Abstract P4-14-06: Neoadjuvant chemotherapy with docetaxel, carboplatin and weekly trastuzumab (TCH) is active in HER2-positive early breast cancer: Results after a median follow-up of over 4 years

Background:

HER2-positive breast cancer is known to carry an adverse prognosis compared to HER2-negative disease, which may however be compensated by the use of HER2-targeted agents. Therefore, most patients with HER2-positive disease larger than 5 mm receive chemotherapy and trastuzumab. Data from adjuvant trials have shown that the combination of docetaxel, carboplatin and weekly trastuzumab (TCH) is well tolerated and as effective as anthracycline containing regimes. Previous investigations on neoadjuvant treatment with TCH showed pCR-rates in the range of 40%, however, survival data have not yet been presented. Here we present 4-year follow-up data for a cohort of 51 patients treated with neoadjuvant TCH.

Methods:

We treated 51 patients with operable HER2-positive breast cancer with a neoadjuvant schedule of docetaxel (75 mg/m2) and carboplatin (AUC 6) q3w and trastuzumab (2(4)mg/kg) q1w. Lymph node involvement was verified by SLNB or core-cut-biopsy. Patients were diagnosed at a mean age of 55 years, 68.6% had ER positive tumors, 39.2% presented with grade 3 disease and 49% of patients were node-positive. Patients were monitored every two cycles by ultrasound. After 6 cycles of chemotherapy all patients had surgery. Axillary dissection was performed in case of positive lymph node status prior to TCH. After surgery trastuzumab was continued q3w up to one year.

Results:

In 50 patients TCH could be administered as planned without dose reductions or delays. One patient suffered from an allergic reaction on taxane after the second cycle, resulting in replacement by gemcitabine. Side effects were mild, no grade III/IV toxicities occurred and no case of cardiomyopathia was observed. 21 (41.18%) patients achieved a pCR, 18 (72.0%) patients converted from cN+ to ypN0. Outcome data at a median follow-up of 51.6 months are as follows.

 All patients (n=51)pCR (n=21)N+ (n=25)cN+ → ypN0 (n=18)ER positive (n=35)G3 (n=20)DFS (n/%)42/82.3517/80.9517/68.016/88.8931/88.5716/80.0DDFS (n/%)46/90.219/90.4820/80.017/94.4433/94.2918/90.0OS (n/%)48/94.1821/100.022/88.018/100.034/97.1419/95.0

Conclusion:

Outcome following neoadjuvant TCH as observed in our analysis compares well to outcome data observed in adjuvant trastuzumab trials such as HERA (4-year follow-up; DFS 78.6% and OS 89.3%) or BCIRG006 (36-month follow-up; DFS 82% and OS 91% in the TCH-arm). Particularly among patients with ER positive disease and those experiencing axillary conversion we obseverd an excellent outcome. Importantly, TCH was well tolerated in our cohort. Therefore our data support the use of TCH as neoadjuvant therapy regimen for patients with HER-positive breast cancer. They also strongly encourage the use of docetaxel and carboplatin as chemotherapy backbone in studies investigating the dual blockade with trastuzumab and pertuzumab in the neoadjuvant setting.

Citation Format: Kolberg H-C, Akpolat-Basci L, Stephanou M, Hannig CV, Liedtke C. Neoadjuvant chemotherapy with docetaxel, carboplatin and weekly trastuzumab (TCH) is active in HER2-positive early breast cancer: Results after a median follow-up of over 4 years. [abstract]. In: Proceedings of the Thirty-Eighth Annual CTRC-AACR San Antonio Breast Cancer Symposium: 2015 Dec 8-12; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2016;76(4 Suppl):Abstract nr P4-14-06.

The carbon tetrachloride model in mice

Recently, the need for more standardized operation procedures in experimental liver fibrosis research was suggested due to dramatic changes in European animal welfare rules. Here, we present a short series of standard operation procedures (SOPs) summarizing the most relevant and widely accepted experimental models for the induction of liver injury leading to liver fibrosis. The described procedures are based on the long-term experience of the Collaborative Research Centre 'Organ Fibrosis: From Mechanisms of Injury to Modulation of Disease' (http://www.sfbtrr57.rwth-aachen.de/), which is supported by the German Research Foundation (SFB/TRR57). These SOPs will help to improve standardization of fibrosis models and to increase the comparability of data between different laboratories with the aim of reducing animal experimentation according to the principle that was proposed in 1959 by Russell and Burch as an ethical framework for conducting scientific experiments with animals, namely the replacement, refinement and reduction (3R) principle. In the first section we focus on the carbon tetrachloride (CCl4) model in mice, which is the toxic model of liver fibrosis induction most commonly used worldwide.

