Patient-derived xenograft platform of OSCC: a renewable human bio-bank for preclinical cancer research and a new co-clinical model for treatment optimization

Abstract

Advances in next-generation sequencing and bioinformatics have begun to reveal the complex genetic landscape in human cancer genomes, including oral squamous cell carcinoma (OSCC). Sophisticated preclinical models that fully represent intra- and inter-tumoral heterogeneity are required to understand the molecular diversity of cancer and achieve the goal of personalized therapies. Patient-derived xenograft (PDX) models generated from human tumor samples that can retain the histological and genetic features of their donor tumors have been shown to be the preferred preclinical tool in translational cancer research compared with other conventional preclinical models. Specifically, genetically well-defined PDX models can be applied to accelerate targeted antitumor drug development and biomarker discovery. Recently, we have successfully established and characterized an OSCC PDX panel as part of our tumor bio-bank for translational cancer research. In this paper, we discuss the establishment, characterization, and preclinical applications of the PDX models. In particular, we focus on the classification and applications of the PDX models based on validated annotations, including clinicopathological features, genomic profiles, and pharmacological testing information. We also explore the translational value of this well-annotated PDX panel in the development of co-clinical trials for patient stratification and treatment optimization in the near future. Although various limitations still exist, this preclinical approach should be further tested and improved.

Keywords

patient-derived xenograft models / personalized medicine / co-clinical trial / patient stratification / oral squamous cell carcinoma

Introduction

Oral squamous cell carcinoma (OSCC) is the most common malignant tumor arising from the oral and maxillofacial region; it accounts for approximately 300 400 new cases and 145 400 deaths in 2012 worldwide [ 1]. In the last few decades, the integration of surgical management of the primary tumor followed by postoperative radiotherapy or chemo-radiotherapy has become the standard of care for this disease [ 2]. According to our recently published survival analysis, the five-year overall survival rate of 256 patients with stage III or IVA OSCC reached 61.7% as a result of improvement of radical surgery and functional reconstruction in combined and sequential therapy for OSCC [ 3]. However, the limited efficiency of cytotoxic chemotherapy against this heterogeneous disease has prompted the development of targeted agents for OSCC. Unfortunately, cetuximab, an epidermal growth factor receptor (EGFR) antibody, is the only FDA-approved targeted agent for OSCC; meanwhile, other targeted therapies against known oncogenic alterations, such as BRAF mutation and ALK rearrangement, have rapidly expanded and have been implemented as first-line treatment in various cancers, including melanoma, non-small cell lung cancer, and lymphoma [ 4, 5]. Thus, the translation and application of targeted therapy for OSCC remain unmet.

The recent progress in high-throughput sequencing technologies has enabled a systematic cataloguing of cancer genomes, which in turn stimulated substantial advancement in understanding the disease’s molecular underpinnings and relevant therapeutic molecular classification [ 6, 7]. Currently, the classification of OSCC based on anatomic site, histology, and TNM stage fails to facilitate targeted therapies in the clinical setting. Various predictive biomarkers and molecular features with regard to certain targeted therapies on the basis of HPV status, driver oncogenetic alterations, or global gene expression patterns are under intense investigation [ 8]. However, the incorporation of these biomarkers into the care of patients remains limited, at least in part, by the lack of validation in reliable preclinical models.

Patient-derived tumor xenograft (PDX) models have been gaining popularity as the preclinical model for drug screening, biomarker identification, and tumor biological studies [ 9]. PDX models can serve as a personalized mouse model known as a “mouse avatar,” wherein a tumor is maintained in a cohort of mice and harvested at different time points after drug application for clinical decision making and underlying mechanism studies [ 10]. Recently, we extensively characterized xenograft models at histological, molecular, and pharmacological levels to ensure that they represent the diversity of clinical situations. Whole-exome sequencing, gene-expression microarray, and copy-number assessment were performed to characterize and, more importantly, categorize PDX models based on their genetic feature and potential drug susceptibility. In addition, a PDX model is considered renewable as a tissue resource in terms of quantity; in terms of quality, it is significant for its stable and clinical predictive value [ 11, 12]. In the foreseeable future, our comprehensively annotated PDX platform might be implemented in a co-clinical trial project for patient stratification and treatment optimization.

