PET of Breast Cancer: Detection and Staging Considerations
Dr Hossein Jadvar, MD, PhD, MPH, MBA
The diagnostic utility of fluorodeoxyglucose positron emission tomography (FDG PET) and PET-computed tomography (CT) in the imaging evaluation of women with cancer was highlighted at the 54th annual meeting of the Society of Nuclear Medicine in Washington, DC, by scientific sessions dedicated to this important topic.
A group of investigators from South Korea sought to determine whether there were preoperative FDG PET-CT findings that could predict early recurrence in women with breast cancer.[1] The mean peak standardized uptake value (SUV) of primary tumors with early recurrence was significantly higher (5.4 ± 3.0) than the mean peak SUV of those with negative recurrence (3.9 ± 2.8) during the follow-up period of 23-28 months after surgery. However, interestingly although the mean peak SUV of the axillary lymph nodes was not significantly different between the 2 groups but patients with greater number of axillary lymph nodes showing perceptibly increased FDG uptake were more likely to develop tumor recurrence.
The impact of FDG PET on the preoperative staging of newly diagnosed breast cancer was studied by the University of Pennsylvania group.[2] The mean peak SUV of the primary tumor in patients with axillary metastases was significantly higher than those primaries without associated axillary involvement or distant metastases (4.1 ± 3.5 vs 2.8 ± 2.3, P = .03).[2,3] Furthermore, PET upstaged the disease in 9.2% and detected unsuspected distant metastases in 7.5% of patients. The authors concluded that FDG PET has a role in initial staging and therapy management of newly diagnosed breast cancer. In a corollary study, the Taiwanese investigators found that FDG PET-CT can detect unsuspected metastases in one fourth of patients with locally advanced breast cancer.[4]
The Penn group also investigated the relationship between peak SUV of breast cancer to histopathologic characteristics.[5,6] The SUVmax of the invasive tumor group was significantly higher than the noninvasive group (early scan at about 1 hour after FDG administration: 3.8 ± 3.6 vs 2.4 ± 1.2, P = .02). In addition, the SUVmax was significantly higher in both estrogen receptor negative (ER-) and progesterone receptor negative (PR-) groups compared to the ER+ and PR+ independent of each other. Therefore the authors concluded that SUVmax in breast cancer is significantly influenced by histologic grade, invasiveness, and the hormonal receptor status of the primary tumor. The same group also observed that SUVmax is higher in the primary tumor that in metastases.[7] In another similar investigation from South Korea, the authors reported that FDG uptake in breast cancer is positively correlated with a number of biomarkers such as c-erb B2 (r = 0.41), p53 (r = 0.45), and Ki-67 (r = 0.42) that can partly explain the observed association of high FDG uptake in tumor with poor prognosis.[8]
The prevalence and the malignancy rate of clinically unexpected focal lesions detected by FDG PET-CT were reported by a group of South Korean researchers.[9] The prevalence of focal breast lesions in a group of 2932 women undergoing FDG PET scan and without a known history of breast cancer was 1.3% with a mean lesion size of 13.9 ± 10.1 mm. However, of these focal breast lesions, the cancer rate was 28%. There was a significant difference in SUVmax and size of malignant lesions in comparison to benign lesions (SUVmax: 5.3 ± 5.2 vs 2.5 ± 1.2; size: 21.4 ± 13.0 mm vs 13.6 ± 7.7 mm, respectively). The authors concluded that rate of focal breast lesions is not low and is associated with a modest number of malignancies, and therefore further diagnostic workup should be considered if focal breast lesions are noted on FDG PET. This is similar to the case of focal hypermetabolic thyroid lesions, which may indicate occult thyroid malignancy.
A study from a group of researchers from India evaluated the role of F-18 PET-CT in the detection of bone metastases in patients with advanced breast cancer.[10] In this investigation, 72 patients with stage III/IV breast cancer underwent imaging with F-18 PET-CT, FDG PET-CT, and Tc-99m MDP bone scintigraphy (planar and single-photon emission computed tomography). The sensitivities for detection of osseous metastases were 100%, 44%, and 81%, respectively, for these imaging procedures. The corresponding specificities were 75%, 100%, and 63%, respectively. Therefore it was concluded that a combination of F-18 PET-CT (high sensitivity) and FDG PET-CT (high specificity) scans would be optimal for the assessment of osseous metastases in breast cancer. Another PET-CT study correlated the morphology of bony metastases on CT with the level of FDG uptake on PET.[11] The level of FDG uptake in the osseous metastasizes lesions were, in descending order, osteolytic (94%), mixed pattern (82%), and osteoblastic (61%). Following treatment, 81% of the osteolytic lesions became FDG negative and blastic on CT. Of the initially osteoblastic lesions, 52% became FDG negative and increased in size on CT. The FDG avidity of the bony metastasis was more reflective of response to treatment than changes in CT morphology. A comparative study of FDG PET-CT and Tc-99m bone scintigraphy also arrived at a similar conclusion that the combined metabolic-morphologic diagnostic information contained in FDG PET-CT is more sensitive than bone scintigraphy in the assessment of osseous metastases in breast cancer.[12]
In a Spanish study, the role of FDG PET was studied in 226 patients with elevated serum tumor markers (CEA, CA 15.3) and negative conventional imaging procedures.[13] FDG PET demonstrated a sensitivity, specificity, positive predictive value, and negative predictive value of 94%, 88%, 91%, and 91%, respectively, for detection of recurrence in this clinical setting. PET also led to change in clinical management of 35% of these patients. The Chinese investigators also reported a similar conclusion that FDG PET-CT is useful in the detection of recurrent and metastatic disease in breast cancer with sensitivity, specificity, and positive and negative predictive values of 96%, 80%, 93%, and 89%, respectively.[14] FDG PET-CT was reported to be helpful in detecting unsuspected contralateral breast tumor in 5% of patients with known ipsilateral breast cancer.[15]
The potential diagnostic utility of a positron emission mammography (PEM) imaging system was also reported.[16] In this report, PEM was found to be superior to whole body PET and equal to MRI for detection of breast tumors with lesion SUVmax to background mean ratio as a potential method to discriminate between benign and malignant lesions. Validation of such a system, however, requires additional clinical experience.