Diagnostic Applications of Nuclear Medicine: Breast Cancer
This chapter details the use of radiotracer methods in application to breast cancer. We provide a background on breast cancer incidence and clinical features, followed by an overview of current treatment approaches. We then review the use of radionuclide methods applied to major diagnostic tasks for breast cancer that include detection, staging (locoregional and distant), response to therapy, and surveillance.
KeywordsBreast cancer Diagnosis Staging Restaging Nuclear medicine [18F]FDG PET/CT Prognosis
16 Alpha-18F-fluoro-17 beta-estradiol
- 99mTc-HDP or 99mTc-HMDP
American College of Obstetricians and Gynecologists
American College of Radiology
American Cancer Society
American Joint Committee on Cancer
Breast and Ovarian Analysis of Disease Incidence and Carrier Estimation Algorithm
Breast cancer type 1 susceptibility protein
Breast cancer type 2 susceptibility protein
BRCA mutation carrier prediction model
Breast-specific gamma imaging, also known as molecular breast imaging
Checkpoint kinase 2
X-ray computed tomography
Ductal carcinoma in situ
Gene for the HER2 receptor
Fluorescent in situ hybridization
Human epidermal growth factor receptor 2, also known as receptor tyrosine-protein kinase erbB-2, or HER2/neu
Locally advanced breast cancer
Metastatic breast cancer
Menopausal hormone therapy
Magnetic resonance imaging
Mammalian target of rapamycin
National Comprehensive Cancer Network
Negative predictive value
Tumor protein p53, also known as cellular tumor antigen p53, phosphoprotein p53, tumor suppressor p53, antigen NY-CO-13, or transformation-related protein 53 (TRP53)
Partner and localizer of BRCA2, also known as FANCN
Pathologic complete response
Positron emission mammography
Positron emission tomography
Positron emission tomography/computed tomography
Phosphatase and tensin homolog
Response evaluation criteria in solid tumors
Society of Breast Imaging
Selective estrogen receptor modulator
Sentinel lymph node
Single nucleotide polymorphism
Single-photon emission computed tomography
Single-photon emission computed tomography/computed tomography
Standardized uptake value
Triple-negative breast cancer
AJCC staging system based on parameters “T” (tumor status), “N” (lymph node status) and “M” (distant metastasis status)
Epidemiology of the Tumor
Breast cancer is the most common non-skin cancer and the second leading cause of cancer mortality in women [1, 2]. Breast cancer is largely a disease of women (one in eight lifetime risk) with an estimated 231,840 new cases expected to be diagnosed in the USA in 2015 . The incidence of breast cancer in men is ~100 times less than in women, with approximately 2,350 new cases expected in 2015. A sharp 7% decline in incidence of breast cancer occurred among white women between 2002 and 2003 attributable to a decline in the use of menopausal hormone therapy (MHT) after a 2002 report linked the use of combined estrogen plus progestin MHT with an increased risk of breast cancer and coronary heart disease . More recently (data from 2007 to 2011), the incidence rates have been stable in white women and increased slightly to 0.3% in black women .
There has been a steady decrease in mortality from breast cancer since 1989, attributed to both improvements in early detection, screening programs initiated in the USA in the early 1980s, and treatment [4, 5]. Despite several randomized control trials demonstrating the mortality benefit of screening, there remains controversy on screening guidelines for average-risk women. New 2015 recommendations from the American Cancer Society (ACS) now suggest yearly screening beginning from 45 through 54 and biennial screening from age 55 to age 74 or expected lifetime of less than 10 years . Other organizations including the Society of Breast Imaging (SBI), American College of Obstetricians and Gynecologists (ACOG), and American College of Radiology (ACR) continue to recommend annual screening beginning at age 40. Just over 40,000 women and 400 men are expected to succumb to the disease in 2015. Mortality rates from the most recent data have slightly favored younger more than older women and white more than black women in recent times .
Risk factors for development of breast cancer are due to one or a combination of genetic/familial, reproductive/hormonal, lifestyle, and environmental factors. Further, some of these factors may be classified as modifiable/external or non-modifiable/internal. The most well-established risk factors for female breast cancer include age, family history and genetics, late first pregnancy, and obesity . Besides obesity, other modifiable factors that contribute to breast cancer risk include physical inactivity, use of MHT, long-term heavy smoking, and alcohol consumption [8, 9]. Other well-established non-modifiable risk factors include long menstrual history (early menarche, late menopause), mantle radiation for Hodgkin’s lymphoma, and inherited gene mutations (e.g., BRCA1/BRCA 2). Factors shown to decrease the risk of breast cancer include breastfeeding for at least a year, increased physical activity, and, in high-risk women, decreasing endogenous estrogens by either inhibition of the estrogen receptor using selective estrogen receptor modulators (SERMs) or blocking estrogen synthesis using aromatase inhibitors (AIs). It is estimated that in some populations of average-risk women, lifestyle risk reduction measures may contribute to up to a 30% decrease in overall risk and chemopreventive measures using SERMS and AIs may reduce breast cancer risk by up to 50% [9, 10, 11, 12].
Age is an important risk factor. The highest risk of developing invasive breast cancer by age group occurs in women 70 and over (1 in 15), while the risk in women 39 years and younger is only 1 in 208. The majority of women who develop breast cancer are postmenopausal; however, breast cancer is not uncommon in premenopausal women. Breast cancer in younger women tends to be more aggressive than breast cancer diagnosed for older women and is an important cause of mortality in younger patients. Breast cancer is the leading cause of cancer death in women age 20–59, accounting for over 12,000 deaths annually .