Partial hepatectomy in mice

The surgical procedure of two-thirds partial hepatectomy (PH) in rodents was first described more than 80 years ago by Higgins and Anderson. Nevertheless, this technique is still a state-of-the-art method for the community of liver researchers as it allows the in-depth analysis of signalling pathways involved in liver regeneration and hepatocarcinogenesis. The importance of PH as a key method in experimental hepatology has even increased in the last decade due to the increasing availability of genetically-modified mouse strains. Here, we propose a standard operating procedure (SOP) for the implementation of PH in mice, which is based on our experience of more than 10 years. In particular, the SOP offers all relevant background information on the PH model and provides comprehensive guidelines for planning and performing PH experiments. We provide established recommendations regarding optimal age and gender of animals, use of appropriate anaesthesia and biometric calculation of the experiments. We finally present an easy-to-follow step-by-step description of the complete surgical procedure including required materials, critical steps and postoperative management. This SOP especially takes into account the latest changes in animal welfare rules in the European Union but is still in agreement with current international regulations. In summary, this article provides comprehensive information for the legal application, design and implementation of PH experiments.

Primary Therapy of Patients with Early Breast Cancer: Evidence, Controversies, Consensus: Opinions of German Specialists to the 14th St. Gallen International Breast Cancer Conference 2015 (Vienna 2015)

For the first time, this year's St. Gallen International Consensus Conference on the treatment of patients with primary breast cancer, which takes place every two years, was held not in St. Gallen (Switzerland) but - for logistical reasons - in Vienna (Austria) under its usual name. The 2015 St. Gallen International Consensus Conference was the 14th of its kind. As the international panel of the St. Gallen conference consists of experts from different countries, the consensus mirrors an international cross-section of opinions. From a German perspective, it was considered useful to translate the results of the votes of the St. Gallen conference into practical suggestions, particularly in light of the recently updated treatment guideline of the Gynecologic Oncology Group (AGO-Mamma 2015) in Germany. A German group consisting of 14 breast cancer experts, three of whom are members of the international St. Gallen panel, has therefore provided comments on the results of this year's votes at the 2015 St. Gallen Consensus Conference and their impact on clinical care in Germany. The 14th St. Gallen conference once again focused on surgery of the breast and the axilla, radio-oncologic and systemic treatment options for primary breast cancer depending on tumor biology, and the clinical use of multigene assays. The conference also considered targeted therapies for older and for younger patients, including the diagnosis/treatment of breast cancer during and after pregnancy and the preservation of fertility.

Diethylnitrosamine (DEN)-induced carcinogenic liver injury in mice

The toxic properties of various nitrosamines in animals and humans are well established. The parenteral or oral administration of the smallest quantities of diethylnitrosamine (DEN) or dimethylnitrosamine (DMN) results in severe liver damage. Most prominent are intense neutrophilic infiltration, extensive centrilobular haemorrhagic necrosis, bile duct proliferation, fibrosis, and bridging necrosis that ends in hepatocarcinogenesis. Due to the robustness of the induced hepatic alterations, the application of DEN in rodents has become an attractive experimental model for studies aimed at understanding the pathogenetic alterations underlying the formation of liver cancer, which represents one of the most common malignancies in humans worldwide. However, several studies have shown that the hepatocarcinogenic effects of nitrosamines might vary with the genetic background of the animals, their sex, their age, and other factors that might impact the outcome of experimentation. We present general guidelines for working with DEN, and a detailed protocol that allows the establishment of highly reproducible liver cancer in mice. The outcome of liver injury after the application of DEN in mice, as estimated by the formation of cirrhosis and cancer, appears to be a suitable animal model for the analysis of some aspects and processes that promote the pathogenesis of hepatocellular carcinoma in humans.

Expression of secreted protein acidic and rich in cysteine (SPARC) in breast cancer and response to neoadjuvant chemotherapy

BACKGROUND

Secreted protein acidic and rich in cysteine (SPARC) has been suggested as a new biomarker and therapeutic target in breast cancer, as well as other tumor types.

PATIENTS AND METHODS

We evaluated the frequency of SPARC expression among different molecular breast cancer subtypes and its role for therapy response after neoadjuvant chemotherapy. In this study, pretherapeutic core biopsies of 667 patients from the neoadjuvant GeparTrio trial were evaluated for SPARC expression by immunohistochemistry using a standardized immunoreactive score (IRS).

