Sg Hilsenbeck

Development of Resistance to Targeted Therapies Transforms the Clinically Associated Molecular Profile Subtype of Breast Tumor Xenografts

The effectiveness of therapies targeting specific pathways in breast cancer, such as the estrogen receptor or HER2, is limited because many tumors manifest resistance, either de novo or acquired, during the course of treatment. To investigate molecular mechanisms of resistance, we used two xenograft models of estrogen receptor-positive (ER+) breast cancer, one with and one without HER2 overexpression (MCF7/HER2-18 and MCF7 wt, respectively). Mice with established tumors were assigned to the following treatment groups: estrogen supplementation (E2), estrogen deprivation (ED), ED plus tamoxifen (Tam), all with or without the epidermal growth factor receptor tyrosine kinase inhibitor gefitinib (G). Another group received ED plus the antiestrogen fulvestrant (MCF7 wt only). Tumors with acquired or de novo resistance to these endocrine therapies were profiled for gene expression and compared with tumors in the E2 control group. One class of genes underexpressed in endocrine-resistant tumors (relative to E2-treated tumors) were estrogen inducible in vitro and associated with ER+ human breast cancers (luminal subtype). Another class of genes overexpressed in tumors with acquired resistance in both models represented transcriptional targets of HER2 signaling and was associated with ER-/HER2+ human cancers (ERBB2+ subtype). A third class of genes overexpressed in MCF7/HER2-18 tumors exhibiting de novo resistance to tamoxifen was associated with ER+ human cancers but not with estrogen-regulated genes. Thus, in response to various endocrine therapy regimens, these xenograft breast tumors shut down classic estrogen signaling and activate alternative pathways such as HER2 that contribute to treatment resistance. Over time, the molecular phenotype of breast cancer can change.

Cancer stem cell markers are enriched in normal tissue adjacent to triple negative breast cancer and inversely correlated with DNA repair deficiency

IntroductionWe hypothesized that cells present in normal tissue that bear cancer stem cell markers may represent a cancer cell of origin or a microenvironment primed for tumor development, and that their presence may correlate with the clinically defined subtypes of breast cancer that show increased tumorigenicity and stem cell features.MethodsNormal tissues sampled at least 5 cm from primary tumors (normal adjacent tissue) were obtained from 61 chemotherapy-naive patients with breast cancer treated with mastectomy. Samples were stained simultaneously with immunofluorescence for CD44/CD49f/CD133/2 stem cell markers. We assessed the association between CD44+CD49f+CD133/2+ staining in normal adjacent tissue and breast cancer receptor subtype (defined by the expression of the estrogen (ER), progesterone (PR), or human epidermal growth factor-2 (Her2) receptors). We also examined the correlation between CD44+CD49f+CD133/2+ immunofluorescence and each of two previously published gene signatures, one derived from stem-cell enriched tissue and one from BRCA mutated tissue expected to have defective DNA repair.ResultsPatients with triple negative breast cancer (ER–/PR–/HER2–) expressed CD44+CD49f+CD133/2+ in 9 of 9 normal adjacent tissue samples compared with 7 of 52 ER+ and/or Her2+ tumors (P < 0.001). Further, expression of CD44+CD49f+CD133/2+ by normal adjacent tissue correlated positively with a stem cell-derived tumorigenic signature (P <0.001) and inversely with a defective DNA-repair signature (P <0.001).ConclusionNormal cells bearing cancer stem cell markers are associated with the triple negative receptor subtype of breast cancer. This study suggests stem cell staining and gene expression signatures from normal breast tissues represent novel tissue-based risk biomarkers for triple negative breast cancer. Validation of these results in additional studies of normal tissue from cancer-free women could lay the foundation for future targeted triple negative breast cancer prevention strategies.

PI3 kinase activation and response to trastuzumab or lapatinib in HER-2 overexpressing locally advanced breast cancer (LABC).

