Personalized Cancer Medicine Research Articles

Census of cancer genes from a comprehensive review of cancer gene panels

Background: The field of personalised cancer medicine and the supporting diagnostic tools are growing rapidly. Several research projects and many molecular diagnostic companies have selected relevant or clinically actionable genes for targeted next-generation sequencing (NGS) panels. However, there is no consensus in the scientific literature about which genes these are, and what the logic of their selection is (based on their biological function, actionability, mutational frequency and prognostic/therapeutic relevance). Our aim was to compile a comprehensive list of cancer genes and analyse their function and potential clinical relevance. Methods: We collected commercially available cancer gene panels (focusing on NGS panels for SNV detection) and also cancer gene lists defined in the scientific literature. Our study included 34 different NGS-based cancer panels of 22 institutes or diagnostic companies. Solid, haematology, or pan-cancer indication panels and cancer predisposition panels were included. This list was extended with 11 ‘cancer gene’ lists based on literature search such as COSMIC Cancer Gene Census, The Cancer Genomics Resource List 2014, and driver genes according to Vogelstein et al, 2013. We categorised the union of the genes based on their association with solid or haematology indications, their functions, and we explored actionable targets and related compounds based on data from publicly available databases (like ChEMBL and Genomics of Drug Sensitivity in Cancer). Results: Panel sizes vary between 8 and 710, with the average of 162 genes. More than half of the panels (18/34) contain <100 genes, 9 panels contain 117-160 genes and the rest (7 panels) contain ≥315 genes. The cumulated gene list of these panels and literature-based collections includes 2040 cancer-related genes. TP53 and KIT, the most frequent genes are listed in 32 panels, while ∼500 genes are listed in one panel only. More than 250 of the genes are reported to be relevant in solid tumours only, while about 140 genes in haematology only. The rest is either not specified or is relevant in pan-cancer indications. 41 ‘ADME-TOX’ genes are included, which are important in pharmacogenomics (ABC and CYP genes). Other functional groups can be defined, for example enzymes (600<genes), and within that: kinases (300<genes). More than 300 out of 2040 genes can be linked to clinically relevant compounds according to at least one of the databases investigated. Conclusions: A universal ‘cancer gene’ list, which have been selected by the scientific community so far for targeted re-sequencing, contains about 2000 genes. No comprehensive targeted cancer gene panel is available for molecular diagnostics which could cover all these somatic and germline alterations associated with cancer. It is highly important to acknowledge the functional and clinical relevance of these cancer genes in order to find the optimal molecular test for an individual in a specific clinical situation, to help the interpretation of the test Results and to enhance the delivery of genomically-informed therapeutic decisions. This might also help researchers towards novel biomarker discovery and targeted drug design. Legal entity responsible for the study: NA Funding: National Oncogenomic and Precision Oncotherapy Program Funded by the Hungarian Innovation Agency. Disclosure: All authors have declared no conflicts of interest.

Approaches to identify kinase dependencies in cancer signalling networks.

Cells integrate extracellular signals into appropriate responses through a complex network of biochemical reactions driven by the activity of protein and lipid kinases, among other proteins. In order to understand this complexity, new approaches, both experimental and computational, have recently been developed with the aim to identify regulatory kinases and infer their activation status in the context of their signalling network. Here, we review such approaches with particular focus on those based on phosphoproteomics. Integration of kinase activity measurements inferred from phosphoproteomics data with other 'omics' datasets is starting to be used to identify regulatory nodes in biochemical networks. These methodologies may in the future be used to identify patient-specific targets and thus advance personalised cancer medicine.

Abstract 1745: Multifocal prostate cancer has high degree of genomic heterogeneity

