Citations

Abstract Abstract #6074 Objective: To analyze blood-serum biomarkers (HER2, EGFR, uPA, and TIMP-1) for prediction of response to treatment in conjunction with Study 96-32-55, a multicenter phase II trial assessing the efficacy and tolerability of two irinotecan schedules in anthracycline- or taxane-refractory MBC patients.
 Methods: MBC patients who experienced disease progression after one to three chemotherapy regimens, including at least one anthracycline- or taxane-based regimen, were randomly assigned to irinotecan in 6-week cycles comprising 100 mg/m2 weekly for 4 weeks, then a 2-week rest or 240 mg/m2 every 3 weeks. During this study the serum samples from each patient were collected at a possibility of three time points; prior to treatment, at the first occurrence of response, and at completion of or withdrawal from treatment. Response prediction was analyzed using predetermined elevated/non-elevated serum biomarker cutoffs. Percentage changes from baseline to first response and disease progression were analyzed.
 Results: In the weekly arm, the objective response (CR+PR) rate was 23% (95% CI, 13% to 37%), and in the every-3-weeks arm, the objective response rate was 14% (95% CI, 6% to 26%). Serum biomarker levels were determined for HER2, EGFR, uPA, and TIMP-1 prior to treatment for 91 patients. Of these 91 patients, 17 had serum measurements at first response, and 38 had serum measurements at their completion of study due to disease progression. Only 7 patients had serum-levels collected at all three event-points. The baseline levels of sHER2, EGFR, uPA, and TIMP-1 were not different among responders and non-responders (Fisher's Exact p=0.41, 0.26, 0.68, 0.75). sHER2 level increased by 20.6% from baseline to disease progression (p=0.01). TIMP-1 level was 15.2% lower than baseline at first response (p=0.03), 16.5% higher than baseline at disease progression (p=0.01), and for 7 patients with all three event-point measures, TIMP-1 was 21.8% lower than baseline at first response and was 22.3% higher than baseline at disease progression (p=0.02, 0.04).
 Conclusions: In this study, serum levels of HER2, EGFR, uPA, and TIMP-1 do not appear to be predictive of response. However, interpretation of the data is compromised due to limited availability of serum at all 3 points. Serum HER2 appears higher at disease progression than at baseline. Serum TIMP-1 appears to decrease from baseline to first response, and then appears to increase at disease progression. Citation Information: Cancer Res 2009;69(2 Suppl):Abstract nr 6074.

Abstract Abstract #6037 Background: Tamoxifen (Tam) is biotransformed to the potent antiestrogen, endoxifen, by the CYP2D6 enzyme. We previously demonstrated that patients (pts) receiving adjuvant TAM with impaired CYP2D6 metabolism due to CYP2D6 (*4) and/or concurrent administration of a CYP2D6 inhibitor had a higher risk of recurrence. Other studies suggest CYP2C19*17 (Schroth JCO 2007) and SULT1A1*2 (Nowell JNCI 2002) may be associated with treatment outcome. With extended follow-up, we sought to evaluate the importance of comprehensive CYP2D6 genotyping as well as the potential association between CYP2C19*17 and SULT1A1 copy number with clinical outcome in pts randomized to TAM in NCCTG 89-30-52.
 Methods: Using DNA derived from paraffin embedded sections, CYP2D6 and SULT1A1 genotype were determined using quantitative multiplex PCR and CYP2C19 by sequencing. Pts administered the following CYP2D6 inhibitors (fluoxetine, paroxetine, sertraline, cimetidine, amiodarone, doxepin, ticlopidine and haloperidol) were considered as intermediate (IM) or poor metabolizers (PM) based on the potency of CYP2D6 inhibition. CYP2D6 phenotype was defined as follows: extensive metabolizers (EM) were pts not administered an inhibitor who did not carry a null allele (*3, *4, *6) and who were not homozygous for an IM allele (*10, *17, *41). CYP2D6 IM were either heterozygous for a null allele or homozygous for an IM allele but not administered an inhibitor or CYP2D6 EM administered a weak inhibitor. PM were pts homozygous for a null allele, or any patient administered a potent inhibitor. The association between genotype or CYP2D6 phenotype and clinical outcome was determined using the log-rank test. Multivariate Cox modeling was performed using traditional prognostic factors.
 Results: The median follow-up of living pts is now 14.5 years. Of 256 pts. randomized to TAM, genotype was determined for CYP2D6 (n=210), CYP2C19*17 (n=170), and SULT1A1 (n=169) with the following allele frequencies: CYP2D6 *3 (.02), *4(.20), *6 (.01), *10 (.03), *17 (.00), *41(.08), and CYP2C19*17 (.22). The frequency of SULT1A1 copy number alleles (CNA) was 1 (4%), 2 (67%), 3 (20%), and 4+ (9%). 14/227 patients (6%) were administered an inhibitor. A multivariate analysis accounting for nodal status and tumor size demonstrated that compared to CYP2D6 EM, CYP2D6 PM had significantly shorter time to recurrence (TTR) (HR 4.0, p=0.001) and DFS (HR 2.0, p=0.02) and CYP2D6 IM tended to have shorter TTR (HR 1.8, p=0.08) and DFS (HR 1.4, p=0.10). DFS did not differ by SULT1A1 copy number (p=0.62) or CYP2C19 *17 (p=0.47)
 Conclusion: Long term follow-up of pts in NCCTG 89-30-52 confirms the importance of CYP2D6 metabolism and further demonstrates the importance of comprehensive genotyping and phenotyping with CYP2D6 inhibitors. We could not identify an association between CYP2C19*17 or SULT1A1 copy number with recurrence but further evaluation is needed in larger cohorts. Citation Information: Cancer Res 2009;69(2 Suppl):Abstract nr 6037.