Establishment and characterization of OSCC PDX models as a renewable resource in a tumor bio-bank for translational cancer research

Translational cancer research relies significantly on the application of cancer tissue to explore the mechanisms of carcinogenesis and validate proposed potential biomarkers. We have established an open-source tumor bio-bank that includes tumor samples from more than 4000 cancer patients, blood samples from more than 2000 cancer patients, and saliva samples from more than 1000 cancer patients in Ninth People’s Hospital, Shanghai Jiao Tong University School of Medicine (http://bmf.shsmu.edu.cn/OMNDB/page/home/home.jsp). To further integrate clinical cancer research with bioinformatics and provide renewable resources for translational cancer research, we also established and characterized a panel of OSCC PDX models.

The workflow of the PDX model establishment is shown in Fig. 1. OSCC tumor tissue was collected and subcutaneously implanted into the immune-compromised mice. The clinical features, such as anatomic site, histology, TNM stage, and HPV status of the patients were documented for annotation of the PDXs. When the size of the xenograft derived from the patient (called P0) reached 1000–1500 mm³ in a mouse, the xenograft was surgically excised and small tumor fragments were re-transplanted to the nude mice (called P1, P2, P3…) for tissue expansion. To date, we have developed a panel of 62 PDX models derived from different OSCC patients. To further annotate these PDX models at the genetic level, we have conducted whole-exome sequencing, gene expression microarray, and genome-wide SNP arrays to facilitate further preclinical studies.

During the characterization of PDX models, the fidelity of the model in serial transplantation should be evaluated (Fig. 1). The histopathologic and immunohistochemical features revealed by HE and IHC staining of the PDXs showed high concordance between PDXs and the corresponding patient’s tumor (Fig. 2). Mutation status, CNV profiles, and gene expression profiles are also required during the evaluation between the xenograft and the original tumor it was derived from. We establish that PDX shares a similar histology and comparable genetic profiles with its parental tumor over several passages.

Comprehensive annotation and classification of the PDX panel that can represent the molecular heterogeneity of OSCC

Recent advancement in high-throughput techniques in human cancer genomes substantially increased knowledge on the molecular underpinning of cancer biology and provided potential targeted therapeutic approaches [ 13]. In OSCC, incorporating the genetic profile into the clinical setting will allow us to move beyond HPV status as the only prognostic biomarker and will facilitate novel clinical trial designs [ 14]. As a result, deciphering the complex molecular heterogeneity in tumors is expected to lead the targeted drug development to more individualized therapies. Aside from genetic alterations, head and neck squamous cell carcinomas (HNSCC) can also be classified based on the gene expression profile that can reflect the overall biological characteristics of the subtype and can be used to generate hypotheses for targeted therapeutics [ 15, 16]. To this end, the quest for personalized medicine will become easy if a comprehensive annotated PDX panel that encompasses the full breadth of molecular heterogeneity is established and incorporated into the clinical setting.

To investigate the comprehensive molecular landscape, the integrative genetic profiles of the PDX models must be analyzed. For example, we show the mutational status of previously reported significantly mutated genes in 23 of our PDX models. Notably, TP53 mutations were detected in 17 of the 23 PDX models. Notch1, Notch2, and CDKN2A were also frequently mutated in the PDX panel (Fig. 3). In addition, 12 PDX models analyzed in this panel demonstrated mutations in the PTP family (Fig. 3). The gene expression data from PDX models can also be utilized to classify these models into distinct subtypes based on recently reported gene signatures [ 17]. If the gene expression pattern of the PDX models can spread across different molecular subtypes, then this scenario may also prove the inter-patient molecular heterogeneity they represent in OSCC.

To further annotate the PDX model at the pharmacological level, chemo-sensitivity experiments on standard-of-care cytotoxic agents in OSCC (docetaxel, cisplatin, and 5-fluorouracil) were also conducted for several models. Consequently, the comprehensive annotations comprising clinical, genetic, and pharmacological profiles of the PDX models were documented in the database of our bio-bank.

Preclinical application of classified PDX models in drug development and cancer research

In a prospective open-label phase III study of docetaxel, cisplatin, and fluorouracil (TPF) induction chemotherapy in patients with resectable OSCC, we found that TPF induction did not improve survival when compared with up-front surgery in patients with resectable stage III or IVA OSCC [ 18]. In addition, most novel targeted agents against recurrent oncogenetic alterations in HNSCC, such as EFGR/HER2 inhibitor, PI3K/mTOR inhibitor, and IFG-1R mAb, are still under clinical trial; in addition, the response to a certain therapy varies among different patients because of the molecular heterogeneity of HNSCC [ 19]. Numerous studies have proven the predictive value of PDX models in drug development and biomarker discovery, including HNSCC [ 20, 21].

