Nuclear Imaging for the Diagnosis of Cardiac Amyloidosis in 2021

Cardiac amyloidosis is caused by the deposition of misfolded protein fibrils into the extracellular space of the heart. The diagnosis of cardiac amyloidosis remains challenging because of the heterogeneous manifestations of the disease. There are many different types of amyloidosis with light-chain (AL) amyloidosis and transthyretin (ATTR) amyloidosis being the most common types of cardiac amyloidosis. Endomyocardial biopsy is considered the gold standard for diagnosing cardiac amyloidosis and differentiating amyloid subtypes, but its use is limited because of the invasive nature of the procedure, with risks for complications and the need for specialized training and centers to perform the procedure. Radionuclide cardiac imaging has recently become the most commonly performed test for the diagnosis of ATTR amyloidosis but is of limited value for the diagnosis of AL amyloidosis. Positron emission tomography has been increasingly used for the diagnosis of cardiac amyloidosis and its applications are expected to expand in the future. Imaging protocols are under refinement to achieve better quantification of the disease burden and prediction of prognosis.


Introduction
Systemic amyloidosis is a multisystem disorder characterized by the formation and deposition of mis-folded protein fibrils which can result in multi-organ failure and death [1,2]. This condition is associated with significant disease burden with increasing incidence and prevalence worldwide over the past decades [3][4][5]. Studies have shown that at least 20 out of one million UK residents are estimated to have systemic amyloidosis with 65% being light-chain (AL) amyloidosis. The prevalence of wild type ATTR amyloidosis is estimated to be 10-25% in people over the age of 80 [6][7][8][9].
Cardiac amyloidosis is defined as a group of disorders that involve the deposition of amyloid protein in the cardiac tissue, leading to myocardial dysfunction [10]. Due to the increasing awareness of the disease, improved life expectancy, and advancements in diagnostic pathways, cardiac amyloidosis is currently diagnosed more frequently than in the past, with AL and ATTR amyloidosis being the most common types [3]. One populationbased study focusing on Medicare beneficiaries in the United States revealed that between 2000 and 2012 the prevalence of cardiac amyloidosis increased from 8 to 17 cases per 100,000 person-year and the incidence increased from 18 to 55 cases per 100,000 personyear [11]. It is estimated that by 2050 there will be almost 25 million cases of wild-type ATTR globally [12,13].

Imaging Techniques and Radiotracers
Nuclear imaging for cardiac amyloidosis includes cardiac scintigraphy which adopts radiotracers from bone scintigraphy and the technique of Positron Emission Tomography (PET) using targeted tracers for amyloid specific proteins ( Figure 1). Nuclear imaging can offer direct visualization of disease activity and semi-quantification of the amyloid burden by calculating the ratio between concentration of the radiotracer in a specific volume of tissue and the concentration if the radiotracers are uniformly distributed. The ratio is also known as Standardized Uptake Value (SUV) [28]. Retention index (RI), calculated to assess the retention of radiotracers in myocardium over certain time interval, can be another important method measuring amyloid deposition quantitatively in PET imaging [29][30][31].

Cardiac Scintigraphy
Technetium (Tc)-labeled radiotracers from phosphate derivatives which are common bone scan agents have been investigated for the diagnosis of amyloidosis since the 1970s [32]. Cardiac scintigraphy utilizing those radiotracers has become more popular and widely used in clinical practice to assist the diagnosis of cardiac amyloidosis. A largesized multi-center study including 1217 patients with suspected cardiac amyloidosis supported that cardiac scintigraphy with 99m Tc-diphosphono-1,2-propanodicarboxylic acid ( 99m Tc-DPD), 99m Tc-pyrophosphate ( 99m Tc-PYP), or 99m Tc-hydroxymethylene diphosphonate ( 99m Tc-HMDP) had a 100% specificity and positive predictive value for ATTR cardiac amyloidosis if there are combined findings of grade 2 or 3 myocardial radiotracer uptake on cardiac scintigraphy and the absence of monoclonal protein in serum or urine [33]. Systematic review also confirmed the accuracy of scintigraphy diagnosing ATTR cardiac amyloidosis with both sensitivity and specificity above 90% [34]. Although it remains unclear how those bone-seeking agents can differentiate the types of amyloidosis, theories have postulated that the higher calcium containing compounds in ATTR cardiac amyloidosis, the unique characteristics of amyloidogenic fibrils, and the more indolent clinical course of ATTR amyloidosis allowing larger amount of amyloid protein to accumulate before the onset of symptoms might play a role [35]. However, the results of cardiac scintigraphy can be affected by multiple factors such as rib fracture and valvular/annular calcifications [36]. Single-Photon Emission Computed Tomography (SPECT) technique can add a three-dimensional visualization to planar scintigraphy as well as more detailed and accurate assessment of radiotracer uptake in the myocardium wall as opposed to the blood pool [37]. Currently, 99m Tc-DPD and 99m Tc-PYP are the two most commonly used and studied radiotracers for the diagnosis of ATTR cardiac amyloidosis.

