Sample parameters affecting the clinical relevance of RNA biomarkers in translational breast cancer research

RACGAP1 promotes lung cancer cell proliferation through the PI3K/AKT signaling pathway

We aimed to investigate the expression and clinic significance of Rac GTPase Activating Protein 1 (RACGAP1) in human lung adenocarcinoma (LUAD). Online database analysis revealed a significant increase in RACGAP1 mRNA expression among 26 types of tumor tissues, including LUAD tissues. Online database and tissue microarray analyses indicated that RACGAP1 expression was significantly upregulated in LUAD tissues. Genetic variation analysis identified four different genetic variations of RACGAPs in LUAD. Moreover, online database analysis showed that RACGAP1 upregulation was correlated with shorter survival in patients with LUAD. After silencing RACGAP1 expression in A549 cells using siRNA and assessing its protein levels via Western blotting, we found that RACGAP1 knockdown inhibited cell growth and induced apoptosis determined using the Cell Counting Kit-8 assay, colony formation assay, and flow cytometry. Mechanistically, western blot analysis indicated that Bax expression increased, whereas Bcl-2 expression decreased. Moreover, RACGAP1 knockdown attenuated PI3K/AKT pathway activation in lung cancer cells. Taken together, our findings showed that RACGAP1 was overexpressed in LUAD tissues and played an important role in lung cancer by increasing cell growth through the PI3K/AKT signaling pathway. This study suggests recommends evaluating RACGAP1 in clinical settings as a novel biomarker and potential therapeutic target for lung cancer.

HLA-J, a Non-Pseudogene as a New Prognostic Marker for Therapy Response and Survival in Breast Cancer

The human leukocyte antigen (HLA) genes are cell-surface proteins, essential for immune cell interaction. HLA-G is known for their high immunosuppressive effect and its potential as predictive marker in breast cancer. However, nothing is known about the HLA-J and its immunosuppressive, prognostic and predictive features, as it is assumed to be a "pseudogene" by in silico sequence interpretation. HLA-J, ESR1, ERBB2, KRT5 and KRT20 mRNA expression were analysed in 29 fresh frozen breast cancer biopsies and their corresponding resectates obtained from patients treated with neoadjuvant chemotherapy (NACT). mRNA was analysed with gene specific TaqMan-based Primer/Probe sets and normalized to Calmodulin 2. All breast cancer samples did express HLA-J and frequently increased HLA-J mRNA levels after NACT. HLA-J mRNA was significantly associated with overexpression of the ESR1 mRNA status (Spearman ρ 0,5679; p = 0.0090) and KRT5 mRNA (Spearman ρ 0,6121; p = 0.0041) in breast cancer core biopsies and dominated in luminal B subtype. Kaplan Meier analysis revealed that an increase of HLA-J mRNA expression after NACT had worse progression free survival (p = 0,0096), indicating a counterreaction of tumor tissues presumably to prevent elimination by enhanced immune infiltration induced by NACT. This counterreaction is associated with worse prognosis. To our knowledge this is the first study identifying HLA-J as a new predictive marker in breast cancer being involved in immune evasion mechanisms.

Model-Based Integration Analysis Revealed Presence of Novel Prognostic miRNA Targets and Important Cancer Driver Genes in Triple-Negative Breast Cancers

Background: miRNAs (microRNAs) play a key role in triple-negative breast cancer (TNBC) progression, and its heterogeneity at the expression, pathological and clinical levels. Stratification of breast cancer subtypes on the basis of genomics and transcriptomics profiling, along with the known biomarkers’ receptor status, has revealed the existence of subgroups known to have diverse clinical outcomes. Recently, several studies have analysed expression profiles of matched mRNA and miRNA to investigate the underlying heterogeneity of TNBC and the potential role of miRNA as a biomarker within cancers. However, the miRNA-mRNA regulatory network within TNBC has yet to be understood. Results and Findings: We performed model-based integrated analysis of miRNA and mRNA expression profiles on breast cancer, primarily focusing on triple-negative, to identify subtype-specific signatures involved in oncogenic pathways and their potential role in patient survival outcome. Using univariate and multivariate Cox analysis, we identified 25 unique miRNAs associated with the prognosis of overall survival (OS) and distant metastases-free survival (DMFS) with “risky” and “protective” outcomes. The association of these prognostic miRNAs with subtype-specific mRNA genes was established to investigate their potential regulatory role in the canonical pathways using anti-correlation analysis. The analysis showed that miRNAs contribute to the positive regulation of known breast cancer driver genes as well as the activation of respective oncogenic pathway during disease formation. Further analysis on the “risk associated” miRNAs group revealed significant regulation of critical pathways such as cell growth, voltage-gated ion channel function, ion transport and cell-to-cell signalling. Conclusion: The study findings provide new insights into the potential role of miRNAs in TNBC disease progression through the activation of key oncogenic pathways. The results showed previously unreported subtype-specific prognostic miRNAs associated with clinical outcome that may be used for further clinical evaluation.