Abstract Prostate cancer is the most common cancer type among men in the Western world. Most prostate cancers are multifocal with individual tumors harboring different aggressiveness. In recent genomics studies, the multifocal nature of prostate cancer has been investigated only in small sample cohorts. Here we have investigated the intra-organ mutational spectra of multiple tumors from a large cohort of prostate cancer patients. From radical prostatectomies performed between 2010 and 2012 at Oslo University Hospital, 43 prostatectomy specimens with multiple and clearly separated tumors were identified based on histology. From each of these prostatectomy specimens, DNA from frozen samples from 2-3 different tumor foci and corresponding normal tissue samples were analyzed by high-coverage whole-exome sequencing, adding up to a total of 159 samples. We identified 3093 somatic substitutions, insertions and deletions, with an average of 27 alterations per tumor sample. Both known and novel significantly mutated genes were identified and their distribution among different tumor foci from the same prostate was examined. Mutations in genes such as SPOP, MED12, and FOXA1 have previously been identified in prostate cancer and were also found to be frequently mutated in this cohort. However, the same mutations were rarely found in multiple tumor foci within the same prostate. In fact, for 13 out of the 43 examined patients there were no common mutations among tumors from the same prostate. For 12 patients, we found only one overlapping mutation among tumor foci. Whereas the overall list of mutated genes (n = 2101) overlap significantly with the 594 genes of the Cancer Gene Census (100/594; p=9.6e-06), this was not the case for 142 genes with overlapping mutations among different tumor foci (2/594; p=0.63). This indicates that the overlapping mutations are not typical cancer driving genes. To conclude, results from exome sequencing of multifocal prostate cancer show a large degree of heterogeneity in genomic alterations between different tumor foci within the same prostate. With very few common inter-foci mutations, an implementation of genome-based personalized prostate cancer medicine will require sampling of all tumor foci to tailor optimal treatment. Citation Format: Marthe Løvf, Sen Zhao, Ulrika Axcrona, Bjarne Johannesen, Andreas M. Hoff, Anne Cathrine Bakken, Kristina Totland Carm, Ola Myklebost, Leonardo A. Meza-Zepeda, Agnes K. Lie, Karol Axcrona, Ragnhild A. Lothe, Rolf I. Skotheim. Multifocal prostate cancer has high degree of genomic heterogeneity [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 1745. doi:10.1158/1538-7445.AM2017-1745

Abstract 2813: An interactome analysis for personalized chemotherapy using PDX/NOG models of non-small cell lung cancer

Abstract Lung cancer is one of the most common malignant diseases in the world, and its prognosis is generally poor. It is crucial to elucidate the biological mechanisms underlying lung cancer and accelerate the development of new treatment strategies. Xenografts derived from engrafting fresh surgical specimens directly into immunodeficient mice have recently enabled the development of more relevant in vivo models of human cancers. These patient-derived xenograft (PDX) models, established by the direct transfer of tumor tissue, retain similar morphologies, heterogeneities, and molecular signatures as the original cancers, and, thus, may be used in promising personalized medicine for cancer. We previously reported the rapid and efficient establishment of PDXs using super immunodeficient NOG mice (PDX/NOG model). In the present study, we analyzed the gene expression and cancer-stroma interaction profiles of PDXs established from NSCLC patients. We also discussed the possibility of individual PDX/NOG model simulations for personalized cancer chemotherapy. Ten NSCLC lines of PDX/NOG (Age 43-78 years; 8 men and 2 women; 7 adenocarcinoma, 1 adenosquamous carcinoma, 1 squamous carcinoma, and 1 large cell carcinoma) were presented. In these cases, clinical information regarding chemotherapy for donor patients was retrieved where possible. Genome sequencing and comprehensive analyses of tumor-stroma interactions (CAncer-STromal INteractome analysis, CASTIN) were performed on mRNA. CASTIN showed tumor-stroma interactions in PDX/NOG comprehensively and quantitatively at the gene expression level by distinguishing gene arrangements in human tissue (Cancer) from mouse tissue (Stroma) as signal strengths (ligand dependency (%) / receptor dependency (%)). Interactions of EGF(cancer)-EGFR(stroma) were observed from 1.2 (15%/0%) to 11.5 (94%/80%) as widely-distributed. Interactions of VEGFA(cancer)-KDR(stroma) were observed from 49.1 (31%/51%) to 301.7 (83%/100%) as closely-distributed. These interactions of EGF-EGFR or VEGFA-KDR in PDX/NOG closely reflected the clinical effectiveness of an EGFR inhibitor (Cetuximab) or VEGF-A inhibitor (Bevacizumab). The CASTIN results of PDX/NOG appear to be reliable for clinical simulations of chemotherapy and will definitely assist in the selection of the most sensitive anti-cancer drug. The fast and efficient establishment of individual PDXs will contribute to personalized anti-cancer therapies. Citation Format: Tsuyoshi Chijiwa, Daisuke Komura, Mizuha Haraguchi, Akira Noguchi, Hidemitsu Sato, Hiroaki Ito, Haruhiko Nakayama, Makoto Katayama, Naoki Miyao, Naruaki Matsui, Yuichi Tateishi, Hiroshi Suemizu, Yoshiyasu Nakamura, Daisuke Furukawa, Takayuki Isagawa, Hiroto Katoh, Shumpei Ishikawa, Masato Nakamura, Yohei Miyagi. An interactome analysis for personalized chemotherapy using PDX/NOG models of non-small cell lung cancer [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 2813. doi:10.1158/1538-7445.AM2017-2813

Colorectal Cancer Cell Line Proteomes Are Representative of Primary Tumors and Predict Drug Sensitivity

