Drug discovery leaders roundtable

Andrew A. Parsons, Vice President Preclinical Drug Development, GlaxoSmithKline

Dr. Parsons has led the Preclinical Development function in the CEDD since its creation. He was previously the Head of Preclinical Development for the Neurology and Gastrointestinal (NGI) Center of Excellence for Drug Discovery (CEDD) and a member of the Executive Leadership Team. Dr. Parsons joined SmithKline Beecham in 1991 as a lead biologist on the Migraine Program and played a key role in the identification and development of Frovatriptan and Tonabersat. He has worked in a number of therapy areas within drug discovery and has led teams that progressed numerous compounds in development. He was also previously the Chairman of the Imitrex (Sumatriptan) International Scientific Advisory Board. Dr. Parsons graduated with a BSc, MSc and PhD in Pharmacology, University of Manchester and is a qualified NLP practitioner and Cognitive Therapist. He worked as a post-doctoral researcher at the Institute of Physiology, University of Munich, Germany. He has authored more than 90 peer-reviewed publications.

Steve Street, Vice President, Head of Research Centres of Emphasis, Head of WRD Continuous Improvement, Pfizer

Steve Street joined Pfizer in 1985 and held a variety of roles within the Chemistry Department in Sandwich UK, before being appointed Head of Chemistry for Sandwich Discovery in 2001. Three years later, at the start of 2004, Steve moved to establish and lead the Chemistry discipline across Worldwide Discovery. Steve held this role through to early 2007 when he was delighted to be asked to take on a new role as Head of the Research Centers of Emphasis, covering a range of activities supporting Worldwide Research including External Research, BioImaging, Computational Sciences and High Throughput Screening. In April 2008, Steve was asked to become Head of Continuous Improvement across all of Pfizer Worldwide Research and Development where his focus was uniquely on improving project and portfolio survival and cycle times.

William Strohl, Vice President of Biologics Research, Centocor R&D, a division of Johnson & Johnson Pharmaceutical Research & Development

Dr. William Strohl received his PhD in Microbiology from Louisiana State University, and worked as a guest researcher at the GBF in Braunschweig. He rose to Full Professor in the Department of Microbiology and the Program of Biochemistry at Ohio State University before moving to Merck in 1997 to head up Natural Products Microbiology. From 2001 to 2008, Dr. Strohl was a leader in Merck’s efforts to develop therapeutic monoclonal antibodies, as well as in-licensing of therapeutic targets and technologies associated with monoclonal antibodies. Dr. Strohl was the scientific leader in the acquisition and integration of Abmaxis and GlycoFi into the Merck Biologics organisation. In 2008, Dr. Strohl was named leader of Antibody Drug Discovery at Centocor (J&J) and in 2009, named as VP and Head, Biologics Research, in the J&J Biotechnology Center of Excellence. Dr. Strohl has over 100 publications, several patents, edited two books and recently wrote a book on therapeutic antibody engineering, to be published in 2012.

Eckhard von Keutz, Senior Vice President, Head Global Early Development, Bayer HealthCare

Eckhard von Keutz directs the global activities of all preclinical development functions (Toxicology, Safety Pharmacology, DMPK, Animal Management). He has the final say in all matters of nonclinical safety for Bayer HealthCare and is the speaker for Toxicology for the entire Bayer group. Eckhard has functional and operational responsibility for all preclinical units at Bayer HealthCare. He has experimental experience in conducting a broad range of toxicological studies, broad experience in designing preclinical / toxicological testing strategies, broad regulatory expertise, broad expertise in toxicological risk assessment and risk management. He has long-lasting managerial experience in supervising department and institutes with up to 540 employees and managing inter-disciplinary teams. He is a lecturer at the University of Duisburg-Essen, University of Bonn and University (Charite) of Berlin and former member of the Scientific Committee (ECETOC), member of the Drug Safety Executive Council (DSEC) and Chairman of the Advisory Board of the Fraunhofer Institute of Toxicology and Experimental Medicine (Hannover, Germany).

External economic pressures have been identified as the major driver for the pharmaceutical outsourcing market. Over and above the fiscal advantages of adopting this strategy, what other benefits and indeed risks do you see associated with this approach?

Steve Street: We definitely began our out – sourcing efforts based on economics and the fiscal benefits. Subsequently, we have realised that we gain benefits in terms of flexibility, diversity and opportunity. Flexibility in terms of where and when we scale up or down on investments, diversity in terms of access to outstanding and committed scientists across the world and opportunity in terms of more and different technologies, projects and markets.

