Eric P. Olsen

Androgen receptor as a mediator and biomarker of radioresistance in triple-negative breast cancer

Increased rates of locoregional recurrence have been observed in triple-negative breast cancer despite chemotherapy and radiation therapy. Thus, approaches that combine therapies for radiosensitization in triple-negative breast cancer are critically needed. We characterized the radiation therapy response of 21 breast cancer cell lines and paired this radiation response data with high-throughput drug screen data to identify androgen receptor as a top target for radiosensitization. Our radiosensitizer screen nominated bicalutamide as the drug most effective in treating radiation therapy-resistant breast cancer cell lines. We subsequently evaluated the expression of androgen receptor in >2100 human breast tumor samples and 51 breast cancer cell lines and found significant heterogeneity in androgen receptor expression with enrichment at the protein and RNA level in triple-negative breast cancer. There was a strong correlation between androgen receptor RNA and protein expression across all breast cancer subtypes (R2u2009=u20090.72, pu2009<u20090.01). In patients with triple-negative breast cancer, expression of androgen receptor above the median was associated with increased risk of locoregional recurrence after radiation therapy (hazard ratio for locoregional recurrence 2.9–3.2)) in two independent data sets, but there was no difference in locoregional recurrence in triple-negative breast cancer patients not treated with radiation therapy when stratified by androgen receptor expression. In multivariable analysis, androgen receptor expression was most significantly associated with worse local recurrence-free survival after radiation therapy (hazard ratio of 3.58) suggesting that androgen receptor expression may be a biomarker of radiation response in triple-negative breast cancer. Inhibition of androgen receptor with MDV3100 (enzalutamide) induced radiation sensitivity (enhancement ratios of 1.22–1.60) in androgen receptor-positive triple-negative breast cancer lines, but did not affect androgen receptor-negative triple-negative breast cancer or estrogen-receptor-positive, androgen receptor-negative breast cancer cell lines. androgen receptor inhibition with MDV3100 significantly radiosensitized triple-negative breast cancer xenografts in mouse models and markedly delayed tumor doubling/tripling time and tumor weight. Radiosensitization was at least partially dependent on impaired dsDNA break repair mediated by DNA protein kinase catalytic subunit. Our results implicate androgen receptor as a mediator of radioresistance in breast cancer and identify androgen receptor inhibition as a potentially effective strategy for the treatment of androgen receptor-positive radioresistant tumors.Radiation: Drug target found for radioresistanceDrugs that block the androgen receptor (AR) protein can help make radiation-resistant breast tumors susceptible to ionizing therapy. Corey Speers from the University of Michigan, Ann Arbor, USA, and colleagues characterized the radiation sensitivity of 21 breast cancer cell lines and then paired the response data with the findings of a high-throughput drug screen to identify a medication called bicalutamide — an AR inhibitor — as one of the most potent agents for overcoming radiation resistance. The researchers then measured the expression of AR in more than 2100 human breast tumor samples and 51 breast cancer cell lines, and found that patients with triple-negative breast cancer, especially those who relapsed after radiation, had elevated levels of AR. Cell experiments with enzalutamide, a newer generation AR-blocking drug, confirmed that targeting AR helps reverse radiation resistance.

Abstract 3396: Optimization of an assay for the detection of PD-L1 by immunohistochemistry in formalin-fixed, paraffin-embedded human tissue and cell lines