CTRC-AACR San Antonio Breast Cancer Symposium: 2008 AbstractsnnAbstract #34 nnBackground: Activating mutations of PI3 kinase ( PIK3CA ) and PTEN loss may be associated with trastuzumab resistance. Trastuzumab, a HER2 humanized monoclonal antibody, and lapatinib, an EGFR/HER2 tyrosine kinase inhibitor are both established treatments. Greater understanding of the cellular response to trastuzumab or lapatinib is needed to tailor targeted therapy for individual patients and identify those less likely to benefit. Material and Methods: We performed two sequential neoadjuvant clinical trials in HER-2 overexpressing LABC: 40 patients received weekly trastuzumab at standard doses given initially as a single agent for the first 3 weeks, then in combination with 3-weekly docetaxel for 12 weeks (T), while 49 patients received lapatinib as a single agent (1,500 mg daily, orally) for 6weeks then the combination of 3-weekly trastuzumab/docetaxel for 12 weeks, before primary surgery (L). Sequential core biopsies of the primary breast tumors were taken at initial, weeks 1 and 3 after the first dose of trastuzumab, and at initial, weeks 2, 4, and 6 after lapatinib. Apoptosis, Ki67 proliferation rate, and PTEN were assessed by immunohistochemistry. Low PTEN was defined as Allred score of <3. Genomic DNA (10-100ng) was sequenced using the BigDye Terminator Cycle Sequencing Kit (Applied Biosystems) and an ABI 3730 automated capillary sequencer. Two sample and paired sample comparisons were performed using nonparametric tests. Results: There was a significant decrease in clinical tumor size after three weeks of trastuzumab (n=35, median=-20%), and six weeks of lapatinib (n=49, median=-74%) compared to pre-therapy (p<0.001). At surgery, pathologic complete response was observed in 38% in patients on upfront T and 70% patient on L. There was a significant increase in apoptosis (median=3.5% to 4.7%, p=0.006) within one week after trastuzumab, with no significant change in Ki67 at any of the time point. Lapatinib was associated with a no significant increase in apoptosis but a significant decrease in Ki67 at week 2, 4, and 6 of therapy (p<0.001). Cases with low PTEN or PIK3CA mutations were significantly less likely to have a pathologic complete response to T (p<0.005). Howver, low PTEN or PIK3CA mutations was not significantly associated with pathologic resistance to L. Conclusions: Activation of PI3 kinase pathway is associated with trastuzumab but not lapatinib resistance. Lapatinib may affect signalling through the Ras/Raf/MAPK/ERK pathway, inhibiting cell division. Low PTEN expression was not associated with lapatinib resistance, and may explain the clinical efficacy of lapatinib in trastuzumab-resistant patients, supporting clinical trials for the combination of both agents.nnCitation Information: Cancer Res 2009;69(2 Suppl):Abstract nr 34.

High IGF-IR Activity in Triple-Negative Breast Cancer Cell Lines Correlates with Sensitivity to IGF-IR Inhibitor BMS-754807 in This Subtype of Human Breast Cancer.

