18F-Florbetapir Binds Specifically to Myocardial Light Chain and Transthyretin Amyloid Deposits: Autoradiography Study

Abstract

Background: (18)F-florbetapir is a promising imaging biomarker for cardiac light chain amyloidosis (AL) and transthyretin amyloidosis (ATTR). Our aim, using human autopsy myocardial specimens, was to test the hypothesis that (18)F-florbetapir binds specifically to myocardial AL and ATTR amyloid deposits.

Methods and results: We studied myocardial sections from 30 subjects with autopsy-documented AL (n=10), ATTR (n=10), and nonamyloid controls (n=10) using (18)F-florbetapir and cold florbetapir compound and digital autoradiography. Total and nonspecific binding of (18)F-florbetapir was determined using the maximum signal intensity values. Specific binding of (18)F-florbetapir was calculated by subtracting nonspecific from total binding measurements (in decays per minute/mm(2), DPM mm(2)) and was compared with cardiac structure and function on echocardiography and the histological extent of amyloid deposits. Diffuse or focally increased (18)F-florbetapir uptake was noted in all AL and ATTR samples and in none of the control samples. Compared with control samples, mean (18)F-florbetapir-specific uptake was significantly higher in the amyloid samples (0.94±0.43 versus 2.00±0.58 DPM/mm(2); P<0.001), and in the AL compared with the ATTR samples (2.48±0.40 versus 1.52±0.22 DPM/mm(2); P<0.001). The samples from subjects with atypical echocardiographic features of amyloidosis showed quantitatively more intense (18)F-florbetapir-specific uptake compared with control samples (1.50±0.17 versus 0.94±0.43 DPM/mm(2); P=0.004), despite smaller amyloid extent than in subjects with typical echocardiograms.

Conclusions: (18)F-florbetapir specifically binds to myocardial AL and ATTR deposits in humans and offers the potential to screen for the 2 most common types of myocardial amyloid.

Keywords: amyloid; autoradiography; echocardiography; florbetapir; pathology; positron-emission tomography; radioisotopes.

© 2015 American Heart Association, Inc.

Figures

Figure 1. ROI

A large region of interest (ROI) was drawn on the 18F-florbetapir autoradiography image to measure the total (A, 18F-florbetapir) and nonspecific (B, unlabeled florbetapir + labeled 18F-florbetapir) binding.

Figure 2. Histology and corresponding 18F-florbetapir autoradiography images in myocardial samples with and without amyloidosis

Sulfated Alcian blue stained images (left, amyloid staining green and myocytes staining yellow) and digital autoradiography images of myocardial samples with extensive ATTR (a), moderate diffuse AL (b), and non-amyloid control (c- myocytes in pink and fibrosis in blue). The middle and the right columns depict total and non-specific 18F-florbetapir binding respectively (amyloid staining dark). The 18F-florbetapir images are displayed using the same window level. The red arrows in (a) point to a focus of malignant melanoma in the heart showing no 18F-florbetapir specific uptake on the autoradiography image.

Figure 3. Distribution of 18F-florbetapir specific binding in AL, ATTR, and control samples

18F-florbetapir specific binding was significantly lower in control samples compared to the AL and to the ATTR samples. The myocardial samples from subjects with atypical echocardiograms and low amyloid extent (red) showed higher specific binding compared to the mean value of controls, but lower specific binding compared to the respective AL and the ATTR samples. DPM = decays per minute.

Figure 4. Distribution of (A) the histological extent of amyloid and (B) 18F-florbetapir specific binding normalized to extent of amyloid in AL and, ATTR samples

These box and whiskers plots show the distribution of the extent of amyloid and 18F-florbetapir specific binding in AL and ATTR samples. The whiskers indicate the range of values for extent of amyloid. Despite a trend to lower extent of amyloid in the AL samples, the mean 18F-florbetapir specific binding normalized to the extent of myocardial amyloid was significantly higher in AL samples compared to ATTR samples.

Figure 4. Distribution of (A) the histological extent of amyloid and (B) 18F-florbetapir specific binding normalized to extent of amyloid in AL and, ATTR samples

These box and whiskers plots show the distribution of the extent of amyloid and 18F-florbetapir specific binding in AL and ATTR samples. The whiskers indicate the range of values for extent of amyloid. Despite a trend to lower extent of amyloid in the AL samples, the mean 18F-florbetapir specific binding normalized to the extent of myocardial amyloid was significantly higher in AL samples compared to ATTR samples.

Figure 5. Imaging ATTR (A) and AL (B) amyloidosis: Echocardiography, histology and molecular imaging

Echocardiograms (left, 2D-parasternal long axis view), sulfated Alcian blue stained images (middle, amyloid staining green and myocytes staining yellow) and digital autoradiography (right, amyloid staining dark) images of myocardial samples of subjects with mild moderate and severe ATTR (A) and AL (B) amyloidosis. These images demonstrate that increased left ventricular (LV) wall thickness is evident at advanced stages of the disease, while increased 18F-florbetapir total binding on autoradiography is seen even with small amounts of amyloid before an overt increase in LV wall thickness. The 18F-florbetapir images are displayed using the same window level.

Figure 5. Imaging ATTR (A) and AL (B) amyloidosis: Echocardiography, histology and molecular imaging

Echocardiograms (left, 2D-parasternal long axis view), sulfated Alcian blue stained images (middle, amyloid staining green and myocytes staining yellow) and digital autoradiography (right, amyloid staining dark) images of myocardial samples of subjects with mild moderate and severe ATTR (A) and AL (B) amyloidosis. These images demonstrate that increased left ventricular (LV) wall thickness is evident at advanced stages of the disease, while increased 18F-florbetapir total binding on autoradiography is seen even with small amounts of amyloid before an overt increase in LV wall thickness. The 18F-florbetapir images are displayed using the same window level.