Dawna Dylewski

Genomic characterization of explant tumorgraft models derived from fresh patient tumor tissue

BackgroundThere is resurgence within drug and biomarker development communities for the use of primary tumorgraft models as improved predictors of patient tumor response to novel therapeutic strategies. Despite perceived advantages over cell line derived xenograft models, there is limited data comparing the genotype and phenotype of tumorgrafts to the donor patient tumor, limiting the determination of molecular relevance of the tumorgraft model. This report directly compares the genomic characteristics of patient tumors and the derived tumorgraft models, including gene expression, and oncogenic mutation status.MethodsFresh tumor tissues from 182 cancer patients were implanted subcutaneously into immune-compromised mice for the development of primary patient tumorgraft models. Histological assessment was performed on both patient tumors and the resulting tumorgraft models. Somatic mutations in key oncogenes and gene expression levels of resulting tumorgrafts were compared to the matched patient tumors using the OncoCarta (Sequenom, San Diego, CA) and human gene microarray (Affymetrix, Santa Clara, CA) platforms respectively. The genomic stability of the established tumorgrafts was assessed across serial in vivo generations in a representative subset of models. The genomes of patient tumors that formed tumorgrafts were compared to those that did not to identify the possible molecular basis to successful engraftment or rejection.ResultsFresh tumor tissues from 182 cancer patients were implanted into immune-compromised mice with forty-nine tumorgraft models that have been successfully established, exhibiting strong histological and genomic fidelity to the originating patient tumors. Comparison of the transcriptomes and oncogenic mutations between the tumorgrafts and the matched patient tumors were found to be stable across four tumorgraft generations. Not only did the various tumors retain the differentiation pattern, but supporting stromal elements were preserved. Those genes down-regulated specifically in tumorgrafts were enriched in biological pathways involved in host immune response, consistent with the immune deficiency status of the host. Patient tumors that successfully formed tumorgrafts were enriched for cell signaling, cell cycle, and cytoskeleton pathways and exhibited evidence of reduced immunogenicity.ConclusionsThe preservation of the patient’s tumor genomic profile and tumor microenvironment supports the view that primary patient tumorgrafts provide a relevant model to support the translation of new therapeutic strategies and personalized medicine approaches in oncology.

In vivo Safety and Antitumor Efficacy of Bifunctional Small Hairpin RNAs Specific for the Human Stathmin 1 Oncoprotein

Bifunctional small hairpin RNAs (bi-shRNAs) are functional miRNA/siRNA composites that are optimized for posttranscriptional gene silencing through concurrent mRNA cleavage-dependent and -independent mechanisms (Rao et al., 2010 ). We have generated a novel bi-shRNA using the miR30 scaffold that is highly effective for knockdown of human stathmin (STMN1) mRNA. STMN1 overexpression well documented in human solid cancers correlates with their poor prognosis. Transfection with the bi-shSTMN1-encoding expression plasmid (pbi-shSTMN1) markedly reduced CCL-247 human colorectal cancer and SK-Mel-28 melanoma cell growth in vitro (Rao et al., 2010 ). We now examine in vivo the antitumor efficacy of this RNA interference-based approach with human tumor xenografted athymic mice. A single intratumoral (IT) injection of pbi-shSTMN1 (8 μg) reduced CCL-247 tumor xenograft growth by 44% at 7 days when delivered as a 1,2-dioleoyl-3-trimethyl-ammoniopropane:cholesterol liposomal complex. Extended growth reductions (57% at day 15; p < 0.05) were achieved with three daily treatments of the same construct. STMN1 protein reduction was confirmed by immunoblot analysis. IT treatments with pbi-shSTMN1 similarly inhibited the growth of tumorgrafts derived from low-passage primary melanoma (≥70% reduction for 2 weeks) and abrogated osteosarcoma tumorgraft growth, with the mature bi-shRNA effector molecule detectable for up to 16 days after last injection. Antitumor efficacy was evident for up to 25 days posttreatment in the melanoma tumorgraft model. The maximum tolerated dose by IT injection of >92 μg (Human equivalent dose [HED] of >0.3 mg/kg) in CCL-247 tumor xenograft-bearing athymic mice was ∼10-fold higher than the extrapolated IC(50) of 9 μg (HED of 0.03 mg/kg). Healthy, immunocompetent rats were used as biorelevant models for systemic safety assessments. The observed maximum tolerated dose of <100 μg for intravenously injected pbi-shSTMN1 (mouse equivalent of <26.5 μg; HED of <0.09 mg/kg) confirmed systemic safety of the therapeutic dose, hence supporting early-phase assessments of clinical safety and preliminary efficacy.

