongue. SUVR using SUVmean from sphere VOI seems to be the most optimal method for [18F]Florbetaben PET analysis. For detecting renal amyloidosis, visual and RI assessment may be more helpful than SUV or SUVR.