Developments in Adjuvant and Neoadjuvant Treatment of Breast Cancer

A Report from the 29th Annual San Antonio Breast Cancer Symposium

Developments in Adjuvant and Neoadjuvant Treatment of Breast Cancer: A Report from the 29th Annual San Antonio Breast Cancer Symposium

Lee Schwartzberg, MD, FACP,The West Clinic

December 14-17, 2006San Antonio, Texas

Lee Schwartzberg, MD, FACP,The West Clinic

The San Antonio Breast Cancer Symposium of 2006 brought together over 8,000 people from all corners of the globe for the unifying purpose of sharing the current state-of-the-art on all aspects of breast cancer.  As usual, this years symposium devoted time to significant advances in the entire spectrum of the disease, including prevention, screening, prognostic factors, adjuvant chemotherapy, hormonal therapy and treatment of metastatic disease. Below is a discussion of results concerning adjuvant and neoadjuvant therapies.

Adjuvant Therapy

Breast cancer is now clearly seen as several different diseases molecularly defined even within an apparent uniform histology.  Increasing evidence suggests each of these diseases demands a different approach to treatment for greatest success:  hormone receptor positive (ER+ PR+) tumors requiring predominantly endocrine therapy; HER2/neu + tumors requiring a HER2 targeted agent; and triple negative ER­- PR HER2 or the so-called basal type requiring aggressive chemotherapy.  The use of Herceptin® (trastuzumab) is now firmly established in the treatment of HER2 positive breast cancer, as is the use of dose-dense chemotherapy in high-risk patients. Neoadjuvant treatment is commonly used to downstage primary tumors and identify biomarkers predictive of benefit. Each of these therapeutic options had further exploration at this years SABCS.

HER 2+ Adjuvant Treatment

Eagerly awaited was the second interim analysis of the BCIRG-006 study, a three arm trial comparing Adriamycin® (doxorubicin) and Cytoxan® (cyclophosphamide) followed by sequential Taxotere® (docetaxel) (AC-T) as a control arm, to two experimental arms:  1) AC and sequential Taxotere plus Herceptin for one year (AC-TH) and 2) a novel non-anthracycline regimen, Taxotere and carboplatinum plus Herceptin (TCH) followed by Herceptin for one year.[[1]]( "_ednref1") Patients with positive lymph nodes or who were high-risk, node-negative (29% of the population) were eligible, and over 3,000 patients were randomized.  Median follow up is now 36 months.  The disease-free survival at four years is 83% for AC-TH, 82% for TCH and 77% for AC-T.  The hazard ratio for the addition of Herceptin is 0.61 (AC-TH) or 0.67 (TCH).  No significant difference in the two Herceptin arms exist (p = 0.12).  Overall survival is 92% for AC-TH, HR=0.59; 91% for TCH, HR=0.66; and 86% for ACT.  

Interestingly, the authors reported data separately for the 900 women with lymph node negative (LN-) disease at entry.  AC-TH had a hazard ratio of 0.32 (p = 0.007) and TCH, HR=0.47; p = 0.0096, demonstrating that the benefit of Herceptin in LN- patients is at least as proportionally good as in higher risk patients.  In terms of safety, TCH was the least toxic regimen with significantly less anthracycline-related symptoms as hand-foot syndrome, stomatitis, vomiting, neuropathy, myalgia and nail changes than AC-TH.  Importantly, the lesser degree of cardiac dysfunction seen with TCH in the first interim analysis persisted with only 4 patients experiencing Grade 3/4 LV dysfunction compared to 20 in the AC-TH arm (p = 0.005) and 4 in the AC-T arm. 

Whereas in the first interim analysis the 35% of patients with coamplification of both topo-isomerase II and HER2 showed benefit of AC-TH compared to either TCH or AC-T, more complete and longer follow up now shows a coming together of the curves.  However, for the two-thirds of patients with no co-amplification, there remains a marked improvement for patients in either of the two Herceptin arms.  These interesting results suggest that topo-isomerase II coamplification in the presence of HER2 amplification promotes some form of resistance to Herceptin. Overall, given the equivalent efficacy and reduced toxicity of TCH, this study supports a practice changing paradigm: the use of a non-anthracycline regimen in the adjuvant treatment of HER2 positive breast cancer patients. 

