Prognostic, Predictive, and POC: Biomarkers from Research to Clinic Agenda

Conference Registration, Coffee, and Breakfast Pastries

Molecular Approaches to Biomarker Discovery for Immune Therapies in Oncology
Terri McClanahan, Director, Profiling & Expression, Merck Biologics Discovery, United States of America

Toward a deeper understanding of the molecular features characterizing tumors which respond to immune therapeutic intervention, we have utilized a broad translational platform including both human tissue analysis and immune competent mouse tumor models.  Molecular characterization of mouse syngeneic tumor models which respond to several immunotherapies, including anti-PD-1 and anti-GITR, has allowed us to formulate specific hypotheses that we can test in human tumor samples, to better understand the molecular milieu of a tumor that will respond to immune checkpoint interventional therapies.  Human tumor analyses include profiling of patient tumor biopsies from clinical trials with Anti-PD-1, Pembrolizumab, and correlation to clinical response.   In addition, orthogonal approaches to characterize tumor infiltrating lymphocytes by immunohistochemistry, flow cytometry, sorting and molecular profiling and have been applied to a wide variety of tumor types in order to characterize the underlying molecular features that may predict response to immune therapies.

Coffee Break and Networking

The Use of Biomarkers and Imaging in Early Phase Cardiovascular Studies Prior to Large Outcomes Trials
David Kallend, Vice President, Global Medical Director, The Medicines Company, United States of America

New targets for the treatment of cardiovascular disease have been assessed with pharmacotherapy for several decades. The most successful have undoubtedly been the HMG-CoA reductase inhibitors or statins which lower total cholesterol, LDL-cholesterol and triglyceride and raise HDL-cholesterol. These changes in biomarkers have been associated with significant improvements in cardiovascular mortality and morbidity.   Other lipid targets have been less successful in terms of improvements in cardiovascular outcomes despite significant changes in these biomarkers. Unless a new therapy is a statin, approval is usually dependent on outcomes data which involves a lengthy and costly development. Retrospectively, once outcome benefit is shown for a new therapy then the biomarker may be validated, and may even help other therapies in the class gain a more rapid approval. Imaging modalities have also been used to assess new therapies and these may eventually prove to be a better surrogate than plasma biomarkers, although LDL-C is still the gold standard and may be as reliable as any other biomarker. Currently, these biomarkers can only be used prospectively when making decisions regarding the conduct of large Phase III outcomes trials.  Until these surrogate biomarkers are clearly linked to improvements in outcomes and therefore validated by the regulatory authorities globally as a route to approval of new lipid modifying therapies, the development of new therapies for the majority of lipid targets will remain a lengthy and costly process.

Lunch and Networking

Salivary Epigenetic Biomarkers To Detect Head and Neck Cancers
Chamindie Punyadeera, Associate Professor, Institute of Health and Biomedical Innovation, Queensland University of Technology, Australia

Head and neck squamous cell carcinoma (HNSCC) encompasses a diverse group of aggressive tumors and is the 6th most common cancer with 650,000 new cases per annum worldwide. HNSCC patients, particularly those with a history of smoking, often develop secondary tumors. Currently, there are no diagnostic tests to detect these cancers at an early stage, as such, most patients present with metastatic disease at the time of diagnosis, leading to 5-year survival  of <60%. With an increasing recognition of the link between oral and systemic disease, attention has turned to saliva as an alternative diagnostic medium. Saliva collection is non-invasive, easy sampling with multiple sampling opportunities and is ideal for 3rd world countries. It is well established that tumor cells secrete biomolecules into the saliva. DNA promoter methylation plays a key role in the transcriptional silencing of these tumor suppressor genes, allowing continuous proliferation and subsequent fabrication of neoplasm. We collected resting saliva from HNSCC patients (n=31) and healthy controls (n=20) and interrogated methylation in RASSF1a, TIMP-3, p16, MED-15, two methylation sites using a sensitive methylation-specific PCR (MSP) assay. For the 5 panel, using logistic regression, an area under the curve (AUC) of 0.96 with a sensitivity of 90% and specificity of 90%. We demonstrate that this salivary panel is clinically useful in detecting HNSCC.