Robustness of biomarker determination in breast cancer by RT-qPCR: impact of tumor cell content, DCIS and non-neoplastic breast tissue

Background

Tissue heterogeneity in formalin-fixed paraffin-embedded (FFPE) breast cancer specimens may affect the accuracy of reverse transcription quantitative real-time PCR (RT-qPCR). Herein, we tested the impact of tissue heterogeneity of breast cancer specimen on the RT-qPCR-based gene expression assay MammaTyper®.

Methods

MammaTyper® quantifies the mRNA expression of the four biomarkers ERBB2, ESR1, PGR, and MKI67. Based on pre-defined cut-off values, this molecular in vitro diagnostic assay permits binary marker classification and determination of breast cancer subtypes as defined by St Gallen 2013. In this study, we compared data from whole FFPE sections with data obtained in paired RNA samples after enrichment for invasive carcinoma via macro- or laser-capture micro-dissection.

Results

Compared to whole sections, removal of surrounding adipose tissue by macrodissection generated mean absolute 40-ddCq differences of 0.28–0.32 cycles for all four markers, with ≥90% concordant binary classifications. The mean raw marker Cq values in the adipose tissue were delayed by 6 to 7 cycles compared with the tumor-enriched sections, adding a trivial linear fold change of 1.0078 to 1.0156. Comparison of specimens enriched for invasive tumor with whole sections with as few as 20% tumor cell content resulted in mean absolute differences that remained on average below 0.59 Cq. The mean absolute difference between whole sections containing up to 60% ductal carcinoma in situ (DCIS) and specimens after dissection of DCIS was only 0.16–0.25 cycles, although there was a tendency for higher gene expression in DCIS. Observed variations were related to small size of samples and proximity of values to the limit of detection.

Conclusion

Expression of ESR1, PGR, ERBB2 and MKI67 by MammaTyper® is robust in clinical FFPE samples. Assay performance was unaffected by adipose tissue and was stable in samples with as few as 20% tumor cell content and up to 60% DCIS.

Electronic supplementary material

The online version of this article (10.1186/s13000-018-0760-6) contains supplementary material, which is available to authorized users.

Molecular pathogenesis of triple-negative breast cancer based on microRNA expression signatures: antitumor miR-204-5p targets AP1S3

Triple-negative breast cancer (TNBC) is an aggressive type of cancer associated with a poor prognosis. Identification of novel therapeutic targets in TNBC is urgently needed. Here, we investigated the microRNA (miRNA) expression signature of TNBC using clinical specimens. In total, 104 miRNAs (56 upregulated and 48 downregulated) were significantly dysregulated in TNBC tissues; miR-204-5p showed the most dramatic downregulation. We then examined the antitumor roles of miR-204-5p in breast cancer (BC) cells. Notably, cancer cell migration and invasion were significantly reduced by ectopic expression of miR-204-5p in BC cells. Genome-wide gene expression analysis and in silico database search revealed that 32 genes were putative miR-204-5p targets. High expression of AP1S3, RACGAP1, ELOVL6, and LRRC59 was significantly associated with poor prognosis in patients with BC, and adaptor-related protein complex 1 sigma 3 subunit (AP1S3) was directly regulated by miR-204-5p, as demonstrated by luciferase reporter assays. AP1S3 overexpression was detected in TNBC clinical specimens and enhanced cancer cell aggressiveness. We further analyzed downstream RNA networks regulated by AP1S3 in BC cells. Overall, this miRNA signature is expected to be an effective tool for identification of miRNA-mediated molecular mechanisms of TNBC pathogenesis.

