Eur J Nucl Med Mol Imaging DOI 10.1007/s00259-014-2787-6
ORIGINAL ARTICLE
Imaging cardiac amyloidosis: a pilot study using 18F-florbetapir positron emission tomography Sharmila Dorbala & Divya Vangala & James Semer & Christopher Strader & John R. Bruyere Jr & Marcelo F. Di Carli & Stephen C. Moore & Rodney H Falk
Received: 3 February 2014 / Accepted: 15 April 2014 # Springer-Verlag Berlin Heidelberg 2014
Abstract Purpose Cardiac amyloidosis, a restrictive heart disease with high mortality and morbidity, is underdiagnosed due to limited targeted diagnostic imaging. The primary aim of this study was to evaluate the utility of 18F-florbetapir for imaging cardiac amyloidosis. Methods We performed a pilot study of cardiac 18F-florbetapir PET in 14 subjects: 5 control subjects without amyloidosis and 9 subjects with documented cardiac amyloidosis. Standardized uptake values (SUV) of 18F-florbetapir in the left ventricular (LV) myocardium, blood pool, liver, and vertebral bone were determined. A 18F-florbetapir retention index (RI) was computed. Mean LV myocardial SUVs, target-to-background ratio (TBR, myocardial/blood pool SUV ratio) and myocardial-toliver SUV ratio between 0 and 30 min were calculated. Results Left and right ventricular myocardial uptake of 18Fflorbetapir were noted in all the amyloid subjects and in none S. Dorbala (*) : M. F. Di Carli Noninvasive Cardiovascular Imaging Program, Heart and Vascular Center, Departments of Radiology and Medicine (Cardiology), Brigham and Women’s Hospital, Harvard Medical School, Boston, MA, USA e-mail: [email protected] S. Dorbala : D. Vangala : J. Semer : C. Strader : J. R. Bruyere Jr : M. F. Di Carli : S. C. Moore Division of Nuclear Medicine and Molecular Imaging, Department of Radiology, Brigham and Women’s Hospital, Harvard Medical School, Boston, MA, USA S. Dorbala : R. H. Falk Cardiovascular Division and the Cardiac Amyloidosis Program, Department of Medicine, Brigham and Women’s Hospital, Harvard Medical School, Boston, MA, USA S. Dorbala : D. Vangala : J. Semer : C. Strader : J. R. Bruyere Jr : M. F. Di Carli : S. C. Moore : R. H. Falk Brigham and Women’s Hospital, 70 Francis Street, Shapiro 5th Floor, Room 128, Boston, MA 02115, USA
of the control subjects. The RI, TBR, LV myocardial SUV and LV myocardial to liver SUV ratio were all significantly higher in the amyloidosis subjects than in the control subjects (RI median 0.043 min−1, IQR 0.034 – 0.051 min−1, vs. 0.023 min−1, IQR 0.015 – 0.025 min−1, P=0.002; TBR 1.84, 1.64 – 2.50, vs. 1.26, IQR 0.91 – 1.36, P=0.001; LV myocardial SUV 3.84, IQR 1.87 – 5.65, vs. 1.35, IQR 1.17 – 2.28, P=0.029; ratio of LV myocardial to liver SUV 0.67, IQR 0.44 – 1.64, vs. 0.18, IQR 0.15 – 0.35, P=0.004). The myocardial RI, TBR and myocardial to liver SUV ratio also distinguished the control subjects from subjects with transthyretin and those with light chain amyloid. Conclusion 18F-Florbetapir PET may be a promising technique to image light chain and transthyretin cardiac amyloidosis. Its role in diagnosing amyloid in other organ systems and in assessing response to therapy needs to be further studied. Keywords Amyloidosis . Heart . 18F-Florbetapir . Positron emission tomography . Diagnosis
Introduction Amyloidosis comprises of a large group of disorders characterized by abnormal folding of normal, abnormal or mutant, insoluble extracellular proteins, in the form of a beta pleated structures [1]. Amyloid deposits in the brain cause Alzheimer’s disease, whereas systemic amyloidosis almost always spares the brain. Cardiac amyloidosis is usually a component of systemic amyloidosis due to systemic light chain (AL) and transthyretin (ATTR, wild type or mutant) amyloidosis, although clinical manifestations may be limited to the heart [1]. Subjects with AL and ATTR cardiac amyloidosis have overlapping imaging features on echocardiography and cardiac MRI, but their clinical course and prognosis are distinct [2]. Cardiac AL amyloidosis is characterized by a rapidly progressive clinical course and high mortality, while
Eur J Nucl Med Mol Imaging
ATTR wild-type disease (senile amyloidosis) typically has a slower clinical course with a much longer overall survival. Nevertheless, cardiac involvement is almost invariably the cause of death in both sets of patients [3]. One of the most effective treatment strategies for amyloidosis is reducing the production of the precursor amyloidogenic protein. Halting the synthesis of the amyloid protein may lead to resorption of amyloid deposits and resolution of organ dysfunction [1]. Due to differences in the genesis of AL and ATTR diseases, treatment options are distinct. In the case of systemic AL amyloidosis, novel chemotherapy directed against reducing the production of light chains, as well as stem cell transplantation have considerably improved survival [4]; however, patients with cardiac AL amyloidosis, the subset with the worse prognosis, generally do not tolerate aggressive chemotherapy and stem cell transplantation, limiting treatment options in these patients [5]. Regarding systemic TTR amyloidosis, novel therapies such as TTR stabilizers [6] and small RNA silencing molecules that stabilize the TTR amyloid protein and prevent them from aggregating are undergoing clinical trials [3]. Thus, early detection of cardiac amyloidosis, identification of the type of amyloidosis and institution of specific therapy is critical to improve outcomes. Echocardiography and cardiac MRI are widely used in clinical practice. They rely primarily on imaging increased left ventricular (LV) wall thickness from amyloid deposits, but do not specifically distinguish amyloid from other hypertrophic heart diseases [2]. Amyloid-directed imaging with 123I systemic amyloid protein (SAP) provides a measure of overall systemic amyloid burden, but is not widely available and is not helpful for cardiac imaging due to blood pool uptake resulting in a poor signal to noise ratio in the heart [3, 7]. One recent study has demonstrated the utility of 11C-labeled Pittsburgh Compound B (PiB) in imaging cardiac amyloidosis, but it is only available at PET sites with a cyclotron. Thus, we currently have no widely available targeted imaging techniques to specifically diagnose amyloidosis in the heart. A novel amyloid imaging agent, 18F-florbetapir, structurally distinct from PiB, was recently approved for imaging beta amyloid protein in the brain, with a very high sensitivity for binding to minute amounts of beta amyloid [8, 9]. We therefore studied the feasibility of using 18F-florbetapir in imaging cardiac amyloidosis. The primary aim of this pilot study was to evaluate 18 F-florbetapir in diagnosing cardiac amyloidosis compared to its use in control subjects without amyloidosis. A secondary aim of this study was to determine if 18Fflorbetapir uptake in the heart can help distinguish between AL and ATTR cardiac amyloidosis. We also explored the kinetics of 18F-florbetapir in the liver, lung and bone to understand if it could potentially be used to diagnose amyloidosis in these organ systems.
