In this study, we initially conducted the phantom study to obtain the ICF for quantitative SPECT and to derive PVC factor for recovering true activity concentration in myocardium. Indeed, the unique characteristic of PVC factor curve as a function of the measured Myo/Bp ACR elucidated that in order to accurately compensate for partial volume effect in myocardium, the coupled effect from Bp activity should take into account while the ratio is below 4.0. In our 99mTc-PYP quantitative SPECT data, we found PVC factor varied in a large range from 0.46 to 1.30 across the entire population (Fig. 1b). If only single PVC factor was utilized or no correction at all, erroneous assessment of activity concentration in myocardium can simply occur to impact on the subsequent calculation of quantitative parameters. While the patient specific Myo/Bp ACR was measured to derive the individual’s PVC factor based on the recovery curve, we found SUVmax, SUVmedian, SUVmean and nSUVmax were able to distinguish the ATTR-CM group from groups of AL cardiac amyloidosis and Others. For the same cohorts categorized by Perugini visual scores, SUVmax, SUVmedian, SUVmean were able to distinguish groups of Grade 2 and 3 from Grade 0 and 1, but not for nSUVmax. Part of reasons may be due to amplified variation after the normalization with SUVbone. So far, there has not been a systematic study to evaluate the reproducibility of 99mTc-PYP quantitative SPECT. To our knowledge, this is the first study to provide the relevant information. In our study, the intra- and inter-reproducibility of the quantitative method were excellent as R2 reached 0.902 to 0.978 with only small systematic difference (intra= -0.14 to -0.06; inter=-0.23 to -0.11) between two repeated measurements. The intra- and inter-reproducibility of quantitative SPECT outperformed that of the semi-quantitative method (R2: 0.811–0.861, all p < 0.0267). Consequently, the 99mTc-PYP quantitative SPECT developed in this study can be a reliable method to measure quantitative parameters, and the reproducibility of SUVmax is additionally better than that of nSUVmax.
Because of the non-quantitative fashion to evaluate cardiac uptake in 99mTc-PYP planar images, a strategy to reference uptake in other tissues must be conducted to determine disease stages. The most widely used gage was a visual comparison of myocardium to ribs as reported by Perugini [20]. Although Perugini visual scores can differentiate ATTR-CM from AL cardiac amyloidosis, they have not proven useful in risk stratification for individuals with proven ATTR-CM [21, 30]. The visual interpretation can be relatively too subjective to precisely categorize groups with different risks although obvious increase in regional 99mTc-PYP uptake in myocardium can be an indicator for mortality [31]. It has also been shown that H/CL ratio measured from 99mTc-PYP planar images may provide prognostic value with slightly higher cutoff (1.6) than the diagnostic criterion (1.5) [32]. Technically, the diagnostic certainty of either visual scores or semi-quantitative measurement can be degraded if intense extra-cardiac uptake exists to impact on the evaluation of both myocardial and bone uptakes [33]. It has been proposed that absolute quantitation of myocardial uptake using quantitative SPECT should help overcome these shortcomings. In our study, we demonstrated that the quantitative SPECT can be useful not only to differentiate myocardial uptake from blood-pool or bone overlay as a rescue to visual interpretation, but also to quantitatively and objectively measure the burn of amyloid deposit in myocardium as demonstrated.
Previous studies reported that nuclear scintigraphy with 99mTc-labeled phosphates tracers (e.g., 99mTc-DPD, 99mTc-PYP, 99m Tc-HMDP) can be utilized as an outstanding non-invasive imaging tool to distinguish AL cardiac amyloidosis from ATTR-CM with excellent diagnosis performance [17–19]. Nonetheless, images produced by these three tracers are not actually identical for the diagnostic purpose [34]. It has been reported that in patients with ATTR-CM, 99mTc-DPD scintigraphy can show deposit of amyloid in gluteal, shoulder, chest and abdominal wall regions beyond myocardium, and 99mTcHMDP scintigraphy can show extensive retention in lungs, whereas 99mTc-PYP normally doesn’t show extra-cardiac amyloid infiltration [35] [36] [37]. The capability of 99mTc-PYP to distinguish ATTR from AL amyloidosis is mainly based on two unique mechanisms: 1) 99mTc-PYP binds more intensively to TTR amyloid fibers which contains higher related calcium compounds and 2) 99mTc-PYP only reveals in affected tissues with a long period of amyloid accumulation. As compared to the quantitative SPECT study with other amyloid tracers (e.g., 99mTc-DPD), we recognized the reversed pattern of increased SUVmax in the group of grade = 3 vs the group of grade = 2 by Perugini visual scores [24]. This pattern may provoke the additional value in prognosis for ATTR-CM using 99mTc-PYP quantitative SPECT. Recently quantitative PET with bone scan agent, 18Fluorine-labeled sodium fluoride (18F-NaF), has not been proved to be useful as a single-photon radiopharmaceutical in evaluating ATTR-CM [38]. Quantitative PET with β-amyloid specific imaging tracers such as 18F-Florbetapir, 18F-Flutemetamol and 11C-PIB enables the quantitative scheme to evaluate cardiac ATTR amyloidosis [14–16]. However, this PET quantitative imaging tool for diagnosis and with a potential in prognosis of AL amyloidosis is not yet ready for routine clinical utilization [39]. As 99mTc-labeled phosphates tracers are widely available, the usage of quantitative SPECT as an accurate measurement technique is not only capable in diagnosis for ATTR-CM, but also the assessment for prognosis that will still need larger studies to conform this potential. Moreover, it provides a quantitative tool to monitor the disease progression for individual with ATTR mutation carriers from family history who not yet presents clinically relevant symptoms. It also enables quantitative assessment of treatment response to proven therapy as well as helpful in conducting trials of new therapeutic agents.
Study Limitations
In this study, there was only limited sample size for the ATTR-CM (n = 6) group was available. This limitation restricted to further statistically differentiate the group of Perugini visual scores = 3 from the group of Perugini visual scores = 2 by using the quantitative parameter, SUVmax although the pattern was observable as shown in our data. Future study should focus to resolve this limitation by increasing the sample size of ATTR group. Another limitation is that no prognosis data were available. Whether SUVmax or the heterogeneity of SUVmax can provide better prognosis than Perugini visual scores or HC/L ratio cannot be answered by this study. Other related technical limitation may be addressed by SPECT and CT data acquired on separate scanners. When non-translational misregistration between SPECT and CT images (e.g. rotational misregistration) may occur, the current program can’t compensate to correct for the type of error. Nonetheless, in our study, no study subject actually showed non-translational misregistration when careful patient positioning between SPECT and CT scans was carried out.