Cardiac PET perfusion tracers: current status and future directions

Abstract

PET myocardial perfusion imaging (MPI) is increasingly being used for noninvasive detection and evaluation of coronary artery disease. However, the widespread use of PET MPI has been limited by the shortcomings of the current PET perfusion tracers. The availability of these tracers is limited by the need for an onsite ((15)O water and (13)N ammonia) or nearby ((13)N ammonia) cyclotron or commitment to costly generators ((82)Rb). Owing to the short half-lives, such as 76 seconds for (82)Rb, 2.06 minutes for (15)O water, and 9.96 minutes for (13)N ammonia, their use in conjunction with treadmill exercise stress testing is either not possible ((82)Rb and (15)O water) or not practical ((13)N ammonia). Furthermore, the long positron range of (82)Rb makes image resolution suboptimal and its low myocardial extraction limits its defect resolution. In recent years, development of an (18)F-labeled PET perfusion tracer has gathered considerable interest. The longer half-life of (18)F (109 minutes) would make the tracer available as a unit dose from regional cyclotrons and allow use in conjunction with treadmill exercise testing. Furthermore, the short positron range of (18)F would result in better image resolution. Flurpiridaz F 18 is by far the most thoroughly studied in animal models and is the only (18)F-based PET MPI radiotracer currently undergoing clinical evaluation. Preclinical and clinical experience with Flurpiridaz F 18 demonstrated a high myocardial extraction fraction, high image and defect resolution, high myocardial uptake, slow myocardial clearance, and high myocardial-to-background contrast that was stable over time-important properties of an ideal PET MPI radiotracer. Preclinical data from other (18)F-labeled myocardial perfusion tracers are encouraging.

Figure 1

Positron range and end point coordinates are shown for various myocardial perfusion PET tracers. The higher the energy of the emitted positron, the longer it travels away from the source before annihilation and the worse the resolution of the imaged target. In this figure, end point coordinates are similar to point source images obtained from a given PET tracer.

Figure 2

Conceptual effect of radiotracer extraction fraction on detection of myocardial perfusion defect severity. Myocardial uptake of various PET tracers (y-axes) are shown when myocardial blood flow (x-axes) increases by a greater increment in a vascular bed supplied by a normal coronary artery versus a lower increment in the myocardial region supplied by a narrowed coronary artery. The differences in myocardial uptake of various tracers are illustrated for ¹⁵O water (Figures 2a), ¹³N ammonia (Figures 2b), ⁸²RB (Figure 2c) and ¹⁸F flurpiridaz (Figures 2d). See text for detailed explanation

Figure 3

Receiver Operating Curves (ROC) for ¹⁸F flurpiridaz (blue) and ^99mTc labeled SPECT (red) for detection of coronary artery disease in Phase 2 ¹⁸F flurpiridaz multicenter study. The area under the ROC curve for ¹⁸F flurpiridaz was significantly better than that of ^99mTc labeled SPECT (p<0.05).

Figure 4

^99mTc SPECT images (upper rows) and flurpiridaz F 18 PET images (lower rows) from a patient with normal coronary arteries. A false positive reversible inferior defect is present on the ^99mTc SPECT images due to shifting soft-tissue attenuation. The flurpiridaz F 18 PET study, however, provided superior image quality and was normal.

Figure 5

^99mTc SPECT and ¹⁸F flurpiridaz PET images in a patient with significant disease in the left circumflex coronary artery. The overall quality of the ¹⁸F flurpiridaz PET images (lower rows) was superior to the ^99mTc SPECT images (upper rows). ¹⁸F flurpiridaz PET images showed reversible anterolateral wall defects in the distribution of the diseased left circumflex coronary artery, but the ^99mTc SPECT images were normal.