Using a rhabdomyosarcoma patient‐derived xenograft to examine precision medicine approaches and model acquired resistance

Precision (Personalized) medicine has the potential to revolutionize patient health care especially for many cancers where the fundamental disease etiology remains either elusive or has no available therapy. Here we outline a study in alveolar rhabdomyosarcoma, in which we use gene expression profiling and a series of drug prediction algorithms combined with a matched patient‐derived xenograft (PDX) model to test bioinformatically predicted therapies.

Anoikis-resistant subpopulations of human osteosarcoma display significant chemoresistance and are sensitive to targeted epigenetic therapies predicted by expression profiling

BackgroundOsteosarcoma (OS) is the most common type of solid bone cancer, with latent metastasis being a typical mode of disease progression and a major contributor to poor prognosis. For this to occur, cells must resist anoikis and be able to recapitulate tumorigenesis in a foreign microenvironment. Finding novel approaches to treat osteosarcoma and target those cell subpopulations that possess the ability to resist anoikis and contribute to metastatic disease is imperative. Here we investigate anchorage-independent (AI) cell growth as a model to better characterize anoikis resistance in human osteosarcoma while using an expression profiling approach to identify and test targetable signaling pathways.MethodsEstablished human OS cell lines and patient-derived human OS cell isolates were subjected to growth in either adherent or AI conditions using Ultra-Low Attachment plates in identical media conditions. Growth rate was assessed using cell doubling times and chemoresistance was assessed by determining cell viability in response to a serial dilution of either doxorubicin or cisplatin. Gene expression differences were examined using quantitative reverse-transcription PCR and microarray with principal component and pathway analysis. In-vivo OS xenografts were generated by either subcutaneous or intratibial injection of adherent or AI human OS cells into athymic nude mice. Statistical significance was determined using student’s t-tests with significance set at α = 0.05.ResultsWe show that AI growth results in a global gene expression profile change accompanied by significant chemoresistance (up to 75 fold, p < 0.05). AI cells demonstrate alteration of key mediators of mesenchymal differentiation (β-catenin, Runx2), stemness (Sox2), proliferation (c-myc, Akt), and epigenetic regulation (HDAC class 1). AI cells were equally tumorigenic as their adherent counterparts, but showed a significantly decreased rate of growth in-vitro and in-vivo (p < 0.05). Treatment with the pan-histone deacetylase inhibitor vorinostat and the DNA methyltransferase inhibitor 5-azacytidine mitigated AI growth, while 5-azacytidine sensitized anoikis-resistant cells to doxorubicin (p < 0.05).ConclusionsThese data demonstrate remarkable plasticity in anoikis-resistant human osteosarcoma subpopulations accompanied by a rapid development of chemoresistance and altered growth rates mirroring the early stages of latent metastasis. Targeting epigenetic regulation of this process may be a viable therapeutic strategy.

Abstract B50: Targeting KRASG12D colorectal cancer with combined inhibition of PI3K/mTOR and MAPK signaling

Compounds validated in preclinical models frequently exhibit limited efficacy when transitioned into human trials. As a result, many patients are stratified into treatment regimens that have little impact on their disease. With the goal of establishing preclinical models that can more accurately predict tumor responses, there has been a recent emphasis to develop patient-derived xenograft (PDX) models. We strove to develop new PDX models of colorectal cancer (CRC) to test the feasibility of targeted therapy using small molecule inhibitors. Following transplantation of patient tumor specimens into athymic nude mice, sixteen PDX models were successfully established. Common somatic mutations were determined using custom gene panels and targeted with appropriate inhibitors against PI3K/AKT, mTOR and/or MAPK signaling. In four independent models, single agent therapies against PI3K and mTOR had little impact on tumor growth. Conversely, robust declines in tumor burden were universally observed in all models with MEK inhibition with the exception of one, a KRAS G12D mutant model. Within this model, tumor response was achieved only with dual inhibition of mTOR and MAPK signaling. Given the unmet clinical need to treat tumors with aberrant KRAS signaling, these results encourage further investigation into combination treatment strategies with these small molecule inhibitors. Citation Format: Danielle Burgenske, David Monsma, Dawna Dylewski, Stephanie Scott, Aaron Sayfie, Donald Kim, Martin Luchtefeld, Katie Martin, Paul Stephenson, Galen Hostetter, Nadav Dujovny, Jeffrey MacKeigan. Targeting KRASG12D colorectal cancer with combined inhibition of PI3K/mTOR and MAPK signaling. [abstract]. In: Proceedings of the AACR Special Conference: Targeting the PI3K-mTOR Network in Cancer; Sep 14-17, 2014; Philadelphia, PA. Philadelphia (PA): AACR; Mol Cancer Ther 2015;14(7 Suppl):Abstract nr B50.

