Search

1520 Background: 25 years of population-based screening have led to a sustained increase in the incidence of breast cancer, largely due to an increase in the rate of node-negative disease. A concom...

561 Background: Histologic grade is one of the strongest prognostic factors for breast cancer patients. In this study we assessed the prognostic value of the 70-gene prognosis signature (MammaPrint), validated for node-negative and -positive patients, for grade (G) 1-3 breast cancer. Methods: 1630 tumor samples of pT1-3 N0-1 breast cancer patients from 7 previously reported studies were analyzed and classified by MammaPrint as low or high risk. 10-year breast cancer-specific survival (BCSS) and distant disease-free survival (DDFS) was analyzed using Kaplan-Meier survival curves and multivariate Cox regression analyses. Results: Of 1,630 patients 392 (24%) were classified as G1 (21% 70-gene high risk), 646 (40%) as G2 (42% high risk) and 592 (36%) as G3 (86% high risk), respectively. At 10 years of follow-up, the 70-gene signature accurately distinguished low-risk from high-risk tumors for all histologic grades: within G1 tumors, absolute BCSS was 91% for 70-gene low risk and 82% for 70-gene high risk pati...

535 Background: According to the NCCN consensus guidelines molecular profiling for breast cancer prognosis may be used in patients with ER-positive, Her2-negative and LN-negative disease. Current proposed NCCN clinical risk assessment suggests adjuvant treatment for the majority of these patients. The 70-gene profile (MammaPrint) is validated as an independent prognostic indicator for patients with lymph node-negative and positive disease. Prognosis prediction by MammaPrint may be more suitable to indicate who needs adjuvant chemotherapy in addition to endocrine therapy. Methods: 566 tumor samples of women with ER-positive, Her2-negative and LN- negative breast cancer from 5 previously reported studies were analyzed and classified by MammaPrint as good or poor prognosis. 10-year breast cancer-specific survival (BCSS) was analyzed according to MammaPrint and the NCCN guidelines. Results: Median follow-up was 3.5 years (range 0.1–21.1). 380 of 566 patients (67%) were classified as good and 186 (33%) as poor prognosis by MammaPrint. Using the NCCN guidelines, 7% were classified as low and 93% as high risk respectively. 349 patients (62%) received no adjuvant treatment, 17% received hormonal treatment only, 2% chemotherapy only and 20% both respectively. At 10 years, BCSS was 91% vs. 67% for the good and poor prognosis groups (HR 4.0 [95%CI 2.0–7.9], p [Table: see text]

11090 Background: Adjuvant! is a web-based tool that calculates individualized 10-year survival probability and estimates the benefit of adjuvant systemic therapy based on age, co-morbidity, tumor size, tumor grade, number of involved lymph nodes and estrogen-receptor status. The Adjuvant! model was constructed using observed overall survival (OS) for women diagnosed with breast cancer between 1988 and 1992 recorded in the US SEER registry (surveillance, epidemiology and end results) and has not yet been validated in any European series. It is of interest to see whether intrinsic characteristics of tumors and survival of Dutch patients may differ from those in the US. Methods: For this validation all patients who were treated at the Netherlands Cancer Institute for primary breast cancer between 1987 and 1998 were selected according to the following criteria: T1–3 tumor, N0–3, M0, definitive primary surgery, complete axillary staging, and no prior malignancies. Follow-up for OS was complete for >99% of the patients until at least February 2007 (median follow-up 11.7 years). Clinicopathological characteristics as mentioned above and adjuvant treatment data were entered into Adjuvant! version 8.0 for each individual, blinded to the clinical outcome, to calculate predicted 10-year outcomes. Results: In this hospital-based cohort of 5381 patients, all known prognostic factors were significant predictors of overall survival. Across all patients, the 10-year OS was excellently predicted by Adjuvant! (predicted OS - observed OS=0.2%; p=n.s.). Also, for most relevant clinicopathological subgroups the differences between predicted and observed OS were within 2%. Adjuvant underestimated the OS in patients with tubular type of breast cancer (predicted - observed = -6.8; p [Table: see text]

