Multidisciplinary frontiers in breast cancer management

A surgeon's perspective


  • S. Eva Singletary MD

    Corresponding author
    1. Department of Surgical Oncology, The University of Texas M. D. Anderson Cancer Center, Houston, Texas
    • Department of Surgical Oncology, The University of Texas M. D. Anderson Cancer Center, 1515 Holcombe Boulevard, Box 444, Houston TX 77030-4095
    Search for more papers by this author
    • Fax: (713) 792-2225


The current paradigm of breast cancer management may be altered significantly over the coming years by the adoption of new treatment schema and devices outside of the surgical arena. New advances in breast cancer imaging will improve our ability to detect early-stage disease but also will assist in monitoring treatment outcomes and support the development of nonsurgical ablation techniques. These advances, some already in use, include a 3-dimensional adaptation of digital mammography, color Doppler ultrasonography that can visualize neovascularization in growing tumors, contrast-enhanced magnetic resonance imaging with improved accuracy for the detection of occult cancers, a specialized approach to positron emission tomography designed for use on the breast, and the development of nanoparticle contrast agents that can be visualized with near-infrared light. Systemic therapy, which revolutionized breast cancer management in the last half of the 20th century, is being reconceptualized, with attention turning to adjusting the timing of chemotherapy. Dose-dense regimens are being tested, and there also is interest in so-called metronomic chemotherapy in which very low doses are given on a very frequent schedule, resulting in reduced toxicity and treatment outcomes that reflect an antiangiogenic mode of action. Finally, the possibility of a breast cancer vaccine continues to intrigue and excite physicians and patients alike, with the promise of enlisting the body's own immune system to seek out and destroy cancer cells and/or prevent the development of future disease. It will be important for surgeons to stay aware of all developments that may improve the care of their patients and to be true surgical oncologists rather than merely surgical technicians. Cancer 2007. © 2007 American Cancer Society.

For hundreds of years, surgery was the major weapon in the fight against breast cancer. Now, with our growing understanding of the biology of breast tumorigenesis and the mechanisms of metastasis, surgeons work as part of multidisciplinary teams to provide the best care for their patients. This involves a reassessment of the most appropriate role of surgeons. Careers are being rethought as surgeons retool their approach to be aware of all aspects of care that may affect their patients and to be true surgical oncologists rather than merely surgical technicians.

Based on 5 decades of research in the last half of the 20th century, the current paradigm of breast cancer management can be summarized briefly as follows: 1) Widespread use of screening mammography means that breast tumors usually are very small (<2 cm) when first detected. 2) For these small tumors, minimally invasive surgical techniques coupled with radiotherapy can be used for primary treatment. 3) Systemic therapy improves long-term outcomes for the majority of patients, although most types of therapy are associated with significant side effects. 4) The risk of future breast cancer (recurrence or new primary) in high-risk patients can be reduced significantly through long-term treatment with pharmacologic agents such as tamoxifen, raloxifene, and aromatase inhibitors. Some significant side effects may be associated with these treatments.

New treatment schema and devices outside of the surgical arena may alter this paradigm significantly as we progress into the 21st century. Innovative approaches to imaging will facilitate earlier and more accurate detection of breast cancer, support the development of nonsurgical ablation techniques, and assist in monitoring early response to systemic therapy. After many years of extensive research into optimal types and amounts of systemic therapy agents, attention also is being focused on the timing of treatments, with thought-provoking research in the areas of dose-dense and metronomic chemotherapy. The development of a vaccine against breast cancer has the potential to provide both patient-specific treatment and disease prevention without the major side effects observed with current approaches. These exciting frontiers of research are the subject of this review.

New Frontiers in Breast Cancer Imaging

The widespread use of screening mammography has had a significant impact on the breast cancer landscape. Where once it was common for tumors to be detected only after they were relatively advanced, most tumors are now quite small when initially discovered, with a reduced probability of axillary involvement. However, there are downsides to mammography. Tumors still are missed with discouraging frequency, and results are especially poor in women with dense breast tissue. Although the concomitant use of ultrasonography has made it easier to distinguish solid masses from cysts, it is difficult to identify malignant versus benign growths, resulting in false-positive reports and unnecessary biopsies. More precise estimates of size are increasingly important for new ablation techniques and for measuring response to neoadjuvant chemotherapy. New approaches that are now being used or developed include: 1) refinements of imaging techniques already in use, 2) functional imaging techniques, and 3) techniques using nanobiotechnology.