Materials and methods This was a prospective case control pilot study of 14 subjects, 9 subjects with definite cardiac amyloidosis and 5 control subjects without amyloidosis. Amyloid subjects were recruited from the Brigham and Women’s Hospital Cardiac Amyloidosis Program. Three control subjects were recruited from web advertisements and two control subjects with known nonischemic heart failure without amyloidosis were recruited from our general cardiology service. Cardiac amyloidosis was diagnosed based on endomyocardial biopsy (in seven patients) or a typical amyloid cardiac morphology (wall thickness measurements of >11 mm, bright echogenic myocardium) and evidence of diastolic dysfunction (mitral inflow parameters, pulmonary venous Doppler information, or tissue Doppler imaging at the mitral septal and lateral annulus) with histological confirmation of amyloidosis from extracardiac structures (in two patients). The type of amyloidosis (AL or ATTR) was classified in all subjects based on immunohistochemistry or, in equivocal cases, by mass spectrometry. Subjects with claustrophobia or unable to lie flat for 60 min were excluded. This study was approved by the Partners Human Research Committee and listed on clinical trials.gov (NCT01683825). Each study subject provided written informed consent. Positron emission tomography The scan protocol included a cardiac 3-D mode PET/CT scan (Discovery ST; GE, Waukesha, WI) acquired in a list mode with arms by the side for 60 min. A 60-min scan duration was chosen because the radiotracer distribution in the heart, lung and liver were not previously reported and a long scan allowed us to identify the appropriate time period for future clinical imaging. The scan was terminated prematurely at 20 min into the acquisition in one subject due to severe nausea (AL amyloidosis and on chemotherapy). No specific dietary preparation (no restriction on oral intake) or withholding of medications was required for this protocol. Following a lowdose (10 mA, 120 kVp, free tidal breathing) scout scan to localize the heart, a low-dose CT transmission scan (10 mA, 120 kVp, free tidal breathing) was obtained over the heart for measurement and correction of attenuation. Following the scout and the transmission scans, the emission scan acquisition was started simultaneously with intravenous injection of 6 mCi of 18F-florbetapir and was continued for 60 min. The estimated whole-body radiation dose to the subjects from the 18 F-florbetapir scan was about 5 mSv [10]. The list-mode images were reconstructed into dynamic images using 38 frames (13 frames of 5 s each, 6 frames of 10 s each, 4 frames of 30 s each, 6 frames of 60 s each, 8 frames of 300 s each, and 1 frame of 600 s). The images were reconstructed using a matrix size of 128 × 128, ordered-
B
M 57
41
51
74
43.88
3
2.4
3,003
39
0
M
430
32
–
2
26.5
50
179.6
Pulmonary artery systolic pressure (mm Hg)
5.5
1.45
E/A
3.5
55
A (cm)
0.23
80
E (cm)
E/e’
166.4
426
Deceleration time (ms) e’ (cm)
55
21
LV mass(gm)
–
–
35
0.21
–
–
N/A
16.05
4.25
237
68
620
40
4.3
4.9
2.4
2.4
117
91
2
8.5
1.1
0.4 3.2
0.9
10.2
9.2
37.55
97.31
LV end diastolic diameter (cm) LV end systolic diameter (cm) Ejection fraction (%)
Left atrial volume index (mL/m2) Interventricular septum (cm) Posterior wall (cm)
2-D echocardiographic features
N/A
0.12
6
214.4
1.37
51
70
147
63
2.7
3.7
1.2
1.2
29
14.43
4
316
1.23
47
58
651
45
3.4
4.4
1.5
1.4
40
32
7.5
–
–
160
628
55
3.8
5.4
1.9
30
9.09
6.05
388.1
0.6
91
55
220
70
2.3
4.4
1.3
1.3
16.75
0.127 10.68
M 23.0
1.34
157
≥60
1
1
29.8
M
Kappa/lambda ratio ratio 45.18
9,413
52
0
– 28
1
3
24.5
M
Lambda
1
26.0
M
17
1
27.5
M
0.02
1
28.0
M
Kappa
43
M 34.2
Troponin T
Pro BNP
43
2,641
eGFR
BNP
Laboratory values
0
3
33.8
F
N/A
11.75
8
118
1.74
54
94
192
55
2.4
3.4
1.4
1.7
23.85
1.08
18.1
19.5
0.13
1,063
45
–
1
22.1
F
46.4
23.33
3
126
2.03
34
70
159
50
2.8
3.0
1.3
1.7
24.09
0.49
66.4
32.5
0.17
3,755
≥60
0
3
21.0
F
30
24.56
5
199.6
1.79
68
123
168
65
2.3
3.1
1.5
1.5
26.86
0.15
17.4
2.59
0.16
10,188
54
0
2
22.1
F
31
26.3
3
126
2.31
34
79
164
70
1.2
2.5
1.8
1.6
19.21
0.18
1.14
20.4
0.03
9,288
33
0
3
33.8
F
0
65
3
69
H
CADa
68
P
NYHA class
78
D
25.5
77
T
BMI (kg/m2)
52
G
African American Caucasian Caucasian Caucasian African American Caucasian Caucasian Caucasian Caucasian Caucasian Caucasian Caucasian Caucasian Caucasian
47
B
M
64
D
Sex
74
R
AL
Race
40
S
H
C
O
ATTR
Control
Age (years)
Variable
Table 1 Echocardiographic features of the study subjects
Eur J Nucl Med Mol Imaging
0.052 –b
0.046
subsets expectation maximization (two iterations, 21 subsets) and postfilter full-width at half-maximum 4.48 mm. Static images were then reconstructed with a slice separation of 3.27 mm from data acquired between 10 min (to avoid blood pool activity and improve delineation of the myocardium from blood pool tracer activity) and 60 min after the start of the scan. These images were reoriented into standard cardiac planes (short axis, vertical long axis and horizontal long axis images) for interpretation. Quantitative image analysis
RI could not be computed as imaging was terminated at 20 min due to severe nausea
Coronary artery disease on invasive or CT angiography a
RI
b
0.045 0.024 0.025 0.023
0.020
0.039
0.032
0.031
0.039
0.073
0.042
1+
4+
1+
0
2+
0
2+
0 0
1+ 1+
0 0
2+ 1+
0
2+
0
0
0
0
0
0
0
0
0
0
0 Lung
Atrial
2+
2+
4+
2+
4+
2+
1+
2+ 2+
4+ 3+
2+ 3+
1+ 2+
2+
3+
3+
0
0 0
0 0
0
0
0 Right ventricle
Moderate Moderate
Left ventricle
0
Right ventricular hypertrophy Right atrial enlargement 18 F-Florbetapir uptake
0
0
H O C
Control Variable
Table 1 (continued)
0
Moderate None Moderate Moderate Moderate
Mild
Severe
Mild
Mild
0 1 1 1 1 1
1
1
1
P G R S B
M
ATTR
D
B
AL
T
D
H
Eur J Nucl Med Mol Imaging
The dynamic 18F-florbetapir images were exported as transaxial and short axis dynamic frames and analyzed using the program Osirix (version 5.7), a custom 4-D viewer plug-in and region of interest (ROI) enhancement tool. The LV myocardial and blood pool standardized uptake values (SUV) were determined as the decay-corrected tissue concentration of radioactivity divided by the total injected radioactivity per body weight in grams. On the short axis DICOM reconstructed images the LV myocardial boundaries were traced over the entire LV myocardial volume at the base, mid ventricle and apex (except for the LV apex segment). A ROI in the center of the LV near the base was used to determine the blood pool SUVs; care was taken when drawing this ROI to minimize count spillover from the myocardium by drawing the ROI boundary with at least a pixel separation from the endocardial surface. Tissue SUVs were computed in the liver using a standard sized region of 6 cm2 in the right lobe of the liver, and a 1-cm2 region in the basal and apical regions of the left lung, and in the vertebral bodies within the field of view. Primary and secondary outcome variables The target to background ratio (TBR) and myocardial retention index (RI) were the primary outcome variables in this study. The TBR was calculated as the ratio of the mean LV myocardial SUV to the blood pool SUV between 10 and 30 min. The 18Fflorbetapir myocardial RI was computed as the mean LV myocardial tissue radiotracer concentration between 10 and 30 min after injection of 18F-Florbetapir, divided by the integral of the blood pool 18F-florbetapir time–activity curve from 0 to 20 min after injection (note that 20 min is the mid-point of the time interval from 10 to 30 min). The mean LV myocardial SUV value and the mean LV myocardium to liver SUV ratio were the secondary variables. Myocardial retention of 18F-florbetapir over time 0 – 30 min was computed as the mean LV SUV in each dynamic frame of the image divided by the integral of the blood pool 18F-florbetapir time–activity curve at that time point. Statistical analysis Continuous variables are shown as means ± standard deviation and compared using a t test when appropriate. Categorical
Eur J Nucl Med Mol Imaging
variables are shown as percentages. The primary and secondary outcome variables were not normally distributed and are reported as medians (IQR) and the differences (between the control and amyloid subjects) were compared using the Kruskal-Wallis test and the Mann-Whitney U post hoc test for pair-wise comparisons (control vs. ATTR, control vs. AL and AL vs. ATTR). A P value of right ventricular uptake of the
Table 1. At the time of the study, one AL subject was in clinical and hematological remission and had been in NYHA class I for 4 years, two were undergoing their chemotherapy and two were after chemotherapy awaiting stem cell transplantation. Three of the four ATTR subjects were on therapy for TTR stabilization (two on doxycycline and one on tafamidis). Echocardiograms were acquired for clinical reasons (within a median 2.5 days of the PET scan) in all the amyloid subjects and in the two control subjects with heart failure. In most of the amyloid subjects, the left and right ventricles were thick, the atria enlarged and diastolic function impaired with restrictive physiology. Mean LV mass was significantly higher in the ATTR subjects than in the AL subjects (473±282 vs. 181±25 g, P=0.05).