Abstract 361: Illuminating the effects of tissue degradation to improve the management of tissues used in cancer research or clinical applications

Proceedings: AACR Annual Meeting 2014; April 5-9, 2014; San Diego, CA The collection and preservation of tissues from surgery to the lab affects the quality and value of research and/or healthcare applications for cancer patients. To this end we are further defining best practices in the management of surgically resected tissues through the analysis of a systems biology model. In the first steps of this process we defined an experimental animal model to assess gene expression signatures of biological pathways related to preservation intervals. Athymic nude murine tissues (kidney, liver, brain, and PDX-models) were excised at baseline and immediately aliquots of tissue were incubated at varying intervals (t=0, 0.5, 1, 3, 6, 9 hours, temperature during intervals was maintained at 37°C). All tissues were preserved at -80C and total RNA was extracted from these tissues using a uniform RNA isolation kit (RNeasy Mini Kit from Qiagen, Inc.) by the same technician. The RNA Integrity Number (RIN) was obtained using the Agilents BioAnalyzer 2100 and RNA concentration using the Nanodrop 8000. Subsequently, cDNA was synthesized using Thermo Scientifics RevertAid First Strand cDNA Synthesis Kit, and qPCR analysis of each tissue was performed using the Qiagen RT2 Profiler PCR Arrays including the Cell Death PathwayFinder (murine origin liver, kidney, brain) and Hypoxia Signaling Pathway (human origin PDX-model) arrays. Results show a variation of transcripts s that were upregulated (Bax, Bcl2, Fos, Egr1, Tnfrsf10b) or downregulated (Ctss, Hmox1, Epo, Snca) with increased length of tissue incubation and tissue-specific patterns of regulation were observed. Furthermore, these changes in gene expression did not correlate with a decrease in RIN scores, which raises questions about the suitability of RIN as a comprehensive assessment of RNA quality at the transcript level. These findings have important implications for cancer research, namely that tissues that have not been stabilized within several minutes of excision from the host might have undergone degradation of RNA templates that reflect tissue management-related biological activity versus disease-related transcriptome. In future work, we plan to compare macroanalytes such as microRNAs, non-coding RNAs and proteins across different tissue types with an assumption that gene signatures and biological effects will differ per tissue type. We will use results from our experimental biopreservation model to continue the emphasis on transcriptome changes within the context of biological relevance. Citation Format: Scott D. Jewell, Eric Collins, John Beck, David Monsma, Dawna Dylewski, Andrew Borgman, Mary Winn, Galen Hostetter. Illuminating the effects of tissue degradation to improve the management of tissues used in cancer research or clinical applications. [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 361. doi:10.1158/1538-7445.AM2014-361

Abstract A70: Chemotherapy-resistant subpopulations in a tumor-initiating cell model of human osteosarcoma