Approximately 30% of breast cancer cases are familial and associated with mutations in breast cancer susceptibility genes. Of these, up to 30% are due to high-penetrance mutations in the BRCA1, BRCA2, PTEN, and p53 genes, and these patients face substantially increased lifetime risk of breast cancer, between 40% and 85% over the average-risk women (~12%). Mutations in genes related to DNA repair, such as PALB2 and CHEK2, have also been shown to be associated with moderate risk [13, 14, 15]. Several risk assessment models have been developed to estimate the breast cancer risk or probability of carrying a susceptibility genetic mutation, most notably the Gail, Claus, BRCAPRO, Breast and Ovarian Analysis of Disease Incidence and Carrier Estimation Algorithm (BODAICEA), and Tyrer-Cuzick models [16, 17, 18, 19, 20]. These models take into account various sets of risk factors to estimate an individual’s risk, including genetic mutations, family history, history of chest irradiation and personal history of ductal carcinoma in situ, ovarian cancer, lobular neoplasia, or atypical ductal hyperplasia, which are used to inform provider and patient decision-making regarding an appropriate screening timetable. The ACR/SBI recommends annual screening with mammography and breast MRI for women with a lifetime risk of greater than 20% for developing breast cancer beginning by age 30, for known high-penetrance gene mutations, 10 years earlier than the youngest affected relative or 8 years after chest radiation .
There are other common genetic variants, single nucleotide polymorphisms (SNPs), that confer a much smaller increased risk, but may in combination account for a more substantially increased risk of development of breast cancer in the remainder of familial breast cancer patients who do not have an identifiable high-risk or moderate-risk genetic mutation. Currently, there are more than 90 risk loci identified through genome-wide-associated studies, a technique used to identify genetic variations associated with breast cancer, with estimated hundreds more yet to be discovered that might explain significant proportion of familial and sporadic breast cancer risk [22, 23].
Underlying Breast Cancer Biology Changes
Breast cancers arise from the epithelial cells in the breast cancer ducts . The most important histologic feature in the primary tumor is the presence or absence of tumor invasion beyond the ductal architecture. Tumors that have not broken through the outer boundaries of ductal structures are termed noninvasive (ductal carcinoma in situ, DCIS) and rarely metastasize . Invasive cancers are those that have broken through tissue boundaries at the time of diagnosis and have the potential to metastasize both through lymphatic and hematologic spread. Other histologic features of note include the architectural pattern (ductal, lobular, or mixed), nuclear grade (low, intermediate, or high), and the presence or absence of lymphovascular invasion . In addition to morphological analyses, biopsy material is routinely assayed for phenotypic features that include the expression of hormone receptors (estrogen receptor [ER] and progesterone receptor [PR]) and HER2/neu (also known as cerbB2), typically by immunohistochemistry or fluorescent in situ hybridization (FISH).
Four major subtypes of invasive breast cance r used clinically (Adapted from )
Standard immunochemical results and cancer grade
ER+ PR+ HER+, usually low grade
Best prognosis, low Ki-67 level
ER+ PR+ HER−, usually intermediate to high grade
Generally more proliferative (high Ki-67 levels) with less marked hormonal receptor expression than luminal A tumors, approximately 30% are HER2 positive
ER− PR− HER+, usually midgrade to high grade
Prognosis much improved with directed therapy (e.g., trastuzumab); 30–40% of tumors also express ERs and PRs
ER− PR− HER−, high grade
Often synonymous with triple negative
Measuring levels of certain cellular biomarkers within tumors, e.g., Ki-67, a marker of cellular proliferation, can clarify the molecular portrait of a particular tumor and also aid in prediction of tumor aggressiveness and behavior. Molecules and receptors may also serve as targets for specific therapy, e.g., the HER2 receptor is targeted with the drug trastuzumab . Some predictive expression panels are now approved and used in clinical practice . For example, the 21-gene assay used in the Oncotype DX (Genomic Health Inc.) panel is now routinely used to determine which patients with early-stage ER-positive breast cancer may benefit from adjuvant chemotherapy in addition to endocrine therapy [27, 35, 36].