Targeting Cancer Stem Cells Using avß3 Integrin as Oncology Biomarker
David Cheresh, Distinguished Professor of Pathology , University of California-San Diego, United States of America

Tumor cells, with stem-like properties, are highly aggressive and often display drug resistance. Here, we reveal that integrin avß3 serves as a marker of breast, lung and pancreatic carcinomas with stem-like properties that are highly resistant to receptor tyrosine kinase inhibitors such as erlotinib. This was observed in vitro and in mice bearing patient-derived tumor xenografts or in clinical specimens from lung cancer patients that had progressed on erlotinib.   Mechanistically, avß3, in the unligated state, recruits KRAS and RalB to the tumor cell plasma membrane, leading to the activation of TBK-1/NFkB.  In fact, avß3 expression and the resulting KRAS/RalB/NFkB pathway were both necessary and sufficient to promote tumor initiation, anchorage-independence, self-renewal, and erlotinib resistance.  Pharmacological targeting this pathway with Bortezomib reversed both tumor stemness and erlotinib resistance. These findings not only identify a?ß3 as a marker/driver of carcinoma stemness but they reveal a therapeutic strategy to sensitize such tumors to RTK inhibition.

Liquid Biopsies in the Era of Precision Medicine
Filip Janku, Associate Professor, Center Medical Director, Clinical and Translational Research Center, The University of Texas MD Anderson Cancer Center, United States of America

Molecular profiling is usually performed on DNA from archival tumor samples obtained during routine therapeutic or diagnostic procedures, which do not necessarily reflect the fact that the molecular profile might change over time. The plasticity of molecular profiles includes emergence of secondary molecular aberrations, which can ultimately drive resistance to treatment. To detect such changes new techniques capable of monitoring molecular profile in the real time need to be implemented. Since sequential or multiple biopsies can hardly be used in routine clinical care because of logistical, financial and ethical barriers there is an unmet for developing new, noninvasive techniques for garnering the greatest amount of information at multiple time points from the least amount of available biologic material. Cell-free (cf) DNA is released to the circulation from cells undergoing apoptosis, necroptosis and active secretion and can be isolated from plasma, serum or urine samples of patients with advanced cancers. In addition, unlike tissue biopsies, obtaining samples of cfDNA (liquid biopsies) is a noninvasive approach, which can be used at multiple time points, with less risk to patients at a lower cost.  Molecular profiling of cfDNA offers an attractive approach for molecular diagnosis, monitoring of treatment effects and detection of molecular mechanisms of resistance.

Deciphering Signaling Networks and Mining for Biomarkers in Genetically Engineered Models of Cancer
Serguei Kozlov, Leidos Biomedical Research, Inc. Principal Scientist, Center for Advancer Preclinical Research, NIH, United States of America

In a quest of improved, more clinically relevant in vivo testing systems mimicking cancer disease, genetically and biologically engineered murine (GEM) models are rapidly gaining reputation of a better predictive platform for discovery of prognostic/diagnostic biomarkers and preclinical drug evaluation. NCI Center for Advanced Preclinical Research (NCI-CAPR) – a broad scope translational initiative tasked with establishing scalable capabilities and best practices in preclinical drug development arena – is developing and managing a portfolio of more than a dozen GEM models faithfully reproducing clinical pictures of pancreatic adenocarcinoma, serous epithelial ovarian carcinoma, non-small cell lung cancer, metastatic melanoma, astrocytoma, and other malignancy types. NCI-CAPR has established rigorous workflows to apply these GEM models in a variety of experimental paradigms aimed at mining systems biology in these animals to pinpoint novel mechanistic aspects of disease progression, or identify informative cancer biomarkers. Such studies are designed to either examine the disease progression from initiation to terminal illness in experimental animals left naïve for therapeutic intervention or treated with a spectrum of oncology compounds, including both standard-of-care and experimental candidate therapeutics. A summary of recent and ongoing experiments will be presented to illustrate the concept of cancer GEM applicability for biomarker data mining and therapeutic discovery.

