Transthyretin amyloid cardiomyopathy disease burden quantified using ^99mTc-pyrophosphate SPECT/CT: volumetric parameters versus SUVmax ratio at 1 and 3 hours

Transthyretin amyloid cardiomyopathy (ATTR-CM) is an increasingly recognized cause of heart failure (HF) resulting from the myocardial deposition of misfolded protein fibrils.¹ Disease-modifying therapies for ATTR-CM, ATTR stabilizer and gene silencing pharmacotherapy have recently been developed.^2,3 This has resulted in a clinical need for early detection and accurate quantitation of disease burden, prognosis, and response to treatment. ATTR-CM can be non-invasively diagnosed in patients without light-chain cardiac amyloidosis using bone scintigraphy with technetium-99m pyrophosphate (^99mTc-PYP), ^99mTc-3,3-diphosphono-1,2-propanodicarboxylic acid (DPD), or ^99mTc-hydroxymethylenediphosphonate.^4,5 Recently, bone scintigraphy has been actively investigated not only for diagnosis, but also for assessments of cardiac function, disease burden, prognosis, and response to therapy.^6‐19

Various quantitative parameters derived from ^99mTc-PYP scintigraphic images using three-dimensional single-photon emission computed tomography (SPECT) are useful for assessing the ATTR-CM disease burden and prognosis.^6‐14 Accurate quantitation also enables interval monitoring of disease progression and responses to treatment. However, an optimal method has not been established. Uptake parameters such as the standardized uptake value (SUV), the SUV ratio (normalized SUV and the SUV retention index), as well as volumetric parameters have been investigated using images acquired at various timing. Most studies have investigated ^99mTc-PYP SPECT images acquired 3 hours after radiotracer injection to minimize the influence of relatively high blood pool activity at 1 hour.^6‐11 Blood pool activity is particularly important for patients without ATTR-CM because it can result in false-positive diagnoses. However, the optimal method for diagnosis, including cohorts without ATTR-CM, is not necessarily the same as that for assessing disease burden.¹⁴ This should be determined based on cohorts with ATTR-CM.^6,11,17‐20 Moreover, images acquired at 1 hour after radiotracer injection have a unique feature that is lacking in images acquired at 3 hours; myocardial abnormal uptake on bone scintigraphic images peaks at 1 hour, then slowly declines.²⁰

The aim of this study was to define whether ^99mTc-PYP volumetric parameters or SUVmax ratios are more effective for quantifying ATTR-CM disease burden at 1 and 3 hours after radiotracer injection.

This study has two important messages. The volumetric parameters derived from ^99mTc-PYP SPECT/CT, CPV and CPA quantified the ATTR-CM disease burden more accurately than SUVmax and SUVmax ratios. The performance of the volumetric parameters was equivalent between 1 and 3 hours after injecting ^99mTc-PYP.

Both CPV and CPA correlated more closely than the myocardial SUVmax and SUVmax ratios with all four prognostic factors (Table 3). One reason for this could be that the volumetric parameters reflect information about internal abnormal regions, whereas SUVmax and SUVmax ratios represent small areas. Furthermore, SUVmax is affected by various technical and physiological factors. Therefore, volumetric parameters could be useful imaging biomarkers for quantifying the ATTR-CM disease burden. Similarly, volumetric parameters derived from ¹⁸F-fluorodeoxyglucose positron emission tomography (¹⁸F-FDG PET) are more effective than SUVmax as prognostic predictors in cancer.²² Volumetric parameters were more effective than the H/CL ratio, which is valuable and widely applied, but has inherent limitations due to being based on two-dimensional images. The H/CL ratio did not correlate with LVEF, GLS, and troponin T (Table 3). Vranian et al. also showed that the H/CL ratio did not correlate with LVEF and GLS in patients with ATTR-CM.²⁹ We previously showed that the H/CL ratio did not correlate with LVEF in patients with ATTR-CM.¹⁰

We found here that CPV and CPA significantly correlated with the four prognostic factors, LVEF, GLS, ECV, and troponin T. Therefore, CPV and CPA should play objective and important roles in prognostic evaluation. An echocardiographic LVEF reduced to < 50% predicts mortality in patients with wild type ATTR-CM (ATTRwt-CM).²⁵ Furthermore, GLS is an independent predictor of all-cause mortality (hazard ratio [HR]: 1.15 per 1% decrease) and a more effective prognostic factor than all other echocardiographic parameters, including LVEF, in patients with HF and reduced LVEF.²⁶ Impaired GLS is similarly a high risk factor for cardiovascular morbidity and mortality in patients with HF and preserved LVEF.²⁷ Furthermore, the ECV determined by CMR has been validated for measuring the amyloid burden and it is an independent predictor of prognosis (HR: 1.155 per 3% increase).²⁸ Bone scintigraphy can be an alternative modality for patients who cannot undergo contrast-enhanced CMR. High-sensitivity cardiac troponin T is used to assess prognosis in patients with ATTRwt-CM.^11,25

Other studies similarly support the notion that ^99mTc-PYP scintigraphy is a useful modality for evaluating prognosis. A volumetric parameter on ^99mTc-PYP SPECT/CT images acquired 3 hours after injection and ECV correlated (r² = .76, P = .001, n = 11).⁷ Even CPV and CPA derived from ^99mTc-PYP SPECT without CT significantly correlated with LVEF and CMR parameters, and had prognostic value and low interobserver variability.^8,9 However, studies without CT did not correct attenuation or scatter. They acquired images only after 3 hours and applied filtered back projection. A multicenter study found that even the two-dimensional H/CL ratio of ^99mTc-PYP scintigraphy can deliver prognostic information.³⁰

