AI Resistance Research Articles

S642 Assessing the Quality of AI Responses and Resistance to Sycophancy in Providing Patient-Centered Medical Advice on Gastroesophageal Reflux Disease

S642 Assessing the Quality of AI Responses and Resistance to Sycophancy in Providing Patient-Centered Medical Advice on Gastroesophageal Reflux Disease

Exploring Antibiotic Resistance Through Artificial Intelligence: A Novel Perspective

Microbial resistance has long been linked to antibiotic resistance, a serious worldwide health issue. But as technology advances quickly in this day and age, a related phenomenon in the field of artificial intelligence [AI] is beginning to take shape. In order to better understand the idea of "antibiotic resistance" in the context of artificial intelligence [AI], this study will compare and contrast the evolution of bacterial resistance with potential obstacles in the design and implementation of intelligent systems. The increasing prevalence of AI systems across several industries highlights the striking similarities between their capacity to adapt and withstand hostile attacks and changing surroundings, and the biological resistance mechanisms seen in bacteria. This study explores the causes behind AI resistance, looking at how data drift, adversarial manipulations, and changing user behavior might cause machine learning systems to lose their effectiveness over time. The paper also examines the ethical ramifications of AI resistance, addressing issues with biases, unforeseen outcomes, and the influence of intelligent systems on society that are resistant to change or intervention. The area of antibiotic stewardship in medicine serves as an inspiration for the paper's discussion of potential mitigation techniques for AI resistance. Through the identification of parallels between AI resistance and antibiotic resistance in bacteria, this study adds to a better comprehension of the difficulties pertaining to the long-term viability and efficiency of intelligent systems. Since AI will continue to be a major influence on the future, it is critical to address the problem of "antibiotic resistance" in this context in order to ensure that AI is developed responsibly and ethically.

Overexpression of BQ323636.1 contributes to anastrozole resistance in AR+ve/ER+ve breast cancer.

Aromatase inhibitors (Ais) are used as adjuvant endocrine therapy for oestrogen receptor-positive (ER+ve) post-menopausal breast cancer patients. Ais, by inhibiting the enzyme aromatase, block the conversion of androgen to oestrogen, reducing oestrogen levels. Resistance to Ais limits their clinical utilisation. Here, we show that overexpression of BQ323636.1 (BQ), a novel splice variant of nuclear co-repressor NCOR2, is associated with resistance to the non-steroidal aromatase inhibitor anastrozole in ER+ve post-menopausal breast cancer. Mechanistic study indicates that BQ overexpression enhances androgen receptor (AR) activity and in the presence of anastrozole, causes hyper-activation of AR signalling, which unexpectedly enhanced cell proliferation, through increased expression of CDK2, CDK4, and CCNE1. BQ overexpression reverses the effect of anastrozole in ER+ve breast cancer in an AR-dependent manner, whilst co-treatment with the AR antagonist bicalutamide recovered its therapeutic effect both in vitro and in vivo. Thus, for BQ-overexpressing breast cancer, targeting AR can combat anastrozole resistance. Clinical study of 268 primary breast cancer samples of ER+ve patients who had been treated with non-steroidal Ais showed 32.5% (38/117) of cases with combined high nuclear expression of BQ and AR, which were found to be significantly associated with Ai resistance. Non-steroidal Ai-treated patients with high nuclear expression of both BQ and AR had poorer overall, disease-specific, and disease-free survival. These findings suggest the importance of assessing BQ and AR expression status in the primary ER+ve breast tumour prior to Ai treatment. This may save patients from inappropriate treatment and enable effective therapy to be given at an early stage. © 2023 The Authors. The Journal of Pathology published by John Wiley & Sons Ltd on behalf of The Pathological Society of Great Britain and Ireland.

Physicians’ acceptance of extended or intermittent adjuvant endocrine therapy (AET): A survey on breast cancer (BC) physicians.

e12515 Background: The results from the GIM4, SALSA, and SOLE trials are potentially practice-changing, but it was unknown whether physicians would accept these trial results and adjust their AET prescriptions for postmenopausal women with BC. Methods: A cross-sectional survey involving 77 BC physicians in China was conducted. The survey questions were developed based on the GIM4, SALSA, and SOLE trial results, as well as referred to the latest NCCN guidelines. Statistical analyses were performed using SPSS version 28.0. Results: Almost half (48%) of the respondents were breast surgeons, 19% were breast specialists, 8% were medical oncologists, and 2% were general surgeons. Forty-five percent of the respondents had a clinical experience of ≥15 years, and the mean ± SD number of BC patients managed annually was 762.10 ± 1,254.50. For pN0, tumor size ≤0.5cm, postmenopausal patients, respondents with ≥15 years of clinical experience were more likely to prescribe a longer duration of AET as compared to those with <15 of clinical experience (p=0.025 and p=0.038 for patients receiving prior 4-5 years of tamoxifen [TAM] and prior 5 years of aromatase inhibitor [AI], respectively). More than half (57.1%) did not agree with the hypothesis that AI resistance can be reversed by intermittent letrozole treatment (the SOLE trial hypothesis), but 50.6% considered intermittent letrozole as an acceptable option, especially for patients with ≤3 positive nodes (79.5%), tumor grade 1 (84.6%), tumor size ≤2cm (82.1%), HER2 negative disease (74.4%). Conclusions: A high proportion of the BC physician respondents would prescribe extended AET with AI beyond 5 years for postmenopausal women with BC, especially for those with higher risk. Intermittent letrozole was considered as an acceptable option for low-risk patients.[Table: see text]

Enhanced IFNα Signaling Promotes Ligand-Independent Activation of ERα to Promote Aromatase Inhibitor Resistance in Breast Cancer.