Programmed Death-1 (PD-1) is a member of the CD28 family of receptors and plays a role in cellular immune response. It is expressed on activated T-cells and B-cells and binds to two ligands, Programmed Death Ligand 1 (PD-L1) and Programmed Death Ligand 2 (PD-L2). PD-1 negatively regulates T-cell functions through interactions with its primary ligand PD-L1, which is variably expressed by a variety of tumors, including melanoma, glioblastoma, and carcinomas of the lung, breast, colon, ovary, and renal cells. Many tumor cells express PD-L1 and as a result, the interaction between PD-1 and PD-L1 has been demonstrated to be an important target for therapeutic intervention. Anti-PD-1 therapeutics such as pembrolizumab and nivolumab have demonstrated clinical activity in melanoma, non-small cell lung cancer and other types of cancer. The purpose of this study was to investigate 7 commercially available anti-PD-L1 antibodies in human tissue and cell lines. Antibodies included 1 mouse monoclonal (clone 29E.2a3), 2 rabbit monoclonals (EPR1161(2) and E1L3N), 3 rabbit polyclonals, and 1 goat polyclonal from 6 manufacturers. An initial screen in human tonsil was performed to ensure appropriate sensitivity and specificity in staining of the crypt epithelium. Only 2 antibodies yielded staining that matched the expected specificity and were selected for further evaluation: rabbit clone EPR1161(2) from Abcam and rabbit clone E1L3N from Cell Signaling. Antibody specificity was evaluated in a positive cell line (MDA-MB-231) and negative cell line (MCF-7). Both antibodies demonstrated appropriate specificity in these cell lines. Antibody sensitivity was evaluated in a variety of human tumor tissues provided by Mosaic Laboratories’ tissue bank. PD-L1 localization was expected to be membranous and cytoplasmic in tonsil crypt epithelium, inflammatory cells and tumor cells. Immunohistochemistry with both antibodies demonstrated staining in tumor cells, lymphocytes and macrophages. Immunohistochemistry with clone E1L3N often demonstrated stronger staining in a greater percentage of cancer cells than EPR1161(2). Of interest, clone E1L3N demonstrated specific staining of some peripheral nerve fibers that was not observed with clone EPR1161(2). In conclusion, 2 commercially available antibodies, rabbit clone EPR1161(2) from Abcam and rabbit clone E1L3N from Cell Signaling are appropriate for immunohistochemistry, of which rabbit clone E1L3N demonstrates greater sensitivity in tumor tissue. Citation Format: Lisa M. Dauffenbach, Gela C. Sia, Patricia A. Cash, Sherif K. Girees, Jianping Zheng, Ryan S. Lim, Eric P. Olsen, Christopher A. Kerfoot. Optimization of an assay for the detection of PD-L1 by immunohistochemistry in formalin-fixed, paraffin-embedded human tissue and cell lines. [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 3396. doi:10.1158/1538-7445.AM2015-3396

Abstract 3218: Maternal embryonic leucine zipper kinase (MELK) confers radioresistance in triple-negative breast cancers (TNBC) through a nonhomologous end joining (NHEJ)-mediated pathway

Background: Increased rates of locoregional recurrence have been observed in TNBC despite the use of radiation (RT); therefore, approaches that result in radiosensitizaton in TNBC may improve clinical outcomes. Despite some controversy as to the role of MELK kinase as an oncogene, we previously established its clear role in conferring radioresistance in TNBC. Here we extend those studies to understand the mechanism of conferred radioresistance. Methods: Clonogenic survival assays were used to quantify the degree of radiosensitivity after MELK inhibition. Mass spectrometry was utilized to identify proteins associated with MELK. NHEJ assays were performed using linearized pEYFP vector and quantitated by FACS. qRT-PCR was used to determine MELK RNA expression. Rad51 and γH2AX foci counting were performed to evaluate homologous recombination (HR). Neutral comet assay was used to quantitate dsDNA break repair. Laser microirradiation studies were performed to determine the location and kinetics of protein shuttle to and from sites of dsDNA breaks, including MELK, Ku70 and Ku80 proteins. Fluorescent microscopy was performed for localization of endogenous MELK protein post RT. To study the effect of MELK kinase activity on Ku70-Ku80 complex formation and recruitment at the DNA damage site, cells were lysed and immunoprecipitation was performed using MELK, Ku70 or Ku80-specific antibodies. Results: We previously demonstrated that genomic or pharmacologic inhibition of MELK confers significant radiosensitization. Mechanistic studies were undertaken to understand the pathways critical to MELK-mediated radioresistance. When MELK is genomically or pharmacologically inhibited, NHEJ and HR reporter assays demonstrate that this radiosensitization is driven through impaired NHEJ at 4, 16, and 24 hrs after RT and not mediated by HR in multiple TNBC cell lines. Tandem mass spectrometry studies of tagged MELK protein identify Ku70 and Ku80 as direct interactors with MELK protein, which is confirmed by Co-IP. Laser microirradiation studies confirm that MELK, Ku70 and Ku80 co-localize to sites of dsDNA breaks, and that MELK kinase function is required to stabilize the Ku70/Ku80 complex at these sites. Comet assay confirms that rapid dissolution of the Ku70/Ku80 complex when MELK is inhibited or functionally dead is not a result of accelerated dsDNA break repair. Conclusion: MELK kinase function is critical in conferring radioresistance in TNBC and inhibition of function confers radiosensitivity through an NHEJ-mediated pathway. MELK also stabilizes Ku70/Ku80 proteins at sites of dsDNA breaks and allows for more efficient repair of breaks induced by ionizing RT. These results further support the rationale for developing clinical strategies to inhibit MELK in combination with RT treatment as a novel radiosensitizing strategy in TNBC. Citation Format: Shyam Nyati, Ben Chandler, Eric Olsen, Leah Moubadder, Meilan Liu, Meleah Cameron, Kari Wilder-Romans, Theodore S. Lawrence, Powel H. Brown, Fellix Y. Fang, Lori J. Pierce, Corey Speers. Maternal embryonic leucine zipper kinase (MELK) confers radioresistance in triple-negative breast cancers (TNBC) through a nonhomologous end joining (NHEJ)-mediated pathway [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 3218.