Refinements of imaging techniques already in use


Nearly 25% of women who are called back for additional imaging after a mammogram will turn out to have false-positive results.1 At the other extreme, as many as 20% of women who present with breast cancer will have had a negative mammogram within the preceding year.2 One of the reasons for these problems in mammographic interpretation has to do with what researchers at Massachusetts General Hospital call structured noise—anomalous areas of structural overlap resulting when the 3-dimensional breast is viewed in 2 dimensions.3, 4 To address this difficulty, those researchers developed tomosynthesis, a 3-dimensional adaptation of standard digital mammography that compares with standard mammography in the same way that a computed tomography (CT) scan compares with normal x-rays. The x-ray tube, rather than producing images from 2 fixed angles, moves in a 50-degree arc about the breast, taking 11 low-dose pictures from various angles that are then pieced together by computer. The procedure requires only 1 breast compression, and the total radiation from the multiple images is smaller than that used for standard mammography. As with a CT scan, the radiologist can slice down visually through the breast, so that structures that may overlap on a standard mammogram can be separated out. Initial trials have indicated increased sensitivity and specificity with tomosynthesis compared with standard digital mammography—patient call-back is reduced significantly, and the positive predictive value is increased.5

Color Doppler ultrasonography.

Color Doppler ultrasonography (CD-US) is a special US technique that evaluates blood flow through blood vessels. US waves bounce off of solid objects, including blood cells. When the cells are moving, there is a change in pitch (the Doppler effect) that can be captured and converted by a computer into an array of colors that visualize the speed and direction of the blood flow.

New tumors cannot grow to more than 1 to 2 mm3 without the recruitment of new capillary blood vessels.6 This neovascularization may be useful for distinguishing between benign and malignant lesions and has been the subject of intense interest as a potential prognostic factor.7 CD-US is being studied as a low-cost, noninvasive technology for quantifying neovascularization.

In a 1995 study, Lee and colleagues8 examined the accuracy of blood flow detected by CD-US as an indicator of malignancy. If blood flow was detected within or at the margin of a lesion, then it was characterized provisionally as malignant. Comparisons with histologic results indicated that CD-US showed 94.5% sensitivity in detecting malignant breast tumors, but had a specificity of only 40.1%, indicating a high rate of false-positive diagnoses. Those investigators recommended using CD-US as a supplement when solid lesions are observed by mammography or standard sonography.

Similar results were reported by Chao and colleagues,9 who measured the numbers of vessels and the systolic peak flow velocity as indicators of vascularization. In 1124 breast lesions, they detected vascularity in 85.6% of benign lesions and in 95.3% of malignant tumors. Although the average counts of vessel numbers and values of flow indices in the malignant tumors were significantly larger than in the benign lesions, those authors concluded that there was too much overlap for this difference to be useful clinically.

Watermann and coworkers10 conducted a retrospective study examining vascularization in early breast cancer measured by CD-US as a prognostic factor for overall survival. Over a 10-year period, 170 women with primary breast cancer were examined with CD-US. The researchers divided the patients into groups according to the number of tumor blood vessels, as defined by direct contact with the tumor and constant visibility over time. For patient with from 1 to 5 vessels, from 6 to 10 vessels, and >10 vessels, the 5-year overall survival rates were 0.88, 0.92, and 0.56, respectively. Multivariate analysis indicated that the number of tumor blood vessels remained an independent prognostic factor when corrected for age, tumor size, lymph node status, tumor grade, hormone receptor status, and histologic type.

Boonjunwetwat and colleagues11 conducted a small study to test the hypothesis that vascularity measured by CD-US could be used to predict response to neoadjuvant chemotherapy. They observed that tumors with no/limited vascularity demonstrated by CD-US tended to show an increased response (25–33%) compared with tumors that showed hypervascularity (13–17%), although these differences were not significant.