RESULTS

An increased SPARC expression (IRS ≥6) was observed in 26% of all tumors. In triple-negative tumors, SPARC expression was increased in 37% of tumors, compared with other molecular subtypes (23% HR+/HER2-, 29% HR+/HER2+ and 22% HR-/HER2+; P = 0.038). Increased SPARC expression was associated with an increased pathological complete response (pCR) rate of 27%, compared with 15% in tumors with low SPARC expression (P < 0.001). In the triple-negative subgroup, pCR rates were 47% in tumors with high SPARC expression, compared with 26% in tumors with low SPARC expression (P = 0.032). In multivariable analysis, SPARC was independently predictive in the overall population (P = 0.010) as well as the triple-negative subgroup (P = 0.036).

CONCLUSIONS

SPARC is frequently expressed in breast cancer with triple-negative breast cancer revealing the highest expression rate. High SPARC expression of the primary tumor is associated with a higher chance of achieving a pathological complete remission after TAC or TAC-NX chemotherapy. As SPARC is an albumin-binding protein and might mediate intratumoral accumulation of albumin bound drugs, SPARC should be further evaluated as a predictive marker especially for response to albumin-bound drugs like nab-paclitaxel.

CLINICAL TRIAL NUMBER

NCT00544765.

Abstract P4-04-06: Systematic analysis of molecular subgroups and age in patients with breast cancer

Background: Breast cancer (BC) molecular subgroups show significant differences with regard to prognosis. Similarly, young age at diagnosis is significantly associated with an unfavorable prognosis.

Methods: Our aim was to analyze the association between BC molecular subtypes and age at diagnosis. Publically available gene expression data (Affymetrix U133A) of 3488 BC patients was analyzed using MAS5.0 and pooled into a single database. Age information was available for 2604 of the patients (74.7%). Of a total of 579 cases with TNBC, 394 samples with most comparable array data were selected to avoid batch effects as previosly described. Age information was available for 309 of these 394 patients (78.4%). Significance Analysis of Microarrays (SAM) was performed to identify genes with differential expression according to age at diagnosis in these 309 patients with TNBC.

Results: Patients &lt;40 years significantly more often presented with triple negative BC compared to patients 40-50 and &gt;50 (34.8 vs. 25.4 vs. 17.3%, respectively). In contrast, a luminal A subtype was diagnosed less often in patients &lt; 40 yrs (20.6 vs. 31.9 vs. 41.1%, respectively). Although we observed significant differences for EFS according to i) age at diagnosis and ii) molecular subtype, EFS differences between patients of distinct age were only modest after stratification for molecular subtype. Among patients with TNBC, 134 genes were upregulated and 627 genes were downregulated in patients &lt; 40 yrs compared to patients ≥ 40 yrs (median FDR &lt; 4%). Several of the downregulated genes in young TNBC patients belong to a metagene for the molecular-apocrine subtype of TNBC and several upregulated genes represent cell cycle genes. However, the effects for individual genes were rather small. Conclusion: The frequency of molecular subgroups differs with age at diagnosis. Young age at diagnosis is significantly associated with adverse prognosis. Within the group of TNBC patients young age is associated with high proliferation and a basal-like phenotype and inversely correlated with the molecular-apocrine subtype of TNBC.

Citation Information: Cancer Res 2013;73(24 Suppl): Abstract nr P4-04-06.

Abstract P5-04-02: Biopsy of metastases impacts treatment choice and patient outcome in breast cancer – Final results of the WSG/DETECT PRIMET study

Background: Changes in tumor biology (e.g., hormone receptor (HR) / HER2 status or grading) between primary tumor (PT) and metastatic tissue (MT) could impact outcome and treatment choice following first recurrence in breast cancer (BC).

Methods: PRIMET is a prospectively planned, retrospective multicenter quality assurance study comparing BC phenotype in tissue from PT, involved lymph nodes (LN) of primary disease, and disease recurrence (DR). PRIMET comprises 635 patients from WSG and DETECT trial groups (11 centers), whose BC was diagnosed between 1980 and 2010; follow-up continued until mid-2012. Patients with unilateral primary BC suffering subsequent local-regional and / or distant DR (LDR / DDR) were included. Clinical data including ER, PR, HER2, and grade were obtained from a systematic chart review in PT and DR; in two centers, these factors were also measured in LN by central pathology. Dependence of post-recurrence survival (PRS) on changes in tumor biological factors was analyzed.