Simple SummaryInterferon alpha (IFNα) signaling is highly upregulated in ER+ breast cancers that become resistant to estrogen deprivation therapy. This study uncovers how enhanced (IFNα)/JAK-STAT signaling directly influences estrogen receptor (ERα) activation in the absence of estrogen. We found that inhibiting IFNα signaling downregulates expression and activation of ERα. In addition, STAT1 and ERα directly interact and regulate a key interferon-stimulated gene, IFITM1, in AI-resistant breast cancer cells. We demonstrate that crosstalk occurs between IFNα and ERα pathways, which contributes to aggression and survival of AI-resistant breast cancer, thus representing a novel mechanism of acquired AI resistance.Aromatase inhibitors (AIs) reduce estrogen levels up to 98% as the standard practice to treat postmenopausal women with estrogen receptor-positive (ER+) breast cancer. However, approximately 30% of ER+ breast cancers develop resistance to treatment. Enhanced interferon-alpha (IFNα) signaling is upregulated in breast cancers resistant to AIs, which drives expression of a key regulator of survival, interferon-induced transmembrane protein 1 (IFITM1). However, how upregulated IFNα signaling mediates AI resistance is unknown. In this study, we utilized MCF-7:5C cells, a breast cancer cell model of AI resistance, and demonstrate that these cells exhibit enhanced IFNα signaling and ligand-independent activation of the estrogen receptor (ERα). Experiments demonstrated that STAT1, the mediator of intracellular signaling for IFNα, can interact directly with ERα. Notably, inhibition of IFNα signaling significantly reduced ERα protein expression and ER-regulated genes. In addition, loss of ERα suppressed IFITM1 expression, which was associated with cell death. Notably, chromatin immunoprecipitation experiments validated that both ERα and STAT1 associate with ERE sequences in the IFITM1 promoter. Overall, hyperactivation of IFNα signaling enhances ligand-independent activation of ERα, which promotes ER-regulated, and interferon stimulated gene expression to promote survival in AI-resistant breast cancer cells.

Abstract 3432: Evolving genomes of breast cancer aromatase inhibitor resistance models

Abstract Acquired endocrine resistance is a severe barrier in the anti-hormonal treatment of breast cancer and is responsible for its failure in about half of the treated patients. In vitro models of long-term estrogen deprived breast cancer cells serve as laboratory tools to explore new therapeutic strategies. The models MCF-7:5C and MCF-7:2A mimic clinical AI resistance but regress with E2 due to the reconfiguration of survival signaling known as E2-inducible apoptosis (5C rapid, 2A delayed reaction). Alterations of stress related pathways including the accumulation of endoplasmic reticulum stress, oxidative stress, and inflammatory stress that occur prior to E2-induced apoptosis (Ariazi et al. 2011; Fan et al. 2013, 2015; Sweeney et al. 2014) and specific miRNA expression profiles (Hoppe et al Oncotarget 2016) have been linked with these resistance and E2-inducible apoptosis phenotypes. Little is known on their underlying genetic variability. Here we compare the mutational profiles of 5C and 2A clones with those of the estradiol stimulated growth sensitive MCF-7:WS8 and standard MCF-7 (ATCC) cells. Whole genome libraries were sequenced on the Illumina HiSeq platform (2x150bp reads). Alignments were manually inspected (bcftools mpileup) to avoid false negative variant calling in poorly (<20x) covered regions. Loci with at least 10x coverage and absence of any variant reads were classified as wildtype. The coding genome positions defined as mutant or wildtype in all four cell types served as a basis to select candidate mutations for the verification of MCF-7 variants via comparison to the COSMIC database (http://cancer.sanger.ac.uk/cosmic) and identification of novel mutations by variant calling (GATK HaplotypeCaller). Putative protein effects of variants were predicted by the SnpEff tool. We verified 421 point mutations in MCF-7 (COSMIC) of which 203 mutations were lost in our models, with additional specific losses of 4, 4, and 18 mutations in WS8, 2A and 5C, respectively. In 5C cells new mutations affected 261 genes with 31 high impact mutations based on splice site alterations in 7 genes (i.e. CERS2, SEMA3A, TDRD7, ATP5H) and gain of stop codons in 24 genes (i.e. CACNA1S, NOTCH4, SUN1, JPH3, SERPINB2). Similarly, in 2A cells there were 12 high impact mutations among 122 affected genes (e.g. ARHGEF5, ARHGAP35, ARAP2). We will report on the integrative genomic analysis including larger structural variants (short indels and CNVs) towards the evolution of these models for a comprehensive understanding of the biology of E2-inducible apoptosis. Citation Format: Reiner Hoppe, Siarhei Kandabarau, Ping Fan, V Craig Jordan, Hiltrud B. Brauch. Evolving genomes of breast cancer aromatase inhibitor resistance models [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2018; 2018 Apr 14-18; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2018;78(13 Suppl):Abstract nr 3432.

