Found in translation

The development of biologic therapies has significantly shifted the treatment paradigm in medicine, particularly in oncology. Groundbreaking research over the past 20 years has sparked new ways of thinking about disease pathology and opened the door for novel targeted therapies that have truly revolutionised medical care. In more recent years, we have observed a second paradigm shift through translational science, in which researchers are taking insights gleaned from the bedside to understand how these breakthrough biologics can be used in even more effective ways. This approach, also known as translational medicine, promises to help deliver personalised healthcare through more effective solutions tailored to populations with unmet medical needs.

Translational science offers an approach for mitigating some of the high chance of failure associated with the research and development process by allowing drug developers to uncover the pathogenesis of diseases and define new therapeutic targets for treatment. The field of oncology has readily integrated translational science, though this approach is integral to drug development in other therapeutic areas. It is a key component to developing new and innovative therapeutic approaches that improve patient outcomes while complementing a company’s lifecycle management.

At MedImmune, we have integrated translational science into our R&D process, and that has allowed us to make significant progress in drug development not only in oncology, but also in other therapeutic areas. We understand that successfully applying translational science to research and development planning requires more than adopting a new process. It requires creating and fostering a different mindset throughout your organisation – one that encourages greater collaboration across multiple disciplinary functions. Translational science is about transferring knowledge and insights from ‘bench to bedside’ and back again. For this to happen, the preclinical and clinical development teams must be amenable to greater alignment and more open, ongoing dialogue and data sharing.

The following are some insights we at MedImmune have gleaned from our experience in successfully making translational science pervasive throughout the organisation. These are important tips to follow for any bioentrepreneur who is interested in employing translational science strategies to improve R&D processes and reduce risk in earlystage research.

Foster a more collaborative mindset

Preclinical scientists must be more involved in providing the data to support the clinical hypothesis for Phase I and Phase II development, rather than just focusing on IND filing and then ‘throwing it over the wall’ to the clinical development team. This means expanding beyond activity-based pharmacology models to deliver a biomarker strategy that can be used in early clinical trials for demonstrating that a drug candidate engages the target (pharmaco – dynamic marker), affects the target pathway (proof of mechanism) and also affects the disease (proof of principal).

Use preclinical biomarker research to guide development

By focusing on target discovery to guide clinical development in patients, translational science enables us to take promising scientific findings out of the laboratory and into clinical research. Once a target is nominated for therapeutic evaluation, we can create a detailed package of target expression in normal and disease tissues, disease linkage and association with clinical outcomes to support the design of that target. At MedImmune, we match the target and the technology to the design of our biologic portfolio. We have found, for instance, that an antibody can be enhanced for action by a-fucosylating the antibody or, if there is expression on disease and normal tissue, then the antibody can be mutated to be devoid of antibody effector function (antibody dependent cell cytotoxicity). Antibodies can also be designed to have extended half life if there is a good therapeutic index.

Translational science is also about selecting the appropriate patients to achieve proof of concept. Critical factors for targeting the molecular causes of disease and defining the patient population most likely to respond to therapy include: molecular / genetic variation identification, diagnostic tests to better predict response and biomarkers to address the right drug and right target. These biomarkers assess safety, pharmacokinetics / pharmacodynamics relationships and proof of mechanism and principle. Discriminatory efficacy and toxicity biomarkers, for example, may improve our ability to eliminate or redesign suboptimal compounds earlier in development or to more precisely define the target population. Any translational clinical development should be able to answer why a certain tumour type / patient population is being evaluated, the prevalence of the target in the patient population and what evidence links that target to driving disease.

Based on pharmacokinetic and pharma – codynamic analysis in preclinical and clinical studies, Phase I starting dose and dose escalation can be predicted for testing, and support the recommendation for Phase II dose(s) to be used in clinical testing. This type of testing can ensure more rigorous definitions of the optimal dose of a candidate drug and may allow for fewer dose escalations in early clinical studies. Identifying biomarker endpoints also assists with internal development decisions such as whether the drug is exhibiting the desired pharmacologic effect on target engagement (pharmacodynamic, PD) and the downstream target pathway (proof of mechanism, PoM) and that this occurs within a dose that has an acceptable safety profile. Phase I testing also can determine dose expansion, including selecting the indication and, when appropriate, determining an acceptable mechanism to segment patient populations for targeted clinical trials. At MedImmune, the development program for MEDI-575 – an antibody to platelet-derived growth factor alpha (PDGFRα) is an exemplary case of this approach. In preclinical studies, MEDI-575 has been shown to inhibit signalling from PDGFRα on cancer cells and supportive stromal cancer-associated fibroblasts, which results in tumour growth inhibition. However, due to high target selectivity, this binding does not impact PDGFRβ, the inhibition of which has been associated with significant clinical toxicities. This selective binding potentially has significant implication in both brain and lung cancer.

The question we ask at this stage is who is the best patient population (disease and subset) for the drug to demonstrate a biological effect on the disease (proof of principle, PoP) and efficacy (proof of concept, PoC)? This means challenging traditional understanding of clinical success of the drug, which relies heavily on the concept of external validity and ability to project the results to the overall population.