We assumed that PDX models can be classified based on the validated driver oncogenetic annotations. A well-recognized role in the pathogenesis of HNSCC has been attributed to the aberrant activation of the PI3K signaling pathway resulting from mutations or amplification/deletion of PI3K pathway-related genes, such as PIK3CA or PTEN [ 2, 19]. The assumption of PI3K pathway activation can also be validated in the RNA and protein level in the PDX model. Thus, PDX models can be classified as “PI3K pathway activated” models and “PI3K pathway driven” models (Fig. 1). Hence, PI3K pathway-targeted drug screening and predictive biomarker discovery can be comparatively conducted in two subtypes of PDX models to allow for a more detailed molecular analysis of predictive biomarker discovery. Similarly, other driver oncogenes can be identified in the PDX panel based on genetic profiles, such as “EGFR amplified/mutated” models, “FGFR1 amplified” models, and “c-MET amplified” models.

However, these unsystematic classified PDX models are not exclusive, which means a PDX model can be both a “PI3K pathway activated” model and a “c-MET amplified” model based on the validated genetic annotations. As a consequence, considering the genetic background, the classified PDX model harboring other genetic disorders will provide more information during targeted drug evaluation, especially during resistant mechanism studies. From the translational point of view, once the concurrent genetic patterns in the PDX model are comparatively analyzed in patients, PDX models (more than one model may be needed) can provide therapeutic suggestions to the corresponding patient with a similar concurrent genetic pattern.

In the meantime, clinical and pharmacological annotations can also be implemented in the classification of the PDX panel. The PDX panel can be classified based on studies focusing on etiology (HPV infection, smoking status, primary or recurrent tumor), histology (differentiation status), and drug resistance (cisplatin, 5-FU, docetaxel chemo-sensitivity). According to the clinical annotation, PDX models defined as “recurrent tumor” or “metastatic tumor” may show distinct molecular patterns compared with those defined as “primary tumor.” Interestingly, “recurrent tumor” or “metastatic tumor” is frequently observed after chemotherapy or radiotherapy in the treatment of OSCC. If the pharmacological annotation confirms the chemo-sensitivity of the “recurrent tumor” or “metastatic tumor” PDX models, then these PDX models are of particular interest to researchers in characterizing cytotoxic therapy resistance mechanisms.

Our studies also illustrated how PDX models can facilitate the development of cancer stem cell (CSC) therapeutics in the treatment of OSCC [ 22, 23]. We demonstrated the therapeutic potential of the c-Met inhibitor that preferentially targets CSC using our PDX models [ 23]. PDX models that can mimic the lineage hierarchy of cells within the tumor may have significant implications on the understanding of molecular mechanisms in CSC and could facilitate CSC-directed clinical trial designs.

Clinical trial design and personalized medicine: PDX models in the development of co-clinical trials

The incorporation of next-generation sequencing (NGS) into clinical oncology facilitated the development of predictive biomarker-directed clinical trials with targeted therapy against certain altered genes or pathways in each cancer patient [ 24]. Consequently, novel clinical trial designs are required, such as trials evaluating targeted agents based on the genetic alterations in enrolled patients; these trials are also called “basket trials.” Despite the significant benefits of recently published basket trials, one of the greatest challenges is that tumors harboring multiple genetic alterations demonstrate limited efficiency in single-drug targeted therapy [ 25]. Even if functional validated and targetable genetic alterations are classified as driver alterations, the relevance of other concurrent alterations in cancer remains inconclusive [ 26]. The concept of co-clinical trial is paralleled trials that integrate GEMMs, which represent the patients enrolled in a clinical trial for drug evaluation and further clinical trial design [ 27]. Such a co-clinical system can help identify the genetic features indicative of drug resistance and evaluate potential drug combinations to overcome such resistance. With regard to the heterogeneity of OSCC, a molecular-annotated PDX panel may provide comprehensive and predictive information on targeted drug selection when compared with GEMMs.