610 Background: Magnetic Resonance Imaging (MRI) of the breast shows superior ability to visualize the extent of invasive breast cancer compared to conventional breast imaging. Nonetheless, MRI may under- or overestimate the extent of invasive disease, and the ability of MRI to depict components of disease around the primary invasive tumor is not well established. The purpose of this study was to precisely correlate MRI findings with histopathologic findings in breast cancer patients and to establish the incidence and quantity of surrounding MRI occult disease in breast cancer patients that are scheduled for breast-conserving therapy (BCT). Methods: Patients were prospectively included if they had biopsy-proven invasive breast cancer and the choice of treatment was BCT after pre-operative MRI. Pathology findings were spatially reconstructed and correlated with preoperative MRI. Tumors were stratified by absence or presence of an extensive intraductal component (EIC- or EIC+). The largest diameter of the MRI-visible lesion was compared with the largest diameter of the primary invasive tumor at pathology. Distances (mm) between the edge of the MRI-visible lesion and surrounding subclinical tumor foci (i.e., DCIS, invasive foci) were measured. At various distances from the edge of the MRI-visible tumor, the incidence of disease was determined. Results: 53 patients with 54 breast tumors were included. 42 tumors were EIC- and 12 were EIC+. The mean size (± SD) of the primary invasive tumor was 18.1 ± 7.5 mm on MRI and 19.5 ± 8.2 mm on pathology (Pearson's correlation coefficient: 0.75). The MRI-visible lesion was larger than or equal to the primary invasive tumor on pathology in 21 (39%) cases. Underestimation of the primary invasive tumor occurred up to 7 mm from the edge of the MRI-visible lesion. Beyond 10 mm, subclinical tumor foci were found in 48% off all tumors, in 36% and 92% of EIC- and EIC+ tumors (p < 0.001). Beyond 20 mm these rates were 19%, 7% and 67%, respectively (p < 0.001). Conclusions: Disease around MRI-visible lesions may be more extensive than assumed prior to treatment, especially in EIC+ tumors. This may have consequences for treatment margins in MRI-guided therapy of localized breast cancer. No significant financial relationships to disclose.

1525 Background: Since the advent of screening, breast cancer incidence has significantly increased in the United States as well as other countries. SEER data from the US shows that the increase is largely in localized disease without a concomitant or significant decrease in regionalized disease. In addition, the increase has occurred essentially in women over 50. We asked whether molecular characterization of tumors would shed light on the types of tumors detected in screening and in locally advanced breast cancers (LABC). Methods: We identified two groups of patients from European studies before and after the introduction of population based screened (in 1995) where the 70-gene prognosis test (MammaPrint) results were available. The first source is the European Validation Study (EVS) (Buyse et al, J Natl Cancer Inst. 2006). The second was a prospective implementation trial of MammaPrint in the Netherlands (RASTER study) (Bueno de Mesquita et al, Lancet Oncol. 2007). The source of LABC patients comes from the multicenter I-SPY TRIAL (CALGB 150007/150012) who underwent routine screening (prior mammogram within 2 years of diagnosis). Results: In women age 50–60, the fraction of cancers that were MammaPrint good prognosis were 40% and 60% prior to and after the introduction of screening, respectively. For women 50–60 who were actually undergoing screening, the fraction of tumors that were MammaPrint good prognosis was 67%. For patients aged 30–40 with stage I and II lymph node-negative disease, who did not undergo screening in either period, the fraction of MammaPrint good prognosis tumors did not change (29% and 31%, for EVS and Raster, respectively). For patients with LABC, from the I-SPY TRIAL, the fraction of tumors that are MammaPrint good prognosis is 7% and 29% for women aged 30–40 and 50–60, respectively. Of the LABC patients undergoing screening, 80% presented clinically during the interval between routine mammograms. Conclusions: Current screening programs are increasing the burden of low-risk cancers. Screening programs should consider including molecular profiles at the time of diagnosis to reduce overtreatment. The majority of LABCs have high-risk molecular profiles and do not present as screen detected cancers. New strategies are needed for early detection of LABC. [Table: see text]