BACKGROUND: Evidence implicates insulin-like growth factor-I (IGF-I) signaling in the development and progression of breast cancer. Within the last few years several drugs targeting IGF-I receptor (IGF-IR) have entered clinical trials and are showing promising early results. BMS-754807 is a new small molecule inhibitor of IGF-IR in Phase 1 clinical trials.METHODS: Minimal and maximal IC50 to BMS-754807 for monolayer proliferation was determined for a panel of 30 breast cancer cell lines. Comparative gene expression analysis was performed using publicly available gene expression among the most resistant and sensitive cell lines. Q-RT-PCR was used to validate gene expression differences. Using the IGF gene signature, we scored each expression profile in a panel of cell lines and tumorgrafts, for correlation with the IGF-induced patterns. IGF-IR and pY-IGF-IR levels were determined in human tumorgrafts by immunohistochemistry.RESULTS: Among the different tumor cell lines, sensitivity to BMS-754807 varied widely from 0.1µM to 25µM. When defining cell lines as sensitive or resistant based on the median IC50, there was a strong enrichment for triple negative (TN; ERa-/PR-/HER2-) breast cancer cell lines in the sensitive group (11/15), while luminal breast cancer cell lines were generally resistant (11/15). We identified 136 differentially expressed genes between sensitive and resistance cell lines. Sensitive breast cancer cell lines preferentially express genes such as CAV1, and CAV2 while resistant cell lines were associated with high expression of ErbB3, and SPDEF. As luminal breast cancer cell lines were generally resistant to BMS-754807, we examined the role of estrogen and ERa on the sensitivity to BMS-754807. ERa mRNA levels strongly correlated with resistance to BMS-754807. Consistent with this, estrogen-withdrawal sensitized luminal breast cancer cells to BMS-754807. Previously, we developed an IGF-I gene signature that was derived from breast cancer cells stimulated with IGF-I, and reported that the signature strongly correlated with the TN subtype of human breast cancer. Consistent with this observation, we found that the IGF-I signature was high in TN breast cancer cell lines, and the IGF-I signature correlated with sensitivity to BMS-754807. To examine this further, we measured IGF-IR and pY-IGF-IR levels in four recently developed tumorgraft models of TN human breast cancer and found strong reactivity in half of the models. Interestingly, the tumorgraft (MC1) with the highest IGF-IR levels and activity showed the strongest enrichment for the IGF-I gene signature.CONCLUSIONS: In summary, our data indicates that the IGF-I pathway is highly active in TN breast cancer and this study provides the preclinical rationale for targeting IGF-IR in this subtype of human breast cancer. Citation Information: Cancer Res 2009;69(24 Suppl):Abstract nr 1132.

BRCA1 gene expression signature predicts for anthracycline-chemosensitivity in triple-negative breast cancer.

CTRC-AACR San Antonio Breast Cancer Symposium: 2008 AbstractsnnAbstract #6039 nnBackground: We used a previously published gene expression signature that can identify tumors from BRCA1 mutation carriers to evaluate its predictive value in triple-negative breast cancer as a marker for chemosensitivity to anthracycline-based chemotherapy. We proposed that based on preclinical evidence suggesting that BRCA1-deficient breast cancer cells are sensitive to DNA damaging drugs such as cisplatin and anthracyclines this gene expression profile may identify tumors with anthracycline chemosensitivity. Two previously published studies defined a gene expression signature associated with BRCA1 germline mutation.(1,2) In these studies, sporadic tumors were misclassified as BRCA1 tumors and further analysis revealed methylation of the BRCA1 promoter region and decreased BRCA1 gene expression. This finding suggests the possibility of identifying sporadic tumors with decreased BRCA1 activity. Methods: We selected from our database of a locally advanced breast cancer neoadjuvant trial all cases of triple negative breast cancer that received 4 cycles of doxorubicin/cyclophosphamide(AC, 60/200 mg/m2, every 3 weeks) prior to surgery. Pathologic response to chemotherapy was disappearance of all invasive cancer or microscopic residual disease. Tumoral gene expression profile previously obtained using Affymetrix U133A Chip was analyzed for an optimal set of 100 most differentially expressed genes distinguishing BRCA1 and sporadic triple negative tumors according to the previously identified gene signature by vant Veer et al.1 We performed unsupervised clustering to determine if this signature could classify a subtype of triple-negative tumors with BRCAness and to test our hypothesis that BRCA1-like tumors are more sensitive to AC. We then performed a supervised analysis to determine the most differentially expressed genes that could prospectively identify triple-negative sporadic tumors with “BRCAness” and tumors from BRCA1 germline carriers that are sensitive to anthracyclines. Results: Of the 66 patients enrolled in our neoadjuvant trial, 12 patients tumors were triple negative and received preoperative AC. By unsupervised clustering, the gene expression pattern associated with BRCA1 cancers subdivided these sporadic cancers in to two groups: Group A(6/7 pathologic responders), and group B(5/5 non-pathologic responders). By supervised analysis, the most differentially overexpressed gene from the BRCA1 profile for AC sensitivity was YWHAH(14-3-3 eta polypeptide), while DKK3(Inhibitor of Wnt and Notch signaling pathway) and RPL23A were most overexpressed in all cases with adriamycin-resistance(p<0.01). Discussion: Triple negative sporadic breast cancer displaying “BRCAness” appear to be sensitive to AC chemotherapy. YWHAH, DKK3, and RPL23A are differentially expressed in anthracycline-sensitive versus resistant tumors. These three genes can potentially identify triple-negative breast cancers that exhibit “BRCAness” and sensitivity to DNA-damaging chemotherapy such as cisplatin, anthracycline, or PARP inhibitors.nnCitation Information: Cancer Res 2009;69(2 Suppl):Abstract nr 6039.