Abstract Abstract #57 Background: ABCSG 8 is the only prospective trial examining the value of switching patients (pts) to an aromatase inhibitor after tamoxifen (Tam) in which randomization occurred prior to therapy. This trial demonstrated that 2 years (yrs) of Tam followed by 3 yrs of anastrozole (TAM to A) was superior to 5 yrs of Tam (Jakesz SABC 2005). Tam is metabolically activated by CYP2D6 to endoxifen and pts with impaired CYP2D6 metabolism have a higher risk of recurrence (Goetz JCO 2005). The HOXB13/IL17BR (H/I) index (Ma Cancer Cell 2004) is associated with Tam resistance while a 5-gene molecular grade index (MGI) distinguishes outcome in grade II tumors. H/I and MGI are complementary prognostic factors that define low (low MGI), intermediate (low H/I, high MGI) and high risk groups (high H/I and MGI) (Ma Clin Cancer Res 2008). We sought to validate these biomarkers in ABCSG Trial 8, and hypothesized that they may predict the benefit of sequencing TAM to A versus Tam alone.
 Methods: From 1996-2004, ABCSG 8 randomized 3,901 pts with ER positive breast cancer. The primary endpoint was event-free survival. Assuming 40% of pts had reduced CYP2D6 metabolism, we designed a nested case control study in which 110 Tam treated cases matched to 2 controls would yield 80% power to detect a relative risk of 2.3 for pts with reduced CYP2D6 metabolism. Additionally, we sought to perform an exploratory analysis of the benefit of switching from Tam to A in pts with impaired vs non-impaired CYP2D6 metabolism. DNA and RNA derived from paraffin embedded sections were used to genotype for CYP2D6 and to obtain H/I and MGI RT-PCR profiles.
 Findings: Thus far, CYP2D6 genotyping is complete in 345 pts corresponding to 53 cases/106 controls (Tam) and 62 cases/124 controls (Tam to A). DNA and RNA yields were excellent with a CYP2D6 genotype call rate of >95%. CYP2D6 allele frequencies were as expected: *3 (.01), *4, (.23), *6 (.01), *10 (.02), *17 (.00), and *41 (.10). 44% of pts had reduced CYP2D6 metabolism (at least one CYP2D6 null allele or intermediate metabolizer) and 8% were CYP2D6 poor metabolizers. For H/I and MGI, 340 pts were profiled with 64%, 20% and 16% in the low, intermediate, and high risk groups, respectively. Tissue sectioning/genotyping for additional cases/controls is ongoing.
 Discussion: This is the first retrospective analysis of a large Tam versus Tam to A trial to evaluate the association between 1) CYP2D6, HOXB13/IL17BR, and MGI with TAM treatment outcome and 2) the predictive benefit of these biomarkers in pts randomized to switch from Tam to AI versus Tam alone. Final results will be presented at the meeting. Citation Information: Cancer Res 2009;69(2 Suppl):Abstract nr 57.