tracer in the ATTR subjects and in the AL subjects. The images in the control subject are scaled up to show the myocardial boundaries. AL light chain amyloidosis, ATTR transthyretin amyloidosis
Eur J Nucl Med Mol Imaging
Qualitative and quantitative imaging of relative uptake of 18F-florbetapir by the myocardium The cardiac image quality was excellent in most subjects with increased LV myocardial 18F-florbetapir uptake in all the amyloid subjects and in none of the control subjects (Table 1, Fig. 1). LV myocardial 18F-florbetapir uptake was diffuse and uniform in all but the one AL amyloid subject (subject AL-T in disease remission showed greater septal uptake). The LV myocardium was well delineated from background in four of the five AL subjects and in three of the four ATTR subjects. All the amyloid subjects and none of the control subjects showed right ventricular 18F-florbetapir uptake. All but one of the amyloid subjects and none of the control subjects demonstrated atrial radiotracer uptake. On polar plot analysis
Fig. 2 LV myocardial and blood pool SUVs in the control, AL and ATTR subjects and an AL subject in remission. The mean LV myocardial and blood pool SUVs are very similar after the first few minutes in the control subjects. In contrast, the AL and ATTR amyloid subjects show higher LV myocardial SUVs after the initial 5 min which remain higher than the blood pool SUV till the end of the acquisition. In the last panel,
of the amyloid subjects, the distribution of 18F-florbetapir in the anterior, septal, inferior, lateral and apical myocardial segments of the myocardium was uniform. Myocardial 18F-florbetapir imaging to differentiate amyloid from control subjects On quantitative analysis, the blood pool activity of 18Fflorbetapir peaked within the first 2 min in all amyloid subjects and control subjects (Fig. 2). The LV myocardial signal in the amyloid subjects was greater than the blood pool activity at about 4 min after injection of 18F-florbetapir and remained so until the end of the study. The median LV myocardial retention of 18F-florbetapir was significantly higher in the amyloid subjects than in the control
the LV myocardial SUVs in an AL subject in remission are seen to be significantly lower than the LV myocardial SUVs in subjects with active AL, and show a minimal difference from the blood pool SUVs, suggesting that uptake of 18F-florbetapir may be related to disease activity. LV left ventricular, Myo myocardial, SUV standardized uptake value, AL light chain amyloidosis, ATTR transthyretin amyloidosis
Eur J Nucl Med Mol Imaging
subjects (Fig. 3). The median RI was 0.043 min−1 (IQR 0.034 – 0.051 min−1) in the amyloid subjects compared to 0.023 min−1 (IQR 0.015 – 0.025 min−1, P=0.002) in the control subjects (Fig. 4). There was no correlation between the myocardial mass and RI (R=−0.17, P=0.64). The median TBR was also significantly higher in the amyloid subjects than in the control subjects (1.84, IQR 1.64 – 2.50, vs. 1.26, IQR 0.91 – 1.36, P=0.001). The median LV myocardial SUV (3.84, IQR 1.87 – 5.65, vs. 1.35, IQR 1.17 – 2.28, P=0.029), and the median ratio of the LV myocardial to liver SUV (0.67, IQR 0.44 – 1.64, vs. 0.18, IQR 0.15 – 0.35, P=0.004) were also significantly higher in the amyloid subjects than in the control subjects. Overall, a myocardial RI of >0.025 min−1 and a TBR of >1.45 identified all the amyloid subjects and none of the control subjects.
Myocardial 18F-florbetapir imaging to differentiate control subjects from AL and ATTR subjects The myocardial RI, TBR and myocardial to liver SUV ratio also distinguished the control subjects from both the ATTR and the AL amyloid subjects (Table 2; Figs. 4 and 5). The myocardial RI was significantly higher in the AL and the ATTR subjects than in the control subjects. Interestingly, despite a significantly lower myocardial mass in the AL than in the TTR amyloid subjects, the median RI tended to be higher in the AL group (P=0.057).
Fig. 4 Distribution of the LV myocardial retention index (RI) in control subjects and amyloid subjects. The RI is significantly higher in the AL amyloid subjects and there is a trend to a higher RI in the TTR amyloid subjects compared to the AL amyloid subjects. None of the control subjects showed a RI of >0.025 min−1. AL light chain amyloidosis, ATTR transthyretin amyloidosis
Table 2 Primary and secondary outcome variables in control subjects and amyloid subjects. Values are median (IQR)
Fig. 3 Myocardial retention of 18F-florbetapir in the control subjects and amyloid subjects. The mean myocardial retention of 18F-florbetapir was higher in the amyloid subjects than in the control subjects. The error bars represent ranges
Variable
Control (N=5)
Amyloid (N=9)
P value
LV RI
0.043 (0.034 – 0.051) 1.84 (1.64 – 2.50)
0.002
LV TBR
0.023 (0.015 – 0.024) 1.26 (0.91 – 1.36)
LV SUV
1.35 (1.17 – 2.28)
3.84 (1.87 – 5.65)
0.03
LV to liver SUV ratio 0.18 (0.15 – 0.35)
0.67 (0.44 – 1.64)
0.004
0.001
LV = left ventricular myocardium; IQR = interquartile range; RI = retention index; TBR = target to background ratio (LV myocardium to blood pool SUV ratio); SUV = specific uptake value. * Mann Whitney U test, P < 0.05 vs. control
Eur J Nucl Med Mol Imaging
Quantitative myocardial 18F-florbetapir imaging of the liver, bone and lungs
Fig. 5 Distribution of 18F-florbetapir myocardium to blood pool SUV ratio (target to background ratio) in control subjects and amyloid subjects. The ratio is significantly higher in the amyloid subjects than in the control subjects . None of the control subjects showed a TBR of >1.45. SUV standardized uptake value
Fig. 6 18F-Florbetapir time–activity curves in the blood pool, liver, bone and lungs in the control subjects and AL and ATTR subjects. Radiotracer activity in the lungs (blue) follows the blood pool activity (red) while the radiotracer accumulates in the liver (green) over the first 6 – 8 min and
Intense 18F-florbetapir uptake was noted in the liver and bones (vertebrae) of all amyloid and control subjects. Hepatic 18Fflorbetapir activity peaked at about 6 – 8 min after injection and remained elevated for the duration of the scan, consistent with known hepatobiliary excretion of this tracer (Fig. 6). It did not affect the cardiac image quality in amyloid subjects but it significantly limited myocardial visualization in control subjects. Bone 18F-florbetapir activity peaked at about 6 – 8 min after injection and remained elevated for the duration of the scan (Fig. 6). In the lungs, 18F-florbetapir cleared rapidly, almost in parallel to the clearance of the blood pool activity, in all control subjects and all but one of the amyloid subjects (Fig. 6). One subject with AL amyloidosis of the lungs (clinical and CT diagnosis) and pleural effusions demonstrated diffusely increased bilateral pulmonary radiotracer
then remains high for at least 1 h. After the first 6 – 8 min, radiotracer activity the lungs (blue) and bone (purple) are low and parallel the blood pool activity (red). AL light chain amyloidosis, ATTR transthyretin amyloidosis, SUV standardized uptake value
Eur J Nucl Med Mol Imaging
uptake, while another subject with ATTR congestive heart failure and pleural effusions did not demonstrate increased pulmonary radiotracer uptake (Fig. 7).
Discussion We tested a promising new method, 18F-florbetapir PET, for the first time in patients with cardiac amyloidosis, and found that it appears to hold potential for imaging cardiac amyloidosis. All the amyloid subjects and none of the control subjects demonstrated diffuse and uniform left and right ventricular 18F-florbetapir uptake. 18Fflorbetapir myocardial RI, TBR and myocardial to liver
A. AL Lung Amyloid
Fig. 7 18F-Florbetapir imaging in cardiac amyloidosis and pleural effusion. 18F-Florbetapir uptake (left columns), CT scans (middle column), and fused PET and CT images (right column) in two amyloid subjects with pleural effusions are shown. The CT scans obtained for attenuation correction demonstrate right-sided pleural effusion in both subjects (middle column). While the AL amyloid subject with pleural effusion (a) shows intense radiotracer uptake in the lungs, limiting visualization of myocardial
SUV ratio were significantly higher in cardiac amyloid subjects than in control subjects. Overall myocardial RI tended to be higher in AL subjects than in ATTR subjects but none of the indices tested (RI, LV myocardial SUV values, TBR or LV myocardium to liver SUV ratio) clearly distinguished AL from ATTR amyloidosis. Importantly, in this pilot study, 18F-florbetapir was useful for imaging both AL and ATTR cardiac amyloidosis, and a RI of >0.025 min−1 and a TBR of >1.45 identified all AL and TTR amyloid subjects and none of the control subjects. 18 F-Florbetapir was originally developed to image beta amyloid deposits in the brain and its radiotracer distribution was studied, but imaging was not focused on the heart [8, 9].