Proteomics holds promise for individualizing cancer treatment. We analyzed to what extent the proteomic landscape of human colorectal cancer (CRC) is maintained in established CRC cell lines and the utility of proteomics for predicting therapeutic responses. Proteomic and transcriptomic analyses were performed on 44 CRC cell lines, compared against primary CRCs (n=95) and normal tissues (n=60), and integrated with genomic and drug sensitivity data. Cell lines mirrored the proteomic aberrations of primary tumors, in particular for intrinsic programs. Tumor relationships of protein expression with DNA copy number aberrations and signatures of post-transcriptional regulation were recapitulated in cell lines. The 5 proteomic subtypes previously identified in tumors were represented among cell lines. Nonetheless, systematic differences between cell line and tumor proteomes were apparent, attributable to stroma, extrinsic signaling, and growth conditions. Contribution of tumor stroma obscured signatures of DNA mismatch repair identified in cell lines with a hypermutation phenotype. Global proteomic data showed improved utility for predicting both known drug-target relationships and overall drug sensitivity as compared with genomic or transcriptomic measurements. Inhibition of targetable proteins associated with drug responses further identified corresponding synergistic or antagonistic drug combinations. Our data provide evidence for CRC proteomic subtype-specific drug responses. Proteomes of established CRC cell line are representative of primary tumors. Proteomic data tend to exhibit improved prediction of drug sensitivity as compared with genomic and transcriptomic profiles. Our integrative proteogenomic analysis highlights the potential of proteome profiling to inform personalized cancer medicine.

Determining risk of severe gastrointestinal toxicity based on pretreatment gut microbial community in patients receiving cancer treatment: a new predictive strategy in the quest for personalized cancer medicine.

Currently, our ability to accurately predict a patient's risk of developing severe gastrointestinal toxicity from their cancer treatment is limited. Risk stratification continues to rely on traditional patient-related and treatment-related factors including age, ethnicity, sex, comorbidities, genetics, agent, dose and schedule. Although informative, these crude measures continue to underestimate toxicity risk, and hence alternative methods of risk prediction must be investigated. Given the increasing focus on the gut microbiome in driving disease, this review will provide an overview of the current literature proposing the gut microbiome as a novel predictive tool for treatment-induced gastrointestinal toxicity. Predictive gut microbial phenotypes have been identified for gastrointestinal toxicity induced by radiation and the checkpoint blocker, Ipilimumab. Each study employed slightly different methods of gut microbiome assessment; however, in all cases, separation of toxic versus nontoxic patients was achieved. No studies have investigated chemotherapy-induced gastrointestinal toxicity. The gut microbiome offers an exciting new method of risk stratification for gastrointestinal toxicity. This would enable identification of high-risk patients prior to treatment, enabling tailored treatment regimens based on personalized risk assessment and the proactive provision of supportive care measures. Based on the plasticity of the gut microbiome, methods of risk mitigation may be investigated.

WINTHER: An international study to select rational therapeutics based on the analysis of matched tumor and normal biopsies in subjects with advanced malignancies.

TPS11625 Background: Today, personalized cancer medicine implies matching the patient’s tumor genomic characteristics with molecularly and immune targeted agents. Although there are an increasing number of DNA aberrations that can now be matched to a cognate therapy, some patients do not display such druggable oncogene drivers. Methods: WINTHER is an open non-randomized study involving 6 cancer centers in France, Spain, Israel, Canada and USA applying genomic and also transcriptomic assays to guide treatment decisions. The novelty of the WINTHER approach lies in the use of tumor and matched normal tissue biopsies together and an algorithm for predicting efficacy of therapies. The aim is to provide a rational therapeutic choice for all of the patients enrolled in the study whether or not they harbor actionable DNA alterations. The study endpoint is the comparison of the progression-free-survival (PFS) under the WINTHER selected therapy to the PFS of the last therapeutic line. Patients included have refractory metastatic cancer of any histological type, with at least one prior therapeutic regimen and performance status of 0 to 1. Patients who have received a matched treatment based on a molecular anomaly as their immediate prior therapy were excluded. After consent, patients undergo a tumor and histologically-matched normal tissue biopsy. Extracted DNA and RNA of both tumor and normal from frozen tissues at the local center under common standard operating procedures are sent to centralized laboratories for omics investigations. DNA is investigated at Foundation Medicine Inc. and RNA at Gustave Roussy using Agilent technology. For RNA, the WINTHER algorithm is applied on the differential RNA expression data between tumor and normal tissues and establishes the list of drugs with the presumed higher score of efficacy for each patient. Patients with actionable genomic events enter in ARM A, and patients without any druggable anomaly of the DNA enter in ARM B and are treated using the WINTHER algorithm RNA-based treatment decision tool. To date, the trial has recruited 303 patients. Clinical trial information: NCT01856296.