Results: Data from 635 patients (including 592 cM0, of whom 46% had LDR only) were available for analysis. Median follow-up in patients alive at analysis was 101 months. Considering cM0 patients, median overall survival (OS) was 176 months; median recurrence-free survival (RFS) was 48 months (DDR present: 45 months; LDR only: 50 months). Median PRS was 59 months (DDR present: 45 months; LDR only: 127 months). In patients with first DR within 18 months, median PRS was 29 months, in others 79 months. HR status in PT/MT was: 61.5% (+/+), 13.2% (+/-), 5.5% (-/+) 19.8% (-/-). Of the HR “switches” in either direction with LN biopsy available, about half already occurred in lymph nodes. HER2 status in PT/MT was: 14.6% (+/+), 6.7% (+/-), 14.9% (-/+) 63.8% (-/-). With LN biopsy available, most losses of HER2 overexpression were already observed in LN tissue, whereas acquired HER2 overexpression was observed in about half of LN biopsies. Triple negative (TN: HR-, HER2-) percentages were 74.4% (non-TN/non-TN), 9.0% (non-TN/TN), 6.1% (TN/non-TN), 10.5% (TN/TN).

Compared to HR+/+, loss of HR+ status (HR+/-) was significantly associated with poorer PRS (hazard ratio: 1.62; p = 0.01). Significantly better PRS was associated with a switch from G3 to G1/2 (hazard ratio: 0.47; p = 0.02). Tumors that switched to TN or that lost HER2 overexpression showed trends toward poorer PRS. Persistent TN was associated with poorer PRS than other combinations.

Among patients with DDR, metastasis in bone only was associated with better PRS than primary or visceral (CNS, lung, liver, etc.) metastasis. Among patients with visceral metastasis, negative HR status in metastasis was associated with poorer survival than in HR+/+ not only for HR-/- (p = 0.02), but also for HR+/- (p = 0.04).

Conclusions:

Tumor biology of primary and metastatic tissue differed in a substantial fraction of patients (HR: 19%; HER2: 22%, TN: 18%); more than half of all changes occurred already in LN. Status changes particularly loss of HR+ status, had significant prognostic impact. We can expect a switch in HR or HER2 status (or both) in about 38% of metastatic tissue biopsies, with presumably important clinical therapeutic consequences, in particular regarding targeted therapies.

Citation Information: Cancer Res 2013;73(24 Suppl): Abstract nr P5-04-02.

Abstract P3-06-05: Expression of SPARC in human breast cancer and its predictive value in the GeparTrio neoadjuvant trial

Background: Secreted protein acidic and rich in cysteine (SPARC) is an albumin-binding protein and associated with poor prognosis in multiple cancers.

The aim of this analysis was to determine the frequency of SPARC expression among different molecular breast cancer subtypes and to evaluate its predictive value for therapy response after neoadjuvant anthracycline/taxane based chemotherapy (CTX) in participants of the GeparTrio trial.

Methods: We evaluated tumoral SPARC expression by immunohistochemistry on tissue microarrays (TMAs) constructed from formalin-fixed paraffin-embedded (FFPE) pre-treatment core biopsies from 667 patients (pts) of the GeparTrio trial. The details of the GeparTrio study design are described elsewhere (von Minckwitz, JNCI 2008). Cutoffs for SPARC expression were determined using the web-based software Cutoff Finder (http://molpath.charite.de/cutoff/).

Results: SPARC protein expression was measurable by IHC on the TMAs and 176 (26.4 %) of 667 tumors were SPARC positive. SPARC expression was increased in pts with triple-negative breast cancer (TNBC) compared to hormone receptor or HER2 positive subtypes (p = 0.039).

SPARC positivity was associated with an increased pCR rate in the overall population (p &lt; 0.001). SPARC negative tumors had a pCR rate of 15%, which increased to 27% in SPARC positive tumors. Similarly, in TNBC the pCR rate increased from 26% in SPARC negative tumors to 47% in SPARC positive tumors (p = 0.032).

In multivariate logistic regression analysis adjusted for standard clinicopathological factors, SPARC was independently predictive in the overall population (p = 0.010) as well as the subgroups of pts with TNBC (p = 0.036).

Conclusions: SPARC is expressed in all biological breast cancer subtypes with TNBC revealing the highest expression rate. Our data suggest that SPARC expression may provide predictive information for response to neoadjuvant CTX.