Abstract 3993: MicroRNA analysis of aromatase inhibitor-resistant breast cancer cells reveals upregulation of a unique miRNA cluster on chromosome 14

Abstract Estrogen signaling has been shown to play a large role in the development and progression of breast tumors that express the estrogen receptor (ERα). Thus, endocrine therapies that either inhibit ERα action (tamoxifen or fulvestrant) or inhibit estrogen synthesis (aromatase inhibitors) have been increasingly valuable for the management of ER+ breast cancer. A frequent obstacle, however, is acquired resistance to endocrine therapies, often characterized by aberrant ERα expression and heightened growth factor signaling. Another aspect of acquired endocrine resistance that is now being examined is the role of microRNA (miRNA). miRNAs are 18-22 base pair RNA molecules whose primary role is to inhibit expression of mRNA targets by inhibition of translation or by degradation of the mRNA. Here, we have conducted next generation sequencing of the miRNA from three breast cancer cell lines - one ER+ line that is estrogen-dependent and AI-sensitive (MCF-7), and two ER+ long-term estrogen deprived (LTED) models of AI resistance (MCF-7:5C and MCF-7:2A). We found that in LTED/AI-resistant MCF-7:5C and MCF-7:2A cells there were over 330 and 240 miRNAs differentially expressed, respectively, compared to AI-sensitive MCF-7 breast cancer cells. Initially, we used qPCR to validate the expression of miRNAs that that were differentially expressed in both LTED cell lines vs. MCF-7. Notably, LTED cell lines had upregulated miR-181a (> 4 to 8-fold) and downregulated miR-221/222 expression compared to MCF-7. Elevated miR-181a has been shown to be associated with more aggressive breast cancers whereas miR-221/222 play a role in tamoxifen resistance through targeting ERα. Further bioinformatic analysis also revealed that the LTED cells had very unique expression profiles compared to one another, indicating that they have adapted to an estrogen-depleted environment through potentially different mechanisms. In MCF-7:5C cells, which have lost expression of the progesterone receptor (PR), a miRNA cluster on chromosome 14 is upregulated. This upregulated cluster, which encodes miR-127, miR-654, miR-889 and others, has been shown to be associated with Luminal B (ER+/PR-) breast cancer, consistent with the hormone receptor status of the MCF-7:5C cells. The oncogenic, estrogen-responsive miR-17/92 cluster on chromosome 13 is upregulated in the MCF-7:2A cells and these miRNAs target proteins including PTEN, p21, and Bim. Overall, these findings reveal altered expression of miRNAs in AI-resistant breast cancer cells and they suggest that targeting these miRNAs might be an effective strategy to reverse AI resistance in breast cancer patients. Citation Format: Erin L. Hayes, Jennifer Knapp, Joan Lewis-Wambi. MicroRNA analysis of aromatase inhibitor-resistant breast cancer cells reveals upregulation of a unique miRNA cluster on chromosome 14. [abstract]. In: Proceedings of the 106th Annual Meeting of the American Association for Cancer Research; 2015 Apr 18-22; Philadelphia, PA. Philadelphia (PA): AACR; Cancer Res 2015;75(15 Suppl):Abstract nr 3993. doi:10.1158/1538-7445.AM2015-3993

Abstract P3-05-24: Adaptation to AI therapy in breast cancer can induce dynamic alterations in ER activity resulting in estrogen independent metastatic tumours

Abstract Acquired resistance to aromatase inhibitor therapy is a major clinical problem in the treatment of breast cancer. The detailed mechanisms of how tumour cells develop this resistance remain unclear. Here, discovery of pro-survival adaptations to the estrogen receptor (ER) signalling pathway in response to AI treatment is reported. Global ChIPseq analysis shows that the ER retains binding activity in AI resistant cells under steroid-depleted conditions. In AI treated patients, evidence of steroid independent adaptive ER signalling was demonstrated by the ER target gene Early Growth Response 3 (EGR3). Expression of EGR3 decreased initially upon AI treatment but subsequently recovered following as little as 12 weeks of AI treatment. In vitro data indicates that this increased expression of EGR3 may enhance cell growth and motility of tumour cells. Finally, evidence from established metastatic tumours suggests that the ER signalling network may undergo further adaptations with disease progression as EGR3 expression is routinely lost in the established metastatic tumour. Overall, these data provide evidence of a dynamic ER response to AI treatment which may provide vital clues for overcoming the clinical issue of AI resistance. Citation Format: Damir Vareslija, Jean McBryan, Ailis Fagan, Aisling Redmond, Yuan Hao, Arnold DK Hill, Leonie S Young. Adaptation to AI therapy in breast cancer can induce dynamic alterations in ER activity resulting in estrogen independent metastatic tumours [abstract]. In: Proceedings of the Thirty-Seventh Annual CTRC-AACR San Antonio Breast Cancer Symposium: 2014 Dec 9-13; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2015;75(9 Suppl):Abstract nr P3-05-24.

