Translational medicine paradigm: hip or hype?

The process of developing new therapies took a different turn with the completion of the Human Genome Project, yielding unprecedented information on potential targets. While the tremendous effort of organizing this information in a rationale fashion took off – assisted by systems biology and other disciplines – the impact on drug development is more evident today than ever before. Besides the emergence of molecular targeted therapies, what could be the key aspects of R&D that resulted in the redesign of drug development as a process and placed translational medicine on the map?

There are at least two general paths leading to the creation of new drugs (Fig. 1). In the first scenario one starts with a medical opportunity identified apriori followed by in-depth research of molecular pathophysiology. This results in a database encompassing molecular pathways and putative targets; further, drug discovery or design yields therapeutic candidates that commence the process of drug development from bench to preclinical IND-enabling studies and then clinic. Interestingly, nowadays, our ability to generate investigational drugs starting from druggable targets (by high throughput screening or biomolecule design) outweighs the rate of knowledge-generation in regard to molecular pathogenesis of complex diseases such as cancer. Thus, an alternate scenario emerged: this is initiated by in-depth evaluation of molecular pathways likely to be involved in key physiologic or pathologic processes without precise medical indications set apriori. Generations of a repository of cellular and molecular biology information thus yields putative targets. Via drug design or discovery processes, this results, in turn, in investigational drugs. Evidently, there is a gap between the knowledge base in support of targets, pathways and investigational drugs on one side and the medical applicability on the other side. Addressing that gap is key to the success of the drug development process from appropriately designing the first in man study to increasing the likelihood of success through clinical development or making a strategic ‘no-go’ decision early on, before pivotal, expensive clinical trials are started. If we overlay now the fact that apparently, similar syndromes may have significantly different underlying molecular pathogenesis, this clearly becomes a key challenge in the era of molecular targeted therapies. In simple terms, translational medicine aims to address this gap and guide drug development to where the opportunity is; its role is more and more prominent because of the molecular complexity of diseases we attempt to tackle – such as cancer – and the great prospect of levering molecular pathogenesis information to optimize drug development.

Translational medicine has quite broad, heterogenic technology base and mission, reflected by the diversity of synonyms: experimental or exploratory medicine, translational research or translational sciences. That is primarily because there are several components of translational medicine, deployed in a manner depending on 1) the development stage of an investigational drug; 2) whether the drug candidate is a first in class or follow on product and finally, 3) the specifics of the R&D process in a given organization. Below, we will briefly discuss translational medicine as a tool to optimize drug development, in context of the scenario described in Fig. 1B.

There are two major components of translation with key impact on defining the development strategy and medical opportunity: first, the exploratory component which by definition involves modeling, experimentation using the investigational drug; secondly, the stratified medicine concept built on the modern understanding and application of biomarkers as pivotal helpers of modern drug development. A sound development strategy (what diseases and indications to pursue in clinical trials, what dosing approach, how to position it relative to the standard of care, what to measure and why) needs to factor in significant information complementing data sets in direct support of the investigational drug per se. There are several categories of biomarkers that are being considered in the translational process: biomarkers in direct support of exploratory pharmacology (‘on target’ effect markers), those that parallel the expression of targets and may predict clinical responsiveness (defining inclusion criteria or ‘stratified medicine’ approach), ‘off target’ markers predicting toxicities and markers that are associated with clinical response. There is a differential emphasis on various types of the biomarkers in a manner depending on the development stage. This defines the ‘flavor’ of translational support to a drug development program, but can become a double-edged sword: when used effectively and appropriately, it expedites a program to a go/no-go decision point based on relevant data set and overall, builds value in a program; nevertheless, when used suboptimally, it either delays the program or may waste considerable resources. Thus, stage-appropriate biomarker research and use needs to be considered in order to successfully use the translational concept (Fig. 2).

Biomarkers linked to ‘on-target’ effect become important during non-clinical development stage and are in direct support of exploratory pharmacology. These are markers that show quantitative or qualitative variation when the drug hits the intended target. The most important source for ‘on target’ biomarkers is the in-depth understanding of the mechanism of action of the investigational drug class as it pertains to the signaling pathways or receptors targeted. Lead optimization and selection can be greatly expedited by using ‘on target’ biomarkers, facilitating correlation with biological and clinical outcome. In addition, such ‘on target’ biomarkers can be translated to man and used in the design of early stage clinical trials, as endpoints.