Embrace novel clinical designs to inform development strategies

Clinical teams need to collectively embrace novel clinical designs and evaluate the correlative studies as rigorously as the clinical endpoints for safety and efficacy. Strong preclinical and/or clinical disease linkage in patient samples and preclinical pharmacology models will be critical in defining the right patient population for a specific treatment, while predictive toxicology models will better inform the safety assessments and risk-tobenefit ratio early in the R&D program. The translational approach of having clinical studies that are designed to learn and provide more than pass/fail answers based on safety and efficacy is emerging. These changes place substantial pressure on initial Phase I trials. The number of variables to be integrated into trial design such as parallel biomarkers, eligibility criteria for subject enrichment, cohorts to observe pharmacodynamic endpoints at nontoxic doses, different schedules and combinations of drugs make the initial stages of clinical research increasingly complex. Without a committed team to design the study, invest in the biomarker assays and generate the data in real time instead of retrospectively, translational-driven clinical trials will not succeed.

Develop ‘go/no go’ criteria early based on biomarker endpoints

In a translational science approach, Phase I research focuses on testing the clinical hypothesis and providing answers for deciding whether or not to proceed to Phase II/III trials. Assessing biomarker data in preclinical and early clinical studies allows us to understand which leads to follow up and which to discard. If compounds are not engaging the target, affecting the pathway or exhibiting an acceptable safety profile, then compounds can be stopped in Phase I clinical trials before we make any further investments. The business incentive for this approach is clear. The amount of money companies spend on research and development quadrupled in the past decade with the staggering 90 per cent of this cost driven by Phase III trials3. This investment, however, does not guarantee success and only one in 5,000 to 10,000 compounds synthesised ends up gaining regulatory approvals4. The high risk and fear of setbacks in later stages of drug development means reluctance from companies to pursue clinical research. For some smaller companies this means facing bank – ruptcy and for patients, potentially fewer beneficial drugs reaching the market.

The translational science approach, through evaluation of biomarker data in preclinical and early clinical studies, helps to mitigate this risk. By having focused development plans with robust data packages and decision points, only compounds with a high probability of improving patient care will be advanced to late stage clinical trials. The idea of not advancing compounds from Phase I to Phase II clinical studies based on biomarker endpoints is new and takes a science driven team to execute the ‘no go’ decision.

Conclusion

Scientific advancements in drug discovery and medicine have come a long way. Investigators are continuing to uncover greater insights into disease pathology and treatment. Integrating translational science into the R&D process is critical to providing companies with a more targeted method for developing the next generation of medicines that address a host of public health needs. Answering key questions about disease pathways as well as the pharmacokinetic and pharmacodynamics of an agent can help developers make better decisions about which compounds to move through development based on sound scientific data. Translational science offers a rigorous approach that has the potential to make the industry more productive by preventing compounds with a low likelihood of providing benefit to patients from advancing to largescale, expensive Phase II and Phase III clinical studies. Additionally, by defining the best suited patient for a treatment, this has the potential to provide benefit to patients and payers. Considering much of the cost of developing new pharmaceuticals is attributed to treatments that fail in clinical development, defining targets early and even predicting clinical trial failures could improve R&D success and reduce overall development costs.

References

1. Karapetis CS, Khambata-Ford S, Jonker DJ, et al. K-ras Mutations and Benefit from Cetuximab in Advanced Colorectal Cancer N Engl J Med 2008;359:1757-65.

2. U.S. Food & Drug Administration. FDA Drug Safety Communication: Reduced effectiveness of Plavix (clopidogrel) in patients who are poor metabolizers of the drug. http://www.fda.gov/Drugs/ DrugSafety/PostmarketDrugSafetyInformationforP atientsandProviders/ucm203888.htm. Accessed on August 17, 2011#

3. Roy, A. How the FDA Stifles New Cures, Part II: 90% of Clinical Trial Costs are Incurred in Phase III. Forbes. 2012. http://www.forbes.com/sites/aroy/2012/ 04/25/how-the-fda-stifles-new-cures-part-ii-90-ofclinical- trial-costs-are-incurred-in-phase-iii/

4. www.nctimes.com/business/article_d9a6fee5- bec2-596b-b3c1-851940972e27.html

About the author

Theresa LaVallee, PhD has over a dozen years of drug development experience focusing in the areas of oncology, inflamm – atory and angiogenesis diseases. Currently, she is Senior Director of Translational Medicine Oncology at MedImmune, LLC where she is responsible for working in a cross functional team with research and clinical to develop and implement the translational strategy for the oncology programs in the pipeline as well as, advance minimally invasive approaches to biomarker development. Prior to joining MedImmune, Theresa worked with EntreMed where she held positions of increasing responsibility and lead the mechanism of action studies for several oncology drug candidates in Phase I and II clinical trials with the goal of selecting clinical indications, drug combinations and correlative study endpoints. Theresa trained in the areas of angiogenesis, cell biology and molecular immunology under the direction of Drs. Tom Maciag at the American Red Cross and Sherie Morrison at UCLA. Her work has resulted in oral presentations at many international meetings and publications in top tier journals.