Risk Assessment and Individualized Therapy: A Report from the 2008 San Antonio Breast Cancer Symposium

The San Antonio Breast Cancer Symposium (SABCS) marked its 31st anniversary with its 2008 meeting, which was the first SABCS presented in collaboration by the CTRC (Cancer Therapy and Research Center at the University of Texas Health Science Center), AACR (American Association for Cancer Research), and the Baylor College of Medicine.

Risk Assessment

The importance of establishing specific risks of developing breast cancer is to gain understanding of how best to monitor women so that breast cancer may be detected and treated in its earliest stages while minimizing false-positives and over-screening. Studies have suggested that risk of breast cancer is higher among women with particular types of benign breast disease; however the risks among varying benign diagnoses has not been well defined.

To explore the relationship between benign breast disease and breast cancer among women under the age of 50, researchers at the Mayo Clinic evaluated information from more than 4,000 women with benign breast disease.1 Two percent had been diagnosed with atypical hyperplasia, 26% had been diagnosed with proliferative disease without atypia, and 72% had been diagnosed with non-proliferative disease. Data from these women was compared to data derived from the NCI Surveillance, Epidemiology, and End Results (SEER) database.
During 20 years of follow-up, 326 of the women were diagnosed with breast cancer.

  • Compared with women in the general population, the risk ratio of the development of breast cancer among women diagnosed with benign breast disease was 1.2 for nonproliferative diseases, 2.02 for proliferative disease without atypia, and 6.92 for atypical hyperplasia (P=0.001).
  • Ipsilateral breast cancer represented 65% of breast cancer diagnoses within the first five years, whereas contralateral breast cancer represented 52% of breast cancer at five or more years of follow-up. (P=0.03).
  • Increased involution resulted in significant reductions in breast cancer risk among all three histologies.

This study confirms previous reports of an elevated risk of breast cancer among women with benign breast disease, with the greatest risk among women diagnosed with atypical hyperplasia. Women under the age of 50 who have atypical hyperplasia of the breast are nearly seven times more likely to develop breast cancer than women in the general population, irrespective of family history. With this marked increased risk, these patients may gain paramount benefit from diligent screening measures.

Treatment with hormone therapy for relief of menopausal symptoms has also been demonstrated as a risk factor for breast cancer. At this year’s SABCS, additional information from the Women’s Health Initiative (WHI) firmly established that postmenopausal hormone therapy with combined estrogen plus progestin does indeed increase a woman’s risk of developing breast cancer.2 According to another study, however, breast cancers that develop in women who have used estrogen plus progestin tend to be less deadly than breast cancers in other women.3

Researchers used updated information from the WHI clinical trial of estrogen plus progestin, as well as information from another component of the WHI (the WHI Observational Study) to further explore the relationship between estrogen plus progestin and risk of breast cancer.4

  • The results confirmed that users of estrogen plus progestin were more likely than nonusers to develop breast cancer. This increased risk declined markedly and fairly rapidly, however, once women stopped using hormones. Patients’ risks were at baseline at approximately two years following discontinuation of hormone therapy.

These analyses of the WHI data support the claim that the recent decreases in breast cancer incidence in the United States may be due to a reduction in the number of women using postmenopausal hormones. It also provides information to women and physicians about what happens to breast cancer risk after hormone cessation.

In a second presentation, researchers assessed the impact on breast cancer survival of hormone use prior to breast cancer diagnosis. Using information from the California Teachers Study, researchers examined breast cancer survival among 2,783 postmenopausal women with breast cancer.5

  • After accounting for other factors that may influence breast cancer survival (such as stage of disease and general health), women who used estrogen plus progestin before their breast cancer diagnosis were 47% less likely to die of breast cancer. There was a suggestion that estrogen alone may also decrease risk, but the effect was smaller and not statistically significant.

This study suggests that although breast cancer risk is significantly increased among women who take estrogen plus progestin, these breast cancers have a tendency to be less deadly than breast cancers that develop in women who have never used postmenopausal hormones.