Staging Categories and Prognostic Stratification
T0 – no primary tumor found
Tis – carcinoma in situ
T1 – invasive cancer <2 cm in greatest dimension
T1mic – microinvasion ≦0.1 cm in greatest dimension
T1a – tumor >0.1 cm but ≦0.5 cm in greatest dimension
T1b – tumor >0.5 cm but ≦1 cm in greatest dimension
T1c – tumor >1 cm but ≦2 cm in greatest dimension
T2 – tumor >2 cm but ≦5 cm in greatest dimension
T3 – tumor >5 cm in greatest dimension
T4 – tumor of any size with direct extension to chest wall or skin
Regional lymph nodes (N) (clinical staging)a
N0 – no regional nodal metastases
N1 – metastasis in mobile ipsilateral axillary nodes
N2 – metastasis to fixed ipsilateral axillary nodes (N2a) or ipsilateral internal mammary nodes in the absence of ipsilateral axillary lymph node metastasis (N2b)
N3 – metastases in ipsilateral infraclavicular nodes and axillary nodes (N3a) or ipsilateral internal mammary nodes and axillary nodes (N3b) or metastases to ipsilateral supraclavicular nodes (N3c)
Regional lymph nodes (N) (pathologic staging)b
pNX – regional lymph nodes cannot be assessed (e.g., previously removed or not removed for pathologic study)
pN0 – no regional lymph node metastasis histologically, no additional examination for isolated tumor cells
pN0(i−) – no regional lymph node metastasis histologically, negative immunohistochemical staining
pN0(i+) – isolated tumor cells identified histologically or by positive immunohistochemical staining, no cluster >0.2 mm
pN0(mol−) – no regional lymph node metastasis histologically, negative molecular findings
pN0(mol+) – no regional lymph node metastasis histologically, positive molecular findings
pN1 – metastasis in one to three axillary lymph nodes and/or in internal mammary nodes with microscopic disease detected by sentinel lymph node dissection but not clinically apparenta
pN1mi – micrometastasis (>0.2 mm, none >2.0 mm)
pN1a – metastasis in one to three axillary lymph nodes
pN1b – metastasis in internal mammary nodes with microscopic disease detected by sentinel lymph node dissection but not clinically apparenta
pN1c – metastasis in one to three axillary lymph nodes and in internal mammary lymph nodesc with microscopic disease detected by sentinel lymph node dissection but not clinically apparenta
pN2 – metastasis in four to nine axillary lymph nodes or in clinically apparenta internal mammary lymph nodes in the absence of axillary lymph node metastasis
pN2a – metastasis in four to nine axillary lymph nodes (at least one tumor deposit >2.0 mm)
pN2b – metastasis in clinically apparenta internal mammary lymph nodes in the absence of axillary lymph node metastasis
pN3 – metastasis in ten or more axillary lymph nodes or in infraclavicular lymph nodes or in clinically apparenta ipsilateral internal mammary lymph nodes in the presence of one or more positive axillary lymph nodes or in more than three axillary lymph nodes with clinically negative microscopic metastasis in internal mammary lymph nodes or in ipsilateral supraclavicular lymph nodes
pN3a – metastasis in ten or more axillary lymph nodes (at least one tumor deposit >2.0 mm) or metastasis to the infraclavicular lymph nodes
pN3b – metastasis in clinically apparenta ipsilateral internal mammary lymph nodes in the presence of one or more positive axillary lymph nodes or in more than three axillary lymph nodes and in internal mammary lymph nodes with microscopic disease detected by sentinel lymph node dissection but not clinically apparenta
pN3c – metastasis in ipsilateral supraclavicular lymph nodes
Distant metastases (M)
MX – distant metastases cannot be assessed
M0 – no distant metastases
M1 – distant metastases
Alternate staging system for invasive breast cancer
Any T, any N, M1
Any T, N3M0
Breast Cancer Detection and Diagnosis, Modalities Other Than Nuclear Medicine for the Tumor
Mammography is the primary imaging modality for breast cancer screening, detection, and diagnosis . The average earlier stage and smaller size of lesions detected by mammographic screening, versus symptomatic and/or palpable lesions, reflects the success of mammographic screening, which has been shown to reduce breast cancer mortality . Full-field digital mammography has largely replaced screen-film mammography. And the recent addition of digital breast tomosynthesis has been shown to reduce false-positive rates and increase cancer detection [41, 42]. Both ultrasound and MRI are important adjuncts to x-ray mammography for detection, diagnosis, characterization, and determination of the extent of breast cancer and are routinely utilized in these roles [43, 44].
MRI has been shown to be useful for screening in high-risk patients, i.e., women with greater than 20% lifetime risk of breast cancer. In a meta-analysis of 14 studies of high-risk women conducted between 2000 and 2011, MRI was shown to have higher sensitivity for malignancy (84.6%) than mammography (38.6%) or ultrasound (39.6%) .
The American Cancer Society recommendations have incorporated MRI for screening in high-risk patients . Mammography, ultrasound, and breast MRI can all be used to direct tissue sampling by needle biopsy for breast cancer diagnosis.
Randomized control trials have shown no survival or quality of life benefit when annual mammography has been compared with intensive imaging or laboratory surveillance in asymptomatic women with early-stage breast cancer .
With regard to staging and metastatic workup for asymptomatic women with a history of early-stage breast cancer, stage I, IIA, and IIB, no further imaging is indicated beyond diagnostic mammography (with adjunct use of US and MRI) given the low chance of distant metastases and the higher chance of false-positive findings . For later stage breast cancer patients, stage III and above current guidelines recommend chest CT, abdomen +/− pelvis CT or MR, bone imaging (bone scintigraphy (BS) or Na18F PET/CT), and optional [18F]FDG PET/CT. Bone scan and [18F]FDG PET/CT are the most common nuclear imaging methods used in staging and metastatic workup. These will be discussed later. Again, anatomic imaging modalities are considered the gold standard with chest and abdominopelvic CT, breast MR, and brain MR performed as directed by physical examination and clinical assessment.
Nuclear Imaging for Primary Detection and Diagnosis
Radiotracer methods for primary breast cancer detection and diagnosis have not as yet found widespread use despite some important advantages over anatomic imaging, namely, the potential for more specific targeting of breast cancer tissue, good tumor to background contrast, and uptake independent of breast density [49, 50, 51, 52, 53]. 99mTc-sestamibi is the most common single-photon radiopharmaceutical used in both scintimammography, imaging of the breast using traditional nuclear medicine gamma cameras, and breast-specific gamma imaging (BSGI, also called molecular breast imaging ), imaging with dedicated breast small field-of-view gamma cameras. The radiopharmaceutical is taken up and retained differentially in breast tumors from normal breast tissue. The specific mechanism is unknown, but appears to depend on factors relating to regional blood flow, plasma and mitochondrial membrane potential, angiogenesis, and tissue metabolism [54, 55, 56]. Scintimammography, however, has a limited ability to detect lesions less than 1 cm and also has lower sensitivity for non-palpable lesions. BSGI can overcome some of these limitations using dedicated devices that provide high spatial resolution using dedicated breast imaging devices [49, 51, 57]. The disadvantage of both techniques is that they require relatively long imaging times and subject the patient to a substantially increased radiation dose (15–30 times) compared with mammography [57, 58]. At present, they are not indicated for routine screening/primary detection.