FoundationOne as a relevant tool for comprehensive genomic profiling and assessment of tumor mutation burden in the era of precision oncology in India.

e23096 Background: Comprehensive genomic profiling (CGP) is gaining acceptability globally, but clinical experience in developing countries like India is limited. CGP identifies genomic alterations (GA), with tumor mutation burden (TMB) and microsatellite status (MSI), revealing therapeutic options such as targeted inhibitors and immunotherapies. We sought to evaluate the mutation frequency and actionability across tumors. Methods: Metastatic and/or refractory patients (referred to Personalized Cancer Medicine Clinic) underwent CGP analysis, including calculation of TMB and MSI, using a targeted NGS panel (FoundationOne, 53 samples; FoundationOne Heme, 4 samples). This panel detects all relevant classes of GA: base substitutions, small indels, rearrangements and copy number changes. Mutation frequencies were compared with the larger Foundation database. TMB status was reported as low (≤5 mutations/Mb), intermediate (6-19 mut/Mb) or high (≥20 mut/Mb). Results: The most common tumor types were lung (23%), breast (14%) and sarcoma (12%); other tumor types, including unknown primary constituted the rest (51%). Most samples were from metastatic sites (60%). Oncogenic GA were found in 131 genes across all tumor subtypes and affected major pathways: apoptosis/cell cycle (31%), PI3K (14%), transcriptional regulation (13%), and receptor tyrosine kinases (10%). Among these GA, 38 were considered actionable and were distributed across 43 (75%) samples. Therapies with FDA approval for the tumor type analyzed were indicated for 18 samples; an additional 25 samples had GA associated with therapies FDA approved for another indication. More than 1 actionable GA was identified in 24/43 (56%). TMB status was low in 36 (63%), intermediate in 19 (33%) and high in 2 (3.5%). High TMB status correlated with high MSI status (p &lt; 0.001). Trend observed in the mutation frequency was comparable with the larger Foundation database. Conclusions: This is the first study in India showing CGP identified actionable targets associated with FDA approved therapies in approx. 32% of cases. TMB status identified 2/57 samples with high mutation burden for whom immunotherapy might be relevant.

Good Laboratory Standards for Clinical Next-Generation Sequencing Cancer Panel Tests.

Next-generation sequencing (NGS) has recently emerged as an essential component of personalized cancer medicine due to its high throughput and low per-base cost. However, no sufficient guidelines for implementing NGS as a clinical molecular pathology test are established in Korea. To ensure clinical grade quality without inhibiting adoption of NGS, a taskforce team assembled by the Korean Society of Pathologists developed laboratory guidelines for NGS cancer panel testing procedures and requirements for clinical implementation of NGS. This consensus standard proposal consists of two parts: laboratory guidelines and requirements for clinical NGS laboratories. The laboratory guidelines part addressed several important issues across multistep NGS cancer panel tests including choice of gene panel and platform, sample handling, nucleic acid management, sample identity tracking, library preparation, sequencing, analysis and reporting. Requirements for clinical NGS tests were summarized in terms of documentation, validation, quality management, and other required written policies. Together with appropriate pathologist training and international laboratory standards, these laboratory standards would help molecular pathology laboratories to successfully implement NGS cancer panel tests in clinic. In this way, the oncology community would be able to help patients to benefit more from personalized cancer medicine.

Harnessing Preclinical Molecular Imaging to Inform Advances in Personalized Cancer Medicine.

Comprehensive molecular analysis of individual tumors provides great potential for personalized cancer therapy. However, the presence of a particular genetic alteration is often insufficient to predict therapeutic efficacy. Drugs with distinct mechanisms of action can affect the biology of tumors in specific and unique ways. Therefore, assays that can measure drug-induced perturbations of defined functional tumor properties can be highly complementary to genomic analysis. PET provides the capacity to noninvasively measure the dynamics of various tumor biologic processes in vivo. Here, we review the underlying biochemical and biologic basis for a variety of PET tracers and how they may be used to better optimize cancer therapy.