Andrew Parsons: It really depends on the definition of outsourcing. In my opinion, there are a number of different activities that could be captured within a broad definition of outsourcing. These include outsourcing: the transfer of internal activities to an external vendor, off-shoring: the transfer of activities to a different location which is typically a lower cost country, which could be either an internal or external resource, and open sourcing or risk sharing: two or more individual companies sharing investment into one or more activities and sharing the reward.

Open Innovation (OI) is a paradigm originally presented by Henry Chesbrough in 2003. This concept allows development of both internal and external ideas using both internal and external development paths to the market place. The OI paradigm is a framework around building alliances with the mindset to develop win-win business models to identify and develop new products.

Which route you choose to explore for an outsourcing strategy will impact the various risks and also the potential benefits for reaching your corporate goal. Therefore having clear strategic aims on how to spend precious resources is an absolute prerequisite for managing the outsourced environment.

For example, if a strategic aim of a company is to increase the diversity of innovative discovery and development projects by accessing external technologies, a risksharing approach could be more aligned with their goals. Alternatively, if cost saving is a key driver then offshoring or outsourcing might be a better fit.

However, there are some common themes which emerge, whichever of the above outsourcing strategies are undertaken. For example, one of the key benefits of using an outsource model is that it can allow a refocusing of internal investments into the key areas the business.

Outsourcing therefore allows creation of space to really focus on what is core to your business needs, and facilitating the develop – ment of the unique features of your business that allows you to differentiate and succeed. There may also be a number of opportunities that allow investment into building other resources, skills and capabilities that could create in-depth expertise and knowledge that allows value creation. In short, outsourcing strategies provide an opportunity to truly focus internal resources on what brings most value to the business.

Outsourcing also provides access to different skill sets, processes and different beliefs about how to do things and why they are important. Such experiences can really bring major insight and challenge to improve internal processes. In my opinion, this is a major opportunity. Interacting with different companies can bring learning and support an environment of learning and building continual improvements and challenge to internal ways of working. Looking at lessons from the natural world, hybrid vigour is a well known feature of biological systems. Outsourcing provides an opportunity to grow stronger.

There are also a number of risks and challenges that need to be considered and balanced with the opportunity. As an organisation, do I have the right skill sets that allow me to assess, select and develop the appropriate business relationship that provides a win-win for both parties? Each of the key steps of assessment, selection, managing the collaboration and ensuring delivery has a level of skill and capability that is needed by both the seller and the buyer. Am I structuring the most appropriate business contract? Is it flexible? How do I manage the collaboration are all essential parts of maximising value and should be given careful consideration.

Also, financial considerations need to feature in your risk assessment. Does the company have enough cash, or is it able to generate enough revenue / investment to be able to deliver on our collaborative project? The pharmaceutical world is rapidly evolving and a number of major changes can take place without much control. For example, one of the collaborators could be purchased by another company. It might become bankrupt or need to downsize. How is talent managed in your partner? Is there a risk that they will lose the skills you are investing within the collaboration?

All these risks will require some thought and development of a mitigation strategy and there is of course the problem of events. When a reporter asked UK Prime Harold Macmillan what can steer the Government off course, he replied “Events dear boy, events.” In my experience, events can and will always happen, so both parties within the collaboration should predetermine how they should react in such a circumstance.

Eckhard von Keutz: Over and above the fiscal advantage of outsourcing, such a strategy provides the opportunity to gain access to external talent, innovations and new tech – nologies. In times where collaborative efforts between pharmaceutical companies, regulatory agencies and academia are ongoing to for example find better evaluative tools to modernise the medical product development pathway, it appears to be very attractive to build platforms which can be publicly used by various partners thereby reducing the need to individually invest for example in new technologies. Such platforms could also be used to perform experimental studies needed to support the development of new drug candidates.

At present, outsourcing is mainly used to increase external flexibility by keeping the internal resources flat. The goal is to contract out work which is exceeding the in-house resource and avoiding idle capacities in times when the output from research is low. In this context, R&D outsourcing may reduce costs but if managed inappropriately, often results in the opposite effect, thus creating more complexity and frustration resulting in higher staff attrition and loss of tacit knowledge from the organisation. It appears that some mediumsized pharmaceutical companies outsource the majority of their toxicological work. Whilst this creates short-term cost savings, it can also create a knowledge gap within their own R&D organisation, the impact of which some companies are now feeling.