Tumor biomarker conversion between primary and metastatic breast cancer: mRNA assessment and its concordance with immunohistochemistry

Biomarker changes between primary (PT) and metastatic tumor (MT) site may be significant in individualizing treatment strategies and can result from actual clonal evolution, biomarker conversion, or technical limitations of diagnostic tests. This study explored biomarker conversion during breast cancer (BC) progression in 67 patients with different tumor subtypes and metastatic sites via mRNA quantification and subsequently analyzed the concordance between real-time qPCR and immunohistochemistry (IHC). Immunostaining for estrogen receptor (ER), progesterone receptor (PR), HER2, and Ki-67 was performed on formalin-fixed, paraffin-embedded PT and MT tissue sections. RT-qPCR was performed using a multiplex RT-qPCR kit for ESR1, PGR, ERBB2, and MKI67 and the reference genes B2M and CALM2. Subsequent measurement of tumor biomarker mRNA expression to detect conversion revealed significant decreases in ESR1 and PGR mRNA and MKI67 upregulation (all p < 0.001) in MT compared to PT of all tumor subtypes and ERBB2 upregulation in MT from triple-negative PT patients (p = 0.023). Furthermore, ERBB2 mRNA was upregulated in MT brain biopsies, particularly those from triple-negative PTs (p = 0.023). High concordance between RT-qPCR and IHC was observed for ER/ESR1 (81%(κ 0.51) in PT and 84%(κ 0.34) in MT, PR/PGR (70%(κ 0.10) in PT and 78% (κ -0.32) in MT), and for HER2/ERBB2 (100% in PT and 89% in MT). Discordance between mRNA biomarker assessments of PT and MT resulting from receptor conversion calls for dynamic monitoring of BC tumor biomarkers. Overall, RT-qPCR assessment of BC target genes and their mRNA expression is highly concordant with IHC protein analysis in both primary and metastatic tumor.

Technical validation of an RT-qPCR in vitro diagnostic test system for the determination of breast cancer molecular subtypes by quantification of ERBB2, ESR1, PGR and MKI67 mRNA levels from formalin-fixed paraffin-embedded breast tumor specimens

Background

MammaTyper is a novel CE-marked in vitro diagnostic RT-qPCR assay which assigns routinely processed breast cancer specimens into the molecular subtypes Luminal A-like, Luminal B-like (HER2 positive or negative), HER2 positive (non-luminal) and Triple negative (ductal) according to the mRNA expression of ERBB2, ESR1, PGR and MKI67 and the St Gallen consensus surrogate clinical definition. Until now and regarding formalin-fixed, paraffin-embedded material (FFPE), this has been a task mostly accomplished by immunohistochemistry (IHC). However the discrepancy rates of IHC for the four breast cancer biomarkers are frequently under debate, especially for Ki-67 which carries the highest degree of inter- and even intra-observer variability. Herein we describe a series of studies in FFPE specimens which aim to fully validate the analytical performance of the MammaTyper assay, including the site to site reproducibility of the individual marker measurements.

Methods

Tumor RNA was extracted with the novel RNXtract RNA extraction kit. Synthetic RNA was used to assess the sensitivity of the RNXtract kit. DNA and RNA specific qPCR assays were used so as to determine analyte specificity of RNXtract. For the assessment of limit of blank, limit of detection, analytical measurement range and PCR efficiency of the MammaTyper kit serial dilutions of samples were used. Analytical precision studies of MammaTyper were built around two different real time PCR platforms and involved breast tumor samples belonging to different subtypes analyzed across multiple sites and under various stipulated conditions. The MammaTyper assay robustness was tested against RNA input variations, alternative extraction methods and tumor cell content.

Results

Individual assays were linear up to at least 32.33 and 33.56 Cqs (quantification cycles) for the two qPCR platforms tested. PCR efficiency ranged from 99 to 109 %. In qPCR platform 1, estimates for assay specific inter-site standard deviations (SD) were between 0.14 and 0.20 Cqs accompanied by >94 % concordant single marker assignments for all four markers. In platform 2, the inter-site SD estimates were between 0.40 and 0.66 Cqs while the concordance for single marker assignments was >94 % for all four markers. The agreement reached between the two qPCR systems located in one site was 100 % for ERBB2, 96.9 % for ESR1, 97.2 % for PGR and 98.6 % for MKI67. RT-qPCR for individual markers was stable up to a 64-fold dilution for a typical clinical sample. There was no change in assay performance detected at the level of individual markers or subtypes after using different RNA isolation methods. The presence of up to 80 % of surrounding non-tumor tissue including in situ carcinoma did not affect the assay output. Sixteen out of 20 RNXtract eluates yielded more than 50 ng/μl of RNA (average RNA output: 233 ng/μl), whereas DNA contamination per sample was restricted to less than 15 ng/μl. Median recovery rate of RNA extraction was 91.0 %.