Introduction: Tumor-initiating cells (TICs) are a subpopulation of therapy resistant cells in osteosarcoma. To date, TICs have been experimentally defined based on the chosen isolation technique. Common to all isolation methods, TICs demonstrate a high degree of the multi-drug resistance phenotype, are more invasive, metastasize readily, and have increased capacity for tumorigenesis. Given this phenotype TICs represent a viable therapeutic target. Cultivating cells in low-adherence, serum-deprived conditions results in spherical colony formation. Certain growth factors present in serum drive epithelial-to-mesenchymal transition, and typically confound TIC models of carcinoma. The effect of serum on mesenchymal-based sarcoma TIC models is less clear. Recent literature suggests that merely the presence of low-adherence conditions can enrich in TICs. We have shown that osteosarcoma cells cultured in low-adherence conditions in the presence of 10% fetal bovine serum (FBS) form spheres, exhibit an increased expression of genes implicated in developmental programs and stem cells, are more resistant to chemotherapy, and more readily initiate tumors than adherent cells when injected into mice. Methods: The osteosarcoma cell lines 143B, mHOS, and MG-63 were cultured in low-adherence plates using DMEM or MEM media supplemented with 10% FBS. Spheres were routinely grown for 4-6 days between passages and dissociated using 0.05% trypsin. Relative mRNA expression levels of Nanog, Sox2, Oct4 and Axin2 was assessed using quantitative-PCR (SYBR Green). Resistance to chemotherapy (doxorubicin and cisplatin) was determined by comparing both adherent cells and spheres (in low adherence 96-well plates) grown to at least the second passage. After 72 hours IC50 values were calculated using the CellTitre Glo luminescence assay. To assess in vivo tumorgenicity, adherent and sphere cells were dissociated and injected into the flanks of nude mice at a density of 10,000 cells. Upon tumor formation the mice were euthanized and the tumors re-implanted into a second set of mice to test their ability to serially transplant. Results: Sarcospheres can be successfully and reproducibly grown and passaged in low-adherent culture conditions in media supplemented with 10% FBS. Expression profiling demonstrates increased expression of Sox2, Oct4, Nanog, and Axin2, all genes that have been identified to be associated with developmental programs and stem cells. Sarcospheres also displayed increased chemoresistance compared to adherent cultures to both cisplatin (IC50 change of 1.9, 6.9, and 6.2-fold in 143B, mHOS and MG-63, respectively) and more so to doxorubicin (IC50 change of 12.7, 35 and greater than 74-fold in 143B, mHOS, and MG-63, respectively). Additionally, in a pilot subcutaneous xenograft study, using sphere and adherent cells, differential growth was observed within the TIC-enriched populations demonstrating (1) earlier tumor initiation and (2) a lower overall proliferation rate among the primary and serially transplanted tumors confirming a persistence of the TIC phenotype through in vivo passaging. Conclusions: Osteosarcoma cells cultured in low-adherence conditions in the presence of serum display many of the characteristics of putative tumor-initiating cells including robust sphere formation, upregulation of key stemness mediators, enhanced chemoresistance, and in vivo tumor initiation and serial transplantability. Using this system, we aim to further evaluate these stem-like populations in order to elucidate potential therapies targeting this specialized, chemoresistant niche. Citation Format: Jessica M. Foley, Noel R. Monks, Donald J. Scholten, II, David J. Monsma, Dawna Dylewski, Paula J. Davidson, Matthew R. Steensma. Chemotherapy-resistant subpopulations in a tumor-initiating cell model of human osteosarcoma. [abstract]. In: Proceedings of the AACR Special Conference on Pediatric Cancer at the Crossroads: Translating Discovery into Improved Outcomes; Nov 3-6, 2013; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2013;74(20 Suppl):Abstract nr A70.

Abstract B12: Overcoming acquired resistance to vemurafenib using clinically relevant PDX models of melanoma.