William Strohl: First, I want to state that all of my views expressed here are mine and do not necessarily represent those of Johnson & Johnson’s family of pharmaceutical companies. Moreover, my answers are given from the perspective of discovery research and do not necessarily apply to other functional areas.

In any kind of transaction or agreement between your company and another, you have to weigh and balance the various risks and benefits of that agreement on your business. The risks for dependency on outsourcing are significant and real. These include results, reagents or materials that do not meet internal quality control criteria, inconsistent results that cause remakes and thus program delays, possible loss of your partner through acquisition or company failure, and control of intellectual property. One of the key drivers for many of these issues is FTE turnover rate at the external vendor, whereby high turnover rates typically decrease consistency, reliability and data security. Thus, if you significantly increase your reliance on outsourcing, you have to be careful to delineate a strategy that includes having a strong diligence process in place to help with selection of potential partners, multiple outsourcing partners, or back-up strategies, for each potential activity, limitations to what is outsourced so that intellectual property is maintained within your organisation, and close scrutiny and quality analysis of the materials and services that are received from external vendors. I cannot emphasise enough the need to have significant internal resources dedicated to the management, quality assurance and interrelationship with external outsourcing vendors. To that end, it is also useful to help certain key external vendors to improve their services through feedback and advice, resulting in some close-knit relationships that will help you to achieve your goals better and to smooth over rough spots in the relationships. The benefits of a well-managed external services arrangement can make the risks worthwhile. These benefits include a fungible external workforce that does not require benefits and long-term program costs, ability to ride the ebb and flow of budgets by decreasing or increasing externally sourced materials and services, respectively, the ‘forcemultiplier’ effect of increasing capabilities without increasing internal headcount, and tapping external expertise that would be difficult to maintain and/or justify internally. Much of the external resourcing efforts in early discovery rely on the production and use of ‘standard, but made-to-order reagents’ such as oligonucleotides, DNA sequences, genes, recombinant proteins, synthetic peptides, antibodies and so forth, which can be very costly to make internally but can be obtained at reasonable cost from vendors who specialise in each of those services.

The economic consequence of late-stage drug attrition has necessitated an ever increasing requirement for toxicity and safety assessments to be performed at the earliest stages of development. In your opinion, what recent technological and/or methodological approaches hold the most potential for enabling a viable and robust ‘fast to fail’ strategy to be adopted within drug discovery industry?

Eckhard von Keutz: The possibility that toxicological findings may lead to late failures during drug development underlines the necessity for predictive tools which allow an early assessment of the potential liabilities of a new drug candidate. Cardiac toxicity (including arrhythmias), hepatic toxicity and CNS toxicities (neuropsychiatric effects, abuse liability, dependency) have been the leading reasons for drug withdrawals over the last decade. Strategies to reduce predictable toxicities are central to improving the quality and viability of new therapeutic agents. To date, standard nonclinical toxicity studies remain the cornerstone of the prediction of toxicity in humans. However, new approaches as well as the refinement of existing methods are necessary to improve the prediction of potential side effects in humans. Various promising investigative approaches are currently being evaluated for potential screening purposes or use on a case-by-case basis following safety alerts in standard nonclinical toxicity studies. The hope is to find a few simple measures that will tell whether a drug is likely to cause serious side effects. From my perspective, genomics and imaging tech – nologies hold the most potential for detecting potential toxicities early on. It should also be mentioned that new and more sensitive diagnostic tools may also facilitate the development of promising drug candidates that have the potential to address unmet medical needs but were previously thought to be ineligible for development because the current diagnostic tools fall short in their ability to detect early development of injuries in humans.

William Strohl: Since I manage a biologics discovery enterprise, my answer here will be specific for biologics and mostly directed towards monoclonal antibodies (MAbs) and Fc fusion proteins (FcFPs), which have a very different toxicity profile than found with small molecules. MAbs and FcFPs generally bind specifically and with high affinity to a specific target or, at most, to a few highly related targets. Toxicity obtained with these types of drug candidates is almost always target mechanismrelated toxicity rather than off-target toxicity. Thus, understanding target biology and the pros and cons of antagonising or agonising the target, or of depleting or modifying the targeted cell population (for cell surface targets), is key to understanding the safety risks that may be associated with the drug candidate. The greatest point for attrition of MAbs and FcFPs during clinical development is after Phase IIa, during which proof-of-concept for the drug mechanism of action is either obtained or not. This attrition rate could potentially be decreased by a significantly increased emphasis on target biology and translation of that biology to human disease states during the discovery and preclinical development phases. This increased emphasis on target biology in the discovery and preclinical development phases also would likely help to understand, and potentially mitigate, clinical safety risks as well.