Abstract OT3-03-03: Challenges faced across borders to open European academic multicentre projects: The ET-FES program part of the ERA-Net TRANSCAN JTC 2011

Abstract Background The activation of international non-profit clinical trials funded by the European Commission (EC) is challenging given the cross-borders regulations and the need to follow specific timelines according to EC rules. We report here the logistic procedures and challenges faced by 4 academic centres from 4 different countries from the European Union (EU) for the activation of such program in metastatic breast cancer (MBC). Materials and Methods The primary objective of the ET-FES program is to validate the use of a new radiotracer 18F Fluoroestradiol (FES), targeting estrogen receptors, as a tool to better predict endocrine responsiveness in MBC, with PET/CT. The trial is sponsored by EO Galliera (Genoa, Italy) and brings together Italy, Spain, France and Germany. In 10/2012, ET-FES was approved for funding from EC under the Seventh Framework Programme after the first ERA-NET TRANSCAN Joint Transnational Call (JTC) for Proposals (2011) on validation of new biomarkers for personalized cancer medicine. Results The official start of the program was set up on 06/2013 by the Italian Ministry of Health. Time to institutional review board and to competent authority (CA) approvals were 1.5 and 11, 2 and 5, 3 and 16, and 13 and 14+ months in Italy, France, Spain and Germany respectively. Overall, no blocking comment was raised by the ethical committees; only minor clinical and methodological issues were raised in Germany and Spain. Issues from CA were raised in all countries except France (12, 21 and 23 queries in Italy, Spain and Germany respectively), on quality aspects of 18F-FES investigational medicinal product dossier. At the sponsor level, time to final agreement signature with Advanced Accelerator Applications, the 18F-FES manufacturing company, required 13 months. First patient could be enrolled in Italy 14 and 22 months after ethical committee approval and after the official start of the ET-FES project respectively. Conclusions As of May 2016, of 310 patients expected, only 28 have been enrolled from Italy. From a regulatory viewpoint and acknowledging that 18F-FES does not have yet any marketing approval in the EU, the ET-FES program approval process was timely completed at the ethical committee level in Italy, France and Spain. Time to CA approval varied across countries and was timely achieved only in France, due to requirements varying from CA to CA, stressing the serious lack of harmonized procedures although intended by the 2001/20/EC directive. Regarding sponsor's responsibilities (i.e. Italy), the administrative procedures needed to activate this type of EU projects remain critical, appealing for more tolerant time span in order to satisfy all the legal aspects on contracts by public bodies, according to national rules and laws. One needs to be very conscious of these timelines when applying to EU/EC calls, especially when the time allowed for the conduction of research is limited (3 years here), unless jeopardizing the entire multicentre and multinational effort. Citation Format: Brain E, Corradengo D, Oriana N, Piccardo A, Matteucci F, Cortes J, Harbeck N, Würstlein R, Piris A, Alberini J-L, Merlo DF, Degenhardt T, Turbiez I, Madar O, Monti M, Cesaro A, Rivitti E, Rollandi GA, Iacozzi M, Campazzi E, Campora S, Camporese D, Gennari A. Challenges faced across borders to open European academic multicentre projects: The ET-FES program part of the ERA-Net TRANSCAN JTC 2011 [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 OT3-03-03.

Characterizing CD137 upregulation on NK cells in patients receiving monoclonal antibody therapy

In the era of personalized cancer medicine, identifying techniques for effectively matching patients to efficacious treatments is a critical step in the treatment process. The advent of anti-cancer immunotherapies necessitates novel approaches to biomarker identification beyond traditional genomic profiling. One promising approach is incorporation of nomograms into treatment decisions. Nomograms are prediction tools, based on statistical modeling, designed to predict treatment outcomes. As a first step toward developing a nomogram, we conducted analyses to predict CD137 expression of natural killer cells after monoclonal antibody (mAb) treatment. Patient, tumor and immune characteristics were collected from 199 patients with breast cancer (N = 62), head/neck cancers (N = 46) and non-Hodgkin's lymphoma (NHL) (N = 91), who were receiving mAb therapy as part of clinical trials. The difference in CD137 expression before and after mAb therapy was assessed by flow cytometry. To evaluate those who respond to mAb therapy via increased CD137 expression, we applied classification and regression trees (CART), multivariable lasso regression tools and Random Forest. The CD137 expression was significantly different for each cancer type [mean (SD): Breast: 6.6 (6.5); Head/Neck: 11.0 (7.0); NHL: 7.5 (7.1), P < 0.0001]. For breast cancer and NHL, FcR polymorphism and baseline CD137 expression were significant predictors of increased CD137 expression; for head/neck cancer, FcR polymorphism and age were significant predictors of increased expression. Our preliminary results suggest that FcR polymorphism, pre-treatment CD137 expression and age are significant predictors of CD137 upregulation in patients. This study demonstrates that the development of a nomogram for therapy response is feasible. Further work validating our models in an independent cohort will provide the next steps in developing a nomogram that may be used to individualize this therapeutic approach for patients (NCT01114256).

Cancer Immunotherapy and Personalized Medicine: Emerging Technologies and Biomarker-Based Approaches.