CD-US is a relatively low-cost, fast, and noninvasive tool for the assessment of tumor vascularization. Currently, it does not appear to be useful for the definitive differentiation of malignant versus benign lesions, but it may hold promise as a prognostic indicator and, possibly, as a predictor of treatment response.

Contrast-enhanced magnetic resonance imaging.

Magnetic resonance imaging (MRI) of breast cancer involves imaging before and after injection of a gadolinium-based contrast agent, which is taken up by both benign and malignant lesions. The distinction between benign and malignant is based on both morphologic and kinetic criteria. Generally, benign lesions are characterized by smooth edges, whereas malignant lesions show spiculated edges. In addition, malignant lesions tend to take up the contrast agent more quickly than benign lesions, reflecting their increased vascularity.

In 6 large prospective studies that have been conducted since the mid-1990s, contrast-enhanced MRI has shown a sensitivity of 71% to 100% for the detection of invasive breast cancer compared with 16% to 40% for mammography and 16% to 33% for US.12–17 Although MRI originally suffered from an unacceptably low specificity, improved algorithms for combining morphologic and kinetic data have resulted in significantly improved specificity ranging from 81% to 99%. Unlike mammography, MRI does not expose the patient to ionizing radiation, and its accuracy is not affected by breast density.

Despite its improved accuracy, contrast-enhanced MRI is not recommended for general screening because of its expense and limited availability. However, it can be valuable for the detection of the following: cancer that is occult on mammography or sonography, multifocal or multicentric disease in patients with a known primary, residual disease in patients with positive margins after breast-conserving surgery, response to neoadjuvant chemotherapy, unknown primary in patients with axillary lymph node metastases, early disease in high-risk BRCA1/2 patients, and recurrence in patients after completion of radiation therapy.

In the future, contrast-enhanced MRI may become an important tool in the development of minimally invasive ablative therapies for the treatment of primary tumors, such as radiofrequency ablation or laser ablation. Such techniques require accurate imaging for the initial assessment of size and location of the tumor as well as for assessment of the success of the treatment. In a study by Burak and colleagues,18 MRI was used for the follow-up assessment of radiofrequency ablation treatment in patients with breast cancer. In that study, MRI was very effective in detecting viable cells after radiofrequency ablation treatment in tumors that were visible on MRI before treatment. If technical problems can be resolved, contrast-enhanced MRI may also be useful for monitoring the progress of tumor ablation during treatment, something that has not been possible with US.

Functional imaging

Images obtained through positron emission tomography (PET) reflect metabolic and physiologic functions that occur in living cells. A positron-emitting radionuclide is attached to a molecule (typically glucose or fluorodeoxyglucose [FDG]) that is ingested or metabolized at a higher rate in tumor cells compared with normal cells. The labeled molecule is injected into the patient, where it accumulates in actively growing tumors. Positron emission is tracked by computer and used to generate a tomographic image of the tumor.

Whole-body PET is very specific for the detection of malignant breast tumors,19–21 but there are significant downsides to whole-body imaging, including limited spatial resolution and considerable noise from normal tissue, making small lesions difficult to detect.22 Positron emission mammography (PEM) was developed to resolve these issues by using a much smaller device that is restricted to a single breast.

PEM uses a parallel pair of detector heads above and below the breast to bring it closer to the source of emissions and provide a broad angle of coverage. This increases the efficiency in detecting photons, resulting in improved spatial resolution.23, 24 In a small pilot study using PEM (n = 17 patients), Levine and coworkers23 were able to detect both in situ and invasive disease with a reported sensitivity of 86% for the detection of malignancy and a specificity of 91%. Rosen and coworkers24 successfully used PEM for the examination of small breast tumors (<2.5 cm). The PEM device was able to identify several lesions that measured <1 cm, including a 0.4-mm noninvasive papillary carcinoma.