Previously, personalized PDX models have been utilized as an in vivo platform to test therapeutic strategies proposed based on the genetic profiles of the patients’ tumor [ 28, 29]. Unfortunately, technical limitations exist when a PDX model is established as “avatar” in clinical applications; such limitations include (1) the engraftment rate of cancer in mice, (2) the period of serial passage and drug evaluation after detailed molecular annotation of a PDX model, (3) the difference in pharmacokinetics between two species, and (4) the divergence in the definition of drug responsiveness in a PDX model and patient. These limitations need to be overcome before the application of real-time personalized medicine for patients based on PDX models. Here, we discuss the possibility of well-characterized PDX models in a co-clinical setting when coordinate stratification criteria are met between patients and PDX models.

In the co-clinical “basket-like trial” testing a targeted drug against driver mutation X (illustrated as the green background in Fig. 4), all the PDX models with mutation X are selected and implemented in the drug test before the enrollment of patients. The identification of the concurrent genetic feature (illustrated as the pink color in the xenograft in Fig. 4) with X mutation that confer de novo resistance in the PDX models can facilitate the exclusion of patients (pink-colored patients with mutation X). Prior to the clinical trial, the drug-sensitive PDX models should be under prolonged treatment to mimic the course of acquired resistance observed in the ongoing clinical trial. The xenograft in the mice should be obtained according to the variation tendencies of the xenograft growth curve in the PDX model to better comprehend the process and molecular underpinning of the acquired resistance. Once the acquired resistance (defined by the tumor relapse under consecutive treatment, illustrated in xenografts or patients as the red background in Fig. 4) is observed in the PDX model, comparative functional analysis of the xenograft should be implemented to identify the molecular mechanisms in different PDX models. Potential drug combinations are then tested in acquired-drug-resistant PDX models based on the result of the analysis. Then, the most effective drug combination in the PDX model is applied to the patients of the corresponding subtypes who also acquired drug resistance (Fig. 4). The purpose of this co-clinical trial is to identify how concurrent genetic alterations may affect the targeted therapy against the known driver mutation. In this process, both the patients and the PDX model should be well characterized and compared at multiple molecular levels to ensure that the de novo and acquired resistance mechanisms identified in the PDX model can be translated to the same subtype of patients.

In this co-clinical “basket-like trial,” the application of the PDX model representing the patients enrolled in the trial allows for the validation of biomarker strategies and discovery of the mechanisms of resistance that may benefit the corresponding subtype of patients. Moreover, the predictive biomarker and the patient stratification criteria identified from the PDX models provide valuable insights into the design of clinical trials. From the translational viewpoint, a PDX model would probably provide therapeutic recommendations to a genetic match-up patient instead of the patient that the model was derived from.

Conclusions

Previously, PDX models have been reported to improve the accuracy of results during drug screening and predictive biomarker development in pharmaceutical companies and laboratories. The recent integration of high-throughput technologies helped accelerate the development of genetically characterized PDX models that provide significant molecular insights during their application. Our comprehensively annotated and classified PDX panel is expected to provide a novel research platform that may be implemented in a co-clinical trial project for patient stratification and treatment optimization in the near future. This newly emerging concept attempts to bridge the gap among genetic informatics, cancer biology, and targeted drug research, thereby accelerating the translation of genomic and therapeutic discoveries into clinical benefits for cancer patients.

References

[1] Torre LA, Bray F, Siegel RL, Ferlay J, Lortet-Tieulent J, Jemal A. Global cancer statistics, 2012. CA Cancer J Clin 2015; 65(2): 87–108

[2] Leemans CR, Braakhuis BJM, Brakenhoff RH. The molecular biology of head and neck cancer. Nat Rev Cancer 2011; 11(1): 9–22

[3] Zhong L, Sun J, Guo W, Tu W, Zhu H, Ji T, Hu Y, Li J, Ye W, Xu L, He Y, Yang W, Wang Y, Ren G, Yin Q, Cai Y, Yang X, Zhang C, Zhang Z. Survival analysis of 256 patients with oral cancer. Chin J Clin Oncol (Zhongguo Zhong Liu Lin Chuang) 2015; 42(4): 217–221 (in Chinese)

[4] Bonner JA, Harari PM, Giralt J, Azarnia N, Shin DM, Cohen RB, Jones CU, Sur R, Raben D, Jassem J, Ove R, Kies MS, Baselga J, Youssoufian H, Amellal N, Rowinsky EK, Ang KK. Radiotherapy plus cetuximab for squamous-cell carcinoma of the head and neck. N Engl J Med 2006; 354(6): 567–578