Abstract P3-05-25: Efficacy of everolimus on multiple mechanisms of AI-resistance in vitro and xenograft, and characterization of their everolimus-resistance

Abstract Background: mTOR inhibitor, everolimus showed remarkable clinical efficacy and has been considered as a promising agent for breast cancer treatment especially among hormone receptor positive (HR+) postmenopausal advanced breast cancer. Now Affinitor (everolimus) is approved for HR+ breast cancer in Japan. However, definitive biomarker has been unclear so far. In addition, the subsequent treatment after refractory to everolimus has not been defined. In preclinical study, it has been reported there are several resistant models of endocrine therapy and we also reported several types of models established from MCF-7-E10, in which estrogen receptor (ER) activity could be monitored by transfected-ERE-GFP. Those EDR cells are characterized as the followings: 1) Type 1 shows constitutive ER overexpression without estrogen, and PI3K/Akt/mTOR pathway upregulated. 2) Type 2 shows low expression of ER (ERE-GFP negative) and IGF-1R/JNK signaling pathway upregulated (Fujiki, 2014, J Steroid Biochem Mol Biol). 3) Type4, which was cultured under the condition of estrogen depletion and addition of testosterone. It showed using androgen metabolite as ligand and both PI3K/Akt/mTOR and MAPK signaling pathway upregulated (Hanamura, 2013, Breast Cancer Res Tr). Methods: Efficacy of everolimus was analyzed in those cell lines in vitro and in vivo about inhibition of cell proliferation. In vivo study, ovarectomized mice were inoculated with Type1 or Type4. Treatments were performed as single agent letrozole, everolimus, or combination and tumor volume changes were compared. Then everolimus resistant cells were generated from those EDR cells after long term exposure to everolimus in vitro. Though we could not acquire everolimus resistant cell from Type4, Type1 and 2 showed its resistance. Those everolimus resistant (EVR) cells are equivalent to resistance to the combination therapy between AI and everolimus after the first line AI treatment failure. Using those cell lines, we investigated the different mechanism of resistance to everolimus. Results: Everolimus extremely inhibited cell proliferation in each EDR type in vitro and in vivo. Besides, Type4 was more remarkably sensitive than parental MCF-7-E10 under testosterone supplemented condition in vitro. Everolimus or combination treatments reached significant tumor volume reduction around 50 percent after 21 days treatment period, in contrast to increase of tumor volume in placebo or letrozole group. In vitro study of developing resistance to everolimus, Type1-EVR cells no longer responded to everolimus and surprisingly fulvestrant also lost sensitivity to EVR cell in spite of keeping ERE activity. Type1-EVR showed much higher expression of phosphorylation of MAP-kinase than parental cell, however, MEK inhibitor or PI3K inhibitor alone was not effective but combination with fulvestrant significantly effective more than parental cell. In type2, JNK inhibitor was effective for parental cell but EVR cell lost sensitivity to JNK inhibitor, however, Src inhibitor was much more effective in Type2-EVR than in parental cell. Conclusion: Everolimus showed remarkable efficacy to any types of EDR cells, however, the mechanism of acquired resistance to everolimus was likely to be different between each EVR cell. This finding implies that different mechanisms of AI resistance follow different types of everolimus resistance. Citation Format: Mariko Kimura, Natsu Fujiki, Toru Hanamura, Toshifumi Niwa, Yuri Yamaguchi, Takashi Ishikawa, Itaru Endo, Shin-ichi Hayashi. Efficacy of everolimus on multiple mechanisms of AI-resistance in vitro and xenograft, and characterization of their everolimus-resistance [abstract]. In: Proceedings of the Thirty-Seventh Annual CTRC-AACR San Antonio Breast Cancer Symposium: 2014 Dec 9-13; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2015;75(9 Suppl):Abstract nr P3-05-25.

P051 Photoacoustic imaging of breast cancer and histological markers of angiogenesis

Furthermore, LR cell lines proliferated by E1S in dose-dependent manner more than their parental cells, but there was no difference of proliferation when estrone and estradiol treated. STS inhibitor administered with letrozole was inhibitory to proliferation of LR cells. LR cells were also inhibited by SERM and SERD. In mRNA analysis of clinical primary tissues of postmenopausal patients, expression of STS mRNA significantly correlated with those of three OATPs which were induced in LR cell lines. Conclusion: In this study, we first demonstrated that the contribution of E1S to ER-positive breast cancer in the context of AI resistance. Targeting therapy to E1S–STS pathway could present a new choice of treatments to AI-resistant breast cancer patients, especially by combination with AI. Disclosure of Interest: No significant relationships.

Nonhypoxic regulation and role of hypoxia-inducible factor 1 in aromatase inhibitor resistant breast cancer