Abstract P1-10-02: A signature predictive of early vs. late recurrence after radiation treatment (RT) for breast cancer that may inform the biology of early, aggressive recurrences

Purpose: Unmet clinical needs in breast cancer (BC) management include the identification of patients (pts) at high risk to fail locally despite standard local therapy including RT and understanding the biology of these recurrences. We previously reported 1 a RT response signature and here extend those studies to identify a signature predictive of timing of recurrence after completion of RT (before or after 3 years). Methods : Two independent patient cohorts (treated with BCS) from non-randomized clinical trials were used for training and validation. The training cohort included 119 pts with in-breast tumor recurrence and the validation cohort had 25 pts with recurrences. Initial feature selection used Spearman9s rank correlation correlating gene expression (14,806 genes) to recurrence time. Genes with sig. correlation (FDR 2) were used to train an elastic net penalized Poisson regression model. This model was locked and then applied to the validation dataset. Cox regression was used for both univariate and multivariable analyses (UVA and MVA). To identify biological-related concepts, Spearman9s corr. coefficients of recurrence time to gene expression within the training cohort were used to generate a pre-ranked list upon which GSEA pathway analysis was performed. Results : Spearman9s correlation identified 485 genes whose expression was significantly associated with recurrence time (early vs. late). Feature reduction further refined the gene list to 41 genes, which were retained within the signature and locked for further validation. In the training dataset the Spearman9s correlation of the continuous score to recurrence time was 0.852 with a P-value of 1.3x10 -34 and an AUC of 0.92. Application of this early vs late signature to an independent BC validation set accurately identifies pts with early vs. late recurrences (Spearman9s corr.=0.537, p-value Conclusion: In this study we derive a BC-specific RT signature predictive of early vs. late recurrence with biologic relevance and validate this signature for prediction of timing of recurrence in an independent clinical dataset. By identifying pts with tumors likely to recur sooner vs. later this signature has the potential to allow for a furthered understanding of the biology underlying early and late recurrences and has a potential to personalize RT, particularly in patients for whom treatment intensification is needed. 1. Clin Cancer Res. 2015 Aug 15;21(16):3667-77. Citation Format: Speers C, Chang L, Santola A, Liu M, Zhao SG, Chandler B, Olsen E, Bartelink H, Feng FY, Pierce LJ. A signature predictive of early vs. late recurrence after radiation treatment (RT) for breast cancer that may inform the biology of early, aggressive recurrences [abstract]. In: Proceedings of the 2016 San Antonio Breast Cancer Symposium; 2016 Dec 6-10; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2017;77(4 Suppl):Abstract nr P1-10-02.