The second approach that can help increase knowledge of the targeted biology is to initiate biomarker and translational medicine programs earlier, so that potential biomarkers both for efficacy and safety risks can be analysed in Phase I clinical trials. Thus, having biomarker and translational medicine experts in on the discovery process and thinking about how to measure downstream pharmacological effects of the activity of MAb or FcFP candidates in in vitro cell based assays, in in vivo animal models, and ultimately in people, can potentially enhance the quality of data obtained in preclinical development and in Phase I, leading to better informed go/no go decision points, as well as better understood toxicology profiles. This process may also help to stratify patient populations so that you are treating the patients who have the most potential to be helped. This ultimately leads to improved probabilities of success with smaller, more focused patient populations which help everyone from the patients, to the payers, regulators and ultimately the biopharmaceutical industry, who will benefit by understanding the drug better and understanding how to apply it best to disease indications.

Andrew Parsons: Prediction of potential safety or toxicity findings is such a major area of research and investment and I believe there are many exciting themes of research that will provide data to support early termination and provide a fast to fail decision point.

I find the area of in vitro assays particularly important, especially in the rapidly developing area of stem cell biology. Research in these areas will create an opportunity to provide relevant tissue cultures to test the actions of potential new drugs and determine the likelihood of any toxicity signal.

I also believe that we need to keep in mind the purpose of the studies. Drug development is a series of individual studies that allow better understanding of the risks associated with treatment as well as the benefits of new therapies to the patient. It is the perception of the benefit-risk assessment in early drug discovery and development that really makes most impact on early investment decisions. Data to make the most informed discussion may not be available until the agent is in late stage development or even life-cycle management so investment choices will need to be made on incomplete data. Therefore, the advent of new technologies and methodological approaches can provide a data-led forced attrition within early stage development. We may never know whether assays used in this way are truly predictive, but the use of these techniques or approaches may lead to a focusing of resources on projects that are deemed to have the best chance to provide appropriate benefit to patients.

In my opinion, the jury may still be unsure as to whether the new approaches may reduce attrition in terms of whether an agent has the promise of being a differentiated new medicine of significant value to the patient. However, it should limit investment on agents that have higher risks of safety or tolerability concerns.

Steve Street: Attrition in humans should be avoided at all costs. The ultimate place to avoid attrition is before the agent is prepared – in cerebro attrition as opposed to either in vitro or in vivo. To this end, we have become much more expert at analysing historic data, identi – fying themes and translating these trends into strategies and tactics to avoid attrition. Most recently I am very excited about the potential of high throughput (HT) in vitro assays (including assays derived from stem cells) to predict common and important in vivo issues such as hepatic toleration. As soon as we have such predictive HT assays the opportunity to develop structure toxicity relationships becomes real, and so does our ability to design out attrition before we make compounds.

Despite the intensified political and social pressure placed upon the pharmaceutical industry to make available life saving treatments to a wider proportion of the world’s population by reducing or subsidising the cost to the patient, many organisations within the industry are reluctant to acquiesce to these demands citing the prohibitively high costs associated with drug research and development as a barrier. To resolve this issue it has been suggested that extending the life span of drug patents would allow cost recovery over a longer period and hence permit price reduction in the shorter term. Placed in the role of a decisionmaker, how would you evaluate this approach and what alternatives if any would you consider viable?

Andrew Parsons: In our time, there is a tremendous opportunity to make a real change to the way the pharma industry operates within the wider constraints of society and bring much needed medicines to the people who need them. There are many initiatives that are currently challenging the way the industry has traditionally worked.

For example, the sharing of intellectual property has been an anathema within the pharmaceutical industry. The protection of intellectual property has been a major focus and loss of protection can have dramatic impact on the bottom line growth for the company. The so-called patent cliff, or loss of protection on major products is a significant hurdle for many companies today.

Within the wider constraints of some open innovation models, shared intellectual property within a crowd sourcing environment has been found to be a productive approach within the telecommunications and automotive industries. Proctor and Gamble’s ‘Connect + Develop’ programme enables entrepreneurs to review ideas for licence and share ideas with the company.