BCIRG-006, like most other adjuvant Herceptin trials, utilizes maintenance Herceptin to complete one year of therapy after chemotherapy.  The choice of this schedule and duration remains entirely empirical.  In contrast, the small Fin-HER study demonstrated that a 9-week course of Herceptin concurrently with chemotherapy could provide similar proportionally benefit as seen in the larger trials.[[2]]( "_ednref2") ECOG E-2918, another smaller pilot trial, randomized 227 HER2+ patients between short course TH-AC vs. long course TH-AC-H for 52 weeks in lymph-node positive patients.[[3]]( "_ednref3") The primary endpoint was the rate of congestive heart failure between arms. There was no difference in CHF: 3 in the short course and 4 in the long course arm.  Six cardiac events occurred <2 years after treatment initiation.  While the disease-free survival and overall survival were not primary endpoints of the study, at five years 76% of the short course H and 75% of long course H were disease-free, while 89% of short course H and 83% of long course H were alive.  Results were even better in the approximately 80% of patients with 3+ IHC or positive FISH results.  The study suggests that much of the benefit of Herceptin can be derived from a brief course in conjunction with chemotherapy; however, no benefit in terms of long-term CHF reduction was seen with an attenuation of the Herceptin schedule. 

A meta-analysis of the effect of HER2 status in studies comparing an anthracycline containing-arm and a non-anthracycline arm was reported.[[4]]( "_ednref4") Overall, across seven studies, the use of anthracycline provided a 10% advantage in both DFS and OS.  In the subset of HER2+ patients (generally obtained retrospectively), the HR for anthracycline in HER2+ was 0.71 for DFS, p=<0.001 and for overall survival HR=0.73; p=<0.001.  In contrast, in HER2- patients in these trials, the HR was around unity for both DFS and OS.  These results suggest that the majority of benefit from anthracyclines may be in the HER2+ patients, perhaps representing a surrogate marker for topo-II amplification.  Furthermore, it supports the concept that anthracyclines do not necessarily benefit the 75% of patients who are HER2- with the possible exception of the small group with topo-II amplification but HER2 non-overexpressed. 

Dose-Dense Chemotherapy

Dose-dense chemotherapy (DDC), a concept in which total doses and cycle doses of drugs are the same but the cycle length is compressed, is based on a mathematical model predicting better cancer cell kill and outcome.  This notion has been clinically proven in the landmark cooperative group study C9741, reported in final form at SABCS 2005.[[5]]( "_ednref5") Confirmatory studies have been highly anticipated and several were reported at this years meeting.  

The German AGO group reported interim analysis of a randomized Phase III trial comparing standard epirubicin and cyclophosphamide followed by Taxol® (paclitaxel), (EC-T) at three-week intervals to sequential dose dense epirubicin, Taxol, and cyclophosphamide (DD ETC).[[6]]( "_ednref6") This study was not a pure example of dose density as the individual doses in the sequential dose dense arms were also higher than in the standard dose arms, and the number of cycles were not equivalent between arms.  Nevertheless, in this trial of high risk (> 4 positive lymph nodes, 25% HER2+) patients, there was a clear 5-year RFS advantage for the DD ETC arm: 70% compared to 62% for EC-T, HR-0.72, p=0.0079 and OS, 82% vs. 77%, HR 0.76, p=0.29.  Toxicity, however, was greater for the DD ETC arm, including leukopenia, anemia, thrombocytopenia and febrile neutropenia.  Moreover, nonhematologic toxicity was also greater for DD ETC, particularly in regard to infection, stomatitis and pain. 

These results contrast with C-9741 where neutropenic events were actually less common in the dose-dense arm.  It is likely that the use of cyclophosphamide 2500 mg/m2 and epirubicin at 150 mg/m2 in DD ETC caused most of the hematologic toxicity, while multiple previous trials have failed to show benefit of dose escalation of either anthracyclines or cyclophosphamide above standard doses. Therefore, the higher doses may have only increased toxicity but not efficacy. Despite greater toxicity in the AGO trial, the improvement in survival for the dose dense arm is impressive and supports the general concept of dose density in high risk women.  