Abstract Abstract #56 Background: Recent findings from NSABP B-31 suggested that breast cancer (BC) patients (pts) with c-MYC (MYC)/HER2 co-amplification in their primary tumors may benefit more in terms of disease-free survival (DFS) from adjuvant trastuzumab (n=471; p<0.0001, HR 0.24; 13 vs. 51 events) than those pts with only HER2 amplification in their primary tumors (n=1078, p=0.007, HR 0.63; 55 vs. 82 events).
 Purpose: The overall objective of this analysis is to determine the association between MYC/HER2 co-amplification and outcome of pts randomized to receive chemotherapy alone (Arm A) or chemotherapy with concurrent trastuzumab (Arm C) on N9831.
 Patients/Methods: This analysis included 868 pts from the HER2+ NCCTG N9831 Intergroup adjuvant trastuzumab phase III clinical trial on Arms A or C. Fluorescent in situ hybridization (FISH) was performed at a central laboratory, Mayo Clinic Clinical Cytogenetics Laboratory, using dual probe mixtures for HER2 (17q11.2-12/centromere 17) and for MYC (8q24/centromere 8). The HER2 amplification status in whole tissue sections was evaluated to determine eligibility of pts for N9831. Tissue microarrays containing three cores from each patient's tumor were analyzed for MYC amplification status. The gene and centromere copy number as well as the distribution of the signals for each marker were determined in sixty non-overlapping interphase nuclei.
 Results: Of 273 pts with completed FISH analyses, 34 (12%) pts had MYC amplification. The rate of MYC amplification was similar across arms (18/135 [13%] vs. 16/138 [12%]; ChiS p=0.66). Co-amplification of MYC/HER2 was observed in 32 (12%) pts with similar co-amplification rates across arms (17/135 [13%] vs. 15/138 [11%]; ChiS p=0.66). Based on preliminary analyses, MYC amplification and MYC/HER2 co-amplification do not appear to be associated with race, menopausal status, hormone receptor status, nodal status, tumor histology, tumor grade, or tumor size (all ChiS p>0.05); however, MYC amplified pts were significantly younger than MYC non-amplified pts (mean 46.4 vs. 51.0, t-test p=0.02) and MYC/HER2 co-amplified pts were significantly younger than MYC/HER2 non-co-amplified pts (mean 45.9 vs. 51.0, t-test p=0.01).
 Conclusions: In our initial analyses of 239 pts, we observed a lower frequency of MYC/HER2 co-amplification than the previously reported 30% for NSABP B-31. Complete FISH analyses of 868 pts in N9831 and association with DFS will be presented. Citation Information: Cancer Res 2009;69(2 Suppl):Abstract nr 56.

Abstract Abstract #5110 Background: It has been demonstrated that certain chemotherapeutic drugs such as taxanes increases cyclooxygenase-2 (COX-2) levels. Upregulation of COX-2 may be a resistance mechanism of the tumor, or a method to modulate toxic effects of the agent. COX-2 is associated with less differentiated and more aggressive breast cancers, and the expression may be a prognostic indicator of disease. Studies defining the “true” impact of therapy on COX-2 activity are lacking. Hypothesis: Chemotherapy increases COX-2 function in patients with invasive breast cancer. This leads to increased PGE2, VEGF, and downregulation of immune responses. Objective: Patients with stage II or III breast cancer were enrolled in a phase II preoperative chemotherapy trial of docetaxel and carboplatin administered every two weeks (4 cycles). We evaluated circulating PGE2, VEGF, and immune cell phenotype at diagnosis and after chemotherapy. Results: Fifty seven patients were enrolled in the study and 32 were analyzed in the lab, most of them being infiltrating ductal carcinoma. Four had complete response, 20 had partial response, and 8 were non-responders. Due to low numbers of patients analyzed, statistical significance was not achieved in most instances. Nevertheless, intriguing data has been generated that warrants further investigation. VEGF/PGE2 versus clinical response to chemotherapy: 86% of the patients in whom VEGF levels decrease post chemotherapy were responders compared to 60% in which VEGF remained unchanged or increased slightly. Responders had decreased VEGF levels on average, while the non-responders increased. Interestingly, even with low sample size, if both VEGF and PGE2 levels increase post treatment the response rate to chemotherapy is significantly lower (55% versus 88% if both VEGF and PGE2 decrease post treatment, p=0.05). Thus, both VEGF and PGE2 are critical factors in determining response rate. Whether COX-2 activity is driving both factors or whether VEGF is independently regulated is yet to be determined. Clinical Response versus Dendritic cell (DCs) / T-cell data: Data revealed highly important trends that warrant future investigation. Patients with increased B7H4 (an immunosuppressive molecule expressed on tolerizing DCs) were non-responders while patients with decreased B7H4 were responders. In contrast, responders had increased levels of CD80 and CD86 (co-stimulatory molecules expressed on activating DCs). VEGF/PGE2 Change versus Dendritic/T-cell data: When VEGF increased post treatment, B7H4 and FoxP3 (T regulatory cells) increased, while CD80, CD86, and CD8 decreased. The association between VEGF and immune cells post chemotherapy treatment is the first indication that VEGF may regulate immune cell function possibly independent of COX-2 activity. These data suggest that anti-VEGF therapy (which may include COX-2 inhibitor) may not only augment responses to chemotherapy but may also augment immune responses post chemotherapy. Citation Information: Cancer Res 2009;69(2 Suppl):Abstract nr 5110.