P1-08-04: Obesity, Adjuvant Therapy, and Survival Outcomes in Early-Stage Breast Cancer.

BACKGROUND: Obesity has risen to epidemic proportions and is associated with worse breast cancer (BC) prognosis in most studies. However, the effects of obesity according to adjuvant therapy choice are largely unknown. To address this issue, we examined the relationship between body mass index (BMI), adjuvant therapy, and survival outcomes in a large cohort of early-stage BC patients. METHODS: We retrospectively studied patients from the Baylor Breast Center Tumor Bank treated from 1970–1995. Patients were divided into 3 BMI classes: normal/underweight (N, BMI RESULTS: There were 4,368 patients. Median age was 58. 74% were postmenopausal. 72% had stage I-II disease, 28% stage III. 76% were estrogen receptor (ER)-positive, 24% ER-negative. Patients distributed into BMI classes as follows: N 48%, Ov 30%, Ob 22%. Higher BMI was associated with postmenopausal status and increasing age, tumor size, positive lymph nodes, and stage, as well as a higher likelihood of receiving treatment. Median follow-up was 5 years. Kaplan-Meier analysis showed that TTR was significantly shorter in the Ov and Ob groups as compared to the N group (p=0.019), due to distant (p=0.001) rather than local (p=0.970) recurrences. DFS was also significantly worse in the Ov and Ob groups (p=0.002), as was OS (p=0.001). The Table shows the hazard ratios for the various survival outcomes after adjustment for age, tumor size, nodal status, and treatment groups. For all patients, TTR, DFS, and OS were significantly worse in the Ob vs. N groups. TTR and DFS were significantly worse in the chemo treated Ob vs. N groups. DFS and OS were significantly better in the endo treated Ov vs. N groups. DISCUSSION: In this large cohort of BC patients, survival outcomes (TTR, DFS, OS) were significantly worse in the obese group. This remained true after adjustment for multiple factors. Obesity was associated with worse survival outcomes in the chemo treated (CMF) group. Overweight was associated with better survival outcomes in the endo treated (tamoxifen) group. These results confirm and extend the results of previous studies. Further studies to discover the reasons for these differences in outcomes are underway. Citation Information: Cancer Res 2011;71(24 Suppl):Abstract nr P1-08-04.

Hyperprolactinemia-inducing antipsychotics increase breast cancer risk by activating JAK-STAT5 in precancerous lesions