B. TTR-Heart Failure
uptake, the ATTR subject with pleural effusions from heart failure (b) shows no increase in lung radiotracer uptake. Also, the time SUV curves (bottom) demonstrate higher lung than blood pool SUV values after the first 2 min in the subject with pleural amyloidosis (a) but not in the subject with congestive heart failure (b). These findings suggest that florbetapir imaging may be helpful in identifying lung amyloidosis. AL light chain amyloidosis, ATTR transthyretin amyloidosis, SUV standardized uptake value
Eur J Nucl Med Mol Imaging
Our study adds to the prior literature by detailing the distribution of 18F-florbetapir in the heart and lungs of control subjects and amyloid subjects. In control subjects, 18 F-florbetapir activity in the blood pool and myocardium peaked early and cleared quickly to very low levels. In amyloid subjects, the blood pool activity peaked early and cleared quickly, whereas the myocardial 18F-florbetapir uptake was rapid followed by myocardial retention for about 60 min, allowing an excellent TBR. The rapid blood pool clearance combined with the quick metabolism to metabolites that do not bind to beta amyloid [11] may explain the lack of late myocardial accumulation of 18Fflorbetapir allowing injection followed by a 30-min scan. Thus the radiotracer characteristics of 18F-florbetapir appear to be optimal for imaging cardiac amyloidosis. Although the specific reasons for myocardial 18Fflorbetapir uptake in cardiac amyloidosis are not known, binding of 18F-florbetapir or its metabolites to amyloid deposits or blood radioactivity are possible mechanisms. 18 F-Florbetapir selectively binds to beta-amyloid aggregates and beta-amyloid plaques in the brain with the binding intensity quantitatively correlated with the density of beta amyloid in post-mortem studies of Alzheimer’s subjects [10]. High myocardial radiotracer uptake and late retention in amyloid subjects compared to control subjects suggests a correlation with amyloid protein. Higher median 18F-florbetapir RI and myocardial SUV values in AL amyloid subjects suggest greater avidity for AL than for ATTR protein. Given the lower myocardial mass in AL subjects than in ATTR subjects, we speculate that higher radiotracer uptake may reflect not only amyloid mass, but also amyloid disease activity. Comparison with 11C-PiB PET Recently, Antoni et al. [12] demonstrated that 11C-PiB, also a beta amyloid imaging agent for Alzheimer’s disease, is helpful in imaging cardiac amyloidosis. In that study, 11C-PiB uptake was noted in all ten amyloid subjects and in none of five control subjects [12]. Similar results were observed in our study, with the RI for 11C-PiB in amyloidosis subjects being significantly higher than in control subjects (0.054 min−1 vs. 0.025 min−1, P=0.007);,but the authors of the previous study did not specifically compare the RI between TTR subjects and AL subjects. The RI values in our study cannot be directly compared to those obtained by Antoni et al. due to differences in the imaging techniques, patient characteristics and radiotracer used. While 11C-PiB offers the advantage of quantifying myocardial amyloid burden, its use is limited to PET centers with a cyclotron (20-min half-life) and is not FDA-approved. On the other hand, 18F-florbetapir can be quantified, can be used at PET centers without a cyclotron (110-min half-life), and is
clinically available as an FDA-approved brain imaging agent. As our initial results are promising, large multicenter studies in subjects with suspected amyloidosis may be undertaken to advance the management of subjects with this underdiagnosed fatal disease. Comparison with 99mTc-pyrophosphate/99mTc-DPD SPECT 99m
Tc-Labeled pyrophosphate (PYP) [13] and 99mTc-3,3diphosphono-1,2-propanodicarboxylic acid (DPD) [14] SPECT imaging is useful in the diagnosis of ATTR disease. Perugini et al. [14] studied 35 subjects (15 ATTR, 10 AL, 10 controls) and found that heart retention (median 7.3 % vs. 3.8 % vs. 2.9 %) and heart/whole-body retention (10 % vs. 5.4 % vs. 5.4 %) of 99mTcDPD were significantly higher in ATTR subjects (P1.5 had a sensitivity of 97 % and a specificity of 100 % in identifying those with ATTR disease. 99m Tc-PYP and DPD SPECT is widely available and excellent for imaging TTR cardiac amyloidosis, likely due to higher concentrations of calcium-containing products in TTR amyloid [13]. However, a negative test does not exclude the possibility of AL cardiac amyloidosis and absolute quantitation is limited, making assessment of response to therapy challenging. In contrast, 18F-florbetapir imaging appears to image both AL and ATTR amyloid and is quantitative. Indeed, 18F-florbetapir could potentially serve as an excellent amyloid-specific screening imaging test, which if positive can be followed by a myocardial biopsy (for disease typing) or by 99m Tc-PYP/DPD imaging (to confirm ATTR when suspicion of TTR amyloid is high). Clinical implications A definitive diagnosis of cardiac amyloidosis is primarily based on right ventricular endomyocardial biopsy with immunohistochemistry or mass spectroscopy as needed. Other findings such as a typical morphological appearance on echocardiography, cardiac MRI [15] or increased extracellular volume fraction on cardiac MRI [16], with an extracardiac histological diagnosis of amyloidosis, may be helpful in diagnosing cardiac amyloidosis in AL disease, yet not always definitive. Also, current imaging techniques tend to diagnose amyloidosis at an advanced stage and after anatomic changes in cardiac structure have occurred, the reversal of which in response to appropriate therapy is slow and not easily measurable. Quantitative imaging with 18F-florbetapir may potentially identify early cardiac amyloidosis, quantify amyloid disease
Eur J Nucl Med Mol Imaging
burden and help assess changes in response to therapy. 18FFlorbetapir may add to 123I SAP scans [7] by quantifying amyloid burden in the heart, as 123I SAP provides an estimate of whole-body amyloid burden. Radionuclide techniques including 18F-florbetapir may be particularly well suited for imaging subjects with renal disease who cannot undergo contrastenhanced cardiac MRI. Due to rapid clearance from normal lung and bone tissue, 18F-florbetapir may offer the ability to identify amyloidosis of the lungs and bones. It may not, however, be helpful for identifying amyloidosis in the liver, as 18Fflorbetapir is eliminated by the hepatobiliary system. Strengths and limitations To the best of our knowledge this is the first study to investigate the use of 18F-florbetapir for imaging cardiac amyloidosis. However, because of the small number of subjects, we did not calculate sensitivity, specificity and diagnostic accuracy. In addition to healthy volunteers, two control subjects with heart failure were included for comparison, as the test will most likely be used in subjects to evaluate symptoms of heart failure. However, future studies including more subjects and subjects with increased LV wall thickness, restrictive physiology and nonamyloid infiltrative heart diseases are warranted to definitively establish the value of 18F-florbetapir in the diagnosis of cardiac amyloidosis. Also, detailed analyses to differentiate AL from ATTR amyloidosis were limited in this pilot study Conclusion The results of this pilot study suggest that 18F-florbetapir may be a novel PET radiotracer for imaging both AL and ATTR cardiac amyloidosis. 18F-Florbetapir appears to be specific in distinguishing subjects with cardiac amyloidosis from control subjects. Further study is needed to better understand its value in following response to therapy and in imaging amyloidosis in other organ systems. Acknowledgments We are indebted to the subjects who participated in this study. We are appreciative of the research support for this study from the Amyloid Foundation. Dr. Dorbala was supported by NIH NHLBI grant K23HL092299. We are grateful to our colleagues at the Brigham and Women’s Hospital Cardiac Amyloidosis Program. Financial Disclosures Dorbala: Research Grant Astellas Global Pharma Development. Di Carli: Research grant from Gilead Sciences.
References 1. Merlini G, Bellotti V. Molecular mechanisms of amyloidosis. N Engl J Med. 2003;349:583–96. 2. Falk RH. Cardiac amyloidosis: a treatable disease, often overlooked. Circulation. 2011;124:1079–85. 3. Ruberg FL, Berk JL. Transthyretin (TTR) cardiac amyloidosis. Circulation. 2012;126:1286–300. 4. Skinner M, Sanchorawala V, Seldin DC, Dember LM, Falk RH, Berk JL, et al. High-dose melphalan and autologous stem-cell transplantation in patients with AL amyloidosis: an 8-year study. Ann Intern Med. 2004;140:85–93. 5. Kristen AV, Perz JB, Schonland SO, Hegenbart U, Schnabel PA, Kristen JH, et al. Non-invasive predictors of survival in cardiac amyloidosis. Eur J Heart Fail. 2007;9:617–24. 6. Miroy GJ, Lai Z, Lashuel HA, Peterson SA, Strang C, Kelly JW. Inhibiting transthyretin amyloid fibril formation via protein stabilization. Proc Natl Acad Sci U S A. 1996;93: 15051–6. 7. Hawkins PN, Lavender JP, Pepys MB. Evaluation of systemic amyloidosis by scintigraphy with 123I-labeled serum amyloid P component. N Engl J Med. 1990;323:508–13. 8. Yang L, Rieves D, Ganley C. Brain amyloid imaging – FDA approval of florbetapir F18 injection. N Engl J Med. 2012;367:885–7. 9. Clark CM, Schneider JA, Bedell BJ, Beach TG, Bilker WB, Mintun MA, et al. Use of florbetapir-PET for imaging beta-amyloid pathology. JAMA. 2011;305:275–83. 10. Lister-James J, Pontecorvo MJ, Clark C, Joshi AD, Mintun MA, Zhang W, et al. Florbetapir F-18: a histopathologically validated betaamyloid positron emission tomography imaging agent. Semin Nucl Med. 2011;41:300–4. 11. Choi SR, Golding G, Zhuang Z, Zhang W, Lim N, Hefti F, et al. Preclinical properties of 18 F-AV-45: a PET agent for Abeta plaques in the brain. J Nucl Med. 2009;50:1887–94. 12. Antoni G, Lubberink M, Estrada S, Axelsson J, Carlson K, Lindsjo L, et al. In vivo visualization of amyloid deposits in the heart with 11C-PIB and PET. J Nucl Med. 2013;54:213– 20. 13. Bokhari S, Castano A, Pozniakoff T, Deslisle S, Latif F, Maurer MS. (99m)Tc-pyrophosphate scintigraphy for differentiating light-chain cardiac amyloidosis from the transthyretinrelated familial and senile cardiac amyloidoses. Circ Cardiovasc Imaging. 2013;6:195–201. 14. Perugini E, Guidalotti PL, Salvi F, Cooke RM, Pettinato C, Riva L, et al. Noninvasive etiologic diagnosis of cardiac amyloidosis using 99mTc-3,3-diphosphono-1,2-propanodicarboxylic acid scintigraphy. J Am Coll Cardiol. 2005;46:1076–84. 15. Syed IS, Glockner JF, Feng D, Araoz PA, Martinez MW, Edwards WD, et al. Role of cardiac magnetic resonance imaging in the detection of cardiac amyloidosis. JACC Cardiovasc Imaging. 2010;3:155–64. 16. Banypersad SM, Sado DM, Flett AS, Gibbs SD, Pinney JH, Maestrini V, et al. Quantification of myocardial extracellular volume fraction in systemic Al amyloidosis: an equilibrium contrast cardiovascular magnetic resonance study. Circ Cardiovasc Imaging. 2013;6:34–9.