Abstract Abstract #5008 Background: Accurate, individualized risk prediction for breast cancer is lacking. Tissue-based features, such as the terminal duct lobular unit (TDLU), may help to stratify women into different risk levels as breast lobules are the anatomic sites of origin of breast cancer. As women age, these lobular structures should regress, which results in reduced breast cancer risk. Regression, however, does not occur in all women.
 Materials and Methods: We have quantified the extent of lobule regression on a benign breast biopsy in 85 breast cancer cases and 142 age-matched controls from the Mayo Benign Breast Disease Cohort, by determining number of acini/lobule and lobular area. We also calculated Gail model 5-year predicted risks for these women.
 Results: There is a step-wise increase in breast cancer risk with increasing numbers of acini/lobule (p=0.0004). Adjusting for Gail model score, parity, histology, and family history did not attenuate this association.
 
 Lobular area was similarly associated with risk. The Gail model estimates were associated with risk of breast cancer (p=0.03). We examined the individual accuracy of these measures using the concordance (c) statistic. The Gail model c-statistic was 0.60 (95% CI; 0.50-0.70); the acinar count c-statistic was 0.65 (95% CI; 0.54-0.75). Combining acinar count and lobular area, the c-statistic was 0.68 (95% GI; 0.58-0.78). Adding the Gail model to these did not improve the c-statistic.
 
 Discussion: Novel, tissue-based features that reflect the status of a woman's normal breast lobules are strongly associated with breast cancer risk. These features appear to provide more accurate risk assessment than the currently used Gail model. Citation Information: Cancer Res 2009;69(2 Suppl):Abstract nr 5008.

Abstract Abstract #404 Background: The oral HDACi vorinostat (V) sensitizes breast cancer cells to tubulin polymerizing agents and anti-VEGF directed therapies in vitro. We sought to determine the safety and efficacy of V plus paclitaxel (P) and bevacizumab (B) as first-line therapy for MBC.
 Methods: Eligible patients (pts) with measurable disease, ECOG PS 0 or 1, and no prior chemotherapy for MBC, received P (90mg/m2) on days 1, 8, 15 and B (10mg/kg) on days 1,15 every 28 days, plus V 200 mg (N=3) or 300 mg (N=28) BID orally for 3 days given the day before, day of, and day after each P dose. Response was assessed after every 3 cycles (RECIST). The trial was designed to distinguish between a response rate of < 40% vs. > 60% (alpha=0.10, beta=0.10) using Simon's optimal 2-stage design. At least 12 responses were required in the first 28 evaluable pts in the 1st stage and 21 of 41 in the 2nd stage.
 Results: There were no dose-limiting toxicities among the first 6 pts treated at the 1st (200 mg BID; N=3) and 2nd (300 mg BID; N=3) dose levels. Among the 26 evaluable pts treated at the recommended phase II dose of 300mg bid in the 1st stage, 14 patients had a partial response (PR-54%; 90% confidence interval [C.I.] 36%-71%), 6 had stable disease (SD) at 3 months and remain on treatment, 2 had SD as the best response with progressive disease (PD) at 11.8 and 15 months, and 4 had PD as their best response. Additionally, 2 of 3 pts treated at the first V dose level had a PR, indicating PR in 16 of 29 overall (55%, 90% C.I 38%-71%). Only 11 of 29 evaluable pts (38%) have progressed after a median followup of 10.5 months (R-2.3-14.9). Two pts had tumor biopsies and peripheral blood mononuclear cells (PBMC) collected before and 4 hours after the 3rd V dose (300 mg BID) which showed increased acetylation of K69 lysine residue of the chaperone protein Hsp 90, upregulation of Hsp70, and downregulation of AKT by Western Blot, providing evidence of Hsp90 inhibition in tumor and PBMC.
 
