Quo Vadis: PET and Single-Photon Molecular Breast Imaging.

My interest in imaging breast cancer with PET began in the late 1980s with my, and many others’, gradual recognition that the radiotracer 18F-FDG, originally designed for brain imaging, was actually a promising visceral tumor imaging agent (1–3). The early contributions of the late Ludwig Strauss to colorectal cancer imaging with PETwere quite informative at that time (4). Although at that stage of my career I was interested in applying monoclonal antibodies to tumor imaging, the target-to-background ratios achieved with 18F-FDG in a variety of animal studies of multiple tumor types were often higher and occurred sooner after radiotracer injection than when targeting with more specific monoclonal antibodies was used. With access to early-generation whole-body PET imaging in the late 1980s at the University of Michigan, the quickest approach to finding out whether 18F-FDG worked in women with breast cancer was to secure human-use approval and conduct the human studies. The answer was, basically, yes, 18F-FDG PETworks, with limitations. While we began to use 18F-FDG with PET, pioneering studies were being performed in the late 1980s in Finland by Heikki Minn et al., who introduced 18F-FDG for breast cancer imaging by a planar, non-PET, technique in 1989 (5). Using a specially collimated nontomographic g-camera, they studied 17 patients with breast cancer and were able to detect the tumor in 14 (82%), including 6 of 8 known lymph node metastases. 18F-FDG was also able to detect bone metastases and was more sensitive in detecting lytic or mixed lesions than in detecting purely sclerotic lesions. In the assessment of treatment response in 10 patients, increased 18FFDG uptake was consistently associated with disease progression, whereas decreased uptake was often, but not invariably, associated with resolving or stable disease. However, planar imaging is an insensitive technique, and images are limited by low resolution and sensitivity and are certainly not quantitative. Case studies, including ours, reported the feasibility of imaging breast cancer using PET with 18F-FDG in several patients (3,4,6). Subsequently, we systematically evaluated the feasibility of using 18F-FDG PET to image the primary tumor and regional and systemic metastases (7). The 18F-FDG PET method detected 25 of 25 known foci of breast cancer, including primary lesions (10/10), soft-tissue lesions (5/5), and bone metastases (10/10). Four additional nodal lesions that had not previously been identified were also detected. Several of the primary cancers were detected in women with radiographically dense breasts, though these tumors were relatively large (.2 cm). That point is of particular interest in view of the current issues surrounding breast cancer detection in women with radiographically dense breasts. In the early 1990s, we determined that although 18F-FDG uptake in breast cancer was multifactorial, uptake often correlated with overexpression of glucose transporter 1 in viable breast cancer cells (8). A general relationship between viable cell number and degree of 18F-FDG uptake was seen in vivo in animals and humans, suggesting that uptake might be used as a metric of treatment response. With breast cancer being located near the heart, it was possible to perform dynamic PET studies of breast cancer and of the great vessels, allowing for noninvasive dynamic, quantitative assessment of treatment-related changes in 18F-FDG uptake in cancerous and normal tissues. In these studies, uptake could be quantified (unlike what could be achieved with planar imaging) and, with noninvasive imaging of the great vessels, could be analyzed by SUV, compartmental analysis, and Patlak–Gjedde analyses. These imaging tools, combined with a growing interest in neoadjuvant chemotherapy (i.e., in situ chemotherapy of the primary tumor in order to provide treatment sooner and have a biomarker of response) as opposed to chemotherapy after mastectomy or lumpectomy, allowed us to test whether quantitative 18F-FDG PET could serve as an early biomarker of response. In brief, this prospective study showed that multiagent systemic therapy caused a rapid and significant decline in breast cancer 18F-FDG uptake, k3 kinetic rate constants, and Ki (or influx constants) for 18F-FDG as early as 8 d after treatment initiation. Further declines in uptake were apparent after 21, 42, and 63 d of treatment in patients who went on to have a complete or partial response, whereas no significant decline in uptake was seen in nonresponding patients (n 5 3) when examined at 63 d after initiation of treatment. This study also showed that metabolic changes antedated anatomic changes and that the substantial declines in tumor glucose metabolism were apparent in responding patients despite no change in tumor size (9). Follow-up studies reviewed elsewhere in this supplement have supported the general validity of this metabolic monitoring approach (10). For example, in the Translational Breast Cancer Research Consortium study of neoadjuvant therapy, the change in 18F-FDG uptake from day 0 to day 15 had an area under the curve of 0.76 in a group of more than 50 patients. This indicates good, but not perfect, accuracy in separating responders from nonresponders (11). Higher accuracies were reported more recently by Groheux et al. (12). Methods of analysis of quantitative For correspondence or reprints contact: Richard L. Wahl, Mallinckrodt Institute of Radiology, 510 S. Kingshighway Blvd., St. Louis, MO 63110. E-mail: [email protected] COPYRIGHT © 2016 by the Society of Nuclear Medicine and Molecular Imaging, Inc. DOI: 10.2967/jnumed.115.159202