The Gail model (consideration of a woman’s age, reproductive history, family history, and history of breast biopsies) is a clinical tool used to assess a woman’s risk of developing breast cancer and identify high-risk patients for enrollment in prevention studies. Now a genetic-based breast cancer test, OncoVue®, may offer a significantly more accurate estimation of this risk than the Gail Model
OncoVue is a new genetic-based breast cancer risk test that uses a combination of a questionnaire and analysis of buccal cell DNA in order to assess risk. OncoVue analyzes 22 single nucleotide polymorphisms (SNPs) in 19 genes and then combines this information with the Gail Model risk factors.

In order to determine the accuracy of OncoVue, researchers collected DNA from 177 women without breast cancer and 169 women diagnosed with breast cancer between 1997 and 1999 in Marin County, California, an area which is known for higher than average breast cancer incidence and mortality rates.6 All samples were then anonymously coded and sent to the lab for analysis; a blind analysis was then performed.

  • The OncoVue score was 2.4 times more accurate than the Gail model in estimating individual risk.

Overall, OncoVue was significantly more accurate than the Gail Model alone in estimating the individual risk of breast cancer among the Marin county women and may provide important implications for all women who may be at an increased risk of developing the disease.

Individualizing Therapy

Evidence continues to emphasize the fact that each individual breast cancer patient possesses innumerable variables that exist along a continuum ranging from variances on a microscopic level, such as subtle differences in genetic components, to a macroscopic level, such as co-morbid conditions and age. As such, individualizing therapy for every patient remains the logical progression of improving outcomes for this disease.

Oncotype DX®, the reverse transcription polymerase chain reaction (RT-PCR) 21-gene assay, represents a major milestone in the progress of individualized therapy. Oncotype DX is a clinically validated test that quantitatively predicts the likelihood of breast cancer recurrence in women with newly diagnosed, early-stage invasive breast cancer and assesses the benefit that these patients will achieve from chemotherapy. Oncotype DX has been included in the NCCN and ASCO guidelines for patients with estrogen-receptor positive early breast cancer to help guide treatment decisions. The 21-gene assay stratifies patients according to their risk of a recurrence by a Recurrence Score (RS). Patients with a score less than 18 have a low risk of recurrence; those with an RS score between 18-30 have an intermediate risk of recurrence; and those with a score greater than 30 have a high risk of recurrence. Oncotype DX has also demonstrated the ability to determine ER/PR and HER status; these results are now included as part of the assay’s results.

More recently, researchers evaluated the ability of Oncotype DX to predict the risk of distant recurrences among postmenopausal women with hormone-positive breast cancers who are treated with Arimidex® (anastrozole).7 The recent trial included 1,308 hormone receptor-positive tumor blocks that were collected retrospectively from patients in the monotherapy arms of the ATAC (Arimidex, tamoxifen, or both trial), in which postmenopausal patients with early breast cancer were treated with either Arimidex or tamoxifen as monotherapy, or sequential administration of both agents. The primary analysis was the relationship between the rate of distant recurrences and the RS of Oncotype DX.

  • At nine years distant recurrence-free rates among lymph node-negative patients were 96% for patients with a low RS, 88% for those with an intermediate RS, and 75% for patients with a high RS (P<0.001).
  • At nine years distant recurrence-free rates among lymph node-positive patients were 83% for patients with a low RS, 72% for those with an intermediate RS, and 51% for those with a high RS (P<0.001).
  • Tumor size was also an independent predictor of time to distant recurrence.
  • The adjusted hazard ratio of the RS as a variable in determining risk of a distant recurrence was 5.25 (2.84-9.73); P<0.001.

These results indicate that the Recurrence Score (RS) of Oncotype DX is an independent predictor of the risk of distant recurrences among postmenopausal, hormone-positive early breast cancer patients who are treated with Arimidex or tamoxifen (Nolvadex®). Oncotype DX continues to demonstrate its clinical utility in providing information in regards to selecting the most appropriate therapy for individual patients with breast cancer. Studies evaluating Oncotype DX in several different capacities are ongoing.