Abstract 5839: Androgen receptor (AR): A novel target for radiosensitization in triple-negative breast cancers (TNBC)

Purpose: Increased rates of local recurrence (LR) have been observed in TNBC despite chemotherapy and radiation (RT). Thus, approaches that result in radiosensitization in TNBC are critically needed. We characterized the RT response of 21 breast cancer cell (BCC) lines using clonogenic survival assays and paired this with high-throughput drug screen data, identifying AR as a top target for radiosensitization. We demonstrate that AR inhibition confers radiosensitization in vitro and in vivo, identified a biomarker of response, and characterize the mechanism of AR-mediated radiosensitization in TNBC. Materials/Methods: Clonogenic survival assays determined the intrinsic RT sensitivity of 21 BCC lines. IC50 values were determined for 130 clinical compounds and correlation coefficients were calculated using IC50 values and SF-2Gy. Gene and protein expression was measured using RNA Seq and RPPA arrays, respectively, in tumor samples (n=2,061) and BCC lines (n=51). AR function was assessed using gene knockdown (KD) or functional inhibition with anti-androgen drugs. We measured in vivo tumor growth with varying control and treatment groups (16-20 tumors/group). Kaplan-Meier analysis was performed to estimate local control. A Cox proportional hazards model and MVA were used to determine variables associated with LRF survival. Results: Our unbiased drug radiosensitizer screen nominated bicalutamide as an effective drug in treating RT-resistant BCC lines (R2= 0.46, p-value 2000 human breast tumor samples and 51 BCC lines and found heterogeneity in AR expression with strongly correlated expression of protein and RNA levels in TNBC (R2=0.89, p-value Conclusion: Our results implicate AR as a mediator of radioresistance in breast cancer and support the rationale for developing clinical strategies, including clinical trials, to inhibit AR as a novel radiosensitizing target in TNBC. Citation Format: Benjamin C. Chandler, Corey W. Speers, Shuang G. Zhao, Meilan Liu, Kari Wilder-Romans, Eric Olsen, Shyam Nyati, Daniel Spratt, Daniel Wahl, Daniel Hayes, Felix Y. Feng, Lori J. Pierce. Androgen receptor (AR): A novel target for radiosensitization in triple-negative breast cancers (TNBC) [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2017; 2017 Apr 1-5; Washington, DC. Philadelphia (PA): AACR; Cancer Res 2017;77(13 Suppl):Abstract nr 5839. doi:10.1158/1538-7445.AM2017-5839

Abstract 1669: Multiplex immunohistochemical staining of PD-L1, PD-1, CD3, CD4, CD8, CD68, FoxP3, and Ki-67 and image analysis of tumor and invasive margin in human FFPE NSCLC tissue

Evaluating the density of immune cell subtypes and their relative spatial positioning in cancers has become an important tool in understanding response to immune checkpoint inhibitors. The purpose of this study was to investigate protein expression of PD-L1, PD-1, CD3, CD4, CD8, CD68, FoxP3, and Ki-67 with four multiplex immunohistochemical assays (Ki-67+CD8, CD3+CD4+Ki-67, FoxP3+PD-1+CD8, and CD68+PD-L1+CD3) using DAB, red, and green chromogens within formalin-fixed, paraffin embedded non-small cell lung (NSCLC) tissues. The Ki-67+CD8 multiplex was developed to evaluate the density of cytotoxic T cells and the percentage that are proliferating. The CD3+CD4+Ki-67 multiplex characterizes the frequency of CD4+ T cells and evaluate proliferation in this subset. The FoxP3+PD-1+CD8 multiplex evaluates the regulatory T cell subset and evaluates PD-1 expression. The CD68+PD-L1+CD3 multiplex characterizes PD-L1 expression on macrophages and T cells. Cell densities were evaluated in the center of tumor and invasive margin regions, and image analysis was performed to quantitate each stain as a single agent and co-expression within each multiplex assay. Citation Format: Lisa M. Dauffenbach, Gela C. Sia, Jianping Zheng, Natalia Jun, Eric P. Olsen, Christopher A. Kerfoot. Multiplex immunohistochemical staining of PD-L1, PD-1, CD3, CD4, CD8, CD68, FoxP3, and Ki-67 and image analysis of tumor and invasive margin in human FFPE NSCLC tissue [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2017; 2017 Apr 1-5; Washington, DC. Philadelphia (PA): AACR; Cancer Res 2017;77(13 Suppl):Abstract nr 1669. doi:10.1158/1538-7445.AM2017-1669