PET has also been tested in application to breast cancer detection and diagnosis, spurred by the finding of increased 18F-fluorodeoxyglucose ([18F]FDG) uptake in cancer compared to normal breast tissue. [18F]FDG uptake in tumors reflects the rate of glucose metabolism and correlates somewhat with histologic type (higher uptake in ductal vs. lobular), tumor histologic grade, and indices of cellular proliferation (higher uptake with higher levels of proliferation) [59, 60]. Overall, the sensitivity of whole-body [18F]FDG PET in the detection of primary breast cancer ranges 64–96%, specificity of 73–100%, positive predictive value of 81–100%, and negative predictive value of 52–89% . Whole-body [18F]FDG PET however has low sensitivity for small, well-differentiated, and in situ cancers . As with 99mTc-sestamibi scintimammography, this has severely limited its use for primary tumor diagnosis.
Dedicated breast positron emission mammography (PEM) units have been developed , and advantages include higher spatial resolution, shorter imaging time, and reduced attenuation compared with whole-body imaging. PEM units allow a co-registration of anatomic and molecular information similar to PET/CT and have biopsy capability . In a large trial of patients with newly diagnosed early breast cancer, PEM showed improvement in specificity over MRI, and the performance of PEM plus conventional imaging is similar to MRI with regard to cancer detection .
However, PEM limitations include difficulty in imaging posterior lesions, variable [18F]FDG uptake in small tumors, and false-positive findings from prior biopsy [66, 67]. In addition, similar to single-photon techniques, radiation doses are significantly higher than mammography .
Nuclear Imaging for Regional Nodal Staging
Staging is typically divided into regional lymph nodes, especially axillary nodes, and distant or systemic staging of sites beyond regional lymph nodes, especially to visceral organs (lung and liver) and bone. Almost all patients with invasive breast cancer will undergo axillary nodal staging, since the presence or absence of axillary lymph nodes is an important consideration for further treatment after surgery .
The traditional approach to axillary nodal staging was the axillary nodal dissection, where axillary contents were removed and examined; however, the morbidity of this process, including a significant incidence of lymphedema, led to the search for a less morbid but equally accurate approach . This consideration motivated sentinel lymph node (SLN) methods , designed to identify and sample the node(s) that are the first draining nodes, i.e., “sentinels” in the axillary drainage basin for the breast lesion. SLN biopsy is a shorter surgical procedure than full axillary nodal dissection with much less morbidity . If sentinel nodes are negative, then the chance of “downstream” nodal metastases is quite small, and the patient can be spared the full dissection. If the sentinel node shows evidence of metastasis, then an axillary dissection may be needed to determine the extent of axillary spread. SLN biopsy has been shown to be highly accurate compared to axillary dissection [69, 70, 71] and has emerged as the standard of care for axillary nodal sampling in patients with clinically negative axillae . SLN biopsy has been thoroughly tested in large clinical trials and shown to be safe and accurate, with comparable relapse and survival compared to axillary nodal dissection [73, 74]. Further, a large randomized clinical trial showed that completion axillary dissection is not useful for improved locoregional control or survival in women with smaller (T1/T2) tumors, no palpable adenopathy, and one or two positive nodes positive by SLNB without gross ENE .
The importance of sampling internal mammary (IM) nodes has been a point of considerable debate . Studies using SLN mapping and lymphoscintigraphy to sample IM lymph have demonstrated the feasibility of this approach [79, 80, 81]; however, the overall percentage of positive IM nodes is on the order of 5% or less. And although a significant proportion of breast cancers have primary drainage to the IM nodes, approximately 30% of medial tumors and 15% of lateral tumors, it is rare to have IM lymph node metastasis in the absence of axillary nodal metastases . As a result, most practices in the USA do not routinely sample IM nodes. Yao et al.  found that patients with positive axillary nodes and IM nodal drainage, even in the absence of IM nodal sampling, were significantly more likely to relapse and die of breast cancer than those without drainage. Another study found a higher rate of nodal relapse outside the axilla, versus in the axilla, after SLN biopsy . Such studies prompt careful consideration of the approach to SLN mapping in breast cancer and the potential benefit of peri-lesional injection, which have a higher likelihood of drainage demonstrating IM nodal drainage, and lymphoscintigraphy in selected patients to identify extra-axillary drainage .
There was considerable interest in [18F]FDG PET as a noninvasive imaging method to detect axillary nodal spread in newly diagnosed breast cancer and avoid the need for invasive biopsy. However, although early studies showed high sensitivity of increased uptake in axillary nodes in breast cancer, multiple subsequent studies including larger multicenter trials have failed to show sufficient sensitivity of PET in the staging of the axilla in patients with early-stage breast cancer [86, 87] to obviate the need for tissue sampling with SLN mapping and biopsy [88, 89, 90]. Early-stage breast cancer patients, i.e., those with small tumors less than 2–3 cm with clinically negative axillary nodes, represent the majority of breast cancer patients, many clinically silent and diagnosed by screening mammography. Given the low probability of metastatic disease in this population and the risk of false positives with whole-body PET/CT, there is no recommended role for [18F]FDG PET in axillary or clinical staging in patients with early-stage disease [21, 91].