A general problem with PET scans is the lack of anatomic specificity. With the much smaller PEM detector heads, this problem can be resolved readily by running the scan either concurrently or sequentially with mammography without releasing the breast compression.23, 24

PEM is quite new, and it has not yet been compared directly with US, mammography, or MRI to demonstrate that it would offer significant advantages over these older technologies. It also has some limitations. PEM is not effective in detecting invasive lobular carcinoma23 and may generate false-positive results from fat necrosis at the site of old biopsies.24 Additional studies with large patient groups are needed to address these technical issues and confirm the preliminary results reported with PEM in pilot studies.

Nanobiotechnology approaches

Near-infrared radiation is of interest for breast imaging, because these wavelengths lie above the absorption level of native chromophores like hemoglobin and below the absorption level of water, so they are able to penetrate biologic tissue up to a depth of 15 cm. Conventional dyes that absorb in this wavelength tend to be toxic, so workers in the growing field of nanobiotechnology are synthesizing light-absorbing and fluorescent nanoprobes that are sensitive to near-infrared radiation, which can be used to improve contrast and potentially detect the expression of specific genes. Researchers at Rice University in Houston are experimenting with metal nanoshells, composite spherical nanoparticles that consist of a dielectric core covered by a thin gold shell, as a replacement for the dyes.25 The optical resonance of the nanoparticles can be adapted to a wide range of wave lengths by varying the relative dimensions of the core and the shell. These nanoshells were used for molecular imaging in a recent study in which anti-HER2-conjugated nanoshells were used to discriminate between living SKBR-3 breast carcinoma cells and noncancer control cells.26 Our growing knowledge of the specific genes associated with breast cancer development should allow this approach to be fine-tuned to an even greater degree.

Systemic Chemotherapy: Issues in Timing

Systemic therapy currently is recommended for all but the most favorable patients (ie, lymph node negative, primary tumor size <10 mm). Guidelines for the exact type of systemic therapy are based on patient age/menopausal status, histologic status of the axillary lymph nodes, size and histologic subtype of the primary tumor, and level of estrogen receptor expressed by the tumor.

For much of the late 20th century, attempts to improve outcomes from systemic therapy were largely restricted to manipulating the type and dose of drugs used. In general, scheduling was designed to balance the need for frequent treatments that allowed minimal time for tumor regrowth with the need for patient recovery from the cytotoxic effects of the drugs, with most multidrug protocols allowing from 21 days to 28 days of recovery between treatments. Overall duration of standard adjuvant chemotherapy is from 4 months to 6 months.

Beginning in the 1990s, oncologists began exploring the idea that increasing the dose density of chemotherapy may improve outcomes while shortening the course of treatment and yielding a greater overall cell kill. The idea of dose density is to minimize the time given for the cancer to grow between treatments. Instead of the common intervals of from 21 days to 28 days between cycles of chemotherapy, treatments are given at intervals of from 7 days to 14 days, frequently with growth factor supplementation. More recently, attention has turned to so-called metronomic chemotherapy in which low doses of the drugs are given on a very frequent schedule with the hope of inhibiting the angiogenic activity needed for tumor growth.

Dose-dense chemotherapy

The concept of dose-dense chemotherapy, as adapted in the Norton-Simon hypothesis,27 stems from the Gompertzian model of tumor growth, which predicts that, because smaller tumors grow faster, tumor growth between cycles of chemotherapy is more rapid when the cell kill has been greatest. Increasing dose density limits the amount of regrowth that can occur between cycles.

Three large prospective trials have tested this hypothesis with mixed results (Table 1).28–30 The 5-year results from Cancer and Leukemia Group B (CALGB) 9741 Intergroup trial, with an enrollment of 1972 patients, were presented at the 2005 San Antonio Breast Cancer Symposium.28 In this protocol, lymph node-positive breast cancer patients received doxorubicin, paclitaxel, and cyclophosphamide (CTX) (either sequentially or with concurrent doxorubicin and CTX followed by paclitaxel) at intervals of 2 weeks or 3 weeks. Both disease-free survival (DFS) and overall survival (OS) were increased significantly in the 2-week treatment arms compared with the 3-week treatment arms, and no differences were observed as a function of sequential versus concurrent drug treatment.