512 Background: The 70-gene expression profile (MammaPrint) is validated as an independent prognostic indicator for breast cancer patients with T1–2 node-negative and positive disease regardless of estrogen receptor status. Here we present the relationship between MammaPrint outcome and chemotherapy benefit in the adjuvant setting. Methods: We performed a pooled analysis of 1,637 patients with MammaPrint outcomes (T1–2, node-negative and positive invasive breast cancer and median FU 7.1 yrs) to determine the chemotherapy benefit of patients treated with adjuvant chemotherapy in addition to endocrine therapy. Patients were collected from 7 large datasets at multiple institutions across Europe. Results: In this meta-analysis, MammaPrint assigned 772 patients (47%) to “low risk” and 865 (53%) to “high risk”. In total 349 patients were treated with endocrine therapy alone, whereas 226 were treated with both chemo- and endocrine therapy. Patients with poor prognosis MammaPrint profile had a substantial benefit from chemotherapy: At 5 years, distant disease-free survival was improved from 69% to 88% (HR 0.28 (95% CI 0.14–0.56, p [Table: see text]

518 Background: Small breast cancers are often considered as low risk tumors in clinico-pathologic risk assessment and treatment guidelines. However, since the ability to metastasize is an early and inherent genetic property of breast cancer and consequently even small tumors can metastasize, identification of prognostic subgroups for patients with small tumors would help to optimize treatment strategies. In the present study the 70-gene prognosis signature (MammaPrint), validated as independent prognostic indicator for node-negative and positive patients, was used to assess prognosis for pT1 breast cancer. Methods: 965 tumor samples of women with pT1 breast cancer from 7 previously reported studies were analyzed and classified by MammaPrint as good or poor prognosis. 10-year distant disease-free survival (DDFS) and breast cancer-specific survival (BCSS) was analyzed according to the 70-gene prognosis signature separately for pT1a/b (1–10mm) and pT1c (11–20mm) tumors. Results: Median follow-up was 7.1 years (range 0.2–25.2). 526 of 965 patients (55%) were classified as good prognosis and 439 (45%) as poor prognosis by the 70-gene signature. 562 patients (59%) received no adjuvant treatment, 19% received hormonal treatment only, 10% chemotherapy only and 12% both respectively. The 70-gene signature accurately predicted differences in 10-year DDFS (HR 2.7 [1.9–3.9], p [Table: see text]

2531 Background: Preoperative chemotherapy is increasingly employed to treat primary breast cancer, allowing an ‘in vivo chemosensitivity test’. Markers which predict a pathological complete response are urgently needed to refine this strategy. This study was conducted to evaluate the use of gene expression profiling to predict response to neoadjuvant anthracycline- or taxane-based chemotherapy. Methods: Patients with operable or locally advanced HER2-negative breast cancer received preoperative chemotherapy: either dose- dense doxorubicin and cyclophosphamide (ddAC) or capecitabine and docetaxel (CD). Core needle biopsies were taken before treatment and gene expression profiling was performed using 35k oligo microarrays. Results: Gene expression profiles were obtained from pretreatment biopsies of 63 tumors. 27% of the patients achieved a (near) pathologic complete remission (pCR), 40% of the patients had a partial remission and 33% of the patients did not respond to chemotherapy. Based on the gene expression profiles, tumors were assigned to the previously identified “molecular subtypes” luminal, basal-like or ERBB2-like (Sorlie et al., PNAS 98: 10869, 2001). 13 out of 25 patients with a basal-like tumor (52%) achieved a complete remission, whereas for the luminal tumors a pCR was only obtained in 2 out of 29 patients. Using four published gene expression classifiers of response to chemotherapy, a reasonable separation between responders and non-responders could be observed for two of these. We also performed exploratory supervised classification analyses on our dataset to identify a novel classifier. This resulted in a classifier for response to therapy irrespective of the chemotherapy regimen used and a second classifier specifically associated with response to ddAC chemotherapy. We will perform validation of these classifiers in samples from patients that are currently being enrolled in the study. Conclusions: Basal-like tumors have a better response to neoadjuvant chemotherapy as compared to other tumor types. The identification of robust gene expression signatures for better response prediction may require larger patient groups and should probably be established separately for each of the molecular subtypes of breast cancer. No significant financial relationships to disclose.