Conclusions

In this study the performance characteristics of MammaTyper were successfully validated. The various sources of analytical perturbations resulted in negligible variations in individual marker assessments. Therefore, MammaTyper may serve as a technical improvement to current standards for decentralized FFPE-based routine assessment of the commonly used breast cancer biomarkers and for molecular subtyping of breast cancer specimens.

Electronic supplementary material

The online version of this article (doi:10.1186/s12885-016-2476-x) contains supplementary material, which is available to authorized users.

Neural Transdifferentiation: MAPTau Gene Expression in Breast Cancer Cells

BACKGROUND

In tumor cells, aberrant differentiation programs have been described. Several neuronal proteins have been found associated with morphological neuronal-glial changes in breast cancer (BCa). These neuronal proteins have been related to mechanisms that are involved in carcinogenesis; however, this regulation is not well understood. Microtubule-associated protein-tau (MAP-Tau) has been describing in BCa but not its variants. This finding could partly explain the neuronal-glial morphology of BCa cells. Our aim was to determine mRNA expression of MAP-tau variants 2, 4 and 6 in breast cancer cell lines.

MATERIALS AND METHODS

Cultured cell lines MCF-10A, MDA-MB-231, SKBR3 and T47D were observed under phase-contrast microscopy for neural morphology and analyzed for gene expression of MAP-Tau transcript variants 2, 4 and 6 by real-time PCR.

RESULTS

Regarding morphology like neural/glial cells, T47D line shown more cells with these features than MDA-MB-231 and SKBR. In another hand, we found much greater mRNA expression of MAP-Tau transcript variants 2, and to a lesser extent 4 and 6, in T47D cells than the other lines. In conclusion, regulation of MAP- Tau could bring about changes in cytoskeleton, cell morphology and motility; these findings cast further light on neuronal transdifferentiation in BCa.

Biological subtyping of early breast cancer: a study comparing RT-qPCR with immunohistochemistry

The biological subtype of breast cancer influences the selection of systemic therapy. Distinction between luminal A and B cancers depends on consistent assessment of Ki-67, but substantial intra-observer and inter-observer variability exists when immunohistochemistry (IHC) is used. We compared RT-qPCR with IHC in the assessment of Ki-67 and other standard factors used in breast cancer subtyping. RNA was extracted from archival breast tumour tissue of 769 women randomly assigned to the FinHer trial. Cancer ESR1, PGR, ERBB2 and MKI67 mRNA content was quantitated with an RT-qPCR assay. Local pathologists assessed ER, PgR and Ki-67 expression using IHC. HER2 amplification was identified with chromogenic in situ hybridization (CISH) centrally. The results were correlated with distant disease-free survival (DDFS) and overall survival (OS). qPCR-based and IHC-based assessments of ER and PgR showed good concordance. Both low tumour MKI67 mRNA (RT-qPCR) and Ki-67 protein (IHC) levels were prognostic for favourable DDFS [hazard ratio (HR) 0.42, 95 % CI 0.25–0.71, P = 0.001; and HR 0.56, 0.37–0.84, P = 0.005, respectively] and OS. In multivariable analyses, cancer MKI67 mRNA content had independent influence on DDFS (adjusted HR 0.51, 95 % CI 0.29–0.89, P = 0.019) while Ki-67 protein expression had not any influence (P = 0.266) whereas both assessments influenced independently OS. Luminal B patients treated with docetaxel-FEC had more favourable DDFS and OS than those treated with vinorelbine-FEC when the subtype was defined by RT-qPCR (for DDFS, HR 0.52, 95 % CI 0.29–0.94, P = 0.031), but not when defined using IHC. Breast cancer subtypes approximated with RT-qPCR and IHC show good concordance, but cancer MKI67 mRNA content correlated slightly better with DDFS than Ki-67 expression. The findings based on MKI67 mRNA content suggest that patients with luminal B cancer benefit more from docetaxel-FEC than from vinorelbine-FEC.