Table 1. Results of Large Prospective Trials Testing the Efficacy of Dose-dense Chemotherapy in Breast Cancer
ReferenceNo. of patientsMedian follow-up, yDrug regimenDose scheduleOutcomeTreatment toxicity
  1. CALGB indicates Cancer and Leukemia Group B; A, doxorubicin; C, cyclophosphamide; FEC, 5-fluorouracil, epirubicin, cyclophosphamide; Q 2 wk, every 2 weeks; q 3 wk, every 3 weeks; q 1 wk, every week; DFS, disease-free survival; OS, overall survival; RBC, red blood cells.

Hudis et al., 2005 (CALGB 9741 trial)2819725A→paclitaxel→C or AC→paclitaxelq 2 wk vs q 3 wkDFS: 76.8% (q 2 wk) vs 71.7% (q 3 wk; P = .012); OS: 83.0% (q 2 wk) vs 79.5% (q 3 wk; P = .049)q 1 wk vs q 3 wk: Increased no. of packed RBC transfusions, decreased neutropenia and febrile neutropenia
Sparano et al., 2005 (E1199 trial)2950524AC→paclitaxel or AC→docetaxelq 1 wk vs q 3 wkNo significant difference in DFSGrade 3/4 toxicity higher in q 3 wk vs q 1 wk for docetaxel arm only
Venturini et al., 200530121410.4FECq 2 wk vs q 3 wkNo significant difference in OS or DFS after adjustment by multivariate analysisq 2 wk vs q 3 wk: Increased asthenia, bone pain, anemia, thrombocytopenia; decreased leucopenia

The 2 other large trials listed in Table 1 showed no difference in outcomes as a function of dose density. The North American Breast Cancer Intergroup E1199 trial, with an enrollment of 5052 patients, compared outcomes from 1-week treatment intervals with 3-week treatment intervals for concurrent doxorubicin and CTX followed by either paclitaxel or docetaxel.29 The Italian trial reported by Venturini and colleagues30 compared treatment intervals of 3 weeks versus 2 weeks for a regimen of 5-fluorouracil, epirubicin, and CTX.

The differences in outcome among these three trials may stem in part from patient sample differences: Patients in the CALGB 9741 Intergroup trial had more advanced disease (all were lymph node-positive) compared with the other 2 groups. Nonetheless, it is noteworthy that the advantage in DFS seen at 5 years in the 2-week arm versus the 3-week arm in the CALGB 9741 trial was reduced from the figure reported at 3 years (7.1% vs 9.3%, respectively).

Even if there is no significant improvement in long-term outcomes, dose-dense chemotherapy may have other advantages over standard dosing schedules. Citron and colleagues31 reported a lower number of hospital admissions for neutropenia and fever in the dose-dense group of the CALGB 9741 trial, and patients appreciated the significant reduction in the overall treatment time.32

Metronomic chemotherapy

Metronomic chemotherapy is a variation of dose-dense scheduling in which low doses of chemotherapy drugs (each significantly below the maximum tolerated dose [MTD]) are given at frequent, even daily, intervals for extended periods with no prolonged drug-free intervals. The doses are low enough that even the cumulative dose may be lower than the MTD. These low doses result in reduced toxicity and a reduced reliance on growth factor support to recover from the myelosuppression.

Chemotherapy drugs interfere with cell division and affect all dividing cells, including the endothelial cells involved with new blood vessel formation in the growing tumor. Usually, the damaged endothelial cells are repaired or replaced during the 3- to 4-week breaks normally used to recover from the side effects of chemotherapy.33 The use of very short cycles does not allow sufficient time for this to occur. In addition, because the endothelial cells are normal and genetically stable, they are unlikely to acquire drug resistance, a phenomenon that is common in genetically unstable tumor cells. They will continue to respond to the mitosis-inhibiting effects of the chemotherapy through repeated doses in much the same way as myeloid cells do. Myelosuppression occurs repeatedly during multiple cycles of standard chemotherapy even if the tumor itself has mutated into a drug-resistant form.34