Cardiac Scintigraphy
Technetium (Tc)-labeled radiotracers from phosphate derivatives which are common bone scan agents have been investigated for the diagnosis of amyloidosis since the 1970s [32]. Cardiac scintigraphy utilizing those radiotracers has become more popular and widely used in clinical practice to assist the diagnosis of cardiac amyloidosis. A large-sized multi-center study including 1217 patients with suspected cardiac amyloidosis supported that cardiac scintigraphy with 99m Tc-diphosphono-1,2-propanodicarboxylic acid ( 99m Tc-DPD), 99m Tc-pyrophosphate ( 99m Tc-PYP), or 99m Tc-hydroxymethylene diphosphonate ( 99m Tc-HMDP) had a 100% specificity and positive predictive value for ATTR cardiac amyloidosis if there are combined findings of grade 2 or 3 myocardial radiotracer uptake on cardiac scintigraphy and the absence of monoclonal protein in serum or urine [33]. Systematic review also confirmed the accuracy of scintigraphy diagnosing ATTR cardiac amyloidosis with both sensitivity and specificity above 90% [34]. Although it remains unclear how those bone-seeking agents can differentiate the types of amyloidosis, theories have postulated that the higher calcium containing compounds in ATTR cardiac amyloidosis, the unique characteristics of amyloidogenic fibrils, and the more indolent clinical course of ATTR amyloidosis allowing larger amount of amyloid protein to accumulate before the onset of symptoms might play a role [35]. However, the results of cardiac scintigraphy can be affected by multiple factors such as rib fracture and valvular/annular calcifications [36]. Single-Photon Emission Computed Tomography (SPECT) technique can add a three-dimensional visualization to planar scintigraphy as well as more detailed and accurate assessment of radiotracer uptake in the myocardium wall as opposed to the blood pool [37]. Currently, 99m Tc-DPD and 99m Tc-PYP are the two most commonly used and studied radiotracers for the diagnosis of ATTR cardiac amyloidosis.

99m Tc-3,3-diphosphono-1,2-propanodicarboxylic Acid ( 99m Tc-DPD) Scintigraphy
99m Tc-DPD scintigraphy is usually performed as a whole-body planar imaging three hours after the injection of 99m Tc-DPD radiotracer which can be followed by SPECT and non-contrast CT as adjuncts. 99m Tc-DPD scintigraphy has high diagnostic accuracy for ATTR amyloidosis, especially when using the Perugini visual score. According to the Perugini visual score, the degree of radiotracer uptake is visually graded by comparing the radiotracer activity in the heart with its activity in the bones. Grade 0 means no cardiac uptake of radiotracer; grade 1 means that cardiac uptake is mild and less than skeletal uptake; grade 2 means that cardiac uptake is moderate and equals skeletal uptake; grade 3 means cardiac uptake is high and stronger than skeletal uptake. Grade 2 and above are considered as positive scan [38]. However, this protocol can lead to false positive results in patients with AL cardiac amyloidosis [33]. Therefore, AL cardiac amyloidosis needs to be ruled out first with serum free light chains, serum protein electrophoresis, and urine protein electrophoresis before interpreting the results of cardiac scintigraphy. As the visual scoring system highly depends on reader expertise, it performed poorly when assessing the degree of amyloid burden [39]. In addition, Perugini visual score has not been found to have any prognostic significance in the overall survival for patients with cardiac amyloidosis [40].
Researchers have attempted to increase the diagnostic accuracy of 99m Tc-DPD scintigraphy and quantitatively assess amyloid burden by calculating the ratio between retention of radiotracer in the heart and retention of radiotracer in other body parts. Heart/wholebody ratio (H/WB), heart/pelvis ratio and heart/contralateral lung ratios (H/CL) are commonly used in clinical studies [41]. ATTR amyloidosis was found to have a higher H/WB ratio than AL amyloidosis. A study from Australia which enrolled biopsy-proven AL and ATTR cardiac amyloidosis has proposed a cut-off of H/WB ratio >0.091 with sensitivity of 92% and specificity of 88% for the diagnosis of ATTR amyloidosis [42]. Additionally, an increasing H/WB ratio has been shown to correlate with major adverse cardiac events in patients with hereditary ATTR cardiac amyloidosis [43]. Interestingly, researchers also found that 99m Tc-DPD scintigraphy might have a role in the diagnosis of extracardiac AL amyloidosis when cardiac uptake is absent [44].
SPECT/CT has been developed to assist quantification in 99m Tc-DPD Scintigraphy by acquiring peak Standard Uptake Values (SUVs) in the myocardium and offering threedimensional assessment [45]. The cardiac peak SUV can be further normalized with the peak SUV on the bone or soft tissue as SUV retention index [46]. Studies have shown that cardiac SUV and SUV retention index are correlated well with Perugini visual scores and a peak SUV cut-off of 3.1 can separate patients with Perugini grade 2 and 3 clearly from those with Perugini grade 0 and 1 [47]. A recent study has found that the amyloid load in 99m Tc-DPD SPECT/CT has correlated well with strain values in echocardiography and biomarkers such as troponin and NT-proBNP (B-type Natriuretic Peptide) [48]. However, SPECT/CT was still unable to differentiate between patients with Perugini grade 2 and 3, which suggests that quantification of amyloid burden by 99m Tc-DPD SPECT/CT needs further improvement [39].