Urinary microRNAs in Early Diabetic Kidney Disease: Measurements, Context and Clinical Prediction
Christos Argyropoulos, Assistant Professor, Division of Nephrology, University of New Mexico, United States of America

Very little information currently exists about urinary microRNA and their association with development of clinical kidney disease in patients with diabetes. This knowledge gap persists despite microRNA’s involvement in the regulation of pathways that are of biological significance in the development of diabetic kidney disease. Filling in the gap may pave the way to novel therapeutics since the current therapeutic paradigm of blockers of the renin-angiotensin-aldosterone axis, reduces risk of renal disease by 30% and many of the large scale trials ran in the last five years fail to translate to novel therapies.  In this talk I  will describe the associations of urinary microRNA signatures associated with the future development of early diabetic kidney disease (diagnosed with the current gold standard of a positive microalbumin test in the urine). A number of microRNAs manifest differential expression changes in the urine of patients who eventually developed kidney damage v.s. those who did not, while some of these changes were even gender specific. A formal statistical rigorous algorithm for the evidence synthesis of microRNA expression changes and their mRNA targets to provide a context for the understanding of these associations will be presented.  In spite of the strength of the associations only a small number of those are plausibly predictive of the future development of kidney damage. Our group thus developed, and internally validated a microRNA “context aware” signature for the early diagnosis of diabetic kidney disease. The implications for clinical practice and drug discovery will also be discussed.

The Pancreas Lives On: Long-Term Persistence of Pancreatic Function in Type 1 Diabetes
Denise Faustman, Director, Immunobiology Laboratory & Associate Professor of Medicine, Massachusetts General Hospital/Harvard Medical School, United States of America

New findings from the Faustman Immunobiology Lab in Boston and from academic institutions in Europe confirm that the pancreases of people with type 1 diabetes continues to produce insulin for decades. Further, persistence of C-peptide appears to correlate with clinical complications and hypoglycemia. Dr. Faustman will present the latest data from large patient studies correlating persistence of C-peptide secretion with diabetic complications. Assays to detect low levels of C-peptide secretion might add value in type 1 diabetes by serving as a new patient-monitoring tool.

Panel Discussion: Biomarker Classes and Diagnostics Development--Pathways to Commercialization
Greg Jones, Director of Business Development and Technical Affairs, MDxHealth SA, United States of America

This panel will attempt to discuss the pathway to commercialization of an oncology biomarker.  Many oncology biomarkers have been identified as potential tools for screening, prognosis and/or companions to a therapeutic approach.  Whether developed for use in a single laboratory or developed as a commercially available kit, the panel will share their insight and experiences with biomarker development, including opportunities in NGS-based diagnostics.

Panel Chairman:  Greg Jones, MDx Health

Panelists:

Baolin Zhang -- US FDA
Terri McClanahan -- Merck & Co.
Shawn Baker -- AllSeq, Inc.
Jan Groen -- MDx Health
Felix Frueh -- Human Longevity, Inc.

Networking Dinner: Enjoy an Awesome Evening and Network with your Fellow Delegates

Close of Day 1 of the Conference

Morning Coffee and Breakfast Pastries

New Technologies Driving Mobile Point-of-Care (POC) Diagnostics
Sridhar Iyengar, Founder/Director, Misfit Wearables, United States of America

This talk will explore some of the new technologies that are being developed that bridge the traditional levels of care found in clinical settings with the convenience of mobile POC  (mPOC) devices. With new mobile and wearable technologies being introduced into the market, there are significant opportunities for deploying new devices that would arguably have higher usage compliance and provide more continuous data for monitoring the health of the user. Some of the challenges are technical (such as developing stable chemistries or reliable image processing), some are related to the relatively uncertain regulatory pathways, and some are at the intersection of user experience and design. This talk will address how to think about overcoming such challenges and will discuss specific case-studies of new technologies that are successfully being brought to the mPOC market in the diabetes and oncology fields.

Coffee Break and Networking

DermoScreen: Smartphone Based Point-of-Care Screening for Skin Cancer
George Zouridakis, Professor, University of Houston, United States of America

Modern smartphones featuring multicore processors, advanced microchips for graphics, high-resolution cameras, and wireless communication can run sophisticated diagnostic algorithms very efficiently. Such devices empower clinicians to make timely and accurate diagnoses at the point of care, and at the same time patients are treated in a more comfortable and less costly setting alternative to hospitalization. We present smartphone-based digital dermoscopy applications for screening melanoma and other skin lesions that are particularly suitable for diverse low-resourced settings.

