TO THE EDITOR:

We read the article by Vranjesevic et al. (1) with great interest and concur with their conclusions. This article points out an important advantage of ¹⁸F-FDG PET over other techniques: the high contrast that it provides between normal and malignant tissues. This contrast is achieved by minimal uptake of ¹⁸F-FDG in the normal breast, compared with uptake in cancer tissue, which is commonly hypermetabolic. The advantageousness of ¹⁸F-FDG PET is particularly true for normal, dense breast tissue, for which early diagnosis of malignant lesions by conventional imaging modalities such as mammography is difficult.

Breast cancer is the most common cancer in women, with approximately 182,000 new cases diagnosed each year in the United States. Screening with conventional mammography along with thorough physical examination is a sensitive method for the early detection of breast cancer. However, mammography does have several limitations in clinical practice. Mammography is moderately sensitive for detecting breast lesions, but its positive predictive value, 35.8%, is low (2). Diagnosis can also be difficult in young women with dense breasts and in those who have implants or have undergone surgery or irradiation to the breast tissue (2). Other imaging modalities, such as CT and MRI, have high sensitivity but low specificity in diagnosing breast cancer. Scintimammography has been used for the detection of nonpalpable breast lesions with promising results, and a recent metaanalysis of the published data has suggested that scintimammography may be useful as an adjunct to mammography and physical examination in the diagnosis of breast cancer (3). ¹⁸F-FDG PET has been compared with scintimammography and was found superior for identifying involved axillary lymph nodes and equal to scintimammography for identifying the primary breast lesion (4). ¹⁸F-FDG PET is a well-established methodology for differentiating benign from malignant tumors, including breast tumors (5).

Increased breast density is considered an independent risk factor for developing breast cancer. Vranjesevic et al. demonstrated that despite significantly higher ¹⁸F-FDG uptake in dense breasts than in nondense breasts, breast density is unlikely to affect the accuracy of ¹⁸F-FDG PET in detecting breast cancer, as peak standardized uptake value (SUV) in dense breasts never exceeded 1.5 (1). We assessed physiologic uptake in normal breast tissue by qualitative visualization and quantitative SUVs.

Thirty-three patients with histologically proven breast carcinoma were reviewed. The median age was 52.6 y (range, 32–79 y). Twenty-one patients had grade 3 or 4 mammographic density (dense breast), and 12 patients had grade 1 or 2 mammographic density (nondense breast), according to the lexicon criteria of the American College of Radiology. All mammographic studies were obtained within 4–6 wk before the ¹⁸F-FDG PET scan. All patients fasted for a minimum of 4 h and were verified to have a normal serum glucose level (less than 150 mg/dL) before the administration of ¹⁸F-FDG (5.2 MBq [0.14 mCi]/kg) through a peripheral vein. Sequential overlapping emission scans of the neck, chest, abdomen, and pelvis were acquired on a dedicated PET scanner (Allegro; Philips Medical Systems) 60 min after the administration of ¹⁸F-FDG.

SUVs for ¹⁸F-FDG were calculated for the contralateral normal breast in 24 patients with unilateral lesions and for normal breast tissue remote from the malignant lesion in 9 patients with bilateral lesions. For this purpose, 12 × 12 mm (9 pixels) circular regions of interest were placed over the area of highest ¹⁸F-FDG activity in the normal breast tissue. Peak and average SUVs were obtained from the normal areas of the 2 groups and were compared for statistically significant differences using the Student t test.

The peak and the average SUVs were 0.99 ± 0.20 and 0.82 ± 0.19, respectively, for the normal tissue of dense breasts and 0.82 ± 0.23 and 0.66 ± 0.24, respectively, for the nondense breasts. Both the peak and the average SUVs of normal dense breasts were significantly higher than those of normal nondense breasts (P = 0.003 and 0.003, respectively). The highest peak SUVs were 1.3 and 1.2 for the normal dense breast and normal nondense breast, respectively. However, the peak SUVs for both dense and nondense breast tissues were well below 2.5, a widely used cutoff value for malignancy.

Although our SUVs were slightly higher than those of Vranjesevic et al., our results showed a significant difference between the dense and nondense normal breast tissues. Similar to their data, our peak SUVs were well below the cutoff values for malignancy. These data further confirm that although dense breasts may have a higher uptake, the accuracy of ¹⁸F-FDG PET studies in diagnosing malignant breast tumors will not significantly be affected by the varying uptake of ¹⁸F-FDG in surrounding normal tissues.