ORIGINAL ARTICLE
Imaging cardiac amyloidosis: a pilot study using 18F-florbetapir positron emission tomography Sharmila Dorbala & Divya Vangala & James Semer & Christopher Strader & John R. Bruyere Jr & Marcelo F. Di Carli & Stephen C. Moore & Rodney H Falk
Received: 3 February 2014 / Accepted: 15 April 2014 # Springer-Verlag Berlin Heidelberg 2014
Abstract Purpose Cardiac amyloidosis, a restrictive heart disease with high mortality and morbidity, is underdiagnosed due to limited targeted diagnostic imaging. The primary aim of this study was to evaluate the utility of 18F-florbetapir for imaging cardiac amyloidosis. Methods We performed a pilot study of cardiac 18F-florbetapir PET in 14 subjects: 5 control subjects without amyloidosis and 9 subjects with documented cardiac amyloidosis. Standardized uptake values (SUV) of 18F-florbetapir in the left ventricular (LV) myocardium, blood pool, liver, and vertebral bone were determined. A 18F-florbetapir retention index (RI) was computed. Mean LV myocardial SUVs, target-to-background ratio (TBR, myocardial/blood pool SUV ratio) and myocardial-toliver SUV ratio between 0 and 30 min were calculated. Results Left and right ventricular myocardial uptake of 18Fflorbetapir were noted in all the amyloid subjects and in none S. Dorbala (*) : M. F. Di Carli Noninvasive Cardiovascular Imaging Program, Heart and Vascular Center, Departments of Radiology and Medicine (Cardiology), Brigham and Women’s Hospital, Harvard Medical School, Boston, MA, USA e-mail: [email protected] S. Dorbala : D. Vangala : J. Semer : C. Strader : J. R. Bruyere Jr : M. F. Di Carli : S. C. Moore Division of Nuclear Medicine and Molecular Imaging, Department of Radiology, Brigham and Women’s Hospital, Harvard Medical School, Boston, MA, USA S. Dorbala : R. H. Falk Cardiovascular Division and the Cardiac Amyloidosis Program, Department of Medicine, Brigham and Women’s Hospital, Harvard Medical School, Boston, MA, USA S. Dorbala : D. Vangala : J. Semer : C. Strader : J. R. Bruyere Jr : M. F. Di Carli : S. C. Moore : R. H. Falk Brigham and Women’s Hospital, 70 Francis Street, Shapiro 5th Floor, Room 128, Boston, MA 02115, USA
of the control subjects. The RI, TBR, LV myocardial SUV and LV myocardial to liver SUV ratio were all significantly higher in the amyloidosis subjects than in the control subjects (RI median 0.043 min−1, IQR 0.034 – 0.051 min−1, vs. 0.023 min−1, IQR 0.015 – 0.025 min−1, P=0.002; TBR 1.84, 1.64 – 2.50, vs. 1.26, IQR 0.91 – 1.36, P=0.001; LV myocardial SUV 3.84, IQR 1.87 – 5.65, vs. 1.35, IQR 1.17 – 2.28, P=0.029; ratio of LV myocardial to liver SUV 0.67, IQR 0.44 – 1.64, vs. 0.18, IQR 0.15 – 0.35, P=0.004). The myocardial RI, TBR and myocardial to liver SUV ratio also distinguished the control subjects from subjects with transthyretin and those with light chain amyloid. Conclusion 18F-Florbetapir PET may be a promising technique to image light chain and transthyretin cardiac amyloidosis. Its role in diagnosing amyloid in other organ systems and in assessing response to therapy needs to be further studied. Keywords Amyloidosis . Heart . 18F-Florbetapir . Positron emission tomography . Diagnosis
Introduction Amyloidosis comprises of a large group of disorders characterized by abnormal folding of normal, abnormal or mutant, insoluble extracellular proteins, in the form of a beta pleated structures [1]. Amyloid deposits in the brain cause Alzheimer’s disease, whereas systemic amyloidosis almost always spares the brain. Cardiac amyloidosis is usually a component of systemic amyloidosis due to systemic light chain (AL) and transthyretin (ATTR, wild type or mutant) amyloidosis, although clinical manifestations may be limited to the heart [1]. Subjects with AL and ATTR cardiac amyloidosis have overlapping imaging features on echocardiography and cardiac MRI, but their clinical course and prognosis are distinct [2]. Cardiac AL amyloidosis is characterized by a rapidly progressive clinical course and high mortality, while
Eur J Nucl Med Mol Imaging
ATTR wild-type disease (senile amyloidosis) typically has a slower clinical course with a much longer overall survival. Nevertheless, cardiac involvement is almost invariably the cause of death in both sets of patients [3]. One of the most effective treatment strategies for amyloidosis is reducing the production of the precursor amyloidogenic protein. Halting the synthesis of the amyloid protein may lead to resorption of amyloid deposits and resolution of organ dysfunction [1]. Due to differences in the genesis of AL and ATTR diseases, treatment options are distinct. In the case of systemic AL amyloidosis, novel chemotherapy directed against reducing the production of light chains, as well as stem cell transplantation have considerably improved survival [4]; however, patients with cardiac AL amyloidosis, the subset with the worse prognosis, generally do not tolerate aggressive chemotherapy and stem cell transplantation, limiting treatment options in these patients [5]. Regarding systemic TTR amyloidosis, novel therapies such as TTR stabilizers [6] and small RNA silencing molecules that stabilize the TTR amyloid protein and prevent them from aggregating are undergoing clinical trials [3]. Thus, early detection of cardiac amyloidosis, identification of the type of amyloidosis and institution of specific therapy is critical to improve outcomes. Echocardiography and cardiac MRI are widely used in clinical practice. They rely primarily on imaging increased left ventricular (LV) wall thickness from amyloid deposits, but do not specifically distinguish amyloid from other hypertrophic heart diseases [2]. Amyloid-directed imaging with 123I systemic amyloid protein (SAP) provides a measure of overall systemic amyloid burden, but is not widely available and is not helpful for cardiac imaging due to blood pool uptake resulting in a poor signal to noise ratio in the heart [3, 7]. One recent study has demonstrated the utility of 11C-labeled Pittsburgh Compound B (PiB) in imaging cardiac amyloidosis, but it is only available at PET sites with a cyclotron. Thus, we currently have no widely available targeted imaging techniques to specifically diagnose amyloidosis in the heart. A novel amyloid imaging agent, 18F-florbetapir, structurally distinct from PiB, was recently approved for imaging beta amyloid protein in the brain, with a very high sensitivity for binding to minute amounts of beta amyloid [8, 9]. We therefore studied the feasibility of using 18F-florbetapir in imaging cardiac amyloidosis. The primary aim of this pilot study was to evaluate 18 F-florbetapir in diagnosing cardiac amyloidosis compared to its use in control subjects without amyloidosis. A secondary aim of this study was to determine if 18Fflorbetapir uptake in the heart can help distinguish between AL and ATTR cardiac amyloidosis. We also explored the kinetics of 18F-florbetapir in the liver, lung and bone to understand if it could potentially be used to diagnose amyloidosis in these organ systems.
