Dynamic analysis of 11C-PIB PET/CT in amyloid light-chain cardiac amyloidosis

49 Objectives: Patients with cardiac amyloidosis (CA), especially those with amyloid light-chain type (AL-CA), have a poor prognosis. The diagnosis of CA is often in advanced disease stage due to the non-specific clinical symptoms. Unfortunately, the treatment options are limited by then1. Therefore, early detection and accurate diagnosis plays an important role on the AL-CA patient management. Dynamic analysis of PET has attracted wide attention in recent years. By tracking the specific binding process of drugs and target, more accurate and specific pathological manifestations can be obtained. The aim of this study is to investigate the diagnosis performance of dynamic 11C-Pittsburgh B (11C-PIB) PET/CT on AL-CA. Methods: Twenty-one patients (age, 59.4±7.9 years; M/F, 12/9) with biopsy-proven diagnoses of AL-CA were enrolled. Seven subjects (age, 55.7±8.7 years; M/F, 4/3) who had neither history nor symptoms of cardiovascular diseases were recruited as healthy control. All subjects underwent a 30-min 11C-PIB PET/CT dynamic scan (SinoUnion PoleStar m660, China) with a dose of 411.1±89.7 MBq. Static images as early, intermediate and late scan at 5, 15, 25 min after radiotracer injection were reconstructed retrospectively. The left atrium blood pool volume of interest (VOI) was manually delineated by an experienced physician on MIM workstation (MIM Software Inc, USA). The left ventricular myocardium VOI and the left ventricle blood pool VOI were auto-segmented and manually calibrated to calculate kinetic parameters K1-k4 and metabolic rate constant Ki of the two-compartment model using Carimas software (Turku PET Centre, Finland, version 2.10)2. The retention index (RI) derived from the time-activity curve (TAC) was defined as the mean standardized uptake value (SUV) of the myocardial VOI from 10 to 20 min divided by the integral of the mean SUV of the left atrium blood pool VOI from 0 to 15 min. Semiquantitative analysis was performed on the static images at different time points. The standardized uptake value ratio (SUVR) was defined as the mean SUV of the myocardial VOI divided by the mean SUV of the left atrial blood pool VOI. Statistical analysis was performed between AL-CA and control groups. P < 0.05 was considered as significant. Results: The representative 11C-PIB PET/CT images at different static periods were shown in Figure 1. Patient case demonstrated higher SUV and slower wash-out process comparing with the control case, which can be observed by 11C-PIB PET/CT images as well as the corresponding TACs in Figure 2. SUVR in patients were significantly higher than that in controls over the whole acquisition period (early SUVR: 5.64±2.56 vs. 2.40±0.63; intermediate SUVR: 4.96±2.76 vs. 1.43±0.23; late SUVR: 4.56±2.67 vs. 1.34±0.22; with P < 0.001). Dynamic analysis (Figure 3) showed that K1, k2 and RI were significant different between AL-CA group and controls (4.96±1.08 vs. 5.82±0.56, 2.36±0.33 vs. 2.69±0.18, 0.04±0.01 vs. 0.14±0.08, with P=0.023, P=0.002, P<0.0001, respectively). Receiver operating characteristic curve (ROC) of RI on differentiating AL-CA group from controls yielded the sensitivity and specificity both 100% with the cutoff value of 0.515. ROCs of SUVR from intermediate and late static images on differentiating AL-CA group from controls presented the same sensitivity (95.2%) and specificity (100%) when the cutoff values were 1.835 and 1.695 (Table 1). Initial results showed that the kinetic parameters were effective to differentiate AL-CA patients from controls. The significant difference can also be observed by RI, which indicated that RI is an important clinical indicator for the diagnosis of AL-CA. Higher sensitivity of SUVR in the intermediate and late period of dynamic imaging indicated that dynamic imaging is valuable. Conclusions: 11C-PIB PET/CT dynamic imaging may represent a promising noninvasive tool for the early diagnosis of AL-CA. Big cohort study was needed for further investigation.