Another study addressing the use of dose-dense chemotherapy, NCIC MA.21, was the subject of an interim analysis.[[7]]( "_ednref7") This Phase III, three arm trial study  randomized patients to CEF (with oral cyclophosphamide) on a day 1, 8 epirubicin and 5-FU schedule every 28 days, dose-dense EC followed by Taxol at 14 day intervals (DD EC-T), or standard AC-T every 21 days.  High risk node-negative or LN+ patients under the age of 60 were eligible.  About one-third of patients had > 4 positive lymph nodes. The primary finding was the inferiority of AC-T to either dose-dense EC-T or FEC with 3-year RFS, 90% for CEF; 89.5% for EC-T and 85% for AC-T (p=0.001).  No significant difference in HR for CEF vs. DD EC-T was detected. The benefit for CEF or dose-dense EC-T was most pronounced in hormone receptor negative patients.  Toxicity was generally higher for CEF, both in terms of febrile neutropenia and cardiac toxicity. Both epirubicin-containing regimens had higher acute venous thromboembolic events than AC-T and both Taxol-containing arms had much higher any grade sensory neuropathy. Interestingly, CEF was also associated with neuropathy (26.3% vs. approximately 70% for Taxol regimens).  While final results for this trial will be available in the future, the head-to-head comparison of dose density clearly confirms the results of 9741.

Given the emerging benefit of dose-dense chemotherapy, investigators at Memorial Sloan-Kettering Cancer Center wondered about the safety of adding Herceptin to DDC as given in C9741.  An interim analysis reported on 70 HER2+ patients with normal LVEF and no prior cardiac history received DDC with Neulasta® (pegfilgrastim) support and Herceptin for one year starting concurrently with Taxol.[[8]]( "_ednref8") Four patients discontinued treatment for cardiac events: one had clinical CHF and three for asymptomatic LVEF decline at six months after start of chemotherapy.  All three patients with asymptomatic LVEF decline had recovery to normal levels. The authors concluded that Herceptin can be safely added to DDC without any increasing cardiac events.  

Taken together, there is now abundant evidence of the benefit of dose density in improving outcome in the adjuvant treatment of early stage breast cancer.  Many questions remain: are sequential single agents less toxic than combinations?  Can dose density be increased further?  How can short and long-term toxicity be lessened?  What is the incremental benefit of dose-dense Taxol vs. weekly treatments? 

Neoadjuvant Chemotherapy

Perhaps the most eagerly anticipated data at SABCS 2006 was the presentation of Tykerb® (lapatinib) and Taxol in newly diagnosed inflammatory breast cancer.[[9]]( "_ednref9") Tykerb is a novel, oral, small molecule inhibitor of both the EGFR receptor (ERbB1) and HER2/neu.  Recently published results confirm the activity of Tykerb in Herceptin-refractory MBC.[[10]]( "_ednref10") Inflammatory breast cancer (IBC) often overexpresses HER2, and ERbB1 and is a very aggressive form of the disease.  Thirty-five patients with IBC were treated with Tykerb 1500 mg/day for two weeks and then had Taxol weekly added for 12 additional weeks.  After 14 weeks of therapy, patients underwent surgery.  The primary endpoint was pathologic CR rate.  Twenty-six percent of the women enrolled had Stage IV disease. Overall, 3/21 patients had a pCR in breast and lymph nodes, including 3/18 with HER2+ and 0/3 without overexpression of HER2.  Thirty percent of patients responded to Tykerb alone by day 14, and, overall, 78% of patients responded to the combination therapy.  Grade > 3 diarrhea was seen in 60% of patients, but other grade 3 toxicity was uncommon including nausea, vomiting, rash, fatigue and asthenia. Preliminary biomarker analysis suggested response to Tykerb correlated with expression of insulin-like growth factor I or deficiency in the PTEN pathway. Further studies with Tykerb in the neoadjuvant, adjuvant, and metastatic setting are well underway. 

A novel approach to neoadjuvant therapy was presented by the Dana-Farber group.[[11]]( "_ednref11") Based on the observation that most patients with BRCA1 mutations have triple negative (ER-, PR-, HER2-), basal-like breast cancer and that BRCA1 is an enzyme which normally functions to repair double stranded DNA breaks, they reasoned that cisplatinum, of which the mechanism of action involves creating double strand breaks, would be effective in these tumors.  

Twenty-eight patients with clinical Stage II/III triple negative breast cancer received cisplatinum as a single agent at 75 mg/m2 every 21 days for 4 cycles.  Six (22%) achieved pCR and 3 had progressive disease. Both of the women with documented BRCA1 mutation achieved pCR. The toxicity of the single drug was modest. These results point towards a developmental strategy including platinum agents in the treatment of basal type breast cancer and perhaps particularly in those patients with BRCA1 mutations.