Other ideas include patent pooling, which appears to be a growing practice within some industries and it will be interesting to see how this practice develops within the pharma – ceutical industry. However, the times do appear to be changing. For example, GSK has published thousands of drug like structures with potential utility for drug discovery in Malaria1 and joined forces in 2009 with Pfizer to pool products into a new HIV company called ViiV.

The ability to adapt to change in the environment is a key feature of success for organisms and organisations. I believe the industry is starting to make the necessary changes to adapt to the fast pace of change. This is occurring not only in how innovation occurs with increasing opportunities for external collaborations but also in access of innovative medicines to the worldwide market place. Initiatives such as tiered pricing of vaccines brings much needed products to patients where they may be most needed.

Each decision-maker has to assess the wider needs of the patient, the payer and the needs of the organisation to create a viable approach to change. I do not believe there is one size that’s fits all, however there is a need to adapt to the new challenges the industry face and bring new innovative medicines to patients.

William Strohl: Pharmaceutical drug develop – ment is a high risk venture in any case, with something like between six to eight per cent of all small molecules and between 20 and 25 per cent of large molecules entering preclinical development making it to the marketplace. As novel targets and mechanisms of action become more difficult to find, there could be additional downward pressures on these already low numbers. There are currently clinical trials for nearly 400 MAbs and Fc fusion proteins targeting at least 185 different molecular targets, of which only around 25 have been validated clinically. The rest of those targets cover a wide span of disease indications, most of which are life threatening or serious illnesses and many of which may result in drugs being approved in the orphan category. Thus, I would argue that the biotechnology industry is already taking on the risk of treating difficult-to-treat diseases, even in the current fiscal environment. While this is more likely opportunistic rather than altruistic behaviour, it points to an industry that is both willing and able to tackle difficult disease problems. Thus, while the incentives are relatively meagre at this point, my contention is that the industry is already taking a lead on discovering and developing medicines for unmet medical needs.

The concept of lengthening the lifespan of marketed products to improve the likelihood that pharmaceutical companies would use the funding to address serious diseases was at least part of the premise behind the 12 year data exclusivity on innovator biologics that was voted into law as part of the 2009 US Healthcare Reform Act. That 12 year data exclusivity period has already come under political pressure and may eventually be changed. The problem with a legislative approach such as that is that it takes at least 10 years and one billion US dollars or more to develop a new drug. Thus, if a company were to be incentivised to make novel biologics against highly difficult disease targets starting now, by 2020 the legislative environment may have completely changed and the original incentive may have become irrelevant. Also, such legislation would need to be approved by each country in which the drugs were to be sold leading to a very low probability that such an approach might be viable. Thus, in my opinion, it would be unwise for any company to agree to significant price reductions now based on this type of political solution that could affect drug lifetimes in a decade.

I am not sure I have an alternative solution to this important strategic question. One possible solution would be to offer significant tax advantages for discovery of drugs with novel mechanisms, although tax laws for something with a definition as imprecise as this could potentially be difficult to manage. Another alternative would be to modify patent laws in the major countries and regions, including the US, EU, Canada, Australia, Brazil, China, India and Russia, that would give greater protection to drugs addressing novel mechanisms of action or disease indications, but this seems highly unlikely.

Steve Street: Undoubtedly, there will always be voices within the pharmaceutical industry advocating longer periods of patent protection for many reasons. If I were a decision-maker, I would not go this route to achieve wider availability of breakthrough drugs. I am not convinced it would achieve the goal and it may cause other ‘unintended consequences’. On one level, I believe that many pharmaceutical companies would be willing to make life saving treatments more widely available. However, I believe the key is that such action would have to avoid any negative impact – put another way, there would not necessarily have to be an immediate positive impact. So my focus – if I were a decision-maker – would be on the issues to avoid (e.g. negative price pressure in developed markets).

Eckhard von Keutz: Extending the lifespan of drug patents would certainly allow cost recovery over a longer period and hence permit price reduction in the shorter term. It must be kept in mind that discovering and developing an innovative therapeutic principle bears a high development risk and that only few candidates will finally make it. Such few new drugs must finance the entire investments into R&D and the present market exclusivity might simply be too short to get the full compensation. At the same time and equally important, one needs to explore and implement all oppor – tunities to reduce unnecessary hurdles during development and to accelerate the develop – ment as much as possible. This again requires combined efforts between pharmaceutical industry and regulators.