Abstract 1790: Evaluation of UCK2 protein expression as a potential predictive biomarker of RX-3117

Background: A novel, orally bioavailable nucleoside analogue, RX-3117, is a prodrug activated intracellularly by Uridine Cytidine Kinase 2 (UCK2) that is thought to be expressed predominantly in tumor tissue. RX-3117 is currently being evaluated in a Phase Ib/IIa multi-center, open-label clinical study in patients with advanced pancreatic and bladder cancer. In this study, we aimed to determine the relation between UCK2 tissue protein expression and the efficacy of RX-3117 in mice xenograft models and also UCK2 protein expression in a panel of human cancer tissues relative to normal tissue. Methods: The UCK2 protein expression in tumor tissues was analyzed by immunoblotting using clone 22-1 rabbit monoclonal antibody. The validated procedure for the immunohistochemistry (IHC) of UCK2 with clone 22-1 was performed in a panel of human formalin-fixed paraffin-embedded (FFPE) cancer and normal tissues. Results: The immunoblotting protein level of UCK2 normalized to β-actin and corresponding tumor growth inhibition (oral RX-3117 dose of 500mg/kg, TIWK) were 57 and 67% in MiaPaCa2, 30 and -5% in BxPC3, 199 and 92% in Colo-205, 21 and 90% in Caki-1, 2 and 39% in A549, and 146 and 79% in H460, respectively. This data indicates an anti-tumor efficacy trend in a UCK2-dependent manner. The IHC of UCK2 showed that positive staining of UCK2 in cancer tissues was observed in 20/20 bladder cancer tissues (100% frequency), 19/20 CRC tissues (95% frequency), 18/20 NSCLC tissues (90% frequency), and 19/20 pancreatic cancer tissues (95% frequency). Average H-Scores of UCK2 in cancer tissues vs. normal tissues were 104 vs. 9 in lung, 97 vs. 20 in bladder, 67 vs. 41 in pancreas and 39 vs. 21 in colon, respectively. Conclusions: The current data showed a correlation trend between UCK2 protein expression level and degree of antitumor activity of RX-3117 in xenograft models. It also supports a higher UCK2 protein expression level in human cancer tissues compared to their normal tissues. This suggests that RX-3117 activity may be specific to tumor tissue, and quantification of UCK2 expression in human cancer tissues may be useful as a predictive biomarker to select patients for their sensitivity to RX-3117 in future clinical studies. Citation Format: Young Bok B. Lee, Deog J. Kim, Christina George, Chang-ho Ahn, Julie Frank, Reza Mazhari, Lisa M. Dauffenbach, Eric P. Olsen. Evaluation of UCK2 protein expression as a potential predictive biomarker of RX-3117 [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2017; 2017 Apr 1-5; Washington, DC. Philadelphia (PA): AACR; Cancer Res 2017;77(13 Suppl):Abstract nr 1790. doi:10.1158/1538-7445.AM2017-1790

Abstract 1373: IHC and RT-PCR assays for detection of cancer antigen NY-ESO-1 in human tissues