IntroductionAlthough aromatase inhibitors (AIs; for example, letrozole) are highly effective in treating estrogen receptor positive (ER+) breast cancer, a significant percentage of patients either do not respond to AIs or become resistant to them. Previous studies suggest that acquired resistance to AIs involves a switch from dependence on ER signaling to dependence on growth factor-mediated pathways, such as human epidermal growth factor receptor-2 (HER2). However, the role of HER2, and the identity of other relevant factors that may be used as biomarkers or therapeutic targets remain unknown. This study investigated the potential role of transcription factor hypoxia inducible factor 1 (HIF-1) in acquired AI resistance, and its regulation by HER2.MethodsIn vitro studies using AI (letrozole or exemestane)-resistant and AI-sensitive cells were conducted to investigate the regulation and role of HIF-1 in AI resistance. Western blot and RT-PCR analyses were conducted to compare protein and mRNA expression, respectively, of ERα, HER2, and HIF-1α (inducible HIF-1 subunit) in AI-resistant versus AI-sensitive cells. Similar expression analyses were also done, along with chromatin immunoprecipitation (ChIP), to identify previously known HIF-1 target genes, such as breast cancer resistance protein (BCRP), that may also play a role in AI resistance. Letrozole-resistant cells were treated with inhibitors to HER2, kinase pathways, and ERα to elucidate the regulation of HIF-1 and BCRP. Lastly, cells were treated with inhibitors or inducers of HIF-1α to determine its importance.ResultsBasal HIF-1α protein and BCRP mRNA and protein are higher in AI-resistant and HER2-transfected cells than in AI-sensitive, HER2- parental cells under nonhypoxic conditions. HIF-1α expression in AI-resistant cells is likely regulated by HER2 activated-phosphatidylinositide-3-kinase/Akt-protein kinase B/mammalian target of rapamycin (PI3K/Akt/mTOR) pathway, as its expression was inhibited by HER2 inhibitors and kinase pathway inhibitors. Inhibition or upregulation of HIF-1α affects breast cancer cell expression of BCRP; AI responsiveness; and expression of cancer stem cell characteristics, partially through BCRP.ConclusionsOne of the mechanisms of AI resistance may be through regulation of nonhypoxic HIF-1 target genes, such as BCRP, implicated in chemoresistance. Thus, HIF-1 should be explored further for its potential as a biomarker of and therapeutic target.

Abstract PD3-5: Survival benefit conferred by the androgen receptor is lost in aromatase inhibitor treated breast cancer

Abstract Aromatase Inhibitors have proven to be most effective in the treatment of post-menopausal breast cancer. Their mode of action is to inhibit the synthesis of estrogen (estrone) by the aromatase enzyme Cyp 19 thereby blocking ligand-dependent activation of the estrogen receptor. What has not been addressed to date is how cells that are deprived of estrogen, may potentially, adapt to the more androgenic environment resulting from long-term treatment with AI therapy. Research from our lab has identified the homeobox protein, HOXC11, to be an indicator of poor response to endocrine therapy and development of metastasis. To further our understanding of HOXC11 and its role in the development of endocrine-resistance and metastatic spread we undertook an RNA-seq experiment to identify its target genes in resistant breast cancer. This analysis identified PSAP, IFIT1 and HSP90AA1. Both PSAP (an androgen agonist) and HSP90AA1 (AR chaperone) are closely associated with AR which led to further investigation into the role of HOXC11 in the development of steroidal adaptability in Letrozole-resistant breast cancer. We hypothesize that HOXC11 regulated expression of PSAP results in oncogenic activation of AR in an AI resistant setting. Our findings have shown that AI-resistant cell lines in vitro have significantly elevated levels of AR and that loss of HOXC11 results in concommitant decrease in AR mRNA. In AI resistance expression of HOXC11 results in upregulation/stabilization of AR by PSAP thus enabling the tumour to adapt to use androgenic steroids for cell proliferation. The anti-androgen, Bicalutamide, reduces cell proliferation and cell motility in AI resistant cell lines. Survival analysis of AR in a TMA (n = 488) indicates that AR confers a survival benefit in the tamoxifen treated population. This protective effect is diminished in patients receiving AI therapy and is reflected in the altered Hazard Ratio of AR from the total population (HR: 0.485) to the AI treated cohort (HR: 1.197). Secreted PSAP was readily detectable in breast cancer patient serum and associates significantly with expression of HOXC11 in matched patient tissue (∼20). PSAP is associated with poor response to endocrine therapy and metastatic spread of prostate cancer and as it is secreted it could potentially be used to monitor patients on AI who might benefit from dual targeted therapy treatment. Citation Information: Cancer Res 2013;73(24 Suppl): Abstract nr PD3-5.

Mechanisms of primary AI resistance in ER-positive HER2-negative breast cancer.

e11519 Background: Current approaches of neoadjuvant endocrine therapy (NAET) are focusing on biomarker analysis of the diagnostic specimens however inadequate response occurs in 5-10% of patients treated. It has been reported hormone resistance may predict Ki67 level after treatment than ΔKi67. Much less is known about the difference between Ki67 increases and still high level of Ki67 after NAET. A pilot study was conducted to investigate the differences of biomarker changes in short-term NAET (letrozole vs. exemestane). Differences of resistant mechanisms between Ki67 increases and Ki67 still high expression and between two AIs were studied. Methods: Totally 116 samples from 58 postmenopausal women with ER positive HER2 negative stage 1-3 breast cancers comparing LET and EXE were analyzed with pre- and post-treatment mRNA expression of multiple biomarkers using QuantiGene Assay. Each of 29 patients took LET or EXE until the surgery (median 22 days). Compared with Ki67 expression of pre/post treatment, we classified resistance into 2 groups, Ki67 increase after treatment (INC) or Ki67 still high expression (equivalent over 14% in IHC) except those in INC (HIG). Results: Clinical characteristics were well balanced between the two treatment arms. Median % changes of Ki67 were -59% in both LET and EXE. Six cases (10%) were in INC (3/LET and 3/EXE). Among 10 cases of still high Ki67 expression, 6 (10%) were in HIG except 4 cases of Ki67 increases (3/LET and 3/EXE). All 6 cases in INC had increases of ERβ, MAPK, PIK3CA, ALDH1, NFkB, and RANKL and decrease of ERα. Five out of 6 in INC had an increase of HER2 and a decrease of PR. While changes in HIG were various, a nonsignificant trend to predict resistance was found. However all 3 cases whose HER2 increased in HIG had also increases of MAPK and PIK3CA. All 3 receiving EXE had increases of PR. Conclusions: Although both LET and EXE had significant decreases of Ki67, no significant difference was found between two. Both had 20% of primary resistance in terms of Ki67. Tumors which Ki67 increase seem to activate various signaling pathway, which mechanism is class effect of AI. What Ki67 is still high supposed to be nonclass effect.