Lunch and Networking

Targeting Genetic Drivers in Pre-malignancy and Cancer
Scott Lippman, Associate Vice Chancellor, University of California-San Diego, United States of America

The identification of genetic drivers in certain advanced cancers has led to major clinical advances in personalized therapy by targeting these drivers with selective inhibitors (e.g., EGFR and ALK in lung adenocarcinoma). Genetic drivers of cancer progression and metastasis can occur in premalignancy. Cyclin D1 amplification was among the first genetic drivers in head and neck (HN) premalignant lesions and drug targeting of Cyclin D1 produced early promising prevention results. Several retrospective studies suggested that loss of heterozygosity (LOH) at certain loci were genetic drivers in oral premalignant lesions (OPLs). LOH profiles were recently shown to prospectively stratify patients with low-grade oral dysplasia at low risk for progression to oral cancer from those with greater risk. Our recently completed EPOC (Erlotinib Prevention of Oral Cancer) phase III multicenter trial validated specific LOH profiles as genetic drivers. EPOC randomized 150 OPL patients to erlotinib or placebo. While primary endpoint analysis was negative, provocative secondary analyses showed a trend of erlotinib activity during the 12-month therapy (lost after stopping the drug) and that erlotinib-induced grade 2+ skin rash was associated with a highly significant reduction in oral cancer-free survival (vs. patients without rash). EPOC has also provided specimens for targeted genomic studies identifying potential genetic drivers in high-risk OPLs (e.g., PIK3CA). Very recent studies of a) lung carcinogenesis have identified early genetic drivers of squamous cell carcinoma (e.g., 3q26 amplification) and b) HN premalignancy in China have identified oncogenic NOTCH1 driver mutations in OPLs. This presentation will focus on genetic drivers in the context of HN and lung cancer.

Novel Biomarker Development in Down's Syndrome and Other CNS Disorders
William Mobley, Distinguished Professor and Chair, Department of Neurosciences, University of California-San Diego, United States of America

Down syndrome is caused by the presence within the genome of a third copy of chromosome 21.  People with Down syndrome have a number of medical issues that must be confronted.  During recent decades, the problems faced by children with Down syndrome, especially congenital heart disease, have been much more effectively treated.  The result is that average longevity has increased from less than 10 years of age in the early 1900s to more than 60 at this time.  With continuing advances in medical care, longevity is likely to increase further.  While a boon to the wellbeing of people with Down syndrome and their families, the result is that more are now approaching the age at which age-related changes in brain function are ensuing.  Indeed, beyond the problems with cognition experienced by all children with Down syndrome, essentially all of those at age 40 years have the brain pathology characteristic of Alzheimer disease and the majority are demented by age 60.  The remarkable fact is that Down syndrome population thus represents by far the largest in which genetic changes are responsible for Alzheimer disease. We are particularly interested in the biology of Down syndrome, especially as it encompasses Alzheimer disease. Indeed, it is a very real possibility that studies of Alzheimer disease in Down syndrome will elucidate the earliest events in Alzheimer disease in the general population and create opportunities for effective new treatments for this disease.  To define the biology of Alzheimer disease in Down syndrome we need first to elucidate the natural history of brain again in Down syndrome.  We will discuss recent studies in which important new findings regarding natural history have been uncovered.  In addition, we will discuss the possible contributions from additional studies and clinical trials in this important population.

Novel Stromal Biomarkers for Anticancer Therapeutics; Macrophage M1 to M2 Transition; a Dual Epigenetic/PI-3 Kinase Inhibitory Platform
Donald Durden, Pediatric Researcher, University Of California School Of Medicine, United States of America