 Gr 3-4 toxicities included neutropenia (26%), neuropathy (23%), fatigue (16%), thrombosis (10%), vomiting (6%), hypertension (3%), and diarrhea (3%).
 Conclusions: Our preliminary findings indicate that the HDACi V inhibits Hsp90 and downregulates AKT in breast cancer in vivo, and enhances the effectiveness of P plus B in MBC. Accrual is nearly complete and the final efficacy analysis will be presented. Citation Information: Cancer Res 2009;69(2 Suppl):Abstract nr 404.

Abstract Abstract #4037 Background: Mammographic breast density is a strong and consistent risk factor for breast cancer. Women with dense tissue in over 60 to 75% of the breast are at 4 to 6 times greater risk than women with little or no density. However, the biology of breast density is not well understood and studies at the tissue level are limited. We present the first report of a tissue-based study of breast density, sampling mammographically dense and non-dense areas of the breast of healthy women with the goal of examining histologic correlates.
 Materials and Methods: Healthy women volunteers aged 40 to 85 years, with no breast symptoms, normal screening mammogram within 6 months, and no personal history of breast cancer were considered eligible. Women with breast-related symptoms, those on endocrine therapy, or with bleeding tendencies were excluded from participation. Sixty consecutive eligible women were enrolled in the study- results are based on the first 34 participants. Eligible women underwent an ultrasound-guided, core-needle breast biopsy of mammographically dense and non-dense areas of the breast, performed by a single experienced radiologist. Quantitative assessment of breast tissue from dense and non-dense areas was performed for epithelium, stroma, and fat. In addition, we assessed the difference in the extent of lobular involution (complete, partial and none) between dense and non-dense areas. We used means and SD to describe distributions in each tissue type. Signed rank test was used to assess within woman differences between dense and non-dense tissue.
 Results: The mean age of the participants was 50.1 years (range 40 to 79 years). The histological findings are outlined below.
 
 Dense tissue is comprised of increased stroma and epithelium and decreased fat, consistent with prior reports on breast density. The finding of 'no involution' was greater in dense tissue compared to non-dense tissue (24% versus 8.8%); non-dense tissue had greater proportion of complete involution compared to dense tissue (82.5% versus 35.2%).
 Discussion: This report is the first to quantify histology in tissue sampled from dense and non-dense regions of the healthy breast. We find a dramatic difference in tissue composition between dense and non-dense tissue in the breast. Future studies using this tissue to assess molecular markers will further help to characterize pathways through which density may increase breast cancer risk. Citation Information: Cancer Res 2009;69(2 Suppl):Abstract nr 4037.