Purpose of reviewThe vision and strategy for the 21st century treatment of cancer calls for a personalized approach in which therapy selection is designed for each individual patient. While genomics has led the field of personalized cancer medicine over the past several decades by connecting patient-specific DNA mutations with kinase-targeted drugs, the recent discovery that tumors evade immune surveillance has created unique challenges to personalize cancer immunotherapy. In this mini-review we will discuss how personalized medicine has evolved recently to accommodate the emerging era of cancer immunotherapy. Moreover, we will discuss novel platform technologies that have been engineered to address some of the persisting limitations.Recent findingBeginning with early evidence in personalized medicine, we discuss how biomarker-driven approaches to predict clinical success have evolved to account for the heterogeneous tumor ecosystem. In the emerging field of cancer immunotherapy, this challenge requires the use of a novel set of tools, distinct from the classic approach of next-generation genomic sequencing-based strategies. We will introduce new techniques that seek to tailor immunotherapy by re-programming patient-autologous T-cells, and new technologies that are emerging to predict clinical efficacy by mapping infiltration of lymphocytes, and harnessing fully humanized platforms that reconstruct and interrogate immune checkpoint blockade, ex-vivo.SummaryWhile cancer immunotherapy is now leading to durable outcomes in difficult-to-treat cancers, success is highly variable. Developing novel approaches to study cancer immunotherapy, personalize treatment to each patient, and achieve greater outcomes is penultimate to developing sustainable cures in the future. Numerous techniques are now emerging to help guide treatment decisions, which go beyond simple biomarker-driven strategies, and are now we are seeking to interrogate the entirety of the dynamic tumor ecosystem.

Is oral squamous cell carcinoma unique in terms of intra- and inter-tumoral heterogeneity?

Tumoral heterogeneity has become one of the important issues in cancer research and carries outmost weightage in designing personalized cancer medicine. Oral squamous cell carcinoma (OSCC) is present in unique microenvironment of oral cavity attributed to site-specific diversified tissue compositions, microbial flora, diversified carcinogenic attacks, potentially malignant disorders, epithelial turnover rate, saliva, and closeness to external environment. Such microenvironment has potential to modulate the pathogenesis of many oral pathologies including OSCC. We believe that such factors and microenvironment are unique for OSCC and cannot be found in other carcinomas of the body. Methodologies such as multiregional sequencing, single-cell genomics, genomic landscape of bulk tumor sequencing, and so on, will help in understanding OSCC uniqueness in terms of heterogeneity. If explored, it will enable us to know the attributes of heterogeneity and their magnitude in OSCC, which will benefit in designing predictable biomarker and effective therapeutic strategy.

Genomic Data for Personalized Cancer Medicine

Genomic Data for Personalized Cancer Medicine

Highlights of This Issue

BRAF and MEK inhibitors have improved the clinical management of BRAF-mutant melanoma; however, patients often develop drug resistance. MITF is a transcription factor required for melanoma formation. Rose and colleagues demonstrate that clinically approved MAPK inhibitors cause the MITF-dependent upregulation of GPNMB. GPNMB is the target of an antibody-drug conjugate (CDX-011) currently in clinical trials. Combining BRAF and/or MEK inhibitors with CDX-011 enhanced melanoma tumor regression and delayed emergence of acquired resistance when compared with treatment with MAPK pathway inhibitors alone. These findings support the initiation of clinical trials to assess the efficacy of MAPK pathway inhibitors combined with CDX-011 for the treatment of melanoma.Osswald and colleagues develop an in vivo two-photon microscopy approach to study blood brain barrier (BBB) permeability properties, tumor growth kinetics, and response to PI3K/mTOR inhibitors in metastatic melanoma of the brain. A BBB permeable inhibitor, but not an impermeable inhibitor, reduced the growth rate of impermeable metastases. These results provide evidence that nonpermeable brain micro- and macrometastases can effectively be targeted by drugs designed to cross the BBB.Patient-derived xenograft (PDX) models have been proved to be effective in prioritizing therapeutic regimens in personalized cancer medicine; however, establishing and testing transplantable tumorgrafts are expensive and time-consuming. In this study, Roife and colleagues developed an ex vivo drug efficacy testing method, Live Tissue Sensitivity Assay (LTSA), for patients with pancreatic ductal adenocarcionma. The test enables efficient testing of PDX tumor in 96-well plates against standard of care chemotherapeutic agents in 2 to 3 days. This system is highly reflective of patients' responses to conventional therapeutic agents and has significant potential for guiding personalized cancer treatment design.There is a pressing need for biomarkers to assist with the clinical management of brain tumor patients. The majority of WHO grade II–III gliomas express mutant forms of Isocitrate Dehydrogenases (IDH) 1 and 2, resulting in excessive production of D-2-Hydroxyglutarate (D-2HG) in the tumor cells. Kalinina and colleagues examined whether this oncometabolite is shed into the cerebrospinal fluid (CSF) and if it can potentially serve as a biomarker. Using mass spectrometry the authors demonstrated high levels of D-2HG in the CSF of IDH1-mutant glioma patients, suggesting that this assay may be developed into a clinical test for diagnosis, prognosis, and follow-up of glioma patients.