Opposing prognostic roles of nuclear and cytoplasmic RACGAP1 expression in colorectal cancer patients

Rac GTPase activating protein 1 (RACGAP1) plays a regulatory role in initiation of cytokinesis, control of cell growth and differentiation, and tumor malignancy, making it a potential prognostic biomarker. RACGAP1 is present in the nucleus, but a diffuse distribution in the cytoplasm also occurs. The aim of this study was to determine the impact of nuclear and cytoplasmic expression of RACGAP1 on clinical outcome to provide further evidence of a role in colorectal cancer. RACGAP1 expression was analyzed by immunohistochemistry in 166 cancer specimens from primary colorectal cancer patients. The mean follow-up time after surgery was 5.4 years (range, 0.01-13.10 years). The prognostic value of RACGAP1 on overall survival was validated by Kaplan-Meier analysis and Cox regression models. RACGAP1 is expressed in colorectal specimen and is present in both the nucleus and cytoplasm in different amounts. Colorectal cancer patients had opposite prognoses depending on the site of RACGAP1 expression. Patients with high nuclear RACGAP1 expression had poor outcomes, whereas those with high cytoplasmic RACGAP1 expression had favorable prognosis (P = .003 and P = .001, respectively). Patients with low nuclear but high cytoplasmic RACGAP1 expression had better survival compared with those with other combinations (P < .001). We suggest that RACGAP1 expression levels in the nucleus and cytoplasm, determined by immunohistochemical staining, predict opposite clinical outcomes and that both could be independent prognostic markers for colorectal cancer.

The Discovery of Novel Biomarkers Improves Breast Cancer Intrinsic Subtype Prediction and Reconciles the Labels in the METABRIC Data Set

BACKGROUND

The prediction of breast cancer intrinsic subtypes has been introduced as a valuable strategy to determine patient diagnosis and prognosis, and therapy response. The PAM50 method, based on the expression levels of 50 genes, uses a single sample predictor model to assign subtype labels to samples. Intrinsic errors reported within this assay demonstrate the challenge of identifying and understanding the breast cancer groups. In this study, we aim to: a) identify novel biomarkers for subtype individuation by exploring the competence of a newly proposed method named CM1 score, and b) apply an ensemble learning, as opposed to the use of a single classifier, for sample subtype assignment. The overarching objective is to improve class prediction.

METHODS AND FINDINGS

The microarray transcriptome data sets used in this study are: the METABRIC breast cancer data recorded for over 2000 patients, and the public integrated source from ROCK database with 1570 samples. We first computed the CM1 score to identify the probes with highly discriminative patterns of expression across samples of each intrinsic subtype. We further assessed the ability of 42 selected probes on assigning correct subtype labels using 24 different classifiers from the Weka software suite. For comparison, the same method was applied on the list of 50 genes from the PAM50 method.

CONCLUSIONS

The CM1 score portrayed 30 novel biomarkers for predicting breast cancer subtypes, with the confirmation of the role of 12 well-established genes. Intrinsic subtypes assigned using the CM1 list and the ensemble of classifiers are more consistent and homogeneous than the original PAM50 labels. The new subtypes show accurate distributions of current clinical markers ER, PR and HER2, and survival curves in the METABRIC and ROCK data sets. Remarkably, the paradoxical attribution of the original labels reinforces the limitations of employing a single sample classifiers to predict breast cancer intrinsic subtypes.

Clinical significance of RacGAP1 expression at the invasive front of gastric cancer

BACKGROUND

Rac GTPase activating protein 1 (RacGAP1) plays a regulatory role in cell growth, transformation and metastasis. The aim of this study was to clarify the association between RacGAP1 expression and clinical outcome in patients with gastric cancer.

METHODS

A total of 232 gastric cancer patients in our institute who underwent surgery without preoperative treatments were enrolled in this study. We investigated RacGAP1 expression using immunohistochemistry (IHC) and evaluated IHC scores calculated by the percentage of positive cells and intensity and its expression at the invasive front. RACGAP1 expression was also assessed.