Tamoxifen has undoubtedly greatly improved outcomes for hormone receptor-positive breast cancer patients; however, a large proportion of patients treated with tamoxifen ultimately do not respond to the drug. Researchers therefore continue to explore markers that indicate response either prior to therapy or that indicate response early in the course of therapy so that treatment may be switched among non-responders.

Researchers involved in the International Breast Intervention Study ([IBIS]; a trial that examined the effects of tamoxifen in the prevention of breast cancer and involved over 7,000 participants), observed a correlation between breast density reduction and reduced breast cancer risk during the study; based on this observation, they decided to conduct a subpopulation analysis of the participants in the trial.8 During the IBIS study, women underwent a baseline mammogram, as well as mammograms at 18, 36, and 54 months to monitor for the development of breast cancer. The subpopulation evaluated included 120 women who developed breast cancer and 943 controls who did not develop breast cancer. The researchers examined the baseline mammograms, as well as the mammograms that took place 12-18 months after treatment with tamoxifen was initiated. The results indicated that 46% of the women in the tamoxifen group experienced a reduction in breast density of 10% or more. Furthermore, the women whose breast density was reduced by 10% or more experienced a 52% reduction in the risk of breast cancer relative to the control group. Conversely, women whose breast density was not reduced by 10% had a non-significant 8% reduction in breast cancer risk.

The researchers concluded that the changes in breast density induced after 12-18 months of treatment with tamoxifen can serve as a biomarker of the impact of tamoxifen on the reduction of breast cancer risk. They assert that changes in breast density may serve as an early indicator of treatment efficacy. Furthermore, these changes in density could even be used to determine which women are benefiting from treatment with tamoxifen and which might need a different risk-reduction approach.

Researchers from the Mayo Clinic also evaluated responsiveness to tamoxifen; however, they tested variances with in the CYP2D6 gene. The CYP2D6 gene codes for an enzyme that breaks down tamoxifen into endoxifen, an active metabolite of tamoxifen. Therefore, certain deficiencies of CYP2D6 result in impaired metabolism of tamoxifen, ultimately associated with a higher risk of recurrence among those treated with the drug. A commercially available test is able to test for deficiencies in CYP2D6; however, the test is currently not widely used as results from studies evaluating the association between CYP2D6 deficiencies and outcomes with tamoxifen among breast cancer patients have not been entirely consistent. Among postmenopausal women with estrogen receptor-positive breast cancer, those with certain variations in the CYP2D6 gene derived little benefit from tamoxifen; based on these findings, researchers at the Mayo Clinic recommend CYP2D6 testing for postmenopausal women being considered for adjuvant tamoxifen therapy.

To further explore the potential relationship between deficiencies in the CYP2D6 gene and outcomes achieved with tamoxifen among women with hormone-positive breast cancer, the researchers conducted a retrospective analysis of a subgroup of patients from the Austrian Breast and Colorectal Study Group-8 (ABCSG-8) trial. The ABCSG-8 was a prospective clinical trial conducted between 1996-2004 in which 3,901 women with hormone-positive breast cancer were randomized prior to treatment to two years of adjuvant tamoxifen followed by three years of Arimidex® (anastrozole) or five years of adjuvant tamoxifen. The current analysis evaluated recurrence rates of poor metabolizers of tamoxifen, intermediate metabolizers of tamoxifen, and extensive metabolizers of tamoxifen according to CYP2D6 status.

  • At five years poor metabolizers who were randomized to tamoxifen had a 3.8-fold increased risk of recurrence compared with extensive metabolizers of tamoxifen.
  • However, poor metabolizers who were randomized to switch to Arimidex after two years of tamoxifen did not experience an increased recurrence risk at years three and five.

Based on these findings, the researchers recommend CYP2D6 testing for postmenopausal women being considered for adjuvant tamoxifen therapy following discussion with the patient regarding the test. It also appears that poor metabolizers who do not recur within two years of adjuvant tamoxifen can achieve significant benefit from switching to a different treatment approach.

Another genetic variance, chromosome 17 polysomy, appears to determine which patients will benefit from anthracycline therapy; however, further study is necessary regarding this potential biomarker as only initial studies have been conducted.