Development of resistance is a significant clinical problem for virtually all targeted cancer therapies. We have generated a reproducible, patient derived xenograft (PDX) model of acquired vemurafenib resistance to address these challenges. Continuous treatment of V600E melanoma tumors, caused synchronous tumor stasis for approximately 7 weeks, following which, all tumors displayed simultaneous resistance marked by rapid tumor growth. Additionally, this model maintains the resistance phenotype upon serial transplantation, providing a platform for testing rational drug selection. The fidelity of the PDX models was further confirmed using a BRAF V600V tumor which did not respond to vemurafenib. Onset of vemurafenib resistance is accompanied by increased phosphor-ERK signifying re-engagement of the MAPK signaling pathway and supporting MEK as a potential target. MEK inhibition in vemurafenib resistant tumors using PD0325901, resulted in rapid tumor shrinkage and dramatically reduced phosphor-ERK levels. Treatment of resistant tumors with PD0325901 alone, whilst leading to rapid tumor shrinkage, showed significant host toxicity and onset of acquired MEKi resistance. Interestingly, combination of vemurafenib + PD0325901 was non-toxic, and showed dramatic and sustained tumor suppression. Upon cessation of PD0325901 at 70 days the tumors remained undetectable for the duration of the study (>100 days). These data support the use of MEK inhibitors post-development of vemurafenib resistance and demonstrate that combination therapy mitigates systemic MEKi toxicity and results in persistent tumor inhibition/eradication. PDX models of acquired resistance provide a unique opportunity to bridge the gap between patients and the basic in vitro biology. Additionally, this PDX system allows the interrogation of the kinetics involved in the development of resistance by longitudinal tumor tissue sampling. Numerous mechanisms have been identified as potential causes of the resistance phenotype. Many have been identified in vitro but not all have been confirmed in patients. We detected no evidence of increased BRAF copy number or expression, although alternative BRAF splicing was identified in resistant tumors. Using differential gene expression accompanied by pathway and network analysis we identified distinct differences in the PDX tumors at various time points during the development of resistance. In particular, a potential role for interferon signaling in resistant tumors was observed. Furthermore, changes in the metabolic profiles were identified with untreated and resistant tumors favoring glycolytic pathways, whereas growth arrested tumors exhibited a preference for oxidative phosphorylation. In conclusion, these results demonstrate the value of PDX models for contributing to clinical cancer management through the decryption of complex drug resistance mechanisms and accelerating the identification of rationally selected drug combinations for bench to bedside applications. Citation Information: Mol Cancer Ther 2013;12(11 Suppl):B12. Citation Format: Noel R. Monks, David J. Monsma, David M. Cherba, Emily Eugster, Dawna Dylewski, Mary E. Winn, Andrew S. Borgman, Paula J. Davidson, Chelsea A. Peterson, Jose M. Pimiento, Alexander E. Ivliev, Yuri Nikolsky, Marina Bessarabova, Valerie S. Calvert, Mariaelena Pierobon, Emanuel F. Petricoin, Craig P. Webb, Brian J. Nickoloff. Overcoming acquired resistance to vemurafenib using clinically relevant PDX models of melanoma. [abstract]. In: Proceedings of the AACR-NCI-EORTC International Conference: Molecular Targets and Cancer Therapeutics; 2013 Oct 19-23; Boston, MA. Philadelphia (PA): AACR; Mol Cancer Ther 2013;12(11 Suppl):Abstract nr B12.

Abstract 363: Development of computer-aided detection (CAD) tool for liver metastasis micro CT imaging using targeted contrast agent

Proceedings: AACR 103rd Annual Meeting 2012‐‐ Mar 31‐Apr 4, 2012; Chicago, IL Introduction: Preclinical in-vivo micro CT studies of liver metastasis are difficult due to poor inherent soft tissue contrast and the need for highly technical, manual analysis of the data. Research has implicated that Kupffer cells in the liver encapsulate liver metastases providing an opportunity to deliver macrophage-specific contrast agents for the detection of small metastatic lesions. A new, long-acting preclinical CT contrast agent that targets Kupffer cells has been developed that may allow automated detection of liver lesions via CAD software. Method: A pancreatic cancer liver metastasis model was created by surgically implanting human pancreatic cancer cell line (L3.6pl) within the spleen. Mice were injected with the contrast agent and scanned to obtain baseline anatomy of the liver before implantation. The mice were scanned 1 day after implantation and weekly after that for 5 weeks to monitor the liver metastasis progression. The control group followed an identical protocol but with a sham surgery. Liver tissues were harvested and fixed in paraffin blocks after the last scan. Paraffin blocks were scanned using high resolution micro CT before IHC staining. Human Mitochondrial and F4/80 IHC were used to identify L3.6pl and Kupffer cells, respectively. The CT images were compared to the IHC images from the same block to verify that the locations of the contrast agent and the Kupffer cells were related. Once the pattern of contrast agent and metastatic tumors had been identified, CAD software was developed for automatic tumor detection. Results: The contrast agent was evenly distributed throughout the healthy liver tissue within 1 hour post injection. In healthy mice, the homogenous distribution of contrast remained unchanged for at least 6 weeks. In liver metastasis models, the contrast began to concentrate in various areas of the liver within 2 weeks post implantation. As tumors developed and grew, the contrast became highly concentrated on the borders of tumors creating a 3 dimensional outline of the lesion. IHC staining and micro CT imaging of the fixed tissue verified that the tumors are surrounded by Kupffer cells and that the distribution of concentrated contrast agent matched them. Software was able to detect the tumors based on these contrast outlines and compare them over successive weekly scans. Conclusion: Our new imaging method enables automated detection and evaluation of liver metastasis 1 mm or smaller from as early as 2 weeks. In addition to allowing better visualization, it provides new insight into macrophage motility within the liver. CAD software can take advantage of this unique capability to automate data analysis and allow for large scale longitudinal studies. This new imaging method could be a useful tool to facilitate longitudinal imaging of liver metastases in mice and has the potential for translation into clinical practice. 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 363. doi:1538-7445.AM2012-363