Proceedings: AACR 107th Annual Meeting 2016; April 16-20, 2016; New Orleans, LAnnIntroduction: NY-ESO-1 is a cancer testis antigen frequently expressed in a broad range of tumors: lung, breast, ovarian, bladder, liver cancer, melanoma, sarcoma and myeloma. CDX-1401, a fully human monoclonal antibody linked to full length NY-ESO-1, binds to dendritic cell (DC) receptor DEC-205 to stimulate NY-ESO-1 specific CD4 and CD8 responses. Immunohistochemistry (IHC) and quantitative RT-PCR (qRT-PCR) assays were developed to determine NY-ESO-1 expression in human tumors and normal adjacent tissues (NAT). A qRT-PCR assay for the detection of LAGE-1, a cancer testis antigen with significant DNA homology to NY-ESO-1, was also developed but no commercially specific antibody was available for IHC.nnMethods: Formalin fixed, paraffin embedded tumors and NAT were procured from a commercial source or obtained from Mosaic Laboratories; split anonymized samples were analyzed for NY-ESO-1 by IHC at Mosaic and for NY-ESO-1 and LAGE-1 by qRT-PCR at MolecularMD (OR) under IRB-approved protocols allowing in vitro analysis of remnant human samples. The study included: 18 each melanoma and ovarian cancers; 13 each colorectal (CRC), head & neck (HN and 38 NAT. HT1080 fibrosarcoma and SKOV3 ovarian cancer cell lines (ATCC, VA) and normal testis were used as controls. The qRT-PCR assays were validated using in vitro transcribed RNA on NY-ESO-1 and LAGE-1 kits provided by Life Technologies and the NY-ESO-1 IHC was developed using a mouse monoclonal antibody (Sigma, MO).nnResults: The NY-ESO-1 and LAGE-1 qRT-PCR assays had a linear range of 100-1,000,000 copies/reaction with limit of detection between 1-10 copies. An excess of LAGE-1 copies did not impact the specificity of the NY-ESO-1 assay and vice-versa. NY-ESO-1 protein was detected by IHC and NY-ESO-1 transcripts and LAGE-1 transcripts were detected in 9 (12%), 12 (16%) and 16 (21.3%), respectively, of the 75 tumor tissues while only 2 of the 38 normal samples were positive for LAGE-1. When stromal staining was considered in addition to epithelial staining, 16 (21.3%) tumor samples were positive by IHC. NY-ESO-1 protein was more prevalent in NSCLC (38%), ovarian cancer (11%) and melanoma (11%) vs. other tumor types. Positive staining by IHC was achieved in tumor tissues shown positive for NY-ESO-1 and negative for LAGE-1 transcripts. In general, tumor tissues obtained from Mosaic were of better quality, as assessed by the pathologist, and more positive than commercially procured samples.nnConclusions: Two qRT-PCR assays were developed to specifically quantify NY-ESO-1 or LAGE-1 mRNA without cross reaction. The antibody used in IHC assay seemed specific to NY-ESO-1. Inadequate sample quality impacted target prevalence. Implementing these assays in clinical trials may aid in prospectively identifying patients with NY-ESO-1 positive malignancies who could derive benefit from investigational NY-ESO-1 vaccines designed to induce anti-tumor immunity, such as CDX-1401.nnCitation Format: Abdel Halim, Rebecca G. Bagley, Lisa Dauffenbach, Eric P. Olsen, Tibor Keler. IHC and RT-PCR assays for detection of cancer antigen NY-ESO-1 in human tissues. [abstract]. In: Proceedings of the 107th Annual Meeting of the American Association for Cancer Research; 2016 Apr 16-20; New Orleans, LA. Philadelphia (PA): AACR; Cancer Res 2016;76(14 Suppl):Abstract nr 1373.

Abstract 899: Combination of PI3K and MEK inhibitor chemosensitivity in human tumor explants and cell lines using the Mosaic Blue assay and relationship to biomarkers by immunohistochemistry