Abstract P2-14-08: The Pilot Trial of Intermittent Exemestane Therapy for Metastatic Breast Cancer

Abstract Background: While aromatase inhibitors are effective for ER/PR positive breast cancer, resistance frequently develops. EXE is a steroidal AI and has a weak hormone-like activity. Long-term EXE treatment up-regulated amphiregulin-mediated EGFR activation and EXE resistance could result from ER-EGFR cross-talk. In cell culture and animal studies, acquired resistance to EXE was significantly delayed with intermittent use, associated with decreased production of amphiregulin. For first line therapy in MBC, EXE had a median progression free survival (PFS) of 10 mos. As second and third line therapies, PFS decreased to 4.7 mos and 2 mos, respectively. After ≥ 2 of prior endocrine therapy, PFS to EXE ranged between 3.4 mos to 3.7 mos. The rationale for this study was that after long-term estrogen deprivation clinically equivalent to long-term exposure to AI, estrogen may induce apoptosis. With intermittent EXE use, we hope to delay the acquired AI resistance. Methods: Aims were to: 1) assess the PFS and 2) compete response (CR), partial response (PR) and clinical benefit rate [CBR=CR +PR + Stable disease (SD)≥ 6 mos], assessment of toxicity, compliance with medication adherence, and quality of life with assessment of bone health. The study included: 1) ER/PR positive MBC in first, second line/third line hormonal therapy, and one prior chemotherapy for MBC with no prior EXE allowed. EXE was given 25 mg daily × 14 days then 7 days off treatment, each 21-day cycle, imaging was done after every three cycles. Correlatives with estradiol level were drawn at day 1 and day 15 in each cycle. Results: From June 2008 to 2012 we enrolled 33 MBC patients ER/PR positive MBC with a median age of 60 yrs (42–79). At this cut off data date, 2 were inevaluable due to incidental colon cancer and change in treatment choice. 17 patients received treatment as first line (52%), 10 as second or third line hormonal therapy (30%), and 6 who received chemotherapy for MBC (18%). Median PFS and (95% CI) was 6.2 mos (1.9–12.2), 4.2 mos (2.1–10.1), and 3.2 mos (2.1-NR), respectively. For first line patients, there were 3 PRs and 4 with SD>6 mos for a CBR of 7/14 (50%). For second/third line, there were 3 with SD>6 mos for a CBR of 3/9(33%), no response seen in the prior chemotherapy group. The major metastatic sites included bone, liver, lung. 4 patients were Her2+ (13%). Adverse effects including hot flashes (62%), insomnia (31%), nausea (24%), bone pain (17%), fatigue (41%), joint pain (31%) were observed. Median number of cycles delivered is 6 (2–24) for first line, 5.5 (1–15) for second/third line, and 5 (2–9) for prior chemotherapy. Correlatives and QOL assessments are ongoing. Conclusions: Compared to the standard continuous EXE, PFS and CBR is inferior in the intermittent EXE as first line therapy, as second and third line therapy, the median PFS was 4.2 comparable to reported 3.7 mos with the CBR was 33% to 31.5% observed in the EFFECT trial. Interestingly, hot-flash rate was much higher at 62% compared to 11.2% in the EFFECT trial. Arthralgia is comparable at 31%. Patients who already received chemotherapy do not appear to respond favorably to intermittent hormonal therapy. Full response data with correlatives regarding the estradiol levels will be presented. Citation Information: Cancer Res 2012;72(24 Suppl):Abstract nr P2-14-08.

342P - Fulvestrant (FUL) Plus Enzastaurin (ENZA) vs FUL Plus Placebo (PBO) in Aromatase Inhibitor (AI)-Resistant Metastatic Breast Cancer (MBC): A Randomized, Double-Blind, Phase 2 Trial