A Mixed Type 1 and Type 2 Kinase Inhibitor That Overrides FLT3 F691 and D835 Resistance Mutations

Introduction: Mutations in FLT3, a class III receptor tyrosine kinase predominantly expressed on hematopoietic progenitor cells, represent the most common genetic alteration in patients with acute myeloid leukemia (AML). Approximately 25% of patients with AML harbor internal tandem duplications (ITD) within the juxtamembrane domain of FLT3. Quizartinib, the first selective FLT3 tyrosine kinase inhibitor (TKI) to show clinical activity, frequently achieves a composite complete remissions (CRc) rate of approximately 50% in relapsed and refractory FLT3-ITD+AML patients. However, patient samples obtained at the time of relapse with quizartinib revealed resistance-conferring FLT3 mutations at two kinase domain residues (F691L and D835V/Y/F). The crystal structure of the FLT3-quizartinib complex indicates that quizartinib binding relies on essential edge-to-face aromatic interactions with the gatekeeper F691 residue, and with F830 within the DFG motif in the activation loop (A-loop). This reliance makes quizartinib responses critically vulnerable to gatekeeper (F691) and A-loop (D835) substitutions. This has led to the identification of TKIs that have activity against quizartinib-resistant FLT3 mutants, including type 1 inhibitors (e.g. crenolanib, ASP2215) targeting specifically D835 mutants, and type 2 inhibitors (e.g. pexidartinib, ponatinib) active against F691L. Until now, no FLT3 inhibitor has demonstrated potent inhibition of both classes of resistance mutations.Methods: To identify FLT3 inhibitors with activity against both gatekeeper and A-loop mutants, we used a modular approach to systematically alter the constituents of quizartinib to optimize interactions with the mutant proteins guided by X-ray structures and modeling. The compounds were profiled through a panel of in vitro growth assays and in vivo studies using Ba/F3 cells engineered to express FLT3-ITD/F691L and FLT3-ITD/D835Y, and MOLM14 quizartinib-resistant clones (MOLM14/F691L and MOLM14/D835Y) that have acquired the same mutations after chronic exposure to quizartinib. When injected into the tail vein of nude mice, the engineered Ba/F3 cells home to the spleen and proliferate, causing marked splenomegaly, whereas MOLM14 parental cells and MOLM14 resistant clones significantly reduce the overall survival of mice. Ba/F3-induced splenomegaly is directly dependent on the activity of the expressed mutant FLT3 and inhibition of splenomegaly measures target engagement in vivo. Analysis of efficacy in the MOLM14 survival models was done by Kaplan-Meier estimates and the logrank statistic.Results: Structure-guided lead optimization generated a series of FLT3 inhibitors with a mixed type 1 and type 2 binding mode and equal potency against the two classes of resistant mutants. Further optimization to improve pharmaceutical properties and maximize in vivo efficacy led to development candidate PLX3623. PLX3623 potently inhibited the growth of two Ba/F3 cell lines expressing F691L and D835Y mutants, with IC50 values of 0.18 (±0.04) and 0.15 (±0.03) nM, respectively. PLX3623 had a comparable antiproliferative effect on the MOLM14 resistant clones, with IC50values of 0.52 (±0.14) and 0.26 (±0.08) nM for the F691L and D835Y clones, respectively. In vivo PLX3623 treatment resulted in dose-dependent reduction in Ba/F3-induced splenomegaly, achieving >90% inhibition at 3 mg/kg dose. PLX3623 at 3mg/kg increased the length of survival of MOLM14 parental cell recipients by 120%, compared to the 80% increase afforded by 3 mg/kg quizartinib. At 3mg/kg dose, PLX3623 extended the survival of mice that received the two MOLM14 resistant clones by 80%; quizartinib at the same dose provided no survival benefit.Conclusion: Clonal evolution and tumor heterogeneity present a major challenge to targeted therapy and personalized cancer medicine. While acquired mutations individually may be actionable targets, often multiple inhibitors are required to combat the clonal diversity in a refractory setting. PLX3623, a FLT3 inhibitor with a mixed type 1 and type 2 binding mode, is the first TKI that is active against mutations conferring resistance through independent structural mechanisms. The in vitro and in vivo potency and favorable nonclinical safety profile of PLX3623 support its clinical development as a potentially best-in-class therapy to overcome clinical resistance to first generation FLT3 inhibitors. DisclosuresLin:Plexxikon Inc.: Employment. Zhang:Plexxikon Inc.: Employment. Matusow:Plexxikon Inc.: Employment. Mumy:Plexxikon Inc.: Employment. Tsang:Plexxikon Inc.: Employment. Zhang:Plexxikon Inc.: Employment. Powers:Plexxikon Inc.: Employment. Spevak:Plexxikon Inc.: Employment. Severson:Plexxikon Inc.: Employment. Tsai:Plexxikon Inc.: Employment. Bollag:Plexxikon Inc.: Employment. Zhang:Plexxikon Inc.: Employment.