As SPARC is an albumin-binding protein and might mediate intratumoral accumulation of nab-Paclitaxel, prospective analysis of SPARC expression is planned in the GeparSepto trial.

Citation Information: Cancer Res 2012;72(24 Suppl):Abstract nr P3-06-05.

Chemotherapy-Induced Amenorrhea - An Update

Because of the heterogeneity in the definition of chemotherapy-induced amenorrhea (CIA) there are distinct differences in the literature with regard to its incidence as well as its dependence on various influencing factors. The occurrence of CIA varies greatly depending on the applied chemotherapy. The pathogenesis of CIA is especially based on a reduction of ovarian reserves. Various sonographic and biochemical factors can be used to exclude or confirm CIA. This is particularly important when an endocrine therapy with tamoxifen is not possible and the use of aromatase inhibitors is under consideration. CIA and especially the frequently thereby resulting early menopause can lead to pronounced restrictions in the quality of life of the affected patients, not least due to the resulting infertility. On the other hand, various studies have shown that CIA may have a positive prognostic significance. Thus, the identification of measures to prevent CIA (for example, through the use of GnRH analogues) is of particular importance.

Predictive value of HER2 serum levels in patients treated with lapatinib or trastuzumab -- a translational project in the neoadjuvant GeparQuinto trial

BACKGROUND

We were able to demonstrate a predictive value of serum HER2 (sHER2) in patients receiving trastuzumab in the neoadjuvant GeparQuattro trial. However, the role of sHER2 in patients receiving neoadjuvant therapy (NT) with lapatinib is still unclear.

METHODS

The neoadjuvant GeparQuinto trial compared trastuzumab vs lapatinib in addition to chemotherapy in HER2-positive primary breast cancer patients. The sHER2 levels were measured by enzyme-linked immunosorbant assay in 210 patients, of whom 109 (52%) patients received trastuzumab and 101 (48%) lapatinib at three different time points.

RESULTS

Twenty-two percent of patients had elevated baseline sHER2 levels (>15 ng ml⁻¹). A decrease of sHER2 levels (>20%) in the trastuzumab and lapatinib-treated group during NT was seen in 44% and 24% of the patients, an increase of sHER2 levels (>20%) was seen in 6% and 41% of patients, respectively. Higher pre-chemotherapy sHER2 levels were associated with higher pathological complete remission (pCR) rates in the entire study cohort (OR 1.8, 95% CI 1.02-3.2, P=0.043). A decline of sHER2 levels (>20%) during NT was a predictor for pCR in the lapatinib-treated patient group (OR: 11.7, 95% CI 1.3-110, P=0.031).

CONCLUSION

Results of this study demonstrate that sHER2 levels change differently during NT depending on the anti-HER2 treatment strategy. Elevated baseline sHER2 levels (>15 ng ml⁻¹) and a decrease of sHER2 levels (>20%) early after therapy initiation are both relevant criteria to predict response to lapatinib-based treatment.

P1-06-03: Predictive Value of HER2, Topoisomerase-II (Topo-II) and Tissue Inhibitor of Metalloproteinases (TIMP-1) for Efficacy of Taxane-Based Chemotherapy in Intermediate Risk Breast Cancer – Results from the EC-Doc Trial

Background: Despite extensive research, there is still no consensus on optimal predictors for use of taxane-based chemotherapy (cht) in early breast cancer. Some studies have revealed HER2 as a significant predictive marker for efficacy of taxanes and anthracyclines. TIMP-1 and Topo-II are reported to be predictive for anthracycline efficacy. In our previous reports, both Ki-67≥20% and central G3 status emerged as significant predictors for taxane benefit. We have now compared HER2 and Topo-II (as protein expression and gene amplification) and TIMP-1 immunoreactivity as well as factor combinations (HT (HER2/TIMP-1) and 2T (Topo-II/TIMP-1) regarding their predictive value for benefit from taxane-based cht.

Methods: The EC-Doc trial randomized 1950 patients with 1–3 positive LN to 6x CEF/CMF vs. 4xEC-4xDoc. Significantly better DFS and OS favoring EC-Doc have been previously reported (Nitz et al., SABCS 2008). Protein expression and gene amplification data as well central histology/grade were available for 772 patients. Survival analysis was performed using Cox proportional hazards and Kaplan-Meier statistics. Analysis of HER2 survival impact status was prospectively planned.