ABSTRACT Disclosure K. Mansouri: K. Mansouri is an employees of Eli Lilly and Co. L. Cirri: L. Cirri is an employees of Eli Lilly and Co. P. Shi: P. Shi is an employees of Eli Lilly and Co. O. Hamid: O. Hamid is an mployees of Eli Lilly and Co. All other authors have declared no conflicts of interest. AIs are used as first-line treatment for postmenopausal women with hormone-receptor-positive MBC. Overexpression of PKC α has been linked to AI resistance in several studies. We examined whether Enza, a serine/threonine kinase inhibitor that targets PKC, could improve the effect of Ful in pts who progressed following first-line AI treatment for MBC. Postmenopausal pts with hormone-receptor-positive, HER2-negative, locally advanced or MBC who progressed on prior AI received a loading dose of Ful 500 mg IM on Day (D) 1 and 250 mg on D 15 of Cycle (C) 1 and D 1 of each cycle thereafter. Enza 500 mg or PBO was administered orally once daily (QD) or 250 mg twice daily (BID). Primary endpoint was the clinical benefit rate (CBR). Secondary endpoints were response rate (RR), duration of CB, progression-free survival (PFS) and safety. A total of 156 pts was randomly assigned to therapy; 152 received at least 1 dose of study drug (39 BID; 55 QD; 58 PBO). Baseline disease characteristics were balanced across arms. There was no statistically significant difference in CBR between pts in Ful + Enza vs Ful + PBO. There was no statistically significant difference in CBRs, RRs and PFS between pts on QD and BID dosing schedules. Pts on BID dosing had numerically more TEAEs vs those on QD and PBO (61.5%, 43.6%, and 46.6%, respectively) and numerically more Grade 3/4 TEAEs vs QD and PBO (17.9%, 9.1%, and 5.2%, respectively). Most frequent Grade 3/4 TEAEs in the BID arm were fatigue (n [%]) (4 [10.3%]), dyspnea (2 [5.1%]) and nausea (2 [5.1%]). 8 pts died; 5 due to disease, 3 due to AEs: 1 drug-related hepatic dysfunction (Enza QD arm), 1 non-drug-related myocardial infarction and 1 non-drug-related respiratory failure (both in the Enza BID arm). Ful + Enza [QD + BID] N = 94 Ful + PBO N = 58 P-value vs PBO CBR Number of responders, n (%), (95% CI) 41 (43.6) (33.4, 54.2) 24 (44.8) (31.7, 58.5) 0.62 RR, % (95% CI) 5.3 (1.7, 12.0) 5.2 (1.1, 14.4) 0.64 PFS, months (95% CI) 5.2 (3.5, 7.4) 5.5 (3.8, 7.4) 0.59 Median duration of CB, months 9.6 9.7 0.86 CBR = complete response + partial response + stable disease Addition of Enza to Ful does not improve disease outcome in pts with locally advanced or MBC after progression on AI.

Characterising an aromatase inhibitor resistant breast cancer cell line

Results Let-R cells have higher levels of ERa compared to Aros. CyclinD1 basal levels are elevated in Aro cells compared to Let-Rs. Levels of JNK and c-jun are higher in Let-R cells compared to Aros. A highly effective ERa knockdown in Let-R cells was optimized. Conclusions The significant ERa knockdown achieved in the Let-R cells will enable further examination of knockdown effects on protein levels and mRNA expression. Elevated expression of ERa in the Let-R cells observed is consistent with the association between AI resistance and ER hypersensitivity. In resistant cells, the decrease in basal levels of CyclinD1 may be partly responsible for the lack of regulation of classical genes like pS2, previously detected by the lab. The elevated levels of JNK and c-jun observed in Let-R cells supports the theory of crosstalk between the ER and growth factor pathways in resistance. Furthermore, it would be of interest to accompany AI treatments with growth factor inhibitors to further our understanding of endocrine resistance.

Abstract 787: Involvement of HER2, HIF-1, and BCRP in cancer stem cell characteristics of letrozole-resistant breast cancer cells

Abstract Studies by us and others suggest that acquired resistance to aromatase inhibitors (AIs; i.e., letrozole) involves a switch from dependence on ER signaling to dependence on growth factor-mediated pathways, such as epidermal growth factor receptor (EGFR)/HER2. Recent work in our lab has also linked hypoxia inducible factor 1 (HIF-1) and breast cancer resistance protein (BCRP, a stem cell marker) to HER2 and AI resistance. Other studies have associated each of these factors with cancer stem cells (CSCs) and CSCs have been implicated in drug resistance. Therefore, this current study aimed to 1) compare the CSC characteristics of letrozole-resistant (LTLTCa) and parental letrozole-sensitive (MCF-7Ca) breast cancer cells; 2) determine the roles of HER2, BCRP, and HIF-1 on CSC characteristics; and 3) compare the effects of HER2 inhibition and of differentiating agent all-trans retinoic acid (ATRA) on CSC population of the LTLTCa cells. LTLTCa cells were found to have higher expression of CSC characteristics compared to parental MCF-7Ca cells: 9.17%±2.0% vs. 0.02% ±0.01% (p <0.01) measured by side population, and 3.08% vs. 0.53% (p<0.05) by aldehyde dehydrogenase (ALDH) activity. Consistent with previous results, inhibition of HER2, HIF-1, or BCRP resulted in decreased CSC characteristics in LTLTCa cells. Inhibition of HER2 by treatment with 1 μM lapatinib (Lap) decreased side population (from 7.91%±2.10 to 0.76%±0.20) and mammosphere formation (from >200 mammospheres/10,000 cells to ∼50 mammospheres/10,000 cells). Flow cytometry also demonstrated that LTLTCa cells high in both ALDH activity and CD44 positivity, were also high in HER2. Inhibition of HIF-1 (by siRNA) or BCRP (by pharmacological inhibitors or siRNA) significantly decreased mammosphere formation from 406 mammospheres/80,000 cells to 93 and 81 mammospheres/80,000 cells, respectively (p<0.01). Pharmacological inhibition of BCRP also decreased side population. Comparing the effects of Lap and ATRA on LTLTCa cells, both drugs decreased mammosphere formation after 48 h and cell viability after 7 days of treatment (p<0.05 for each). Both drugs also increased ERα and decreased BCRP expression in LTLTCa cells. Lastly, Lap and ATRA each increased the expression ratio of epithelial markers versus mesenchymal markers, but they did this by affecting different proteins. Lap increased expression of epithelial marker E-cadherin, but had no effect on mesenchymal marker N-cadherin. ATRA decreased N-cadherin but had no effect on E-cadherin. Overall, the results from this study suggest that HER2-HIF-1-BCRP signaling in contributes to acquired breast cancer cell resistance to letrozole by 1) increasing the cancer stem cell population, and possibly by 2) promoting epithelial mesenchymal transition (EMT), an indicator of cancer cell invasive and metastatic potential. Citation Format: {Authors}. {Abstract title} [abstract]. In: Proceedings of the 103rd Annual Meeting of the American Association for Cancer Research; 2012 Mar 31-Apr 4; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2012;72(8 Suppl):Abstract nr 787. doi:1538-7445.AM2012-787