Evaluation of a tyrosine kinase peptide microarray for tyrosine kinase inhibitor therapy selection in cancer.

Personalized cancer medicine aims to accurately predict the response of individual patients to targeted therapies, including tyrosine kinase inhibitors (TKIs). Clinical implementation of this concept requires a robust selection tool. Here, using both cancer cell lines and tumor tissue from patients, we evaluated a high-throughput tyrosine kinase peptide substrate array to determine its readiness as a selection tool for TKI therapy. We found linearly increasing phosphorylation signal intensities of peptides representing kinase activity along the kinetic curve of the assay with 7.5–10 μg of lysate protein and up to 400 μM adenosine triphosphate (ATP). Basal kinase activity profiles were reproducible with intra- and inter-experiment coefficients of variation of <15% and <20%, respectively. Evaluation of 14 tumor cell lines and tissues showed similar consistently high phosphorylated peptides in their basal profiles. Incubation of four patient-derived tumor lysates with the TKIs dasatinib, sunitinib, sorafenib and erlotinib primarily caused inhibition of substrates that were highly phosphorylated in the basal profile analyses. Using recombinant Src and Axl kinase, relative substrate specificity was demonstrated for a subset of peptides, as their phosphorylation was reverted by co-incubation with a specific inhibitor. In conclusion, we demonstrated robust technical specifications of this high-throughput tyrosine kinase peptide microarray. These features required as little as 5–7 μg of protein per sample, facilitating clinical implementation as a TKI selection tool. However, currently available peptide substrates can benefit from an enhancement of the differential potential for complex samples such as tumor lysates. We propose that mass spectrometry-based phosphoproteomics may provide such an enhancement by identifying more discriminative peptides.

Theranostic barcoded nanoparticles for personalized cancer medicine.

Personalized medicine promises to revolutionize cancer therapy by matching the most effective treatment to the individual patient. Using a nanoparticle-based system, we predict the therapeutic potency of anticancer medicines in a personalized manner. We carry out the diagnostic stage through a multidrug screen performed inside the tumour, extracting drug activity information with single cell sensitivity. By using 100 nm liposomes, loaded with various cancer drugs and corresponding synthetic DNA barcodes, we find a correlation between the cell viability and the drug it was exposed to, according to the matching barcodes. Based on this screen, we devise a treatment protocol for mice bearing triple-negative breast-cancer tumours, and its results confirm the diagnostic prediction. We show that the use of nanotechnology in cancer care is effective for generating personalized treatment protocols.

Culture and Drug Profiling of Patient Derived Malignant Pleural Effusions for Personalized Cancer Medicine.

IntroductionThe use of patients’ own cancer cells for in vitro selection of the most promising treatment is an attractive concept in personalized medicine. Human carcinoma cells from malignant pleural effusions (MPEs) are suited for this purpose since they have already adapted to the liquid environment in the patient and do not depend on a stromal cell compartment. Aim of this study was to develop a systematic approach for the in-vitro culture of MPEs to analyze the effect of chemotherapeutic as well as targeted drugs.MethodsMPEs from patients with solid tumors were selected for this study. After morphological and molecular characterization, they were cultured in medium supplemented with patient-derived sterile-filtered effusion supernatant. Growth characteristics were monitored in real-time using the xCELLigence system. MPEs were treated with a targeted therapeutic (erlotinib) according to the mutational status or chemotherapeutics based on the recommendation of the oncologists.ResultsWe have established a robust system for the ex-vivo culture of MPEs and the application of drug tests in-vitro. The use of an antibody based magnetic cell separation system for epithelial cells before culture allowed treatment of effusions with only moderate tumor cell proportion. Experiments using drugs and drug-combinations revealed dose-dependent and specific growth inhibitory effects of targeted drugs.ConclusionsWe developed a new approach for the ex-vivo culture of MPEs and the application of drug tests in-vitro using real-time measuring of cell growth, which precisely reproduced the effect of clinically established treatments by standard chemotherapy and targeted drugs. This sets the stage for future studies testing agents against specific targets from genomic profiling of metastatic tumor cells and multiple drug-combinations in a personalized manner.