The PI3K and MEK pathways promote tumor cell survival through a variety of downstream signals and are commonly explored in cancer research. Many agents are in clinical development and new evidence suggests that a combination of these inhibitors may be required to achieve optimal anti-tumor activity. Twenty-one viable colorectal and non-small cell lung cancer tissue explants with matched formalin-fixed, paraffin embedded (FFPE) tissue and 8 cell lines (A431, AGS, HCT-116, HT-29, MCF-7, MDA-MB-468, MKN45, and T47D) were used for this study. Fresh tissues were disaggregated into single cells and exposed to GDC-0941, a PI3K inhibitor, and AZD6244, a MEK inhibitor, both as single agents and in combination in the Mosaic Blue assay. Combination results were compared to single agent results. Samples were classified as resistant or sensitive. Matched FFPE tissues and cells were stained with the following biomarkers to identify predictive relationships: pAKT, p4EBP1, pPRAS40, ppS6 kinase, pS6 ribosomal protein, pERK, PTEN, cMET, EGFR, HER3, Ki-67, and p53. To assess pharmacodynamics changes, the 8 cell lines were also exposed to 1 concentration of each inhibitor both as single agents and in combination. The cells were fixed and paraffin embedded for IHC staining of the following biomarkers to identify pharmacodynamic relationships: pAKT, p4EBP1, pPRAS40, ppS6 kinase, pS6 ribosomal protein, pEGFR, pHER3, pMET, pERK, and Ki-67. The combination of inhibitors demonstrated additive growth inhibition in the explants and cell lines. The average combination IC50 was more potent than single agents, with greater additivity demonstrated in colorectal cancer compared to lung cancer. Predictive and pharmacodynamic biomarkers were identified with both single agent and combination treatments, and differences were observed between the two inhibitors. In summary, this study demonstrated that the Mosaic Blue assay is a useful tool for classifying sensitive and resistant tumor explants and cell lines and also confirmed that dual targeting of PI3K and MEK pathways is a valuable approach to achieving higher therapeutic activity in vitro. Evaluation of biomarker profiles revealed predictive and pharmacodynamic relationships in this setting. Citation Format: Lisa M. Dauffenbach, Gela C. Sia, Patricia A. Cash, Sherif K. Girees, Ryan S. Lim, Jianping Zheng, Eric P. Olsen, Christopher A. Kerfoot. Combination of PI3K and MEK inhibitor chemosensitivity in human tumor explants and cell lines using the Mosaic Blue assay and relationship to biomarkers by immunohistochemistry. [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 899. doi:10.1158/1538-7445.AM2014-899

Abstract 3745: Optimization of an assay for the detection of phosphorylated FAK by immunohistochemistry in formalin-fixed, paraffin-embedded human tissue and cell lines

Proceedings: AACR Annual Meeting 2014; April 5-9, 2014; San Diego, CAnnFocal adhesion kinase (FAK) is a non-receptor protein tyrosine kinase with elevated expression in most human cancers, most notably in invasive metastasis. FAK is involved in cell adhesion, motility, and apoptosis, and it is a target for oncology therapeutics. FAK activation occurs by phosphorylation at Tyr397 in response to integrin clustering caused by cell adhesion. The development of a sensitive and specific immunohistochemical assay for phosphorylated FAK (pFAK) is important for pharmacodynamic biomarker studies. The purpose of this study was to investigate 4 commercially available anti-pFAK antibodies in human tissues and cell lines. Antibodies included 3 rabbit monoclonals (EP2160Y, 141-9, and 31H5L17) and 1 rabbit polyclonal. Antibody specificity was evaluated in a variety of cell lines (BxPC3, HCT-116, LOVO, MDA-MB-231, and MiaPaca2) treated with 2 FAK inhibitors: TAE226 (NVP-TAE226) and PF-562271. Decreased immunohistochemical staining for pFAK was observed in post-treatment samples of cell lines with baseline pFAK expression. TAE226 was more potent at inhibiting pFAK staining. Membrane staining was observed with the rabbit polyclonal and rabbit clone EP2160Y only. Each of the 4 antibodies was tested in human breast cancer tissues using tissue microarrays created by Mosaic Laboratories. The percentage of breast cancer samples that were pFAK positive was 96% with clone EP2160Y, 83% with clone 141-9, 92% with clone 31H5L17 and 96% with the polyclonal. The percentage of cells that were positive within each breast cancer sample ranged from 63% to 83%, depending on the antibody used (standard deviations between 30% and 44%). Immunohistochemistry with the polyclonal and clone EP2160Y antibodies demonstrated membrane and cytoplasmic staining, while clones 141-9 and 31H5L17 demonstrated primarily cytoplasmic staining. Based on reduction of pFAK staining in treated cells and acceptable performance in human cancer tissues, pFAK immunohistochemistry is “fit for purpose” for evaluation of FAK inhibitor activity in pre- and post-treatment clinical trial tissue biopsies.nnCitation Format: Lisa M. Dauffenbach, Gela C. Sia, Patricia A. Cash, Sherif K. Girees, Ryan S. Lim, Jianping Zheng, Eric P. Olsen, Rana Richeh, Christopher A. Kerfoot. Optimization of an assay for the detection of phosphorylated FAK by immunohistochemistry in formalin-fixed, paraffin-embedded human tissue and cell lines. [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 3745. doi:10.1158/1538-7445.AM2014-3745