Final results of the German GEPAR-trio study were the subject of a presentation.[[12]]( "_ednref12") This complex neoadjuvant study assessed response to two cycles of TAC and then assigned patients to further therapy based on their clinical response. Those with minimal response were randomized to continue TAC to complete eight cycles vs. switching after four more cycles of TAC to a combination of Navelbine® (vinorelbine) and Xeloda (NX).  Those with clinical partial response or better to two cycles of TAC were randomized to complete a total of six or eight cycles of the same regimen. This large study enrolled 2,000 women.  Two-thirds were responders to two cycles of TAC.  Interestingly, nonresponders were significantly more likely to have lower grade tumors, 72.6% vs. 61.3%; be HR positive, 68.5% vs. 58.3%; or have locally advanced tumors, 19.4% vs. 13.0%; while there was no difference in response by the HER2 status, age, tumor size or nodal involvement.

For clinical responders after two cycles of TAC, the pCR rate was 21-23.5% after further TAC with no significant benefit of prolonged therapy after six cycles.  In contrast, nonresponders had a 5.7-6.0% pCR rate with no benefit to switching a putatively noncross-resistant regiment of NX.  This large study allowed for multivariate analysis demonstrating that young age, high grade tumor, ductal histology, hormone receptor negativity and a complete clinical response after two cycles of TAC were all predictive of ultimate pathologic complete response.  The lack of benefit from additional drugs in patients with suboptimal response to the three most active agents in breast cancer suggest the need for a different approach, perhaps with biological agents added, in future trials.

Hope for the Future

Two presentations derived from retrospective analyses of large databases give a glimpse of what can be expected in the future of breast cancer.  Sir Richard Peto presented the first results of the Early Breast Cancer Trial 1st Collaborative Group latest five year analysis for 2005-2006.[[13]]( "_ednref13") The Oxford review, as it is known, now comprises over 660,000 patients enrolled in clinical trials since 1984 and provides a broad canvas meta-analysis of the effects of breast cancer treatment.

The group has demonstrated that the large decrease in breast cancer mortality that has occurred since 1990 in women under the age of 70 is due to the accumulated effect of multiple small to moderate improvements in treatment.  Trends emerging from the most recent analysis include the benefit of adding radiation therapy to adjuvant chemotherapy for all lymph node positive patients and the benefit of anthracycline-based chemotherapy vs. CMF.  The 15-year recurrence rate difference for patients <50 years of age who receive chemotherapy is about 12%, particularly prominent in ER-negative patients. The authors predict that the overall deaths from breast cancer in 2010 will be one-half of what they were in 1990 in the United Kingdom and the United States. 

Ravdin and colleagues presented exciting and encouraging data from the most recent analysis from the SEER database utilizing incidence data in invasive breast cancer through 2004.[[14]]( "_ednref14") There was a significant decrease in the number of breast cancer cases diagnosed in 2003 and 2004 compared to an essentially flat line incidence from 1998 to 2002. Virtually all of the decrease was in ER-positive cases.  Overall, there was a 7% decrease in age adjusted breast cancer incidence in 2003 relative to 2002, which translated into almost 14,000 fewer patients diagnosed.  Subgroup analyses demonstrated that almost all of the reduction was accounted for by women > 50 years of age.

What factors underlie this marked decrease?  The authors postulated two leading possibilities: (1) change in mammographic screening levels or (2) change in the use of hormonal therapy.  There was virtually no change in mammographic screening in 2003.  However, the publication of the Womens Health Initiative results in July 2002 led to a dramatic 50% reduction in hormonal therapy usage within 6 months with further decline in hormonal therapy usage by women through 2004.  The prior use and reduction in use was mainly seen in women > 50 years of age.  These trends support the concept that cessation of hormone replacement therapy may keep ER+ tumors from progressing from a pre-clinical state to a clinically apparent one.  Obviously this potentially exciting result will require both more study and more time to assess if 2003 was merely an aberration or hopefully the beginning of a new era of reducing the burden of breast cancer. 

Oncotype DX

Results from continued clinical studies evaluating Oncotype DX in different roles in the setting of adjuvant breast cancer continue to demonstrate its effectiveness in providing more individualized treatment options for patients with early breast cancer.

Oncotype DX is a CLIA-approved reverse-transcription polymerase chain reaction (RT-PCR) test that may be useful in determining which patients with newly diagnosed, Stage I or II, node-negative, estrogen receptor-positive breast cancer require adjuvant chemotherapy in addition to anti-estrogen therapy. Oncotype DX predicts the risk of a patient experiencing a recurrence 10 years following diagnosis utilizing formalin-fixed paraffin-embedded tumor tissue. Oncotype DX comprises a panel of 21 genes (16 cancer-related genes and 5 control genes), derived from 3 studies, to accomplish a recurrence-score algorithm. Oncotype DX also helps predict the likelihood of the benefit of adjuvant chemotherapy in these patients.