BackgroundPsychiatric medications are widely prescribed in the USA. Many antipsychotics cause serum hyperprolactinemia as an adverse side effect; prolactin-Janus kinase 2 (JAK2)-signal transducer and activator of transcription 5 (STAT5) signaling both induces cell differentiation and suppresses apoptosis. It is controversial whether these antipsychotics increase breast cancer risk.MethodsWe investigated the impact of several antipsychotics on mammary tumorigenesis initiated by retrovirus-mediated delivery of either ErbB2 or HRas or by transgenic expression of Wnt-1.ResultsWe found that the two hyperprolactinemia-inducing antipsychotics, risperidone and pimozide, prompted precancerous lesions to progress to cancer while aripiprazole, which did not cause hyperprolactinemia, did not. We observed that risperidone and pimozide (but not aripiprazole) caused precancerous cells to activate STAT5 and suppress apoptosis while exerting no impact on proliferation. Importantly, we demonstrated that these effects of antipsychotics on early lesions required the STAT5 gene function. Furthermore, we showed that only two-week treatment of mice with ruxolitinib, a JAK1/2 inhibitor, blocked STAT5 activation, restored apoptosis, and prevented early lesion progression.ConclusionsHyperprolactinemia-inducing antipsychotics instigate precancerous cells to progress to cancer via JAK/STAT5 to suppress the apoptosis anticancer barrier, and these cancer-promoting effects can be prevented by prophylactic anti-JAK/STAT5 treatment. This preclinical work exposes a potential breast cancer risk from hyperprolactinemia-inducing antipsychotics in certain patients and suggests a chemoprevention regime that is relatively easy to implement compared to the standard 5-year anti-estrogenic treatment in women who have or likely have already developed precancerous lesions while also requiring hyperprolactinemia-inducing antipsychotics.

Abstract B01: B cell lymphoma 9 mediates a cross talk between the canonical Wnt and EGFR signaling in breast cancer

Human ductal carcinoma in situ (DCIS) are the most common type of non-invasive breast cancers. The five-year survival rate for women diagnosed with non-invasive DCIS is 98% while the five-year survival plummets to 83-27% for breast cancers that have become invasive and have spread to distant sites [also referred to as invasive ductal carcinoma (IDC)]. To study DCIS pathobiology and factors that promote their transition to IDC, we have developed a novel in vivo DCIS model, MIND (mouse-intraductal), that involves intraductal injection of epithelial cells derived from primary human DCIS biopsy and surgical samples thus mimicking the entire process of DCIS to IDC transition. As a complementary approach, we have utilized human DCIS/IDC tandem lesions, which are patient DCIS that show a transition to IDC within the same breast. Analysis of RNA and protein at distinct stages of in situ to IDC using both models showed B cell lymphoma-9 (BCL9) up-regulation to be associated with DCIS transition to IDC. BCL9 is a recently identified co-activator of Wnt-stimulated beta-catenin-mediated transcription. Our studies showed that in vivo silencing of BCL9 led to inhibition of DCIS invasion and reversal of EMT. We have also demonstrated a direct binding interaction between BCL9 and beta-catenin and showed suppression of beta-catenin-mediated transcription by BCL9 knockdown. Analysis of patient DCIS samples revealed a significant correlation between high nuclear BCL9 expression and pathologic characteristics associated with DCIS recurrence: Estrogen receptor (ER) negative and Ki67. Furthermore, analysis of the TCGA data showed BCL9 gene to be upregulated in 26% of breast cancers. This is a significant gene alteration when compared to HER2 (ERBB2) gene (19%) and estrogen receptor (ESR1) gene (8%) alterations in breast cancers. Interestingly, a significantly higher proportion of basal like invasive breast cancers compared to luminal breast cancers showed BCL9 amplification suggesting that BCL9 may predispose to the development of basal breast cancers. We have performed an RPPA analysis on our DCIS cell lines KD BCL9 vs. control. This analysis indicated that BCL9 KD showed down-regulation in a number of genes in the EGFR signaling pathway including p-EGFR, p-HER2, p-STAT3, and p-Src. Conclusion: BCL9 is a molecular driver of DCIS invasive progression. The molecular mechanism for BCL99s role in breast cancer progression is through the enhancement in the canonical Wnt and EGFR signaling. Citation Format: Hanan Elsarraj, Hong Yan, Jennifer Knapp, Anna Tsimelzon, Shixia Huang, Andrew Godwin, Sue Hilsenbeck, Dean Edwards, Fariba Behbod. B cell lymphoma 9 mediates a cross talk between the canonical Wnt and EGFR signaling in breast cancer. [abstract]. In: Proceedings of the AACR Special Conference on Advances in Breast Cancer Research; Oct 17-20, 2015; Bellevue, WA. Philadelphia (PA): AACR; Mol Cancer Res 2016;14(2_Suppl):Abstract nr B01.