Nuclear Methods for Staging of Distant Metastases
Systemic evaluation for metastatic disease is recommended in the initial workup for high-risk and later clinical stage (≥stage III) breast cancer as well as for patients with recurrent breast cancer and known metastatic disease . High-risk disease is commonly defined as locally advanced breast cancer (LABC) , including primary tumor larger than 5 cm, skin or chest wall involvement, fixed axillary nodes, positive supraclavicular/infraclavicular/IM nodes, and inflammatory cancer. Patients with LABC have a poor prognosis because of the high incidence of distant metastases during follow-up [92, 93]. In these patients detection of distant metastases is crucial to determine the initial goals of therapy, i.e., curative versus palliative and/or local control, and thus systemic staging is recommended as part of initial evaluation . The bone scan and PET/CT play significant roles in systemic staging in breast cancer patients. Important areas of more distant spread for breast cancer include nodal beds outside the axillary and IM chain, for example, in the mediastinum, the lungs, the liver, and the bones .
The reported sensitivity ranges from 62% to 100% and specificity ranges from 78% to 100% [100, 101]. False positives can be caused by trauma, degenerative changes, and other benign conditions, while false negatives can occur in the presence of predominantly osteolytic metastases, low bone turnover, or in avascular regions (such as necrosis) . Single-photon emission computed tomography (SPECT) has been shown to improve sensitivity and specificity [102, 103, 104], with further improvement in diagnostic confidence shown with SPECT/CT [105, 106, 107, 108], likely due to the reduction of false positives through the morphological appearance CT. SPECT is particularly helpful to characterize solitary lesions in the spine and/or to differentiate bone metastases from degenerative lesions. Salvelli et al. reported the sensitivity and specificity of SPECT to be over 90% .
Besides bony metastatic disease, [18F]FDG PET has proven useful in the general detection of distant metastases for initial staging and restaging of breast cancer patients [95, 137, 138, 139, 140, 141]. A 48-patient pilot study by van der Hoeven et al. suggested that PET was more sensitive than conventional imaging in detecting distant metastases . Recently, a prospective cohort study by Groheux et al. compared conventional imaging staging (bone scan, chest radiography or dedicated CT, liver ultrasound, enhanced CT abdomen and pelvis) with a single session of staging with [18F]FDG PET  in 117 patients with LABC. The study found that not only did PET change the clinical stage in 61 patients (52%), PET identified more distant metastases compared to the conventional imaging (43 vs. 28 patients). In addition, PET/CT was found to outperform conventional imaging in detecting bone, distant lymph node, and liver metastases. Diagnostic chest CT outperformed PET/CT, owing to respiratory motion and partial volume effect reducing the sensitivity of PET for sub-centimeter lung nodules, and relatively more difficult visualization of sub-centimeter lung nodules on the lower dose acquisition CT obtained as part of the PET/CT.
There is ongoing study of the utility of PET/CT in staging intermediate risk breast cancer patients, i.e., patients with large breast cancers (T2) or clinical N1 disease [142, 143]. In a recent prospective study of 254 patients with clinical stage IIA through IIIC (as determined by clinical examination, mammography, breast MRI, and locoregional ultrasound), Groheux et al. found that imaging with [18F]FDG PET/CT revealed distant metastatic disease in 2.3% of patients with clinical stage IIA disease (1/44), 10.7% of patients with stage IIB disease (6/56), 17.5% of patients with stage IIIA disease (11/63), 36.5% of patients with stage IIIB disease (27/74), and 47.1% of patients with stage IIIC disease (8/17) . As expected the yield of metastatic disease increases with increasing clinical stage. Further studies are needed to determine clinical impact and cost-effectiveness of staging evaluation with PET/CT of patients in the intermediate risk category.
Approaches to Breast Cancer Therapy
With exception of very early- and late-stage disease, almost all breast cancer care involves a combination of locoregional and systemic treatments. Locoregional therapy includes surgery and radiotherapy, and systemic therapy includes chemotherapy (e.g., taxanes), endocrine (hormonal, e.g., aromatase inhibitors) therapy, and/or targeted (biologic, e.g., anti-HER2 drugs) therapy.
Primary tumor excision remains a key component of the treatment of newly diagnosed breast cancer . This includes mastectomy, removal of the entire breast, and breast-conserving surgery (BCS) , also known as “lumpectomy” of the cancer . Lumpectomy followed by adjuvant radiotherapy provides low rates of local recurrence and survival similar to mastectomy, but with preservation of the breast . For invasive disease, axillary nodal staging is routinely performed at the time of surgery. The purpose of axillary sampling is largely for diagnosis; the therapeutic value of lymph node removal remains largely unproven . For this purpose, sentinel lymph node has emerged as the preferred method for patients with clinically negative axillae , sparing the patient most of the morbidity, principally lymphedema, of full axillary dissection .
Breast cancer is one of the solid tumors that is often responsive to systemic therapy. Systemic therapy can be cytotoxic chemotherapy and can also involve more targeted and selective treatments. In all but the earliest cancers, adjuvant systemic therapy is used in addition to locoregional therapy to treat residual microscopic disease  and has contributed significantly to reduced breast cancer mortality. Systemic cytotoxic chemotherapy is used in more advanced or aggressive tumors, typically ER- or PR-negative tumors , triple-negative tumors, HER2-positive tumors, and large tumors [41, 48].