Because metronomic chemotherapy is designed primarily to be antiangiogenic instead of cytotoxic, the principal outcomes are expected to differ from those observed with standard chemotherapy: long-term stable disease versus the destruction of existing disease.35 Three small clinical trials have investigated the use of metronomic chemotherapy in patients with advanced breast cancer (Table 2).36–38 In the study by Orlando and colleagues,36 low-dose methotrexate (MTX) and low-dose CTX were administered orally to 153 patients with metastatic breast cancer. At a median follow-up of 23 months, those authors reported that 16% of patients achieved a prolonged clinical benefit (no disease progression for ≥12 months). The other 2 studies analyzed metronomic chemotherapy alone versus in combination with other agents. In a study that involved 114 elderly women with advanced disease, Bottini and colleagues37 reported that letrozole in combination with CTX (50 mg per day) significantly increased the rate of overall response compared with letrozole alone. Burstein and coworkers38 used the same MTX/CTX regimen reported in the study by Orlando et al either alone or in combination with the antiangiogenic agent bevacizumab. The addition of bevacizumab to the metronomic chemotherapy combination resulted in significant increases in overall response and time to progression compared with MTX/CTX alone. However, the percentage of patients achieving stable disease was similar in both groups (41% vs 38%, respectively). All 3 studies reported very favorable toxicity profiles for the drug regimens.

Table 2. Results of Clinical Trials Testing the Efficacy of Metronomic Chemotherapy in Advanced Breast Cancer
ReferencePatient populationMedian follow-up, moDrug regimenOutcome
  1. MBC indicates metastatic breast cancer; MTX, methotrexate; CTX, cyclophosphamide; CR, complete response; PR, partial response; PCB, prolonged clinical benefit (no disease progression for at least 12 months); ER+, estrogen receptor positive; OR, overall response; DFS, disease-free survival; SD, stable disease; TTP, time to progression.

Orlando et al., 200636MBC (n = 153)23MTX 2.5 mg 2 times daily 2 d/wk and CTX 50 mg every dCR, 3.3%; PR, 16.3%; PCB, 15.7%
Bottini et al., 200637Elderly, T2-T4, N0-N1, ER+ (n = 114)24Letrozole 2.5 mg daily vs letrozole and CTX 50 mg dailyOR, 87.7% for letrozole and CTX vs 71.9% for letrozole; DFS, 82% vs 83.5%, respectively
Burstein et al., 200538MBC (n = 55)12CTX 50 mg daily and MTX 2.5 mg twice daily 2 d/wk vs CTX/MTX and bevacizumab 10 mg/kg every 14 dOR, 10% for CTX/MTX vs 29% for CTX/MTX and bevacizumab; SD, 38% vs 41%, respectively; TTP, 2 mo vs 5.5 mo, respectively

Although these results seem promising, the studies reported were very small. If larger controlled trials provide additional evidence for the clinical usefulness of metronomic chemotherapy, then this approach would have several advantages to the patient. In the metronomic drug regimens that have been tested to date, drugs have been given orally, making administration more convenient for the patients. The low toxicity means that there is another option for patients who may be poor candidates for standard chemotherapy because of advanced age or comorbidities or for patients who refuse chemotherapy because of their fear of significant side effects. Because patients are not expected to develop resistance to the drugs used, relatively inexpensive off-patent agents can be used, reducing the cost of treatment.39 Further cost reductions are anticipated from the reduced need for supportive care for treatment side effects. Metronomic chemotherapy targets a different cell population than standard therapies, so combining it with other low-impact modalities, such as targeted therapies, endocrine therapy, or antitumor vaccines, should be possible.

Although all of this may seem attractive, there is an important cautionary note. When antiangiogenic agents were first introduced, they were hailed as a magic bullet that surely would usher in a new era of effective cancer treatment.40 This has not happened yet with these agents, and clinically important results may prove to be equally elusive with this new approach to antiangiogenesis.

Breast Cancer Vaccines

Both patients and physicians continue to be intrigued and excited by the possibility of a breast cancer vaccine that could enlist the body's own immune functions to seek out and destroy cancer cells and/or prevent the development of future disease. Compared with other available therapies for breast cancer, a vaccine would have the advantage of being easy to administer and relatively nontoxic.