Materials and methods This was a prospective case control pilot study of 14 subjects, 9 subjects with definite cardiac amyloidosis and 5 control subjects without amyloidosis. Amyloid subjects were recruited from the Brigham and Women’s Hospital Cardiac Amyloidosis Program. Three control subjects were recruited from web advertisements and two control subjects with known nonischemic heart failure without amyloidosis were recruited from our general cardiology service. Cardiac amyloidosis was diagnosed based on endomyocardial biopsy (in seven patients) or a typical amyloid cardiac morphology (wall thickness measurements of >11 mm, bright echogenic myocardium) and evidence of diastolic dysfunction (mitral inflow parameters, pulmonary venous Doppler information, or tissue Doppler imaging at the mitral septal and lateral annulus) with histological confirmation of amyloidosis from extracardiac structures (in two patients). The type of amyloidosis (AL or ATTR) was classified in all subjects based on immunohistochemistry or, in equivocal cases, by mass spectrometry. Subjects with claustrophobia or unable to lie flat for 60 min were excluded. This study was approved by the Partners Human Research Committee and listed on clinical trials.gov (NCT01683825). Each study subject provided written informed consent. Positron emission tomography The scan protocol included a cardiac 3-D mode PET/CT scan (Discovery ST; GE, Waukesha, WI) acquired in a list mode with arms by the side for 60 min. A 60-min scan duration was chosen because the radiotracer distribution in the heart, lung and liver were not previously reported and a long scan allowed us to identify the appropriate time period for future clinical imaging. The scan was terminated prematurely at 20 min into the acquisition in one subject due to severe nausea (AL amyloidosis and on chemotherapy). No specific dietary preparation (no restriction on oral intake) or withholding of medications was required for this protocol. Following a lowdose (10 mA, 120 kVp, free tidal breathing) scout scan to localize the heart, a low-dose CT transmission scan (10 mA, 120 kVp, free tidal breathing) was obtained over the heart for measurement and correction of attenuation. Following the scout and the transmission scans, the emission scan acquisition was started simultaneously with intravenous injection of 6 mCi of 18F-florbetapir and was continued for 60 min. The estimated whole-body radiation dose to the subjects from the 18 F-florbetapir scan was about 5 mSv [10]. The list-mode images were reconstructed into dynamic images using 38 frames (13 frames of 5 s each, 6 frames of 10 s each, 4 frames of 30 s each, 6 frames of 60 s each, 8 frames of 300 s each, and 1 frame of 600 s). The images were reconstructed using a matrix size of 128 × 128, ordered-
B
M 57
41
51
74
43.88
3
2.4
3,003
39
0
M
430
32
–
2
26.5
50
179.6
Pulmonary artery systolic pressure (mm Hg)
5.5
1.45
E/A
3.5
55
A (cm)
0.23
80
E (cm)
E/e’
166.4
426
Deceleration time (ms) e’ (cm)
55
21
LV mass(gm)
–
–
35
0.21
–
–
N/A
16.05
4.25
237
68
620
40
4.3
4.9
2.4
2.4
117
91
2
8.5
1.1
0.4 3.2
0.9
10.2
9.2
37.55
97.31
LV end diastolic diameter (cm) LV end systolic diameter (cm) Ejection fraction (%)
Left atrial volume index (mL/m2) Interventricular septum (cm) Posterior wall (cm)
2-D echocardiographic features
N/A
0.12
6
214.4
1.37
51
70
147
63
2.7
3.7
1.2
1.2
29
14.43
4
316
1.23
47
58
651
45
3.4
4.4
1.5
1.4
40
32
7.5
–
–
160
628
55
3.8
5.4
1.9
30
9.09
6.05
388.1
0.6
91
55
220
70
2.3
4.4
1.3
1.3
16.75
0.127 10.68
M 23.0
1.34
157
≥60
1
1
29.8
M
Kappa/lambda ratio ratio 45.18
9,413
52
0
– 28
1
3
24.5
M
Lambda
1
26.0
M
17
1
27.5
M
0.02
1
28.0
M
Kappa
43
M 34.2
Troponin T
Pro BNP
43
2,641
eGFR
BNP
Laboratory values
0
3
33.8
F
N/A
11.75
8
118
1.74
54
94
192
55
2.4
3.4
1.4
1.7
23.85
1.08
18.1
19.5
0.13
1,063
45
–
1
22.1
F
46.4
23.33
3
126
2.03
34
70
159
50
2.8
3.0
1.3
1.7
24.09
0.49
66.4
32.5
0.17
3,755
≥60
0
3
21.0
F
30
24.56
5
199.6
1.79
68
123
168
65
2.3
3.1
1.5
1.5
26.86
0.15
17.4
2.59
0.16
10,188
54
0
2
22.1
F
31
26.3
3
126
2.31
34
79
164
70
1.2
2.5
1.8
1.6
19.21
0.18
1.14
20.4
0.03
9,288
33
0
3
33.8
F
0
65
3
69
H
CADa
68
P
NYHA class
78
D
25.5
77
T
BMI (kg/m2)
52
G
African American Caucasian Caucasian Caucasian African American Caucasian Caucasian Caucasian Caucasian Caucasian Caucasian Caucasian Caucasian Caucasian
47
B
M
64
D
Sex
74
R
AL
Race
40
S
H
C
O
ATTR
Control
Age (years)
Variable
Table 1 Echocardiographic features of the study subjects
Eur J Nucl Med Mol Imaging
0.052 –b
0.046
subsets expectation maximization (two iterations, 21 subsets) and postfilter full-width at half-maximum 4.48 mm. Static images were then reconstructed with a slice separation of 3.27 mm from data acquired between 10 min (to avoid blood pool activity and improve delineation of the myocardium from blood pool tracer activity) and 60 min after the start of the scan. These images were reoriented into standard cardiac planes (short axis, vertical long axis and horizontal long axis images) for interpretation. Quantitative image analysis
RI could not be computed as imaging was terminated at 20 min due to severe nausea
Coronary artery disease on invasive or CT angiography a
RI
b
0.045 0.024 0.025 0.023
0.020
0.039
0.032
0.031
0.039
0.073
0.042
1+
4+
1+
0
2+
0
2+
0 0
1+ 1+
0 0
2+ 1+
0
2+
0
0
0
0
0
0
0
0
0
0
0 Lung
Atrial
2+
2+
4+
2+
4+
2+
1+
2+ 2+
4+ 3+
2+ 3+
1+ 2+
2+
3+
3+
0
0 0
0 0
0
0
0 Right ventricle
Moderate Moderate
Left ventricle
0
Right ventricular hypertrophy Right atrial enlargement 18 F-Florbetapir uptake
0
0
H O C
Control Variable
Table 1 (continued)
0
Moderate None Moderate Moderate Moderate
Mild
Severe
Mild
Mild
0 1 1 1 1 1
1
1
1
P G R S B
M
ATTR
D
B
AL
T
D
H
Eur J Nucl Med Mol Imaging
The dynamic 18F-florbetapir images were exported as transaxial and short axis dynamic frames and analyzed using the program Osirix (version 5.7), a custom 4-D viewer plug-in and region of interest (ROI) enhancement tool. The LV myocardial and blood pool standardized uptake values (SUV) were determined as the decay-corrected tissue concentration of radioactivity divided by the total injected radioactivity per body weight in grams. On the short axis DICOM reconstructed images the LV myocardial boundaries were traced over the entire LV myocardial volume at the base, mid ventricle and apex (except for the LV apex segment). A ROI in the center of the LV near the base was used to determine the blood pool SUVs; care was taken when drawing this ROI to minimize count spillover from the myocardium by drawing the ROI boundary with at least a pixel separation from the endocardial surface. Tissue SUVs were computed in the liver using a standard sized region of 6 cm2 in the right lobe of the liver, and a 1-cm2 region in the basal and apical regions of the left lung, and in the vertebral bodies within the field of view. Primary and secondary outcome variables The target to background ratio (TBR) and myocardial retention index (RI) were the primary outcome variables in this study. The TBR was calculated as the ratio of the mean LV myocardial SUV to the blood pool SUV between 10 and 30 min. The 18Fflorbetapir myocardial RI was computed as the mean LV myocardial tissue radiotracer concentration between 10 and 30 min after injection of 18F-Florbetapir, divided by the integral of the blood pool 18F-florbetapir time–activity curve from 0 to 20 min after injection (note that 20 min is the mid-point of the time interval from 10 to 30 min). The mean LV myocardial SUV value and the mean LV myocardium to liver SUV ratio were the secondary variables. Myocardial retention of 18F-florbetapir over time 0 – 30 min was computed as the mean LV SUV in each dynamic frame of the image divided by the integral of the blood pool 18F-florbetapir time–activity curve at that time point. Statistical analysis Continuous variables are shown as means ± standard deviation and compared using a t test when appropriate. Categorical
Eur J Nucl Med Mol Imaging
variables are shown as percentages. The primary and secondary outcome variables were not normally distributed and are reported as medians (IQR) and the differences (between the control and amyloid subjects) were compared using the Kruskal-Wallis test and the Mann-Whitney U post hoc test for pair-wise comparisons (control vs. ATTR, control vs. AL and AL vs. ATTR). A P value of right ventricular uptake of the
Table 1. At the time of the study, one AL subject was in clinical and hematological remission and had been in NYHA class I for 4 years, two were undergoing their chemotherapy and two were after chemotherapy awaiting stem cell transplantation. Three of the four ATTR subjects were on therapy for TTR stabilization (two on doxycycline and one on tafamidis). Echocardiograms were acquired for clinical reasons (within a median 2.5 days of the PET scan) in all the amyloid subjects and in the two control subjects with heart failure. In most of the amyloid subjects, the left and right ventricles were thick, the atria enlarged and diastolic function impaired with restrictive physiology. Mean LV mass was significantly higher in the ATTR subjects than in the AL subjects (473±282 vs. 181±25 g, P=0.05).