[5] Sawyers C. Targeted cancer therapy. Nature 2004; 432(7015): 294–297

[6] Stransky N, Egloff AM, Tward AD, Kostic AD, Cibulskis K, Sivachenko A, Kryukov GV, Lawrence MS, Sougnez C, McKenna A, Shefler E, Ramos AH, Stojanov P, Carter SL, Voet D, Cortés ML, Auclair D, Berger MF, Saksena G, Guiducci C, Onofrio RC, Parkin M, Romkes M, Weissfeld JL, Seethala RR, Wang L, Rangel-Escareño C, Fernandez-Lopez JC, Hidalgo-Miranda A, Melendez-Zajgla J, Winckler W, Ardlie K, Gabriel SB, Meyerson M, Lander ES, Getz G, Golub TR, Garraway LA, Grandis JR. The mutational landscape of head and neck squamous cell carcinoma. Science 2011; 333(6046): 1157–1160

[7] Agrawal N, Frederick MJ, Pickering CR, Bettegowda C, Chang K, Li RJ, Fakhry C, Xie TX, Zhang J, Wang J, Zhang N, El-Naggar AK, Jasser SA, Weinstein JN, Treviño L, Drummond JA, Muzny DM, Wu Y, Wood LD, Hruban RH, Westra WH, Koch WM, Califano JA, Gibbs RA, Sidransky D, Vogelstein B, Velculescu VE, Papadopoulos N, Wheeler DA, Kinzler KW, Myers JN. Exome sequencing of head and neck squamous cell carcinoma reveals inactivating mutations in NOTCH1. Science 2011; 333(6046): 1154–1157

[8] Chung CH, Parker JS, Karaca G, Wu J, Funkhouser WK, Moore D, Butterfoss D, Xiang D, Zanation A, Yin X, Shockley WW, Weissler MC, Dressler LG, Shores CG, Yarbrough WG, Perou CM. Molecular classification of head and neck squamous cell carcinomas using patterns of gene expression. Cancer Cell 2004; 5(5): 489–500

[9] Tentler JJ, Tan AC, Weekes CD, Jimeno A, Leong S, Pitts TM, Arcaroli JJ, Messersmith WA, Eckhardt SG. Patient-derived tumour xenografts as models for oncology drug development. Nat Rev Clin Oncol 2012; 9(6): 338–350

[10] Malaney P, Nicosia SV, Davé V. One mouse, one patient paradigm: New avatars of personalized cancer therapy. Cancer Lett 2014; 344(1): 1–12

[11] Hidalgo M, Amant F, Biankin AV, Budinská E, Byrne AT, Caldas C, Clarke RB, de Jong S, Jonkers J, Mælandsmo GM, Roman-Roman S, Seoane J, Trusolino L, Villanueva A. Patient-derived xenograft models: an emerging platform for translational cancer research. Cancer Discov 2014; 4(9): 998–1013

[12] Zhang X, Claerhout S, Prat A, Dobrolecki LE, Petrovic I, Lai Q, Landis MD, Wiechmann L, Schiff R, Giuliano M, Wong H, Fuqua SW, Contreras A, Gutierrez C, Huang J, Mao S, Pavlick AC, Froehlich AM, Wu MF, Tsimelzon A, Hilsenbeck SG, Chen ES, Zuloaga P, Shaw CA, Rimawi MF, Perou CM, Mills GB, Chang JC, Lewis MT. A renewable tissue resource of phenotypically stable, biologically and ethnically diverse, patient-derived human breast cancer xenograft models. Cancer Res 2013; 73(15): 4885–4897

[13] Vogelstein B, Papadopoulos N, Velculescu VE, Zhou S, Diaz LA Jr, Kinzler KW. Cancer genome landscapes. Science 2013; 339(6127): 1546–1558

[14] Kang H, Kiess A, Chung CH. Emerging biomarkers in head and neck cancer in the era of genomics. Nat Rev Clin Oncol 2015; 12(1): 11–26

[15] De Cecco L, Bossi P, Locati L, Canevari S, Licitra L. Comprehensive gene expression meta-analysis of head and neck squamous cell carcinoma microarray data defines a robust survival predictor. Ann Oncol 2014; 25(8): 1628–1635

[16] Lohavanichbutr P, Méndez E, Holsinger FC, Rue TC, Zhang Y, Houck J, Upton MP, Futran N, Schwartz SM, Wang P, Chen C. A 13-gene signature prognostic of HPV-negative OSCC: discovery and external validation. Clin Cancer Res 2013; 19(5): 1197–1203