Most widely used vaccines (for influenza, hepatitis B, rubella, etc) are prophylactic in nature, that is, they are directed against a specific etiologic agent prior to actual disease exposure. Cancer, of course, is a more complex situation. Although some cancers are associated with specific infectious agents,41 the evidence for most cancers (including breast cancer, as discussed below) is conflicting and difficult to interpret. Thus, most experimental cancer vaccines to date have been designed to be therapeutic, stimulating a cellular immune response against antigens from established tumors.

Basics of immunology

The human immune system is extraordinarily complex, and a detailed description is far beyond the scope of this article. Figure 1 presents a very basic outline that is intended to assist in understanding the experimental approaches that have been used in developing breast cancer vaccines. The figure shows that B cells provide the basis for the humoral response, interacting with foreign antigens in their native form and further differentiating into plasma cells, which secrete specific antibodies, or memory cells, which respond quickly to future exposures. The cellular response involves the interaction of T cells (cytotoxic [CD8+] or helper [CD4+]) with processed foreign antigens presented by the major histocompatibility (MHC) proteins. The activated, cytotoxic T cells release compounds that induce apoptosis in the infected cell and also undergo clonal expansion, generating daughter cells that can recognize and kill other infected cells. The activated helper T cells generate nonspecific responses to the invaders (eg, inflammation) and also activate B cells to begin the production of antibodies.

Figure 1.

Schematic representation of humoral and cellular bases of immunologic mobilization against foreign invaders.

Creating a therapeutic vaccine against breast cancer

For some or all of the reasons listed in Table 3,42–46 it has proven to be extraordinarily difficult to create effective therapeutic cancer vaccines against solid tumors. Nonetheless, a variety of approaches can be used to increase the immune response, largely aimed at selecting appropriate antigens and ensuring their effective presentation to the immune system.

Table 3. Potential Difficulties to Overcome in Constructing a Vaccine Against Solid Tumors*
  • *

    See Campoli et al., 200542; Radford et al., 200543; Molldrem, 200644; Kortylewski et al., 200545; Gillanders and Thalachallour, 200646.

Most tumor antigens are self antigens and only mildly immunogenic
Tumor cells may express low levels of antigen
Antigen may be shielded from immune surveillance
Developing tumor may shed antigens
Only a fraction of tumor cells may express antigens
Tumor cells may express low levels of costimulatory proteins needed to generate a robust immune response
Advanced-stage patients recruited for trials may have large tumors; cells on the interior may not be accessible
Metastatic tumors may release immunosuppressive factors into the microenvironment
Metastatic tumors associated with decreased numbers of peripheral blood lymphocytes and dendritic cell dysfunction

Table 447–50 shows some of the approaches that are being used currently for the development of experimental breast cancer vaccines. Of the 4 studies that are summarized here, 3 have produced promising results in animal studies but have not been tested yet in patients with breast cancer. Two of those studies used DNA vaccines, which involve the in vivo transfection of full-length, cDNA molecules into target cells, where they produces antigens that are processed and presented using normal MHC I- and MHC II-based mechanisms.48, 49 The third animal study used the bacterium Lysteria monocytegenes as a vector to carry overlapping fragments of the HER2/neu gene into the target cells.50 Avigan and colleagues47 used a fused tumor cell/dendritic cell vaccine in a small pilot study with 16 breast cancer patients. Clinical response to the vaccine was apparent in 3 of the 16 patients and included near complete regression of a large chest wall mass with stable disease for 2 years in 1 patient, 50% regression of an adrenal mass and 44% regression of a pulmonary nodule with stable disease for 6 months in a second patient, and stabilization of disease for 3 months in the third patient.

Table 4. New Approaches for Vaccine Development in Breast Cancer
ReferenceApproachRationaleStudy descriptionResults
  1. MHC indicates major histocompatibility; cDNA, combinational DNA; L.m., Lysteria monocytegenes.