tracer in the ATTR subjects and in the AL subjects. The images in the control subject are scaled up to show the myocardial boundaries. AL light chain amyloidosis, ATTR transthyretin amyloidosis
Eur J Nucl Med Mol Imaging
Qualitative and quantitative imaging of relative uptake of 18F-florbetapir by the myocardium The cardiac image quality was excellent in most subjects with increased LV myocardial 18F-florbetapir uptake in all the amyloid subjects and in none of the control subjects (Table 1, Fig. 1). LV myocardial 18F-florbetapir uptake was diffuse and uniform in all but the one AL amyloid subject (subject AL-T in disease remission showed greater septal uptake). The LV myocardium was well delineated from background in four of the five AL subjects and in three of the four ATTR subjects. All the amyloid subjects and none of the control subjects showed right ventricular 18F-florbetapir uptake. All but one of the amyloid subjects and none of the control subjects demonstrated atrial radiotracer uptake. On polar plot analysis
Fig. 2 LV myocardial and blood pool SUVs in the control, AL and ATTR subjects and an AL subject in remission. The mean LV myocardial and blood pool SUVs are very similar after the first few minutes in the control subjects. In contrast, the AL and ATTR amyloid subjects show higher LV myocardial SUVs after the initial 5 min which remain higher than the blood pool SUV till the end of the acquisition. In the last panel,
of the amyloid subjects, the distribution of 18F-florbetapir in the anterior, septal, inferior, lateral and apical myocardial segments of the myocardium was uniform. Myocardial 18F-florbetapir imaging to differentiate amyloid from control subjects On quantitative analysis, the blood pool activity of 18Fflorbetapir peaked within the first 2 min in all amyloid subjects and control subjects (Fig. 2). The LV myocardial signal in the amyloid subjects was greater than the blood pool activity at about 4 min after injection of 18F-florbetapir and remained so until the end of the study. The median LV myocardial retention of 18F-florbetapir was significantly higher in the amyloid subjects than in the control
the LV myocardial SUVs in an AL subject in remission are seen to be significantly lower than the LV myocardial SUVs in subjects with active AL, and show a minimal difference from the blood pool SUVs, suggesting that uptake of 18F-florbetapir may be related to disease activity. LV left ventricular, Myo myocardial, SUV standardized uptake value, AL light chain amyloidosis, ATTR transthyretin amyloidosis
Eur J Nucl Med Mol Imaging
subjects (Fig. 3). The median RI was 0.043 min−1 (IQR 0.034 – 0.051 min−1) in the amyloid subjects compared to 0.023 min−1 (IQR 0.015 – 0.025 min−1, P=0.002) in the control subjects (Fig. 4). There was no correlation between the myocardial mass and RI (R=−0.17, P=0.64). The median TBR was also significantly higher in the amyloid subjects than in the control subjects (1.84, IQR 1.64 – 2.50, vs. 1.26, IQR 0.91 – 1.36, P=0.001). The median LV myocardial SUV (3.84, IQR 1.87 – 5.65, vs. 1.35, IQR 1.17 – 2.28, P=0.029), and the median ratio of the LV myocardial to liver SUV (0.67, IQR 0.44 – 1.64, vs. 0.18, IQR 0.15 – 0.35, P=0.004) were also significantly higher in the amyloid subjects than in the control subjects. Overall, a myocardial RI of >0.025 min−1 and a TBR of >1.45 identified all the amyloid subjects and none of the control subjects.
Myocardial 18F-florbetapir imaging to differentiate control subjects from AL and ATTR subjects The myocardial RI, TBR and myocardial to liver SUV ratio also distinguished the control subjects from both the ATTR and the AL amyloid subjects (Table 2; Figs. 4 and 5). The myocardial RI was significantly higher in the AL and the ATTR subjects than in the control subjects. Interestingly, despite a significantly lower myocardial mass in the AL than in the TTR amyloid subjects, the median RI tended to be higher in the AL group (P=0.057).
Fig. 4 Distribution of the LV myocardial retention index (RI) in control subjects and amyloid subjects. The RI is significantly higher in the AL amyloid subjects and there is a trend to a higher RI in the TTR amyloid subjects compared to the AL amyloid subjects. None of the control subjects showed a RI of >0.025 min−1. AL light chain amyloidosis, ATTR transthyretin amyloidosis
Table 2 Primary and secondary outcome variables in control subjects and amyloid subjects. Values are median (IQR)
Fig. 3 Myocardial retention of 18F-florbetapir in the control subjects and amyloid subjects. The mean myocardial retention of 18F-florbetapir was higher in the amyloid subjects than in the control subjects. The error bars represent ranges
Variable
Control (N=5)
Amyloid (N=9)
P value
LV RI
0.043 (0.034 – 0.051) 1.84 (1.64 – 2.50)
0.002
LV TBR
0.023 (0.015 – 0.024) 1.26 (0.91 – 1.36)
LV SUV
1.35 (1.17 – 2.28)
3.84 (1.87 – 5.65)
0.03
LV to liver SUV ratio 0.18 (0.15 – 0.35)
0.67 (0.44 – 1.64)
0.004
0.001
LV = left ventricular myocardium; IQR = interquartile range; RI = retention index; TBR = target to background ratio (LV myocardium to blood pool SUV ratio); SUV = specific uptake value. * Mann Whitney U test, P < 0.05 vs. control
Eur J Nucl Med Mol Imaging
Quantitative myocardial 18F-florbetapir imaging of the liver, bone and lungs
Fig. 5 Distribution of 18F-florbetapir myocardium to blood pool SUV ratio (target to background ratio) in control subjects and amyloid subjects. The ratio is significantly higher in the amyloid subjects than in the control subjects . None of the control subjects showed a TBR of >1.45. SUV standardized uptake value
Fig. 6 18F-Florbetapir time–activity curves in the blood pool, liver, bone and lungs in the control subjects and AL and ATTR subjects. Radiotracer activity in the lungs (blue) follows the blood pool activity (red) while the radiotracer accumulates in the liver (green) over the first 6 – 8 min and
Intense 18F-florbetapir uptake was noted in the liver and bones (vertebrae) of all amyloid and control subjects. Hepatic 18Fflorbetapir activity peaked at about 6 – 8 min after injection and remained elevated for the duration of the scan, consistent with known hepatobiliary excretion of this tracer (Fig. 6). It did not affect the cardiac image quality in amyloid subjects but it significantly limited myocardial visualization in control subjects. Bone 18F-florbetapir activity peaked at about 6 – 8 min after injection and remained elevated for the duration of the scan (Fig. 6). In the lungs, 18F-florbetapir cleared rapidly, almost in parallel to the clearance of the blood pool activity, in all control subjects and all but one of the amyloid subjects (Fig. 6). One subject with AL amyloidosis of the lungs (clinical and CT diagnosis) and pleural effusions demonstrated diffusely increased bilateral pulmonary radiotracer
then remains high for at least 1 h. After the first 6 – 8 min, radiotracer activity the lungs (blue) and bone (purple) are low and parallel the blood pool activity (red). AL light chain amyloidosis, ATTR transthyretin amyloidosis, SUV standardized uptake value
Eur J Nucl Med Mol Imaging
uptake, while another subject with ATTR congestive heart failure and pleural effusions did not demonstrate increased pulmonary radiotracer uptake (Fig. 7).
Discussion We tested a promising new method, 18F-florbetapir PET, for the first time in patients with cardiac amyloidosis, and found that it appears to hold potential for imaging cardiac amyloidosis. All the amyloid subjects and none of the control subjects demonstrated diffuse and uniform left and right ventricular 18F-florbetapir uptake. 18Fflorbetapir myocardial RI, TBR and myocardial to liver
A. AL Lung Amyloid
Fig. 7 18F-Florbetapir imaging in cardiac amyloidosis and pleural effusion. 18F-Florbetapir uptake (left columns), CT scans (middle column), and fused PET and CT images (right column) in two amyloid subjects with pleural effusions are shown. The CT scans obtained for attenuation correction demonstrate right-sided pleural effusion in both subjects (middle column). While the AL amyloid subject with pleural effusion (a) shows intense radiotracer uptake in the lungs, limiting visualization of myocardial
SUV ratio were significantly higher in cardiac amyloid subjects than in control subjects. Overall myocardial RI tended to be higher in AL subjects than in ATTR subjects but none of the indices tested (RI, LV myocardial SUV values, TBR or LV myocardium to liver SUV ratio) clearly distinguished AL from ATTR amyloidosis. Importantly, in this pilot study, 18F-florbetapir was useful for imaging both AL and ATTR cardiac amyloidosis, and a RI of >0.025 min−1 and a TBR of >1.45 identified all AL and TTR amyloid subjects and none of the control subjects. 18 F-Florbetapir was originally developed to image beta amyloid deposits in the brain and its radiotracer distribution was studied, but imaging was not focused on the heart [8, 9].