Results from previous trials have also demonstrated that Oncotype DX may help predict anti-cancer responses to neoadjuvant chemotherapy with Taxotere® (docetaxel) as well as changing treatment decisions in at least one-quarter of women with early breast cancer. Oncotype DX is already covered by Medicare for eligible patients and continues to be evaluated in clinical trials to further define its role.

According to results recently presented at the 2006 annual San Antonio Breast Cancer Symposium (SABCS), Oncotype DX more accurately predicted  estrogen receptor (ER) status when compared to immunohistochemistry or ligand binding (LB) testing for ER status among node-negative breast cancer patients.[[15]]( "_ednref1") In addition, quantitative analysis by Oncotype DX of ER and progesterone-receptor (PR) status demonstrates differing prognosis and prediction of benefit from Nolvadex® (tamoxifen) in this patient population.[[16]]( "_ednref2")

Estrogen receptor status is an important component in determining optimal treatment for patients with breast cancer. However, the power of different testing methods for ER expression, such as IHC or LB testing varies considerably. 

The first study presented at 2006 SABCS was conducted by researchers affiliated with the National Surgical Adjuvant Breast and Bowel Project (NSABP) to evaluate the effectiveness of results from Oncotype DX in accurately measuring ER expression. This trial included 297 node-negative, ER-positive breast cancer patients who had already received treatment with tamoxifen in the prior NSABP B-14 study. The NSABP B-14 trial had randomized node-negative, ER-positive breast cancer patients to tamoxifen or placebo. 

To test for ER expression, study sites had performed LB testing for ER at the time of patient enrollment, the NSABP Foundation Laboratory performed IHC with the DakoCytomation ER/PR pharmDx, and one of the genes from the 21-gene panel of Oncotype DX was used. The time to a distant recurrence among patients was an endpoint used to measure the accuracy of ER expression among the 3 tests. 

  • Higher ER expression with all three testing methods was associated with a greater time to a distant recurrence.
  • Quantitative measurement from Oncotype DX had the strongest association with time to cancer recurrence of all the three testing methods; p<0.0001 for Oncotype DX; p=0.003 for IHC; p=0.045 for LB testing. 

The researchers concluded that results from Oncotype DX are more accurate in quantifying ER expression than IHC and LB in node-negative breast cancer patients. One of the researchers stated that results from this study suggest Oncotype DX may be the most precise way to measure this widely used biomarker, and could possibly expand how we use the Oncotype DX assay in the future. 

A second study presented at the 2006 SABCS evaluated the prognostic and predictive associations between ER and PR analysis of Oncotype DX. This study included the examination of results from the NSABP-14 trial and the Kaiser Oncotype DX studies.  The Kaiser study included node-negative breast cancer patients in which the association of gene expression and breast cancer death were evaluated.

In ER-positive patients in both the NSABP and Kaiser studies, the level of expression of ER is significantly predictive of benefit from tamoxifen, but is not significantly associated with the prognosis of untreated patients.

The level of PR expression is not significantly predictive of benefit from tamoxifen, but is significantly associated with the prognosis of patients. 

The researchers from this study concluded that quantitative ER and PR expression as measured through Oncotype DX provide different and distinct associations with the prognosis and prediction of the benefit from tamoxifen among patients with node-negative breast cancer.


[[1]]( "_edn1") Slamon D, Eiermann W, Robert N, et al.  BCIRG 006: 2nd interim analysis phase III randomized trial comparing doxorubicin and cyclophosphamide followed by docetaxel (AC-T) with doxorubicin and cyclophosphamide followed by docetaxel and trastuzumab (AC-TH) with docetaxel, carboplatin and trastuzumab (TC) in Her2neu positive early breast cancer patients. 29th Annual San Antonio Breast Cancer Symposium. San Antonio, Texas. 2006; Abstract 52.

[[2]]( "_edn2") Joensuu H, Kellokumpu-Lehtinen PL, Bono P, et al. Adjuvant docetaxel or vinorelbine with or without trastuzumab for breast cancer. New Engl J Med 2006; 354:809-820.

[[3]]( "_edn3") Sledge G, ONeill A, Thor A, et al.  Adjuvant trastuzumab: long-term results of E2198.  29th Annual San Antonio Breast Cancer Symposium. San Antonio, Texas. 2006; Abstract 2075.