Here, we revealed for the first time that myocardial abnormal ^99mTc-PYP uptake quantified by SUVmax was significantly higher at 1 hour than at 3 hours (Table 4). The uptake of ^99mTc-DPD, which is slightly different from that of ^99mTc-PYP,³¹ similarly peaks at 1 hour.²⁰

The present findings supported the notion that the time-efficient acquisition of ^99mTc-PYP SPECT images at 1 hour could accurately quantify the ATTR-CM disease burden. Correlations between prognostic factors and CPV1.4 and CPA1.4 were equivalent between 1 and 3 hours in our small cohort of patient (Table 5). Thus, larger studies are warranted. To determine whether ^99mTc-PYP SPECT images acquired at 1 hour are useful for prognostic evaluation is important for three reasons. Myocardial abnormal uptake peaks at 1 hour, then slowly declines. Laboratory throughput and patient satisfaction were increased by the 1-hour protocol compared with the 3-hour protocol. Some experienced centers have adopted 1-hour acquisitions, particularly in the US for ^99mTc-PYP imaging.³² A close correlation has been determined between ECV and myocardial ^99mTc-PYP SUVmax at 1 hour corrected for blood pool (r² = .83, P = .001, n = 9).¹² The diagnostic ability of ^99mTc-PYP three-dimensional SPECT is comparable between 1 and 3 hours.^33‐35 Delayed acquisition has been reported as a cause of false-negative diagnoses of ATTR-CM.³⁶ Oncological ¹⁸F-FDG PET image acquisition at 1 hour is popular and delayed acquisition is optional. The expert consensus recommendation for image acquisition timing using ^99mTc-PYP was 1 hour in 2019.⁵ This consensus was revised to recommend image acquisition at 2 or 3 hours with 1 hour as optional. The main reason for this was to avoid false-positive diagnoses caused by excessive blood pool activity.³⁷ The blood pool activity is particularly important in patients without ATTR-CM. However, the optimal method for diagnosis, including cohorts without ATTR-CM, is not necessarily the same as that for assessing disease burden,¹⁴ which should be assessed based on cohorts with ATTR-CM.^6,11,17‐20

We included only patients with biopsy that were positive for ATTR amyloid. This is a strength of our original study. Several studies have applied positive criteria for ATTR-CM, such as visual grading scores of 2 or 3 for myocardial uptake (≥ rib uptake). However, some patients with ATTR-CM have focal or absent ^99mTc-PYP uptake,^4,6,10,33 which we identified in some of our patients with EMB-proven ATTR-CM (Figure 2B and Table 2). In addition, several causes of false-negative and false-positive results of bone scintigraphy in patients with suspected ATTR-CM have been identified.³⁶

We evaluated the threshold dependence of correlations and found that the correlation coefficients of CPV1.4 and CPA1.4 were slightly closer than those of CPV1.2 and CPA1.2 for the four prognostic factors (Table 6). The optimal threshold is not necessarily identical between disease burden quantitation and diagnosis. The volumetric evaluation of bone scintigraphy could potentially be an objective marker for the diagnosis of ATTR-CM, especially in patients with focal abnormal uptake. We previously showed that CPV1.2 diagnosed ATTR-CM more effectively than CPV1.4 at 3 hours.¹⁰ That finding and those of images acquired at 1 and 3 hours herein (Figure 4B; Table 4) revealed that the threshold 1.4 × aortic blood pool SUVmax might result in false-negative diagnoses in patients with focal abnormal uptake that might occur during the early stage of ATTR-CM.¹⁰

The present study showed that the SUVmax ratios correlated more closely than myocardial SUVmax with the four prognostic factors (Table 3). This means that accurate measurements of myocardial SUVmax might not directly reflect the disease burden because myocardial SUVmax also depends on inter-organ competition for ^99mTc-PYP uptake. The SUV retention index, which has been mainly investigated using ^99mTc-DPD,^16,18,20 has not been studied in detail using ^99mTc-PYP.¹² Although the SUV retention index has been proposed to account for competition among the myocardium, bone, and soft-tissue radiotracer uptake,^{16 99m}Tc-DPD and ^99mTc-PYP uptake mechanisms slightly differ.³¹ The present study showed that the SUVmax retention index was significantly higher at 1 hour than at 3 hours. Similarly, the SUVpeak retention index using ^99mTc-DPD was significantly higher at 1 hour than at 3 hours.²⁰

Volumetric parameters derived from ^99mTc-PYP SPECT/CT, CPV and CPA quantified the ATTR-CM disease burden more accurately than the SUVmax, SUVmax ratio, and H/CL ratio. The performance was equivalent between 1 and 3 hours after injection. Larger studies are warranted to clarify an optimal nuclear imaging biomarker to manage patients with ATTR-CM.

Volumetric evaluation of ^99mTc-PYP SPECT/CT quantified the ATTR-CM disease burden more accurately than the SUVmax and SUVmax ratios including the SUVmax retention index. Significant correlations between volumetric parameters and prognostic factors were equivalent between images acquired at 1 and 3 hours after ^99mTc-PYP injection, whereas the myocardial SUVmax, retention index, and visual grading score were significantly higher at 1 hour than at 3 hours.