B. TTR-Heart Failure
uptake, the ATTR subject with pleural effusions from heart failure (b) shows no increase in lung radiotracer uptake. Also, the time SUV curves (bottom) demonstrate higher lung than blood pool SUV values after the first 2 min in the subject with pleural amyloidosis (a) but not in the subject with congestive heart failure (b). These findings suggest that florbetapir imaging may be helpful in identifying lung amyloidosis. AL light chain amyloidosis, ATTR transthyretin amyloidosis, SUV standardized uptake value
Eur J Nucl Med Mol Imaging
Our study adds to the prior literature by detailing the distribution of 18F-florbetapir in the heart and lungs of control subjects and amyloid subjects. In control subjects, 18 F-florbetapir activity in the blood pool and myocardium peaked early and cleared quickly to very low levels. In amyloid subjects, the blood pool activity peaked early and cleared quickly, whereas the myocardial 18F-florbetapir uptake was rapid followed by myocardial retention for about 60 min, allowing an excellent TBR. The rapid blood pool clearance combined with the quick metabolism to metabolites that do not bind to beta amyloid [11] may explain the lack of late myocardial accumulation of 18Fflorbetapir allowing injection followed by a 30-min scan. Thus the radiotracer characteristics of 18F-florbetapir appear to be optimal for imaging cardiac amyloidosis. Although the specific reasons for myocardial 18Fflorbetapir uptake in cardiac amyloidosis are not known, binding of 18F-florbetapir or its metabolites to amyloid deposits or blood radioactivity are possible mechanisms. 18 F-Florbetapir selectively binds to beta-amyloid aggregates and beta-amyloid plaques in the brain with the binding intensity quantitatively correlated with the density of beta amyloid in post-mortem studies of Alzheimer’s subjects [10]. High myocardial radiotracer uptake and late retention in amyloid subjects compared to control subjects suggests a correlation with amyloid protein. Higher median 18F-florbetapir RI and myocardial SUV values in AL amyloid subjects suggest greater avidity for AL than for ATTR protein. Given the lower myocardial mass in AL subjects than in ATTR subjects, we speculate that higher radiotracer uptake may reflect not only amyloid mass, but also amyloid disease activity. Comparison with 11C-PiB PET Recently, Antoni et al. [12] demonstrated that 11C-PiB, also a beta amyloid imaging agent for Alzheimer’s disease, is helpful in imaging cardiac amyloidosis. In that study, 11C-PiB uptake was noted in all ten amyloid subjects and in none of five control subjects [12]. Similar results were observed in our study, with the RI for 11C-PiB in amyloidosis subjects being significantly higher than in control subjects (0.054 min−1 vs. 0.025 min−1, P=0.007);,but the authors of the previous study did not specifically compare the RI between TTR subjects and AL subjects. The RI values in our study cannot be directly compared to those obtained by Antoni et al. due to differences in the imaging techniques, patient characteristics and radiotracer used. While 11C-PiB offers the advantage of quantifying myocardial amyloid burden, its use is limited to PET centers with a cyclotron (20-min half-life) and is not FDA-approved. On the other hand, 18F-florbetapir can be quantified, can be used at PET centers without a cyclotron (110-min half-life), and is
clinically available as an FDA-approved brain imaging agent. As our initial results are promising, large multicenter studies in subjects with suspected amyloidosis may be undertaken to advance the management of subjects with this underdiagnosed fatal disease. Comparison with 99mTc-pyrophosphate/99mTc-DPD SPECT 99m
Tc-Labeled pyrophosphate (PYP) [13] and 99mTc-3,3diphosphono-1,2-propanodicarboxylic acid (DPD) [14] SPECT imaging is useful in the diagnosis of ATTR disease. Perugini et al. [14] studied 35 subjects (15 ATTR, 10 AL, 10 controls) and found that heart retention (median 7.3 % vs. 3.8 % vs. 2.9 %) and heart/whole-body retention (10 % vs. 5.4 % vs. 5.4 %) of 99mTcDPD were significantly higher in ATTR subjects (P1.5 had a sensitivity of 97 % and a specificity of 100 % in identifying those with ATTR disease. 99m Tc-PYP and DPD SPECT is widely available and excellent for imaging TTR cardiac amyloidosis, likely due to higher concentrations of calcium-containing products in TTR amyloid [13]. However, a negative test does not exclude the possibility of AL cardiac amyloidosis and absolute quantitation is limited, making assessment of response to therapy challenging. In contrast, 18F-florbetapir imaging appears to image both AL and ATTR amyloid and is quantitative. Indeed, 18F-florbetapir could potentially serve as an excellent amyloid-specific screening imaging test, which if positive can be followed by a myocardial biopsy (for disease typing) or by 99m Tc-PYP/DPD imaging (to confirm ATTR when suspicion of TTR amyloid is high). Clinical implications A definitive diagnosis of cardiac amyloidosis is primarily based on right ventricular endomyocardial biopsy with immunohistochemistry or mass spectroscopy as needed. Other findings such as a typical morphological appearance on echocardiography, cardiac MRI [15] or increased extracellular volume fraction on cardiac MRI [16], with an extracardiac histological diagnosis of amyloidosis, may be helpful in diagnosing cardiac amyloidosis in AL disease, yet not always definitive. Also, current imaging techniques tend to diagnose amyloidosis at an advanced stage and after anatomic changes in cardiac structure have occurred, the reversal of which in response to appropriate therapy is slow and not easily measurable. Quantitative imaging with 18F-florbetapir may potentially identify early cardiac amyloidosis, quantify amyloid disease
Eur J Nucl Med Mol Imaging
burden and help assess changes in response to therapy. 18FFlorbetapir may add to 123I SAP scans [7] by quantifying amyloid burden in the heart, as 123I SAP provides an estimate of whole-body amyloid burden. Radionuclide techniques including 18F-florbetapir may be particularly well suited for imaging subjects with renal disease who cannot undergo contrastenhanced cardiac MRI. Due to rapid clearance from normal lung and bone tissue, 18F-florbetapir may offer the ability to identify amyloidosis of the lungs and bones. It may not, however, be helpful for identifying amyloidosis in the liver, as 18Fflorbetapir is eliminated by the hepatobiliary system. Strengths and limitations To the best of our knowledge this is the first study to investigate the use of 18F-florbetapir for imaging cardiac amyloidosis. However, because of the small number of subjects, we did not calculate sensitivity, specificity and diagnostic accuracy. In addition to healthy volunteers, two control subjects with heart failure were included for comparison, as the test will most likely be used in subjects to evaluate symptoms of heart failure. However, future studies including more subjects and subjects with increased LV wall thickness, restrictive physiology and nonamyloid infiltrative heart diseases are warranted to definitively establish the value of 18F-florbetapir in the diagnosis of cardiac amyloidosis. Also, detailed analyses to differentiate AL from ATTR amyloidosis were limited in this pilot study Conclusion The results of this pilot study suggest that 18F-florbetapir may be a novel PET radiotracer for imaging both AL and ATTR cardiac amyloidosis. 18F-Florbetapir appears to be specific in distinguishing subjects with cardiac amyloidosis from control subjects. Further study is needed to better understand its value in following response to therapy and in imaging amyloidosis in other organ systems. Acknowledgments We are indebted to the subjects who participated in this study. We are appreciative of the research support for this study from the Amyloid Foundation. Dr. Dorbala was supported by NIH NHLBI grant K23HL092299. We are grateful to our colleagues at the Brigham and Women’s Hospital Cardiac Amyloidosis Program. Financial Disclosures Dorbala: Research Grant Astellas Global Pharma Development. Di Carli: Research grant from Gilead Sciences.
References 1. Merlini G, Bellotti V. Molecular mechanisms of amyloidosis. N Engl J Med. 2003;349:583–96. 2. Falk RH. Cardiac amyloidosis: a treatable disease, often overlooked. Circulation. 2011;124:1079–85. 3. Ruberg FL, Berk JL. Transthyretin (TTR) cardiac amyloidosis. Circulation. 2012;126:1286–300. 4. Skinner M, Sanchorawala V, Seldin DC, Dember LM, Falk RH, Berk JL, et al. High-dose melphalan and autologous stem-cell transplantation in patients with AL amyloidosis: an 8-year study. Ann Intern Med. 2004;140:85–93. 5. Kristen AV, Perz JB, Schonland SO, Hegenbart U, Schnabel PA, Kristen JH, et al. Non-invasive predictors of survival in cardiac amyloidosis. Eur J Heart Fail. 2007;9:617–24. 6. Miroy GJ, Lai Z, Lashuel HA, Peterson SA, Strang C, Kelly JW. Inhibiting transthyretin amyloid fibril formation via protein stabilization. Proc Natl Acad Sci U S A. 1996;93: 15051–6. 7. Hawkins PN, Lavender JP, Pepys MB. Evaluation of systemic amyloidosis by scintigraphy with 123I-labeled serum amyloid P component. N Engl J Med. 1990;323:508–13. 8. Yang L, Rieves D, Ganley C. Brain amyloid imaging – FDA approval of florbetapir F18 injection. N Engl J Med. 2012;367:885–7. 9. Clark CM, Schneider JA, Bedell BJ, Beach TG, Bilker WB, Mintun MA, et al. Use of florbetapir-PET for imaging beta-amyloid pathology. JAMA. 2011;305:275–83. 10. Lister-James J, Pontecorvo MJ, Clark C, Joshi AD, Mintun MA, Zhang W, et al. Florbetapir F-18: a histopathologically validated betaamyloid positron emission tomography imaging agent. Semin Nucl Med. 2011;41:300–4. 11. Choi SR, Golding G, Zhuang Z, Zhang W, Lim N, Hefti F, et al. Preclinical properties of 18 F-AV-45: a PET agent for Abeta plaques in the brain. J Nucl Med. 2009;50:1887–94. 12. Antoni G, Lubberink M, Estrada S, Axelsson J, Carlson K, Lindsjo L, et al. In vivo visualization of amyloid deposits in the heart with 11C-PIB and PET. J Nucl Med. 2013;54:213– 20. 13. Bokhari S, Castano A, Pozniakoff T, Deslisle S, Latif F, Maurer MS. (99m)Tc-pyrophosphate scintigraphy for differentiating light-chain cardiac amyloidosis from the transthyretinrelated familial and senile cardiac amyloidoses. Circ Cardiovasc Imaging. 2013;6:195–201. 14. Perugini E, Guidalotti PL, Salvi F, Cooke RM, Pettinato C, Riva L, et al. Noninvasive etiologic diagnosis of cardiac amyloidosis using 99mTc-3,3-diphosphono-1,2-propanodicarboxylic acid scintigraphy. J Am Coll Cardiol. 2005;46:1076–84. 15. Syed IS, Glockner JF, Feng D, Araoz PA, Martinez MW, Edwards WD, et al. Role of cardiac magnetic resonance imaging in the detection of cardiac amyloidosis. JACC Cardiovasc Imaging. 2010;3:155–64. 16. Banypersad SM, Sado DM, Flett AS, Gibbs SD, Pinney JH, Maestrini V, et al. Quantification of myocardial extracellular volume fraction in systemic Al amyloidosis: an equilibrium contrast cardiovascular magnetic resonance study. Circ Cardiovasc Imaging. 2013;6:34–9.
![Examining endogenous dopamine in treated schizophrenia using [¹¹C]-(+)-PHNO positron emission tomography: A pilot study.](/assets/img/hold.png)