[17] De Cecco L, Nicolau M, Giannoccaro M, Daidone MG, Bossi P, Locati L, Licitra L, Canevari S. Head and neck cancer subtypes with biological and clinical relevance: meta-analysis of gene-expression data. Oncotarget 2015; 6(11): 9627–9642

[18] Zhong LP, Zhang CP, Ren GX, Guo W, William WN Jr, Sun J, Zhu HG, Tu WY, Li J, Cai YL, Wang LZ, Fan XD, Wang ZH, Hu YJ, Ji T, Yang WJ, Ye WM, Li J, He Y, Wang YA, Xu LQ, Wang BS, Kies MS, Lee JJ, Myers JN, Zhang ZY. Randomized phase III trial of induction chemotherapy with docetaxel, cisplatin, and fluorouracil followed by surgery versus up-front surgery in locally advanced resectable oral squamous cell carcinoma. J Clin Oncol 2013; 31(6): 744–751

[19] Wen Y, Grandis JR. Emerging drugs for head and neck cancer. Expert Opin Emerg Drugs 2015; 20(2): 313–329

[20] Sen M, Pollock NI, Black J, DeGrave KA, Wheeler S, Freilino ML, Joyce S, Lui VW, Zeng Y, Chiosea SI, Grandis JR. JAK kinase inhibition abrogates STAT3 activation and head and neck squamous cell carcinoma tumor growth. Neoplasia 2015; 17(3): 256–264

[21] Klinghammer K, Raguse JD, Plath T, Albers AE, Joehrens K, Zakarneh A, Brzezicha B, Wulf-Goldenberg A, Keilholz U, Hoffmann J, Fichtner I. A comprehensively characterized large panel of head and neck cancer patient-derived xenografts identifies the mTOR inhibitor everolimus as potential new treatment option. Int J Cancer 2015; 136(12): 2940–2948

[22] Sun S, Wang Z. Head neck squamous cell carcinoma c-Met⁺ cells display cancer stem cell properties and are responsible for cisplatin-resistance and metastasis. Int J Cancer 2011; 129(10): 2337–2348

[23] Sun S, Liu S, Duan SZ, Zhang L, Zhou H, Hu Y, Zhou X, Shi C, Zhou R, Zhang Z. Targeting the c-Met/FZD8 signaling axis eliminates patient-derived cancer stem-like cells in head and neck squamous carcinomas. Cancer Res 2014; 74(24): 7546–7559

[24] Simon R, Roychowdhury S. Implementing personalized cancer genomics in clinical trials. Nat Rev Drug Discov 2013; 12(5): 358–369

[25] Lopez-Chavez A, Thomas A, Rajan A, Raffeld M, Morrow B, Kelly R, Carter CA, Guha U, Killian K, Lau CC, Abdullaev Z, Xi L, Pack S, Meltzer PS, Corless CL, Sandler A, Beadling C, Warrick A, Liewehr DJ, Steinberg SM, Berman A, Doyle A, Szabo E, Wang Y, Giaccone G. Molecular profiling and targeted therapy for advanced thoracic malignancies: a biomarker-derived, multiarm, multihistology phase II basket trial. J Clin Oncol 2015; 33(9): 1000–1007

[26] Dancey JE, Bedard PL, Onetto N, Hudson TJ. The genetic basis for cancer treatment decisions. Cell 2012; 148(3): 409–420

[27] Nardella C, Lunardi A, Patnaik A, Cantley LC, Pandolfi PP. The APL paradigm and the “co-clinical trial” project. Cancer Discov 2011; 1(2): 108–116

[28] Garralda E, Paz K, López-Casas PP, Jones S, Katz A, Kann LM, López-Rios F, Sarno F, Al-Shahrour F, Vasquez D, Bruckheimer E, Angiuoli SV, Calles A, Diaz LA, Velculescu VE, Valencia A, Sidransky D, Hidalgo M. Integrated next-generation sequencing and avatar mouse models for personalized cancer treatment. Clin Cancer Res 2014; 20(9): 2476–2484

[29] Morelli MP, Calvo E, Ordoñez E, Wick MJ, Viqueira BR, Lopez-Casas PP, Bruckheimer E, Calles-Blanco A, Sidransky D, Hidalgo M. Prioritizing phase I treatment options through preclinical testing on personalized tumorgraft. J Clin Oncol 2012; 30(4): e45–e48

RIGHTS & PERMISSIONS

Higher Education Press and Springer-Verlag Berlin Heidelberg