Table 7 Breast Cancer

StudyPatient Number/ CharacteristicsProcedure ProtocolResultsConclusionsComments
Nieweg et al. (1993)
University of Texas
M. D. Anderson Cancer Center, Houston, TX
Prospective study
20 subjects, 11 with primary breast cancer, one with axillary involvement whose primary tumor could not be found, and eight without malignancyRegional FDG PET with blinded visual analysis and TNTR analysisPrimary tumor correctly detected by PET in 10 (91%) of 11 subjects; no false-positives.
Axillary involvement correctly identified in five (100%) of five subjects; no false-negatives or positives.
FDG PET can be used for the detection of breast cancer and lymph node metastasis.Small study sample
Adler et al. (1993)
University Hospitals of Cleveland,
Cleveland, OH
Prospective study
28 subjects with 35 newly discovered, untreated breast massesFDG PET for all and axillary scan for 19 subjects with blinded visual analysis and DUR analysis; surgical or needle aspiration biopsy for all; axillary node dissection for 19 subjectsSensitivity and specificity of PET: 96% and 100% for differentiating benign and malignant masses; 90% and 100% for determining the presence of axillary nodal involvement.
Significant correlation between DUR and histologic grade of malignancy.
FDG PET may be useful as a prognostic indicator for determining treatment strategy.Small study sample
High pretest probability of breast cancer and axillary involvement in study population
Avril et al. (1996b)
Technical University of Munich, Munich, Germany
Prospective study
51 subjects with newly discovered breast lesionsFDG PET with blinded visual analysis and SUV analysis; histologic confirmationSensitivity and specificity for differentiating benign and malignant tumors: 68% to 92% and 97% to 100%, depending upon tumor size and type of image analysis applied.FDG PET may be useful in a subgroup of subjects with indeterminate results on conventional breast imaging.Small study sample
Scheidhauer et al. (1996)
University Hospital,
Köln, Germany
Prospective study
30 subjects with suspected primary or recurrent breast cancerFDG PET with software-automated fusion of transmission and emission images and blinded visual analysisSensitivity, specificity, and accuracy of PET for detecting breast cancer: 91%, 86%, and 90%.
Accuracy of clinical exam and mammography / ultrasound for detecting breast cancer: 73% and 70%.
Accuracy of PET for detecting axillary involvement: 94%
16 of 16 (100%) known metastasis in field of view detected by PET.
FDG PET accurate for detecting tumors and metastasis.
FDG PET allows accurate tumor staging.
Small study sample
High pretest probability of breast cancer in study population
Jannson et al. (1995)
University Hospital,
Uppsala, Sweden
Prospective study
16 subjects with breast cancer, 11 with locally advanced tumors, three with recurrent disease, two with distant metastasisFDG or 11C-MET PET with SUV analysis of PET images; basal and follow-up PET; confirmation by CT, ultrasound, mammography, and clinical findings1st PET follow-up: Reduced tumor tracer uptake in 11 subjects.
2nd PET follow-up: Further reduction of tracer uptake in eight of 11 subjects with reduced tracer uptake on 1st follow-up (PET not done in other 3) and reduction of tracer uptake in two of 3 subjects who initially showed increased tracer uptake.
Conventional evaluation: Response in 12 subjects.
PET may be valuable in predicting response to treatment earlier than is possible with most conventional means.Small study sample
Avril et al. (1996a)
Technical University of Munich, Munich, Germany
Prospective study
51 subjects with suspected breast cancerFDG PET with blinded visual and SUV analysis and; pre-PET clinical examinationSensitivity and specificity in identifying axillary involvement: 58% and 85%, respectively, for clinical examination; 79% and 96%, respectively, for PET when considering all lesions; 94% and 100%, respectively, for PET when considering only lesions >2 cm.
False-negatives in lesions < 1.2 cm., though PET visualized a node of .8 cm.
Extent of disease also determined by PET in 12 (29%) of 41 patients with proven breast cancer.
FDG PET accurate in identifying axillary involvement in patients with lesions of >2 cm, spatial resolution limits detection of smaller lesions.
PET cannot substitute for histopathologic analysis but may alter therapeutic options.
Small study sample
Utech et al. (1996)
University of Illinois College of Medicine and Methodist Medical Center,
Peoria, IL
Prospective study
124 subjects with newly diagnosed breast cancerFDG whole-body PET with blinded DURSensitivity, specificity, and negative predictive value of FDG PET in identifying axillary involvement: 100% (no false negatives in 44 tumor-positive nodes), 75% (20 false-positives in 80 tumor-negative nodes), and 100%, respectively.
Weak correlation found between DUR and tumor size and between DUR and S-phase.
High sensitivity and negative predictive value of FDG PET suggest that axillary dissection may not be necessary in patients without axillary FDG uptake.Positive predictive value and accuracy not provided.
Adler et al. (1997)
University Hospitals of Cleveland,
Cleveland, OH
Prospective study
50 subjects with breast cancerFDG PET with independent blinded visual analysisSensitivity, specificity, positive predictive value, negative predictive value, and accuracy of FDG PET in identifying axillary lymph node involvement: 95%, 66%, 63%, 95%, and 77%, respectively.
False-positives due to extensive sinus histiocytosis and fat replacement in all cases.
Only one false negative result, which was due to excessive adipose tissue in an obese patient.
FDG PET may be used as a screening test for axillary lymph node metastasis.Small study sample

NOTE: TNTR, tumor-to-normal-tissue ratio; DUR, differential uptake ratio; SUV, standardized uptake value; Kpat, Patlak value.

From: Positron Emission Tomography (PET) for Oncologic Applications

Cover of Minnesota Health Technology Assessments
Minnesota Health Technology Assessments [Internet].
Minnesota Health Technology Advisory Committee.
St. Paul (MN): Minnesota Department of Health; 1995-2001.

NCBI Bookshelf. A service of the National Library of Medicine, National Institutes of Health.