In ER-expressing tumors, adjuvant endocrine therapy is used, typically an aromatase inhibitor or tamoxifen . In higher-risk ER-expressing tumors, both chemotherapy and endocrine therapy can be used . RNA-based genomic testing can be used to aid in estimating the risk of distant recurrence and identify patients in this group who may benefit from chemotherapy . Both types of treatment are also primary therapy for metastatic disease . For HER2-positive breast cancer, trastuzumab, a HER2-specific monoclonal antibody has been shown to substantially improve survival . Directed therapy has become a key component of both adjuvant and metastatic systemic therapy for patients with HER2-overexpressing tumors [152, 153]. Pertuzumab, a monoclonal antibody directed against a different epitope than trastuzumab on the HER2 receptor, and lapatinib, a tyrosine kinase inhibitor targeting the HER2 pathway, are both routinely used in therapy regimes for higher-risk HER2-positive disease .
New breast cancer targets and therapy combinations continue to be investigated [154, 155]. Improvement in progression-free survival in women with hormone receptor-positive metastatic breast cancer has been demonstrated with the addition of the mTOR inhibitor everolimus  and more recently with the addition of the drug palbociclib, an inhibitor of cyclin-dependent kinases 4 and 6 .
In ER-negative tumors, chemotherapy is typically an important component of systemic therapy. This is typically combined with HER2-directed therapy in HER2-overexpressing tumors. On the other hand, tumors that are ER/PR/HER2 negative (triple negative, TNBC) are treated primarily using chemotherapy. The ongoing study is particularly important in triple-negative patients where therapies effective in other breast cancer subtypes (endocrine and HER2-targeted therapies) are unlikely to work [158, 159].
For patients with LABC, or those with large tumors who have smaller breasts and/or desire breast conservation, neoadjuvant therapy is considered. The primary purpose of this therapy is to reduce the size to the tumor to better facilitate BCS and local control. However, response to neoadjuvant therapy also provides valuable assessment of the tumor’s response to therapy prior to definitive treatment.
Radiotherapy is commonly used as an adjunct to surgery in locoregional breast cancer treatment . Lumpectomy in the absence of breast radiotherapy leads to unacceptably high rates of local recurrence ; therefore breast radiotherapy is an integral part of breast conservation treatment. In addition to radiation of the breast in patients undergoing breast conservation, adjuvant irradiation can also be used for the chest wall in mastectomy patients, but it is usually reserved for patients with larger tumors or documented chest wall invasion . Adjuvant radiotherapy is also used in the treatment of involved nodal basins and has been shown to enhance survival in those patients with multiple axillary nodal metastases .
Assessing the Efficacy of Treatment(s)
Evaluating the efficacy of systemic treatment is an important diagnostic need for breast cancer. The largest population of patients undergoing systemic therapy for breast cancer receives treatment as adjuvant therapy, after completion of primary surgery. Since there should be no residual macroscopic disease after definitive surgery, imaging is not helpful for response monitoring for adjuvant therapy, except for re-staging upon suspicion of recurrence. Imaging however plays an important role in response evaluation of systemic neoadjuvant therapy (NAT), given prior to definitive surgery primarily in LABC, as well as systemic therapy given as the primary treatment for metastatic, stage IV disease.
In the neoadjuvant setting, studies have demonstrated that the extent of residual breast and axillary disease after treatment is prognostic for both disease-free survival (DFS) and overall survival (OS) [162, 163, 164]. Patients demonstrating complete pathologic response (pCR), defined as no residual invasive tumor at post-therapy surgery histopathology, have improved long-term outcome compared to patients without pCR [162, 164]. NAT offers both the opportunity to assess tumor responsiveness while improving surgical options for patients with locally advanced disease by down-staging the primary tumor.
Sized-based approaches such as physical exam and mammography that rely on changes in size for response evaluation (e.g., RECIST criteria ) do not reflect early changes at the molecular level that may occur prior to changes in tumor size. Functional imaging therefore may be particularly helpful in determining treatment efficacy earlier in the course, allowing for earlier change of therapeutic strategy, thus avoiding unnecessary toxic systemic side effects of ineffective treatments.
One of the earliest studies showing the use of [18F]FDG PET in the assessment of treatment response was performed by Wahl and colleagues in 1993 . The investigators observed significant quantitative differences in the [18F]FDG uptake measured before and after 2 months of therapy (mid-therapy) for responders versus nonresponders. While nonresponders had no significant change, patients who went on to a good response had declines in [18F]FDG SUV of 50% or more. Most subsequent studies evaluating response at mid-therapy found similar quantitative findings (reviewed in [63, 167, 168]. A meta-analysis of 19 studies including over 900 patients by Wang et al.  underscored that imaging earlier in the course of treatment (after the first or second cycle) with cutoffs between 55% and 65% declines from baseline as response criteria offered better correlation with pathologic response of primary tumors.