The pharmaceutical industry has historically been very successful in performing all the tasks associated with the drug discovery process internally. Although outsourcing has and continues to be discussed, what do you envisage is the optimal level of outsourcing? Are there any aspects of outsourcing that have surpassed or failed to meet their intended expectations?

Steve Street: I believe the real benefit of outsourcing is that it drives focus on the core competences that differentiate any organisation. Frequently quoted examples in the pharmaceutical industry of such core comp – etences would be medicinal chemistry and target selection. In these core competencies, I would expect the pharmaceutical industry to maintain most if not all of the capability. In most if not all of the other associated activities, I could imagine the pharmaceutical industry main – taining less and less capability.

Eckhard von Keutz: In the field of preclinical development, there is much advantage in keeping the process until POC (proof of concept) internally. During this phase, which is key to identify the drug candidates with the potential to become ‘winners’ and to eliminate the ‘losers’ prior to any significant expenditure, it is particularly important to have shared accountabilities along the value change to force different functions and groups to manage the inevitable trade-offs in the process. Once a compound has been adequately profiled (DMPK, Tox), the next phase can be easily outsourced. It should be mentioned that the outsourcing strategy is currently changing. Whereas in the past, individual studies have been outsourced because they were exceeding the internal workload today’s approach is to outsource complete phases. In this context, CROs (Contract Research Organisations) are becoming partners for the pharmaceutical industry and not providers of individual services. In the past, outsourced studies were sometimes difficult to manage and a significant resource of frustration. They required adequate internal technical competence as well as proper coordination management to effectively utilise complicated and long-distance collaborative thirdparty relationships. This may change in an environment of mutual trust and when the CROs become true partners of the pharmaceutical industry.

William Strohl: As mentioned in my answer to the first question, the level and type of outsourcing needs to be balanced between risk and benefit. Most large pharma now carry out much of their medicinal chemistry efforts by outsourcing to, or in collaboration with, Asian CROs. The delicate balance that needs to be struck in such arrangements is where the intellectual property is actually conceived and to what degree does the CRO have input on strategy and design. As such CROs become more sophisticated and capable of generating novel intellectual designs that are incorporated into molecules, this question becomes more critical. In the biologics arena, several new CROs have recently sprung up who claim to be able to run discovery and even early development programs from soup to nuts. While this eventually may be a pathway for large pharma and biopharma to take with some programs, it may be a bit early at the current time for some of us in the industry to make such broad-based external investments. The key issues that remain to be completely satisfied for running external programs at CROs are the security of the data and information generated, and how to deal with the intellectual property created in such programs.

Models that may be very attractive in the future as external CROs grow in sophistication from the scientific, business, and intellectual property standpoints include risk sharing, option deals, and/or co-development of molecules, especially through clinical proof-of-concept. These types of deals may eventually result in more molecules with novel mechanisms of action entering the clinic with shared and more broadly spread risk, which ultimately would also result in shared rewards if successful.

Andrew Parsons: The optimal level of resourcing is likely to be different for each company. It really depends on the core expertise of the company and the strategic direction of the company in returning value to patients and shareholders. As already mentioned, the type of outsourcing the company embarks upon is a key consideration.

For many, I would envisage that a balanced approach between internal investment, outsourcing, off-shoring and more collaborative approaches to share the benefits of success would be an ideal mix of different strategies.

In my opinion, having a level of internal core skills and capabilities is an important prerequisite for any outsourcing strategy. Therefore, I believe it is difficult to realise the value of a fully outsourced R&D organisation from target identification to launch and post marketing activities.

Different companies may embark upon a fully outsourced early or late discovery / development platform and therefore the optimal balance will be a reflection of what is available within the external community to partner or collaborate. As the saying goes, it takes two to tango and the balance of skills, capabilities and appetite for creating novel business frameworks within the extended sphere will be a key feature in determining the optimal balance within a network of companies.

The overall aim of the industry must be to improve productivity per dollar investment, with current attrition rates being high in reaching proof of concept in patient trials and product launch the optimal balance will reflect the high risk associated with the major investment needed to bring a new medicine to patients.

Peer-reviewed literature suggests that there are a vast number of targets implicated in diseases which can be investigated for drug discovery purposes. In many cases, the basis of the disease association is questionable from a drug discovery point of view and has implications for target validation. Do you see the pharmaceutical industry de-risking their drug discovery pipelines by focusing on well validated targets which are also likely to be in the pipelines of competitors?