99m Tc-Pyrophosphate ( 99m Tc-PYP) Scintigraphy
Although promising, 99m Tc-DPD is not approved for use by the Food and Drug Administration (FDA) in the United States. Hence, 99m Tc-pyrophosphate ( 99m Tc-PYP) is the only FDA-approved radiotracer in the US to diagnose cardiac amyloidosis [49]. Clinicians usually obtain anterior, lateral and left anterior oblique planar views as well as SPECT imaging following injection of 99m Tc-PYP [50]. The degree of myocardial tracer uptake is graded using the semi-quantitative Perugini visual score and quantitative analysis by obtaining radiotracer activity within a region of interest (ROI) drawn over the heart corrected and its activity in the contralateral side of ROI to calculate a heart-to-contralateral (H/CL) ratio [51]. Unlike the 3-h protocol which is required in 99m Tc-DPD scintigraphy, it has been found that a 1-h protocol in 99m Tc-PYP imaging is comparable to the 3-h protocol for the diagnosis of ATTR cardiac amyloidosis. This translates to a 98% sensitivity and a 96% specificity of planar imaging and SPECT, identical between the 1-h and 3-h protocols [52]. The 1-h protocol reduces cost and time without compromising the diagnostic accuracy of the test and thus it is widely used.
Bokhari et al. found that subjects with ATTR cardiac amyloidosis had a significantly higher cardiac visual score (p < 0.0001) as well as higher H/CL ratio (p < 0.00001) than AL amyloidosis and they concluded that using a H/CL ratio of ≥ 1.5, which is consistent with intensely diffused myocardial tracer retention, had a 97% sensitivity and 100% specificity (p < 0.0001) for identifying ATTR cardiac amyloidosis [35]. In a multicenter study which enrolled 171 participants, 99m Tc-PYP scan demonstrated an overall 91% sensitivity and 92% specificity for detecting ATTR cardiac amyloidosis with area under the curve of 0.960 (95% CI, 0.930-0.981) [53]. It has also been noted that an H/CL ratio ≥ 1.6 predicts lower 5-year survival compared with group of patients with an H/CL ratio ≤ 1.6 (log-rank p = 0.02) [53]. Despite the high accuracy of the 99m Tc-PYP scan visual score and H/CL ratio, the addition of SPECT is still necessary to rule out misclassified cases and distinguish myocardial activity from blood pool uptake [54]. A recent study showed that combining 99m Tc-PYP and Thallium (Tl)-201 may improve diagnostic accuracy of both visual differentiation and H/CL semi-quantification for ATTR amyloidosis [55]. In addition, an integrated approach of utilizing both high sensitivity cardiac troponin T and 99m Tc-PYP scintigraphy can significantly increase diagnostic yield of wild-type ATTR cardiac amyloidosis [56]. A series of studies of 99m Tc-DPD and 99m Tc-PYP scintigraphy published from 2020 to 2021 are listed in Table 1. Statistically significant correlation between DPD uptake and all echocardiographic strain parameters in all regions, as well as the biomarkers of troponin and logarithmic NT-proBNP. High diagnostic accuracy of both visual inspection and semi-quantitative methods of 11 C-PiB PET imaging to distinguish cardiac amyloidosis from controls. The uptake of 11 C-PiB was significantly higher in AL cardiac amyloidosis than ATTR cardiac amyloidosis.
Lee et al. [59] 2020 11 C-PiB 41 chemotherapy-naïve AL cardiac amyloidosis patients were enrolled. Myocardial uptake of 11 C-PiB on PET was compared with endomyocardial biopsy for quantification of amyloid deposit.
The degree of myocardial 11 C-PiB uptake is significantly higher in patients with cardiac amyloidosis and higher degrees of uptake was associated with lowest survival from death, heart transplantation and acute decompensated heart failure.