RESULTS

RacGAP1 expression was observed in the nuclei of gastric cancer cells. Evaluation by IHC score showed no significant correlations with clinicopathological variables except for histological differentiation. In transcriptional analyses, RACGAP1 expression was elevated in diffuse type gastric cancer than intestinal type without a significant difference. We observed significant correlations of RacGAP1 protein expression at the invasive front with older age, tumor size, lymph node metastasis, lymphatic invasion, vascular invasion and advanced stage. Patients with RacGAP1 protein expression at the invasive front had significantly poorer prognosis than those without it (P < 0.0001). In multivariate analysis, lymph node metastasis, distant metastasis and positive RacGAP1 expression at the invasive front were independent prognostic factors (lymph node metastasis: P = 0.0106; distant metastasis: P = 0.0012; RacGAP1: P = 0.0011).

CONCLUSIONS

RacGAP1 expression at the invasive front in gastric cancer was significantly correlated with factors reflecting tumor progression and poor prognosis. Our data suggest that RacGAP1 might play important roles in the progression of gastric cancer.

Preanalytical variables and performance of diagnostic RNA-based gene expression analysis in breast cancer

Prognostic multigene expression assays have become widely available to provide additional information to standard clinical parameters and to support clinicians in treatment decisions. In this study, we analyzed the impact of variations in tissue handling on the diagnostic EndoPredict test results. EndoPredict is a quantitative reverse transcription PCR assay conducted on RNA from formalin-fixed, paraffin-embedded (FFPE) tissue that predicts the likelihood of distant recurrence in patients with ER-positive/HER2-negative breast cancer. In this study, we performed a total of 138 EndoPredict assays to study the effects of preanalytical variables such as time to fixation, fixation time, tumor cell content, and section storage time on the EndoPredict test results. A time to fixation of up to 12 h and fixation of up to 5 days did not affect the results of the gene expression test. Paired samples of FFPE sections with tumor cell content ranging from 15 to 95 % and tumor-enriched samples showed a correlation coefficient of 0.97. Test results of tissue sections that have been stored for 12 months at +4 or +20 °C showed a correlation of 0.99 when compared to results of nonstored sections. In conclusion, preanalytical tissue handling is not a critical factor for diagnostic gene expression analysis with the EndoPredict assay. The test can therefore be easily integrated into the standard workflow of molecular pathology.

Association of VEGF-A splice variant mRNA expression with outcome in bevacizumab-treated patients with metastatic breast cancer

BACKGROUND

The prognostic utility of vascular endothelial growth factor A (VEGF-A) splice variants in patients with advanced breast cancer treated with bevacizumab has not been studied.

PATIENTS AND METHODS

A total of 111 patients with metastatic breast cancer treated with weekly docetaxel or ixabepilone without bevacizumab (cohort A) and 100 treated with weekly paclitaxel and bevacizumab (cohort B) were studied. Formalin-fixed tumors were macrodissected for reverse transcription quantitative polymerase chain reaction relative quantification of VEGF-A165, -189, and -206 isoforms spliced at exon 8 proximal splice site (VEGF-Axxxa) and at exon 8 distal splice site (VEGF-Axxxb).

RESULTS

For high VEGF-Axxxa, the hazard ratios (HRs) for progression were 1.08 (P = .71) in non-bevacizumab-treated patients (cohort A) and 0.66 (P = .22) in bevacizumab-treated patients (cohort B), and the HRs for death were 1.45 (P = .13) and 0.50 (P = .049), respectively. The interaction of VEGF-Axxxa with bevacizumab administration was significant (P = .011) for overall survival (OS). High tissue VEGF-Axxxb was not prognostic in cohort A but was predictive for bevacizumab benefit in cohort B (HR for progression, 0.57 [P = .04]; HR for death, 0.51 [P = .02]). Exploratory analyses done only in cohort B suggested that abundance of VEGFR1 messenger RNA (mRNA) in peripheral blood and low VEGFR2 mRNA in tissue correlated with poor outcome. In multivariate analysis, high tissue mRNA of angiogenic VEGF-Axxxa in the presence of bevacizumab therapy predicted for favorable progression-free survival (HR for progression, 0.39; P = .0227) and OS (HR for death, 0.32; P = .0140).

CONCLUSION

Tissue mRNA expression of angiogenic VEGF-Axxxa isoforms was retrospectively associated with adverse prognosis in the absence of bevacizumab and with favorable outcome when bevacizumab was administered in patients with advanced breast cancer.