Abstract Abstract #4033 Background: Aromatase activity within the breast is a source of estrogen that may cause breast cancer. Mammographic density (MD) is a risk factor for breast cancer whose biologic basis is unknown. Our study compared aromatase expression in tissue from dense and non-dense areas of the breasts of healthy volunteers.
 Methods: Participants were 40+ yrs, had a screening mammogram with visible MD, no history of cancer and were not on endocrine therapy. Ultrasound-guided core biopsies were done within 6 months of mammography to obtain three paired cores from mammographically dense and non-dense regions of the breast. Immunostaining for aromatase expression employed the streptavidin-biotin amplification method using the recently developed 677 mouse monoclonal antibody. Immunoreactivity (IR) was scored in terms of proportion of cells staining positive for aromatase (PPC) (0=<1%, 1=1-25%, 2=26-50%, 3=51+% (and 4=75%+ for adipocytes)) and relative immunointensity (0=none, 1=weak, 2=moderate, 3=intense) for each cell type (stroma, normal ductal epithelium, adipocytes). A composite score weighting the PPC by their relative intensity (range 0-9; 0-12 for adipocytes) was also calculated. The sum of the composite score across the three cell types provided a global assessment of aromatase IR. Repeated measures analyses evaluated differences in aromatase IR for dense compared to non-dense tissue within and across cell types. Parameter estimates (β) indicated the average difference between dense and non-dense IR.
 Results: 18 (37%) of the 49 participants were premenopausal (median age 46 yrs). Summing across cell types, the global composite score showed increased aromatase IR on sections sampled from dense vs. non-dense regions (β=5.3,p<0.001). This global difference reflected increased aromatase IR in the stroma and normal ductal epithelium from dense sections (vs. non-dense). For 42 women with >1% stromal cells on each section, there was evidence for increased IR on dense sections as indicated by the greater PPCs (β=0.9), relative intensity (β=0.7) and composite score (β=2.7)(p's<0.001). Similarly, for 20 women with >1% normal ductal epithelium in both dense and non-dense sections, there was a greater composite score (β=1.4,p=0.004) for cells on dense sections although differences for PPC (β=0.3,p=0.11) and relative intensity (β=0.3,p=0.09) did not reach statistical significance. No differences were seen in IR for adipocytes from the two density regions (46 women). Findings were unchanged with adjustment for covariates, including proportions of each cell type per section.
 Conclusions: There is strong evidence for increased aromatase expression in the stromal and normal ductal epithelium of dense vs. non-dense tissue. These results support ongoing research into mammographic density as a biomarker of effect of aromatase inhibitors. Citation Information: Cancer Res 2009;69(2 Suppl):Abstract nr 4033.

Abstract Abstract #4020 Background: HER2 positive (HER2+) breast cancer accounts for 20% of all breast cancer and is an independent adverse prognostic factor. Breast magnetic resonance imaging (MRI) has been shown to identify additional mammographically occult ipsilateral (I/L) and contralateral (C/L) disease, however indications for preoperative MRI in patients with newly diagnosed breast cancer are not standardized. While invasive lobular cancer, younger patient age, and increased breast density have been suggested as routine indications, the role of preoperative MRI according to HER2 status has not been evaluated.
 Methods: All patients with invasive breast cancer who underwent preoperative MRI between January 2003 and December 2007 were reviewed. Retrospective chart review was performed to collect clinical, radiographic, and pathologic data. The incidence of additional I/L or C/L malignacy was calculated. Additionally, changes in surgical management because of MRI findings were recorded.
 Results: Of the 401 women with invasive breast cancer undergoing preoperative breast MRI, 51 (13%) had HER2+ tumors. The average age at the time of MRI was similar in both groups. HER2+ patients were more likely to have nodal disease, and tumors with higher rates of lymphovascular invasion or extensive intraductal components.
 
 When compared to HER2- patients, more HER2+ patients had additional ipsilateral MRI abnormalities. A second ipsilateral cancer was found in 31% of HER2+ patients compared to 25% of HER2- patients. When analyzed according to ER status, HER2+/ ER- patients were more likely to have an additional site of cancer compared to HER2+/ER+ patients (42% vs 23%). Conversely, among HER2- patients, rates of additional ipsilateral cancer were similar for HER2-/ER+ (26%) and HERr2-/ER- (23%) patients. Of the 216 patients (HER2+ and HER2-) with additional ipsilateral MRI findings (benign or malignant), 125/216 (58%) patients were deemed appropriate breast conservation (BCT) candidates after MRI. However, BCT was only chosen in 84/125 (67%) patients. Additional contralateral disease was documented in 8% of HER2+ versus 5% of HER2- patients.
 Conclusions: HER2 positivity remains and important prognostic factor for women with breast cancer. Indications for routine preoperative MRI are unclear. In patients with newly diagnosed Her2+ invasive breast cancer routine preoperative MRI may be valuable in the subset of patients who are also ER-. Further study is needed to determine the influence of MRI on patient decisions for surgical therapy. Citation Information: Cancer Res 2009;69(2 Suppl):Abstract nr 4020.