Positron Emission Tomography (PET)
Positron emission tomography (PET) scanning is another imaging modality which can help diagnose cardiac amyloidosis [63]. PET imaging offers higher spatial resolution secondary to the decay of positrons and more accurate quantification of amyloid burden by using direct amyloid-binding radioactive tracers [64]. 11 C-Pittsburgh B ( 11 C-PiB) and 18 F-labelled agents (such as 18 F-florbetapir and 18 F-florbetaben) are the two most common classes of radioactive tracers used for this purpose [65]. The tracers were originally developed to bind beta amyloid in the brain of patients with Alzheimer disease but it was reported later that they might have utility in diagnosing cardiac amyloidosis as well [66]. Higher cardiac uptake of both 11 C-PiB and 18 F-labelled agents was constantly observed in both AL cardiac amyloidosis and ATTR cardiac amyloidosis, compared to controls in pilot studies [29,60,[67][68][69][70]. Additionally, the radiotracer activities of both 11 C-PiB and 18 F-labelled agents have been found to be higher in AL cardiac amyloidosis than ATTR amyloidosis. A meta-analysis which combined the results of three pilot PET studies demonstrated that AL amyloidosis has significantly higher radiotracer activities than ATTR amyloidosis and thus PET imaging carries the potential to differentiate between AL and ATTR amyloidosis [71]. Overall, PET imaging for the diagnosis of cardiac amyloidosis is still in the early stages but future development of this technique is anticipated.

11 C-Pittsburgh Compound B PET Imaging
11 C-Pittsburgh compound B ( 11 C-PiB) PET imaging is a well-established technique for detecting β-amyloid in Alzheimer disease [72]. PiB, thioflavin-T, is an amyloid binding dye, and is theoretically able to bind to amyloid fibrils of any type, including amyloid fibrils in the myocardium [73]. A Swedish study included 10 patients with systemic amyloidosis (7 AL, 2 hereditary ATTR, 1 wild-type ATTR) and cardiac involvement, and the results showed increased myocardial 11 C-PiB uptake in all the patients 15-25 min after injection of 11 C-PiB. On the other hand, increased uptake was not seen in the five patients of the control group [29]. A Korean prospective study, which included 22 amyloidosis patients (15 with and 7 without cardiac involvement) and 10 normal controls, calculated the SUV and found significantly higher values in patients with cardiac amyloidosis than the control group (median 3.9 (range 1.7 to 19.9) vs. 1.0 (range 0.8 to 1.2), p < 0.001) [67]. 11 C-PiB PET imaging shows promise for identifying specific types of cardiac amyloidosis, especially for the AL subtype. A dual-center study showed that 11 C-PiB PET imaging had 100% diagnostic accuracy of AL amyloidosis and that the uptake was significantly higher in AL cardiac amyloidosis compared to ATTR cardiac amyloidosis [58]. Researchers from Korea compared 11 C-PiB PET imaging with endomyocardial biopsy in patients with chemotherapy-naive AL cardiac amyloidosis and they found that the degree of the 11 C-PiB uptake on PET image was significantly higher in patients with cardiac amyloidosis and it corresponded well with the extent of amyloid deposition on the biopsy specimens. Patients with higher 11 C-PiB uptake had high risks of composite adverse clinical outcomes including death, requiring heart transplantation and acute decompensated heart failure [59].
A study conducted in Japan revealed that 99m Tc-PYP scintigraphy and 11 C-PiB PET imaging can complement each other. In this study, the combination of positive 11 C-PiB PET and negative 99m Tc-PYP was observed in all AL cardiac amyloidosis and early onset V30M hereditary ATTR cardiac amyloidosis, while the combination of positive 99m Tc-PYP and negative 11 C-PiB PET was consistent in all wild-type ATTR cardiac amyloidosis, as well as the late-onset V30M and non-V30M hereditary ATTR cardiac amyloidosis [57]. However, 11 C-PiB PET imaging is limited by its short half-life of 20 min and the requirement of an onsite cyclotron for its production [74].