[[4]]( "_edn4") Gennari A, Sormani M, Puntoni M, et al.  A Pooled analysis on the interaction between HER-2 expression and responsiveness of breast cancer to adjuvant chemotherapy.  29th Annual San Antonio Breast Cancer Symposium. San Antonio, Texas. 2006; Abstract 41.

[[5]]( "_edn5") Hudis C, Citron M, Berry D, et al.  Five year follow-up of INT C9741: dose dense (DD) chemotherapy (CRx) is safe and effective.  Proceedings of the 28th Annual San Antonio Breast Cancer Symposium, San Antonio Texas. 2005; Abstract 41.

[[6]]( "_edn6") Moebus V, Lueck H, Thomssen C, et al.  Dose-dense sequential chemotherapy with epirubicin (E), paclitaxel (T) and cyclophosphamide (C) (ETC) in comparison to conventional schedule chemotherapy in high-risk breast cancer patients (> 4+ LN). Mature results of an AGO-trial).  29th Annual San Antonio Breast Cancer Symposium. San Antonio, Texas. 2006; Abstract 43.

[7] Burnell M, Levine M, Chapman J, et al.  A randomized trial of CEF versus dose dense EC followed by paclitaxel versus AC followed by paclitaxel in women with node positive or high risk node negative breast cancer, NCIC CTG MA.21: Results of interim analysis.  29th Annual San Antonio Breast Cancer Symposium. San Antonio, Texas. 2006; Abstract 53.

[[8]]( "_edn8") Dang C, Smith K, Fornier M, et al.  Mature cardiac safety results of dose-dense (DD) doxorubicin and cyclophosphamide (AC) followed by paclitaxel (T) with trastuzumab (H) in HER2/neu overexpressed/amplified breast cancer (BCA). 29th Annual San Antonio Breast Cancer Symposium. San Antonio, Texas. 2006; Abstract 2101.

[[9]]( "_edn9") Cristofanilli M,Boussen H, Baselga J, et al.  A phase II combination study of lapatinib and paclitaxel as a neoadjuvant therapy in patients with newly diagnosed inflammatory breast cancer (IBC).  29th Annual San Antonio Breast Cancer Symposium. San Antonio, Texas. 2006; Abstract 1. 

[[10]]( "_edn10") Geyer C, Forster J, Lindquist D, et al.  Lapatinib plus capecitabine for HER2-positive advanced breast cancer.  N Engl J Med 2006 Dec 28;355(26):2733-43.

[[11]]( "_edn11") Garber J, Richardson A, Harris L, et al.  Neoadjuvant cisplatin (CDDP) in triple-negative breast cancer (BC). 29th Annual San Antonio Breast Cancer Symposium. San Antonio, Texas. 2006; Abstract 3074.

[[12]]( "_edn12") v Minckwitz G, Kuemmel S, du Bois A, et al.  Individualized treatment strategies according to in vivo chemosensitivity assessed by response after 2 cycles of neoadjuvant chemotherapy.  Final results of the Gepartrio study of German breast group.  29th Annual San Antonio Breast Cancer Symposium. 2006; Abstract 42.

[13] EBCTCG Secretariat, on Behalf of EBCTCG, University of Oxford, Oxford, United Kingdom.  Highlights from the early breast cancer trialists collaborative group (EBCTCG) 205-2006 worldwide overview.  29th Annual San Antonio Breast Cancer Symposium. San Antonio, Texas. 2006; Abstract 40. 

[[14]]( "_edn14") Ravdin P, Cronin K, Howlander N, et al.  A Sharp decrease in breast cancer incidence in the United States in 2003.  29th Annual San Antonio Breast Cancer Symposium. San Antonio, Texas. 2006; Abstract 5. 

[[15]]( "_edn1") Kim C, Tang G, Baehner F, et al. A Comparison of Estrogen Receptor (ER) Measurement by Three Methods in Node-Negative, Estrogen Receptor (ER)-Positive Breast Cancer: Ligand Binding (LB), Immunohistochemistry (IHC), and Quantitative RT-PCR. Proceedings from the 2006 annual San Antonio Breast Cancer Symposium. San Antonio, Texas. 2006. Abstract #3116.

[16] Baehner F, Habel L, Quesenberry C, et al. Quantitative RT-PCR Analysis of ER and PR by Oncotype DX Indicates Distinct and Different Associations with Prognosis and Prediction of Tamoxifen Benefit. Proceedings from the 2006 annual San Antonio Breast Cancer Symposium. San Antonio, Texas. Oral presentation December 17, 2006. Abstract #45.