Avigan et al., 200447Fusion of tumor cells and dendritic cellsTargets multiple antigens; uses the ability of dendritic cells to efficiently present antigens with MHC class I and II pathwaysSixteen breast cancer patients injected with a dendritic/tumor fusion cell vaccineClinical response in 3 of 16 patients (19%)
Narayanan et al., 200448DNA vaccine based on a single purified antigenFull-length cDNA molecule transfected into target cells where it produces antigens that are processed by MHC I and II mechanismsMice transgenic for human HLA-A2 and CD8 inoculated with vaccine based on human mammoglobin-A cDNAVaccine produced an immunologic response, and was associated with regression of tumors derived from a mammaglobin-A-positive cell line
Chopra et al., 200649DNA vaccine based on whole cell sheared DNA fragments transfected into highly immunogenic fibroblastsWhole tumor cells alone are poorly immunogenic; fibroblasts are effective at presenting antigens and grow easily in cultureMice injected with vaccine derived from modified LM fibroblasts transfected with genomic DNA fragments from a highly aggressive murine breast adenocarcinomaVaccine alone effective as a prophylactic agent; mice with established tumors required pretreatment with paclitaxel to show a survival response
Singh et al., 200550Vaccines using L.m. as a vectorAlthough most bacteria are broken down in phagocytic vacuoles, some L.m. can escape into the cytoplasm, enabling them to replicate and grow within the cell; L.m. antigens can generate both MHC I and II responsesRecombinant strains of L.m. containing 5 overlapping fragments of rat HER2-neu used to vaccinate mice previously injected with a HER2-positive rat breast tumor cell lineInoculation with all 5 strains stopped tumor growth; regression in some animals occurred >1 mo after inoculation

These experimental approaches to developing a therapeutic breast cancer vaccine still are in very early stages of development, and, as noted above, most have not yet moved beyond preclinical trials. It seems likely that, once the development moves into clinical studies, the optimal use of these vaccines will be in patients with early-stage disease or in patients with minimal residual disease after the completion of other therapies.

Prophylactic vaccines: Are some breast cancers attributable to an infectious agent?

Almost 20% of cancers worldwide are attributable to infectious disease, and prophylactic vaccination against the infecting organisms may offer significant protection against the related cancers.41 The search for an infectious etiology for breast cancer has been going on for many years, with most research centering on human papillomavirus, Epstein-Barr virus, and a human equivalent of mouse mammary tumor virus.

Table 551–57 shows that the evidence of a viral etiology for breast cancer, while suggestive, remains largely descriptive and indirect. It is likely that, if such a relationship exists, then it will be much more complex than the simple relationship observed for such conditions as influenza. However, with such a compelling upside—the potential to prevent thousands of new cases of breast cancer each year—the search continues.

Table 5. Current Evidence Implicating Viruses as Etiologic Agents for Breast Cancer
VirusEvidence (Reference)
  1. HPV indicates human papillomavirus; EBV, Epstein-Barr virus; MMTV, mouse mammary tumor virus.

HPVHPV identified in human breast tumors but not in normal breast tissues from women undergoing cosmetic surgery (Kan et al., 200551)
Normal human breast cells in culture are immortalized by infection with HPV (strains 16 and 18) (Band et al., 199052)
EBVIdentification of EBV genes or gene products in breast tumors (Glaser et al., 200453)
In vitro evidence of growth-stimulatory properties of EBV in breast epithelial cells in culture (Speck and Longnecker, 200054)
MMTVMajor etiologic agent of breast cancer in mice (Bittner, 193655; Moore et al., 197956)
DNA sequence with homology to the MMTV ENV gene found in 40% of human breast tumors but in <2% of normal breast tissue samples (Etkind et al., 200057)


During the lifetime of most surgeons practicing today, we have seen breast cancer management evolve dramatically from a paradigm centered on radical surgery to one that involves the synergistic combination of multidisciplinary approaches. During this time, our concept of curing breast cancer also as evolved. In 1970, less than 40 years ago, there was hopeful talk about a single magic bullet that might be found to attack cancer. With our increased knowledge of the molecular biology of tumors, there is now increasing awareness that breast cancer is a complex family of diseases rather than a single entity and is unlikely to yield to any single approach. If recent history has a lesson to teach us, it is that the treatment developments of the next century most likely will lie in reducing the morbidity associated with breast cancer, turning it into a chronic disease that can be treated effectively, with good quality of life achieved over extended periods.