Results: The entire and the investigated study populations did not differ regarding baseline characteristics. After median follow up of 64 months, both DFS (5y 90% vs. 80%, p=0.006) and OS (5y 95% vs. 92%, p=0.022) rates significantly favored EC-Doc vs. CEF in this cohort as well. HER2 over-expression (3+ and/or FISH≥2.0) was reported in 158 tumors (20%), Topo-II aberration (deletion or amplification) was reported in 78 (49.4%) HER2+ and in 83 (13.6%) HER2−negative tumors; 496 tumors were classified as TIMP-1 immunoreactive (65.2%). None of these factors were significantly prognostic for EFS in this collective. Regarding DFS, EC-Doc was strongly superior to FEC in HER2+ tumors (HR=0.29, 95%CI: 0.12−0.7, p=0.006) but not in HER2− tumors (p=0.18). In Topo-II aberrated tumors, the benefit of EC-Doc was remarkably strong (HR=0.28, 95% CI: 0.11−0.69, p=0.006), whereas the benefit was not significant in Topo-II normal tumors (p=0.16), which comprise more than ¾ of the total. In contrast, Topo-II protein overexpression (&gt;10%) was not associated with a stronger benefit in either subgroup. The superiority of EC-Doc to FEC was significant in the larger group of TIMP-1 immunoreactive tumors (HR=0.57, p=0.025) but not in TIMP-1 negative tumors (p=0.14), similar behavior was seen in “HT” and “2T” subgroups (significance with HR about 0.5 in the “+” subgroups). In a multivariate model for DFS including age, tumor size, Ki-67, central grade, HR, HER2, TOPO_II aberration, TIMP-1 status, therapy and interactions of all these factors with therapy arm, the only significant therapy interaction was that of (high) Ki-67 (HR=0.76, 95% CI: 0.59−0.98, p=0.03); significant main effects in this model were age, central grade, and Ki-67.

Conclusions: These data suggest predictive significance for Topo-II aberration, TIMP immunoreactivity and HER2 over-expression as well as a multivariate predictive significance of high Ki-67 for enhanced benefit of taxane-based cht.

Citation Information: Cancer Res 2011;71(24 Suppl):Abstract nr P1-06-03.

PD03-02: Prognostic and Predictive Predictors for Triple Negative Breast Cancer

Background: Both the prognosis and the therapeutic options in triple negative breast cancer (TNBC) are rather limited. Current prognostic gene expression profiles for breast cancer mainly reflect proliferation status and are most useful in ER-positive cancers. The identification of prognostic gene signatures from TNBC cohorts in previous studies was hindered due to relatively small sample sizes.

Materials and Methods: All currently available TNBC gene expression datasets generated on Affymetrix U133 gene chips were assembled. To minimize inter-laboratory variation we analyzed only highly comparable arrays and data set-biased genes were filtered. Supervised analysis was applied to identify a prognostic signature from a finding cohort of 394 TNBC and validation was performed in an independent cohort of 261 TNBC. The genes from the prognostic predictor were analyzed for their correlation to known molecular phenotypes among TNBC.

Results: Two supervised prognostic signatures consisting of 264 and 26 probesets, respectively, were obtained when applying different cutoffs for false discovery rates of 25% and &lt; 3.5% in the finding cohort. In multivariate analysis in the independent validation cohort hazard ratios of 4.03 (95% CI 1.71−9.48; P=0.001) and 4.08 (95% CI 1.79−9.28; P=0.001), respectively, were obtained for the two signatures. When compared to 16 metagenes for previously described molecular phenotypes in TNBC the prognostic signatures displayed highest correlation to metagenes for IL-8/inflammation, VEGF/angiogenesis, and Histones. A subset of genes in the 264-probeset signature was inversely associated with a poor prognosis (29/264=11.0%). Most of these “good prognosis” genes are correlated with immune cell metagenes (21/29=72.4%). In contrast both identified supervised prognostic signatures did not correlate to previously published prognostic signatures (recurrence score, genomic grade index, Amsterdam signature, wound response signature, 7-gene immune response module, stroma derived prognostic predictor, and a medullary like signature). Regarding the response of TNBC to neoadjuvant chemotherapy the predictive value of the B-cell metagene was superior to the 264- and 26-probeset signatures. However combination of the B-cell metagene and the signatures increased the AUC in ROC-analysis from 0.606 to 0.656. Conclusions: The use datasets consisting only of TNBC allows identification supervised prognostic signatures for TNBC which are unrelated to previously known prognostic signatures.

Citation Information: Cancer Res 2011;71(24 Suppl):Abstract nr PD03-02.