The knowledge that breast cancer is a molecularly heterogeneous disease has prompted response evaluation studies within the various molecular subsets of breast cancer . Groheux et al.  and Humbert et al. , in prospective studies including women with HER2-positive breast cancer treated with trastuzumab-containing regimens showed that low residual metabolism post-NAT (SUV <=3.0 after two cycles and <2.1 after one cycle, respectively) strongly correlated with pCR. Similar studies in patients with ER-positive/HER2-negative tumors and triple-negative tumors have also been conducted [173, 174, 175]. ER-positive/HER2-negative tumor type is somewhat less [18F]FDG avid at baseline and less sensitive to chemotherapy than more aggressive subtypes (e.g., TNBC) , and pCR is rarely achieved. Studies have found that for this subtype baseline SUV and changes after one or two cycles of NAT may offer prognostic value in the form of survival rather than pCR [174, 175]. For triple-negative tumors, change in SUV from baseline has been shown to correlate with pCR, which is more closely associated with survival . These studies indicate that assessment of serial PET data to infer breast cancer response to therapy should be considered in light of breast cancer subtype, supported by studies showing that [18F]FDG kinetics may vary by breast cancer subtype .
Although some researchers have shown advantages of obtaining specific kinetic parameters of the tumor over the static SUV measurements , measuring percentage change in SUV from baseline and measuring absolute SUVs at baseline and follow-up time points are the most commonly used parameters in response assessment studies, likely due to ease of calculation and reproducibility. These studies highlight the need for standardization of the timing of interim [18F]FDG PET evaluation, cutoff points, and criteria used in assessing response to therapy for different subtypes of breast cancer [177, 178]. Further, study outcomes may also be influenced by the specific treatment regimes.
Studies of [18F]FDG PET performed after the completion of chemotherapy have shown reduced sensitivity for residual disease – i.e., that the absence of [18F]FDG uptake is not a reliable indicator of complete pathologic response [167, 179, 180, 181]. This is especially true for axillary nodal disease since the sensitivity for residual microscopic disease post-therapy is low. However, recent studies have shown that the presence of [18F]FDG uptake post-therapy is specific and highly predictive of relapse . Therefore, even though [18F]FDG PET may miss small-volume disease post-therapy, the presence or absence of [18F]FDG uptake carries prognostic significance that may be important in directing follow-up.
A particularly difficult clinical problem is the evaluation of response for bone metastases . Given that many bone lesions do not demonstrate a measurable soft tissue component, anatomic imaging using RECIST 1.1 is limited particularly with respect to evaluation of osseous metastases. Just as with detection, PET/CT has shown superiority to bone scan in assessment of treatment efficacy with regard to osseous metastases. Though excellent for detection, BS is problematic for response evaluation as changes will significantly lag behind the bone metastasis response and may even transiently worsen or “flare” in response to successful therapy [188, 189]. This flare phenomenon has also been shown with Na18F PET/CT given the similarity in mechanism [190, 191] and thus has implications for the timing of Na18F PET in response evaluations. In addition, given that [18F]FDG PET is not as sensitive for sclerotic lesions and breast cancer bone metastases may be mixed lytic/sclerotic or sclerotic, Na18F PET may provide a complementary role in assessment of response to therapy and has been shown to impact decision-making in treatment monitoring .
The observed differences between bone scintigraphy and [18F]FDG PET for visualizing bone metastases  led some investigators to hypothesize that [18F]FDG PET might provide a useful and accurate means for assessing bone metastasis response. Stafford and colleagues  showed that changes in [18F]FDG uptake correlated with the clinical assessment of response in bone-dominant MBC as well as with changes in breast cancer tumor markers, which are used clinically to assess bone metastasis response. A follow-up study  by Specht et al. showed that changes in [18F]FDG uptake predicted time to progression, and Tateishi et al.  showed that both metabolic and structural changes by [18F]FDG PET/CT predicted bone metastasis time to progression.
These studies show the significant role [18F]FDG PET may play in assessing treatment response for bone-dominant MBC.
As breast cancer treatment becomes increasingly targeted and individualized, there are increasing demands upon breast cancer diagnosis to help direct the therapeutic approach. The imaging approaches in current clinical use for diagnosing and staging breast cancer – mammography, ultrasound, contrast-enhanced MRI, CT, bone scan, and [18F]FDG PET/CT – provide important information on disease extent and response to treatment, but are relatively nonspecific. Advances are more specific and quantitative, and molecular imaging methods are poised to help provide the kind of diagnostic data that are needed to help take advantage of new therapeutic options for breast cancer. We highlight some selected examples below. These include imaging tumor receptor expression and imaging tumor proliferation to assess early response to treatment.
The expression of tumor receptors and other proteins involved in cellular signaling is particularly relevant to breast cancer treatment. Targets of interest include the estrogen and progesterone receptors (ER, PR), targets for endocrine therapy, and HER2, the target for HER2-directed therapy. These biomarkers have been traditionally measured by in vitro assay of biopsy material ; however, the development of vivo molecular imaging of these targets is setting the stage for more specific guidance on treatment and management of breast cancer. Complementary advantages of imaging to measure receptor expression include its non-invasiveness, the ability to measure receptor expression in the entire disease burden and thus the ability to avoid sampling error that can occur with heterogeneous receptor expression, and the potential for serial studies of in vivo drug effects on the target [197, 198].
PET imaging of the ER using the radiopharmaceutical 16 alpha-18F-fluoro-17 beta-estradiol (18F-FES)  has been validated as a measure of ER expression in breast tumors against ER expression assays of tissue samples by radioligand binding  and immunohistochemistry . Higher 18F-FES SUV has been correlated with response to endocrine therapy [197, 202, 203]. A majority of breast cancers express ER (approximately 70%) and ER status is an initial key determinant in therapeutic strategy . Given that there may be substantial heterogeneity of ER expression across tumor sites [205, 206], functional imaging of the ER receptor with PET/CT appears positioned to provide valuable insights into tumor burden characterization and pharmacodynamic response to targeted therapy at different stages of the disease . Interest in imaging of the progesterone receptor (PR) in vivo using 21-18F-fluoro-16α,17α-[(R)-(1′-α-furylmethylidene)dioxy]-19-norpregn-4-ene-3,20-dione (18F-FFNP) has been increasing  as tumors with ER/PR co-positivity are shown to be more responsive to endocrine therapy .