18 F-Labelled Agents PET Imaging
18 F-labelled PET imaging is another promising imaging technique utilizing fluoride labelled radiotracers, primarily 18 F-florbetapir and 18 F-florbetaben. 18 F-florbetapir and 18 F-florbetaben are both FDA-approved radioactive tracers for Alzheimer's disease with half-lives more than 100 min, which is a potential advantage over 11 C-PiB PET for use in clinical practice. [75,76]. A pilot study in 2014 which enrolled 14 subjects (9 subjects with definite cardiac amyloidosis and 5 control subjects without amyloidosis) found that myocardial retention of 18 F-florbetapir was higher in amyloid subjects, especially in patients with AL cardiac amyloidosis [77]. An autoradiography study using myocardial autopsy sections yielded similar results, showing that 18 F-florbetapir uptake was higher in amyloid samples versus controls but also found higher uptake in the AL groups compared to the ATTR samples [78]. Another pilot study in 2016 indicated that 18 F-florbetaben PET imaging can accurately diagnose and differentiate between cardiac amyloidosis and hypertensive heart disease [69]. The percentage of myocardial 18 F-florbetaben retention was found to be an independent determinant of myocardial dysfunction in cardiac amyloidosis [69]. A pilot study in 2019 involving 22 subjects (5 proven and 17 with clinical suspicion of cardiac amyloidosis) revealed that 18 F-florbetaben-PET could further distinguish between the underlying amyloid types with higher retention in patients with AL amyloidosis than ATTR amyloidosis [79]. A recently published prospective study showed that delayed 18 Fflorbetaben cardiac uptake may distinguish AL cardiac amyloidosis from ATTR amyloidosis given higher mean SUV in patients with AL amyloidosis which was sustained over the whole acquisition period [60]. In addition, amyloid-directed PET can be used to assess therapy response. It has been shown that amyloid burden on PET after treatment with antiinflammatory (AA), anti-myeloma (AL) and TTR-stabilizing (ATTR) therapies correlated well with changes in performance status and serological markers [79]. A series of most recent studies for 11 C-PiB and 18 F-labelled agents PET imaging can be found in Table 1.
Researchers have combined fluoride PET imaging with MRI in patients with cardiac amyloidosis to improve diagnostic accuracy of ATTR amyloidosis [61]. However, while PET imaging can distinguish between cardiac amyloidosis and controls, particularly when using quantitative analysis, it seems to be less sensitive when diagnosing cardiac amyloidosis than the more established nuclear medicine studies with 99m Tc-PYP or 99m Tc-DPD [62,80]. The comparison between cardiac scintigraphy and PET imaging is summarized in Table 2. Our review highlights the importance of nuclear imaging for the diagnosis of cardiac amyloidosis with most updated clinical evidence and covers the most common radiotracers in this field. We not only elaborate on cardiac scintigraphy which is the more established nuclear imaging modality for cardiac amyloidosis, but also include the most recent evidence regarding PET imaging. However, we do acknowledge that other radiotracers, for example 99m Tc-HMDP, are present and may play a role in the diagnosis of cardiac amyloidosis. In addition, we are unable to find large-sized clinical studies to compare between those radiotracers mentioned in this review.

Conclusions
Cardiac scintigraphy with SPECT is the current standard of care for diagnosing patients with ATTR cardiac amyloidosis. However, PET imaging is another promising, noninvasive option for the diagnosis of cardiac amyloidosis and may help distinguish between AL amyloidosis and ATTR amyloidosis. The potential benefit of PET-based radiotracers includes better sensitivity for AL cardiac amyloidosis diagnosis and assessment of response to treatment. Future studies are anticipated.