Macrophages, a major leukocyte population present in tumors, play an essential role in promoting tumor growth by affecting angiogenesis, immune suppression, invasion and metastasis. Despite considerable research efforts, the signal transduction events within macrophages which encode the complex cascade of events required for tumor growth and polarization of macrophages are poorly understood. Our previous findings demonstrate that a4ß1 and avß3 integrin directed cell migration on specific extracellular matrices (ECM) requires a specific kinase-GTPase pair, Syk-Rac2. This pathway was shown to regulate the ECM dependent postnatal angiogenic response with no defect in vascular development. Herein, we have linked this ECM dependent signaling pathway in macrophages to the regulation of M2 macrophage differentiation and to the regulation of tumor growth, invasion and metastasis. Using mouse genetic models, we provide direct evidence that a macrophage specific, a4ß1 integrin dependent Syk-Rac2 signaling axis acts in concert with the p110 isoform (PTEN-PI-3 kinase pathway) to control HIF1a levels, tumor growth and metastasis. We report a novel proteasome/E3 ligase dependent mechanism by which PI-3 kinase regulates HIF1a levels under hypoxic conditions in the stromal compartment. Moreover, treatment with Syk or PI-3 kinase inhibitors demonstrate potent activity in multiple metastatic models. The results define a novel molecular mechanism for the regulation of HIF1a and a macrophage autonomous signaling pathway that is required for alterations in the ECM and the provisional integrin to regulate tumor metastasis and suggest treatment for metastatic disease targeting this pathway in the M2 macrophage compartment. Moreover, we describe a novel chemistry platform scaffold at SignalRx Pharmaceuticals, Inc. from which we have developed a number of potent dual PI-3 kinase/BET brd4 bromodomain inhibitors for cancer therapeutics.

Biodistribution of Antibody-based Drugs Using in vivo Fluorescence Imaging as a Biomarker
Anand Giddabasappa, Principal Scientist, Global Science and Technology, Pfizer, United States of America

Currently, biodistribution of small molecules and biologics are evaluated by PET imaging, autoradiography or pharmacokinetic studies. PET and autoradiography use radiological reagents which need special facilities or a CRO and also have limited shelf life due to radioactive decay.  Conversely, the PK methods are terminal, require large number of animals and evaluation of whole body biodistribution is not easily done due to technical difficulties. Recent developments in optical imaging probes and technologies have given us an opportunity to evaluate biodistribution and targeting of biologic drugs without use of radio-labeled materials. In this presentation I will be discussing the opportunities and challenges of in vivo fluorescence imaging in biodistribution of antibody based drugs.

Use of Isogenic Human iPSCs to Elucidate Redox-mediated Pathways to Parkinson’s, Alzheimer’s, and other Neurodegenerative Disorders
Rajesh Ambasudhan, Research Assistant Professor, Sanford-Burnham Medical Research Institute, United States of America

Dr. Ambasudhan uses human stem cell based "disease-in-a dish" models to study neurodegenerative disorders. Recently, his laboratory developed isogenic patient-derived hiPSC models of Parkinson's disease, to complement transgenic mouse models, and characterized certain redox-mediated protein post-translational modifications that may be an early event in the disease pathogenesis. These studies also led to screens for drugs that can thwart these modifications and may be developed as potential disease-modifying therapies. The recent findings will be presented in the seminar.

Predictive Biomarkers and Point-of-Care Diagnostics for Personalized Medicine
Luke Lee, Director, Global Health Research and Technology (BIGHEART), NUS, United States of America

In this talk, I will present the development of predictive personalized medicine by bridging the gap between bench and bedside: integrated molecular diagnostic systems (iMDx) and integrative microphysiological analysis platforms (iMAPs), which are ideal for predictive and prognostic human biomarkers discovery.  The iMDx is a mobile point-of-care biomarkers detection system: it comprises a self-contained sample preparation from whole blood, multiplexed protein and ultrafast nucleic acid amplification assays on chip with a sample-to-answer readout platform.  The iMAPs is an advanced human organs on chip platform that recapitulates physiologically relevant microenvironments and patient’s clinical conditions for identifying human biomarkers, secretome proteomics, human disease modeling, experimental medicine, drug screening, and toxicology. Additionally, I will also discuss the progress on patient-specific iPSCs-based iMAPs for personalized medicine.  This emerging iMAPs technology has the potential to become an alternative to 2D and 3D cell cultures and animal models.  In summary, I will share my vision for predictive personalized medicine with low-cost precision monitoring and digital recording/analysis of health conditions.

Close of Day 2 of the Conference