Previously, HER2 and topoisomerase II (TOP2A) demonstrated potential in being biomarkers for predicting responses to anthracyclines among breast cancer patients. However, it now appears that chromosome 17 polysomy may have greater predictive power than either HER2 or TOP2A (both of which reside on chromosome 17) in predicting a benefit from anthracycline therapy in this patient population. Due to known anthracycline toxicity, particularly irreversible cardiovascular damage, understanding which patients would benefit from the class of drugs while eliminating exposure among patients who would not benefit is of importance.

Researchers from the United Kingdom and Scotland recently conducted an analysis from the National Epirubicin Adjuvant Trial (NEAT), a trial that ultimately helped establish anthracyclines as a standard of care for early breast cancer. The NEAT trial involved over 2,400 women with early breast cancer who were treated with either Ellence® (epirubicin) plus CMF (cyclophosphamide, methotrexate, and fluorouracil) or CMF. Relapse and overall survival were improved by approximately 30% among women treated with Ellence.

In the current analysis, Dr. Bartlett and colleagues evaluated tumor samples from 1,646 patients involved in the NEAT trial for gene alterations in HER2 and TOP2A as well as chromosome 17 polysomy.

  • Chromosome 17 polysomy was identified in approximately 30% of patients in the NEAT trial and was not prognostic for survival.
  • Chromosome 17 polysomy was associated with a 35-40% reduction in relapses among those treated with Ellence/CMF compared with those treated with CMF only.
  • HER2 and TOP2A were prognostic for survival but not predictive of anthracycline response.

Dr. Poole, the lead investigator of the NEAT, trial stated, “We are close to being able to use this new marker in the clinic to select appropriate therapies in early breast cancer.”

A second analysis to investigate chromosome 17 polysomy as a potential biomarker for predicting benefit from anthracycline therapy in the adjuvant breast cancer setting was also conducted by researchers from Scotland and the United Kingdom.9 The researchers collected data from the two large adjuvant breast cancer trials (BR9601 & MA.5) in which Adriamycin® (doxorubicin)/CMF was compared to CMF alone. Outcomes from both trials were significantly improved among patients who received Adriamycin, further establishing the use of anthracyclines in standard care for breast cancer. Tumor samples from 944 patients were available from the two trials.

  • Chromosome 17 polysomy was identified in approximately 40% of patients and significantly predicted for a decreased overall and disease-free survival (DFS) (P=0.039 and P=0.007, respectively).
  • Chromosome 17 polysomy was significantly associated with increased DFS among patients treated with Adriamycin in both trials.
  • Multivariate Cox regression analysis that included grade, size, and ER-status of tumors, as well as nodal status, polysomy, and HER2 status indicated a highly significant association between chromosome 17 polysomy and the use of Adriamycin for both overall survival and DFS (HR 1.61, C.I. 1.01-2.57, p=0.047 and 1.75 C.I. 1.15-2.64, p=0.008, respectively). There was no significant association observed between HER2 status and treatment in the multivariate analysis.

The researchers stated, “In both trials C17 polysomy predicted for enhanced benefit from A [Adriamycin]. This is confirmed by Multivariate regression analysis showing a significant interaction of C17 as a marker with treatment. Polysomy 17 may reflect either chromosomal instability or polyploidy and further analysis is warranted to test this unifying hypothesis for prediction of benefit from adjuvant A.”

Among women with early breast cancer who undergo a sentinel lymph node (SLN) biopsy, the prognostic value of isolated tumor cells (ITC) or nodal micrometastases (MM) remains unknown, as results from studies evaluating this issue are conflicting. Thus, optimal systemic treatment options for this group of patients, particularly for those with otherwise favorable tumor characteristics, remains inconclusive. Therefore, understanding the prognostic value of ITC and/or MM, as well as understanding outcomes associated with different therapies for these patients is essential in establishing the most appropriate therapeutic choices for this group of patients.