Abstract CT319: A randomized, controlled trial of high dose vs. standard dose vitamin D for aromatase inhibitor-induced arthralgia in breast cancer survivors

Proceedings: AACR Annual Meeting 2014; April 5-9, 2014; San Diego, CAnnAromatase inhibitors (AIs) are the most effective adjuvant anti-hormonal therapy for estrogen receptor positive (ER+) post-menopausal breast cancer patients, with superior disease free survival over tamoxifen. However, approximately half of the women who take this drug will develop significant joint pains, termed Aromatase Inhibitor-Induced Arthralgia (AIA). Though this medicine should be taken for 5 years, the joint pain can be so troublesome that 13-20% may prematurely discontinue it because of the arthralgia, thus sacrificing their best chance of disease free survival. Furthermore, daily quality of life is significantly affected by this chronic pain while taking AI therapy. Nonetheless, neither the etiology nor the treatment of AIA is well-understood within the medical community.nnIt is known that low estrogen levels, as seen on AI therapy, decrease the amount of available, active vitamin D. Low Vitamin D is implicated in non-specific musculoskeletal pain in non-cancer patients, possibly mediated by higher levels of inflammatory cytokines IL-6 and TNF-alpha in a low vitamin D state. Vitamin D replacement has been shown to be helpful for AIA in some small clinical trials. We therefore propose to test the hypothesis that AIA can be effectively prevented by high dose Vitamin D supplementation, and that effective prophylaxis of the problem will lead to improved compliance with AI therapy.nnIn this clinical trial, we will enroll 184 women who are beginning adjuvant AI therapy and randomize half of them to receive high dose vitamin D (50,000 IU Vitamin D3 per week for 12 weeks, followed by 2000 IU Vitamin D3 daily for 40 weeks), and the other half to receive standard dose vitamin D (800 IU Vitamin D3 daily for 52 weeks). We will assess each womans baseline joint pains via a questionnaire (Health Assessment Questionnaire-II). We hypothesize that AIA can be effectively prevented with high dose vitamin D, and this will correlate to increased compliance in the high dose vitamin D arm.. The primary endpoint is development of AIA as measured by a defined change in HAQ-II score between baseline and 12 wks. This randomized Phase 2 study is powered to detect an improvement from an expected 33% incidence of AIA in the standard dose group to 18% or less in the high-dose group (alpha=10%, one-tailed). Finally, as an exploratory objective, we will determine whether those in the standard dose vitamin D arm have a correlation between higher levels of inflammatory cytokines and development of AIA.nnThis study targets a very common cause of pain among breast cancer survivors and aims to offer an effective treatment strategy to alleviate pain and improve quality of life as well as medication compliance.nnCitation Format: Polly A. Niravath, Sue Hilsenbeck, Tao Wang, Mothaffar Rimawi. A randomized, controlled trial of high dose vs. standard dose vitamin D for aromatase inhibitor-induced arthralgia in breast cancer survivors. [abstract]. In: Proceedings of the 105th Annual Meeting of the American Association for Cancer Research; 2014 Apr 5-9; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2014;74(19 Suppl):Abstract nr CT319. doi:10.1158/1538-7445.AM2014-CT319

Abstract ES3-1: Early Phase Trials – What Are They For?

Early in drug development, there are lots of things we need to learn about new agents or new combinations of agents in order to make ‘good’ decisions about whether and how to proceed. The landscape of clinical trials is changing, with greater focus on targeted therapies and biomarker defined patient populations. Trial design, as we will hear in this session, must adapt to these new challenges. However, fundamentals still apply. We still need to demonstrate that drugs are safe and effective. As an introduction to the session, we will review the basic steps in drug development, especially early development, and the basic questions that must be answered at each step, in order to develop evidence for safety and efficacy. Citation Information: Cancer Res 2012;72(24 Suppl):Abstract nr ES3-1.