10507 Background: About 20% of ductal carcinoma in situ (DCIS) lesions diagnosed by stereotactic core needle biopsy (SCNB) are shown to be invasive on postoperative pathology examination. Sentinel lymph node biopsy (SLNB) is an accurate method of evaluating axillary lymph nodes in patients with invasive breast cancer. The aim of this study was to review our experience with lymphatic mapping in patients with a SCNB diagnosis of pure DCIS, to evaluate the DCIS underestimation rate and consequently the risk of lymph node metastases. Methods: Files from 160 patients diagnosed with pure DCIS by SCNB between July 1999 and March 2005 were retrieved from our database. Patients with DCIS were selected for SLNB if there was concern for presence of an invasive component on the basis of size, palpability or imaging. Results: The median age of the study group was 55 years (range 29–85 years) and median DCIS size on mammography was 25 mm (range 4–96 mm). Thirty-six (23%) out of the 160 women underwent a SLNB. Macrometastases in the sentinel node were detected in seven (19%) patients and one (3%) patient was found to have a micrometastasis. Twenty (56%) of these 36 patients had invasive lesions on final pathology. Of the 124 women who did not receive SLNB, 29 (24%) turned out to have invasive lesions on postoperative evaluation. In total, 49/160 (31%) patients with pure DCIS diagnosed by SCNB had invasive breast cancer (range pT1mic-pT2) on final pathology. Finally, 88 patients underwent lymphatic staging by SLNB, basal node sampling or complete axillary lymph node dissection. Nodal involvement was present in 14 (16%) out of these 88 patients with initially diagnosed DCIS: 36% in patients with invasive breast cancer and 2% with pure DCIS on final pathology. Conclusion: Postoperative pathology examination of the specimen demonstrates DCIS underestimation in nearly one third of SCNB diagnosed DCIS patients. SLNB is of benefit for these patients as in 19% of the patients who undergo a SLNB macrometastatic disease in the sentinel node can be found. No significant financial relationships to disclose.

20005 Background: Neoadjuvant chemotherapy is increasingly employed in operable breast cancer. Our initial studies on a cDNA array platform failed to identify gene expression patterns predicting response to neoadjuvant chemotherapy in breast cancer patients (J Clin Oncol 23:3331, 2005). Now we included more patients and used oligo microarrays. Methods: Patients with operable or locally advanced breast cancer were included in a randomized phase II study or received neoadjuvant chemotherapy off protocol. All except 7 patients began chemotherapy with 3 courses of dose-dense adriamycin and cyclophosphamide (ddAC) and response was evaluated by MRI. Patients with a response and a HER2-positive tumor were then randomized between either 3 additional courses of ddAC or six weekly courses of carboplatin, paclitaxel and trastuzumab (CPT). Patients without response were switched to CPT. Patients with HER2-negative tumors were randomized between 3 courses of either ddAC or capecitabine and docetaxel (CD). After evaluation, patients without response were switched to the alternative treatment arm. From all patients 14G core needle biopsies were taken before treatment and total RNA was isolated. Amplified mRNA was labeled and hybridized to 35k human oligo microarrays from our microarray facility. Results: So far, 77 patients have been included into the study. From 48 of these, good quality RNA from tissue with >50% tumor cells was isolated. 43 patients had received ddAC as initial chemotherapy; 32 of these had not been switched to another regimen. In a training set containing 11 pathological complete remissions (pCR) and 9 non-responders (NR) we could separate these groups by using 20 genes in a supervised classification and a 9-step cross validation. These results could be validated in an independent set of 11 samples (6 pCR, 5 NR). From 10 out of 11 samples, response status could be predicted correctly, independent from the treatment regimen. Although ER-positive tumors have a lower pCR rate than ER-negative ones, the steroid hormone receptors were not present in the classifier. Conclusions: We conclude that it should be possible to identify a reliable gene expression profile associated with response to adriamycin based neoadjuvant chemotherapy in breast cancer. No significant financial relationships to disclose.