To explore the prognostic implications of ITC and MM, researchers from the Netherlands conducted a retrospective cohort study referred to as the MIRROR (Relevant and Robust or Rubbish) study, in which data from over 3,200 patients was evaluated.10 The study included data from patients treated for breast cancer from 1997-2005 in hospitals in the Netherlands. All patients had undergone a SLN dissection and were restaged according to the sixth TNM classification. Patients were divided into the three following cohorts:

  • Cohort I (n=838): patients had no presence of ITM or MM and did not receive adjuvant chemotherapy
  • Cohort II (n=832): patients had ITM or MM and did not receive adjuvant chemotherapy
  • Cohort III (n=958): patients had ITC or MM and did receive adjuvant chemotherapy

Results at five years are as follows:

  • Recurrence rates were significantly increased and DFS was significantly reduced among patients with ITM and/or MM compared with those with negative nodes (P<0.05).
  • ITM or MM resulted in an absolute reduction of 9% in DFS compared with patients with negative nodes
  • The presence of either ITM or MM resulted in comparable reductions in five-year DFS.
  • Among patients with ITM and/or MM, adjuvant chemotherapy significantly improved five-year DFS and reduced recurrence rates so that outcomes were comparable to patients with negative nodes.

The researchers stated that it appears that breast cancer patients with ITM and/or MM can gain significant benefit in terms of recurrence reductions and improved DFS with the addition of adjuvant chemotherapy. However, newer systemic therapies have emerged since the initiation of this study in 1997, which may affect outcomes associated with chemotherapy. Nonetheless, the results indicate that both ITM and MM are prognostic indicators among women with early breast cancer.

Researchers from the M. D. Anderson Cancer Center and collaborators in Belgium and Austria also have reported that among women with T1a,bN0M0 breast tumors (less than 0.5 cm), those who are HER2-positive have a fivefold increase in distant relapses compared with those who are HER2-negative.11

This retrospective analysis included data from 965 patients from the M. D. Anderson Cancer Center who were diagnosed between 1990 and 2003. The majority of cancers were hormone-receptor positive (68%), while 10% were HER2-positive (irrespective of hormone status) and 23% were triple-negative cancers.

  • At five years recurrence-free survival was 77.1% for HER2-positive compared with 93.7% for HER2-negative tumors.
  • Distant recurrence-free survival at five years was 86.4% for HER2-positive and 97.2% for HER2-negative tumors.
  • This represents a nearly threefold greater risk for recurrence and a more than fivefold greater risk for a distant recurrence among HER2-positive tumors compared with HER2-negative tumors.

Results from this study raise the possibility of the introduction of Herceptin among patients with node-negative, small HER2-positive cancers. Dr. Gonzalez-Angulo, senior author of the study stated that this finding “shows that patients with HER2-positive tumors 1 cm or less have a significant risk of relapse and should be considered for clinical trials of systemic anti-HER2 adjuvant therapy or, if a clinical trial is not available, adjuvant therapy should be discussed with them.”

Accurate staging is also imperative in order to provide optimal treatment choices in breast cancer. Staging for early breast cancer prior to surgery typically involves mammogram, ultrasound, and biopsy. Retrospective data has indicated that MRI used in this staging process may provide more accurate staging information and therefore may reduce unnecessary mastectomy or reoperation. However, no prospective randomized trials evaluating this issue have been performed until the COMICE trial.

The COMICE trial, conducted by researchers in the United Kingdom, is the first prospective randomized trial evaluating MRI as a loco-regional staging method for women with breast cancer prior to surgery.12 The multicenter trial included 1,623 women who underwent triple staging assessment including mammogram, ultrasound, and biopsy prior to scheduled wide-excision. Patients were then randomized to MRI imaging.

  • The positive predictive value of MRI imaging was 61.8%, and the negative predictive value was 83.7%.
  • There were no differences in survival or quality of life between the two groups of patients; however, one-third of the patients in the group that did not receive MRI expressed anxiety that they had not undergone the additional screening.
  • MRI did not predict for reoperation rates, which were 18.75% for those who underwent MRI compared with 19.33% for those who did not undergo MRI (OR=0.96, p=0.7691).
  • MRI accurately detected additional cancerous lesions in nearly 5% of patients who underwent MRI; however, this did not reach statistical significance.
  • Young age and lobular cancer were the only significant predictors of reoperation rates.