HER2 expression in breast cancer has become an important indicator of prognosis and an increasingly important target for therapy, and imaging probes using both single-photon radionuclides (e.g., 111In-labeled-trastuzumab and 99mTc-ICR12) and positron-emitting radionuclides (e.g., 64Cu-trastuzumab, 64Cu-DOTA-ZHER2:477, 68Ga-trastuzumab F(ab′)2 fragments, 68Ga-ABY-002, and 89Zr-trastuzumab) have been developed and validated [209, 210]. 89Zr-trastuzumab PET/CT has emerged as one of the more favorable HER2 imaging methods due to advantageous radionuclide properties, biodistribution characteristics, and straightforward quantification capabilities of PET compared to single-photon techniques [210, 211]. A recent multicenter trial (ZEPHIR study) including patients with HER2-positive MBC who had progressed after prior trastuzumab therapy and were due to receive trastuzumab emtansine – a recently approved antibody-drug conjugate  – has shown that pretreatment HER2 and metabolic imaging with 89Zr-trastuzumab PET/CT and [18F]FDG PET, followed by metabolic assessment with [18F]FDG PET after one cycle of therapy, had a 100% NPV for therapeutic response after three cycles of therapy and also predicted time to treatment failure . This study paves the way for other studies to assess the utility of combination imaging using PET and other modalities not only in determination of treatment efficacy in a number of clinical settings, such as the neoadjuvant and first-line metastatic settings, but also as a predictive tool in the pre-therapy setting .
Although studies of [18F]FDG PET have shown that tumor glycolysis declines early in the course of treatment [215, 216, 217], it is likely that other pathways more closely tied to cellular growth and death may provide even earlier and more specific indications of therapeutic response. Imaging cellular proliferation in breast tumors is becoming increasingly enticing, especially given the recent approval of a new drug palbociclib, which inhibits cell cycle progression into S phase .
A number of PET tracers are being developed and studied  with much attention focused on thymidine analogs, since thymidine is a substrate for activity of the thymidine salvage pathway for DNA synthesis. 18F has largely replaced the short half-life 11C  for radiolabeling proliferation analogs (new ref see comments), with the most promise shown with 18F-fluorothymidine (FLT). A recent phase II study showed some promise for its use in predicting pCR in the neoadjuvant setting . Work on other proliferation targets such as the sigma-2 receptor ligand is ongoing [222, 223].
- 2.Society, A.C. American Cancer Society. Cancer facts & figures 2015. Atlanta: American Cancer Society; 2015.Google Scholar
- 39.Edge SB, et al. AJCC cancer staging manual. 7th ed. New York: Springer; 2010.Google Scholar
- 48.Network NCC. NCCN clinical practice guidelines in oncology. Breast Cancer. Version 1. 2016. Available at https://www.nccn.org
- 52.Mettler F, Guiberteau M. Essentials of nuclear medicine imaging. 6th ed. St. Louis: Mosby Elsevier; 2012.Google Scholar
- 92.Perez EA, Foo ML, Fulmer JT. Management of locally advanced breast cancer. Oncology (Williston Park). 1997;11(9 Suppl 9):9–17.Google Scholar
- 122.18F-fluoride PET/CT versus 99mTc-MDP scanning for detecting bone metastases: a randomized, multi-center trial to compare two bone imaging techniques. Available from: https://clinicaltrials.gov/ct2/show/NCT00882609
- 123.18F – sodium fluoride pet imaging as a replacement for bone scintigraphy. Available at: https://clinicaltrials.gov/ct2/show/NCT01930812
- 148.Carlson RW, et al. NCCN task force report: adjuvant therapy for breast cancer. J Natl Compr Canc Netw. 2006;4 Suppl 1:S1–26.Google Scholar
- 157.Finn RS, et al. The cyclin-dependent kinase 4/6 inhibitor palbociclib in combination with letrozole versus letrozole alone as first-line treatment of oestrogen receptor-positive, HER2-negative, advanced breast cancer (PALOMA-1/TRIO-18): a randomised phase 2 study. Lancet Oncol. 2015;16(1):25–35.PubMedCrossRefGoogle Scholar
- 187.Lin NU, et al. Phase II study of lapatinib in combination with trastuzumab in patients with human epidermal growth factor receptor 2 – positive metastatic breast cancer: clinical outcomes and predictive value of early [18F]fluorodeoxyglucose positron emission tomography imaging (TBCRC 003). J Clin Oncol. 2015;33(24):2623–31.PubMedPubMedCentralCrossRefGoogle Scholar
- 203.Clark AS. Using nuclear medicine imaging in clinical practice: update on PET to guide treatment of patients with metastatic breast cancer. Oncology (Williston Park). 2014;28(5):424–30.Google Scholar
- 213.Gebhart G, et al. Molecular imaging as a tool to investigate heterogeneity of advanced HER2-positive breast cancer and to predict patient outcome under trastuzumab emtansine (T-DM1): the ZEPHIR trial. Ann Oncol. 2016; 27(4):619–24.Google Scholar
- 214.Clark AS, DeMichele A, Mankoff D. HER2 imaging in the ZEPHIR study. Ann Oncol. 2016; 27(4):555–7.Google Scholar