The second category of biomarkers that are crucial to the development strategy are those linked to the activity or expression of the target or the pathway that the investigational drug is intended to act upon. Such biomarkers can be used for expression analysis across a variety of human tissue samples to define with a higher degree of accuracy, the medical opportunity. Further, such biomarkers may become pivotal to the eligibility criteria at least in early stage clinical trials. In diseases with complex molecular pathogenesis and variable expression profile such as cancer, it may be key to implement early enough biomarker-guided development, to ensure a chance of success for investigational drugs that are capable from pharmacologic standpoint only in subsets of patients. In the simplest scenario, the target itself may become a biomarker.

A third category of biomarkers are those linked to ‘off target’ effects of the investigational drug. Generation of data sets encompassing pharmacogenetics and toxicogenetics information is key to predict liabilities and opportunities during early development process. A wide array of technologies integrating in silico, data-base mining to array technologies in support of genomics, proteomics and ultra sensitive analytical methodologies, stimulated unprecedented research in this area. Many of today’s molecular targeted drugs carry a plethora of effects as opposed to being uniquely specific to the target they were originally developed against. Understanding earlier rather than later the toxicogenetics and pharmacogenetics profile in settings relevant to man, becomes a competitive advantage feature since the implications relative to lead optimization or clinical development strategy. Multi target reactivity of an investigational drug has both benefits and risks: while offering the possibility of co-targeting additional pathways and broadening up or enhancing the efficacy of the drug, it may result in increased toxicity. Again, this reflects the need to generate appropriate information during early development, allowing decision-making (go /no-go, lead optimization, or adjustments to clinical development approach and target disease population).

The concept of stage-appropriate and program-dependent use of translational medicine is important at multiple levels. Generation and use of scientific information such as data sets encompassing biomarkers needs to be channeled in support of complex decisional processes, allowing rapid strategic decision-making. Two categories of decisions are particularly important and should be enabled: ‘no-go’ relative to lead candidates and optimization of clinical development plans (e.g., disease indications, target populations to pursue). To that aim, biomarkers do not need to be fully ‘developed’ (qualified, validated) in the same way diagnostics are, although a successful molecular targeted therapy program may likely result in companion tests or diagnostics. In essence, only very few biomarkers will ever make it as diagnostics or companion tests – requiring considerable development-related expenditure – yet, they can be very effectively used in the drug development decisional process. The need for analytical stringency afforded by full-blown validation of biomarkers can be offset by creative utilization of multiple markers and correlation analyses between PD (pharmacodynamics) biomarkers and clinical endpoints. In fact, this latter aspect becomes a key tenet of translation, with real impact in the early clinical development.

Another dimension is adjusting the nature of translational support depending on the specifics of the program. For example, if a program encompasses a first-in-class investigational drug, translational resources should be primarily focused to define a sound stratified medicine approach (eg. biomarker-guided eligibility criteria; and measurement of biological response). While if a program is directed at ‘follow on’ or second generation compounds, creating toxicogenetics and pharmacogenetics data would enable rationale drug-design and optimization for the purpose of obtaining compounds with improved efficacy and/or diminished toxicity.

Altogether, recent progress in biomedical sciences, the expansion of the potential list of molecular targets and advances in drug development makes translational medicine an integral part of the process. Complementing drug design, discovery, optimization and testing, translational medicine’s responsibility is to guide the development process through generating scientific information with the aim to define and maximize the opportunity in clinic. This incipient paradigm remains to be tested and if successful, it will yield new generations of innovative drugs in under-addressed therapeutic areas, with a potential to bring molecular medicine to new heights.

About Adrian Bot

Adrian Bot, MD, PhD, is Senior Director of Scientific Management and Acting Head of Translational Medicine at MannKind Corporation, in Valencia, California. He obtained his MD at the University of Medicine and Pharmacy in Timisoara, Romania and his PhD in Biomedical Sciences at Mount Sinai School of Medicine in New York.

He previously held appointments at the Scripps Research Institute and Alliance Pharmaceutical Corp. in San Diego. Bot has authored more than 100 publications and patents in oncology, vaccines, drug delivery technologies and immunotherapy. He is the Editor in Chief of the International Reviews of Immunology and has been on the advisory boards of several organizations.

Adrian Bot, MD, PhD, Senior Director, Scientific Management, Translational Medicine (Acting Head), MannKind Corporation, Valencia, CA 91354. Tel: (001) 661 775 5544. Email: abot@mannkindcorp.com