These researchers asserted that, according to these results, additional MRI for loco-regional staging prior to breast-conserving therapy does not appear to warrant the extra cost as no significant differences in survival or reoperation rates were achieved compared with standard staging methods.


At the 2008 SABCS, data presented on the biologic processes of the disease as well as appropriate risk assessment and screening remain crucial to maintaining the momentum already achieved in moving towards overcoming breast cancer. Individualized care has been established as the new approach to therapy, for which tests such as Oncotype DX have become integral in making decisions for individual patients.


1 Ghosh K, Pankratz VS, Reynolds CA et al. Benign breast disease and breast cancer risk in young women. Presented at the San Antonio Breast Cancer Symposium. December 13, 2008. Abstract 62.

2 Chlebowski RT, Kuller L, Anderson G et al. Breast cancer after stopping estrogen plus progestin in the Women’s Health Initiative. Presented at the San Antonio Breast Cancer Symposium. December 13, 2008. Abstract 64.

3 Marshall SF, Chang E, Clarke CA et al. Hormone therapy use before diagnosis and breast cancer survival in the California Teachers Study. Presented at the San Antonio Breast Cancer Symposium. December 13, 2008. Abstract 65.

4 Dalessandri KM, Miike R, Wrensch MR, et al. Validation of OncoVue®, a new individualized breast cancer risk estimator in the Marin County, California adolescent risk study. 31st Annual San Antonio Breast Cancer Symposium. December 10-14, 2008. Abstract 502.

5 Dowsett M, Cuzick J, Wales C, et al. Risk of distant recurrence using Oncotype DX in postmenopausal primary breast cancer patients treated with anastrozole or tamoxifen: a TransATAC study. Program and abstracts of the 31st Annual San Antonio Breast Cancer Symposium; December 10-14, 2008; San Antonio, Texas. Abstract 53.

6 Cuzick J, Warwick J, Pinney L, et al. Change in breast density as a biomarker of breast cancer risk reduction; results from IBIS-1. 31st Annual San Antonio Breast Cancer Symposium. December 10-14, 2008. Abstract 61.

7 Goetz M, Ames M, Gnant M et al. Pharmacogenetic (CYP2D6) and gene express profiles (HOXB13/IL17BR and molecular grade index for prediction of adjuvant endocrine therapy benefit in the ABCSG 8 trial. Presented at the San Antonio Breast Cancer Symposium. December 13, 2008. Abstract 57.

8 Bartlett J, Munro A, Dunn J, et al. Chromosome 17 Polysomy (Ch17) as a Predictor of Anthracycline Response: Emerging Evidence from the UK NEAT Adjuvant Breast Cancer Trial. Presented at the 2008 annual San Antonio Breast Cancer Symposium. Abstract 45.

9 Bartlett J, Desmedt C, Munro A, et al. Chromosome 17 Polysomy: a Unifying Hypothesis Underlying Benefit from Adjuvant Anthracyclines? Presented at the 2008 annual San Antonio Breast Cancer Symposium. Abstract 6059.

10 de Boer M, van Deurzen CH, van Dijck JA, et al. Micrometastases and isolated tumor cells: relevant and robust or rubbish? (MIRROR): preliminary results of the MIRROR study from the Dutch breast cancer trialists’ group (BOOG). 2008 San Antonio Breast Cancer Symposium. Abstract 23.

11 Rakkhit R, Broglio K, Peintinger F, et al. Significant increased recurrence rates among breast cancer patients with HER2-positive, T1a,bNOMO tumors. 2008 San Antonio Breast Cancer Symposium. Abstract number 701.

12 Drew P, et al. The UK NIHR multicentre randomised COMICE trial of MRI planning for breast conserving treatment for breast cancer. Proceedings from the 2008 San Antonio Breast Cancer Symposium. Abstract 51.

Copyright © CancerConsultants. All Rights Reserved.