In the pipeline: entinostat may overcome AI resistance

In the pipeline: entinostat may overcome AI resistance

Abstract 2303: The potential role of nonhypoxic HIF-1 in HER2-mediated aromatase inhibitor resistant breast cancer

Abstract Although aromatase inhibitors (AIs; i.e., letrozole) have been shown to be highly effective in treating estrogen receptor positive (ER+) breast cancer, a significant percentage of patients either do not respond to AIs or become resistant to them. Studies suggest that resistance to AIs involves a switch from dependence on ER signaling to dependence on growth factor-mediated pathways, such as epidermal growth factor receptor (EGFR)/HER2. The mechanism by which HER2 is involved in AI resistance remains unclear. It is, therefore, important to elucidate the HER2-mediated pathway that contributes to AI resistance, and to identify other relevant factors involved that can be used as biomarkers of AI resistance and targets for therapy. One such factor may be HIF-1, a heterodimeric transcription factor made up of an inducible alpha (α) subunit and a constitutively expressed beta (β) subunit. HIF-1 regulates genes important for cell survival, metabolic adaptation, and angiogenesis. Oxygen (O2) tension is a well-known regulator of HIF-1α, but other factors independent of O2 can also regulate it. Previous studies have shown that transfecting HER2 into cells and activation of kinase pathways increase HIF-1α expression and/or activity. Unlike the well-studied role of hypoxia-regulated HIF-1α in a variety of cancers, nonhypoxic regulation of HIF-1α and its role in cancer remains largely unclear. Here we have investigated HIF1α in AI resistance. Preliminary results indicate that basal HIF-1α protein expression is 3-fold (p<0.01) higher in letrozole-resistant, long term letrozole- treated MCF-7Ca (LTLTCa) cells than in their parental MCF-7Ca cells under nonhypoxic conditions. The higher expression in LTLTCa cells is likely due to increased protein synthesis for several reasons. LTLTCa cells do not have higher HIF-1α mRNA levels than MCF-7Ca cells, nor is the mRNA more stable. Also, compared to basal HIF-1α protein that rapidly degrades within 1 h of treatment with protein synthesis inhibitor cycloheximide, HIF-1α protein expression after treatment with HIF-1α stabilizer CoCl2 is increased 2.8-fold (p<0.001) and is resistant to degradation. Also, HIF-1α protein expression is significantly decreased (0.05-fold vs. vehicle, p<0.01) in LTLTCa cells treated with either HER2 inhibitor lapatinib or PI3K/Akt pathway inhibitor LY294002, which indicates a role for HER2 activated-PI3K/Akt pathway. Lastly, breast cancer resistance protein (BCRP) mRNA and protein expression is 3-fold higher in LTLTCa cells vs. MCF-7Ca cells (p <0.01), and its expression levels are increased and decreased by CoCl2 and lapatinib, respectively. Lapatinib, additionally, decreases basal HIF-1α binding to the BCRP promoter. These results suggest that a mechanism of resistance to AIs may be through regulation of nonhypoxic HIF-1 target genes, such as BCRP, an efflux transporter protein and stem cell marker implicated in chemoresistance. Citation Format: {Authors}. {Abstract title} [abstract]. In: Proceedings of the 102nd Annual Meeting of the American Association for Cancer Research; 2011 Apr 2-6; Orlando, FL. Philadelphia (PA): AACR; Cancer Res 2011;71(8 Suppl):Abstract nr 2303. doi:10.1158/1538-7445.AM2011-2303

C116 マイクロチャネル内埋め込み薄膜センサによる流体温度計測(第2報)(温度計測)

Micro-size membrane-type resistance thermometer and thermocouples made of Pt and AI were fabricated by using MEMS process in the present study. The performances were evaluated by measuring the temperature of the fluid in a micro-channel for the purpose of developing a high resolution and low cost temperature sensors for μ-TAS, and to provide fundamental data in developing a database of micro fabrication processes. Several types of sensor pattern were tested in which the spatial resolution and membrane thickness were changed. AI resistance thermometers, which are available to be produced at a lower cost, showed a reasonable performance compared with Pt sensors. The effects of the membrane thickness on the sensor characteristics were examined, and the thickness has its limit value to degrade the sensor reliability.