William Strohl: In a book I recently wrote, I addressed exactly this question for monoclonal antibodies (MAbs) and Fc fusion proteins (FcFPs). The data indicate that there are 185 unique molecular targets currently addressed by about 400 MAbs and FcFPs on the market or in the clinic. Of those 185 targets, about 25 are clinically validated and another 10 – 15 should be clinically validated within the next year or so. Interestingly, of those 400 MAbs and FcFPs in the clinic or on the market, 134 are against just the top 15 targets. For example, there are at least 12 different MAbs against CD20 either marketed or in clinical trials, 11 different MAbs and FcFPs targeting the IL 6 pathway, 11 targeting amyloid-β, 11 targeting VEGF/VEGF-R, 10 targeting TNF-α, and so forth. Not all of these candidates will become commercial successes, and just as it has been for small molecules for many years, I expect that for any given target, first-in-class and best-in-class biologics will take the lion’s share of the markets in most cases. Moreover, about 88 per cent (USD 46 billion) of the current value of marketed MAbs and FcFPs (USD 52 billion) is tied up in only five targets: TNF-α, VEGF, CD20, Her2, and EGFR. While this should change significantly as additional MAbs and FcFPs against a broader array of targets hit the market over the next five years or so, it is also instructive. Some promising targets for which MAbs and FcFPs are working their way through late stage clinical trials include amyloid- β, IL-6, IL-13, IGF-1R, IL-17 and NGF. In each of these cases, multiple companies have viable candidates, so there will be likely be significant competition for those new markets if these new target mechanisms prove medically beneficial and safe.

The flip side to the data mentioned above is that >260 MAbs and FcFPs currently in clinical trials address targets that are not in the top 15 mentioned above. Some of these MAbs and FcFPs target very novel, and intriguing, mechanisms of action while others are addressing orphan indications or difficult-totreat diseases. I think this spells out significant hope that the pharma and biopharma sector is addressing not only the obvious targets but also those that are more novel, and by definition, riskier. This gets back to an early question about funding and risk. I think these data also suggest that the large pharma and biopharma are investing more in novel and innovative science than often given credit for. While I cannot divulge any targets here, I can say that we at the J&J Biotechnology Centre of Excellence are collaborating with our therapeutic area colleagues to target some very novel mechanisms of action in our efforts to treat a broader array of disease indications, and every indication appears to me that we will continue to do so. I am very proud of that fact. As scientists in the pharma and biopharma industries, one of the key things that brings us to work each day is the knowledge that we are doing our part to address unmet medical needs – this is clearly the inspiration for innovation in our industry and as long as there are unmet medical needs, there will be some company, small or large, who will try novel approaches to address those needs. This represents a significant personal factor in our industry that is often lost when talking about budgets, costs, third-party payers, and other more global aspects of the industry.

Steve Street: The pharmaceutical industry has been, and will always be, passionate about derisking their drug discovery pipelines. Linkage of pharmacology to disease is an increasing problem, especially in complicated and poorly understood indications like Alzheimer’s disease. Any well validated target – be it genetic or competitor validated – will always be a focus. In many ways I can foresee a model where target validation becomes much more of an open innovation or pre-competitive activity and the pharmaceutical industry competes on advancing quality differentiated medicines to patients.

Andrew Parsons: I believe the focus for the pharmaceutical industry needs to be on bringing medicines that have value to the market place. This is not only for creating a meaningful difference to the patient it also has to make an economic benefit to society. I do not believe creating small incremental improvements will be a sustainable proposition for R&D organisations over the next 10 to 20 years.

In terms of validated targets that already have available medicines, focusing on these targets may well be an approach that some companies will choose to make as it helps build a scientific rationale to secure investment. How different will my ‘me-too or me-better’ really be? How will it really add value to patients and those who justify the cost of the medicine? These are key questions to address before embarking on a well validated route.

An alternative approach would be to focus on novel areas with high unmet need and limited available therapy. This approach creates a strong rationale in terms of the end user i.e. patients and organisations/agencies that pay for treatment. However, a major issue with this approach is how to select the target in the first place. Over the last 10 years, our ability in the industry to ‘mine data’ and identify key genetic variants that are associated with disease has increased significantly.

Our understanding has significantly increased in many diseases, especially those in which we have a more complete understanding of the underlying molecule mechanisms. A number of rare diseases may qualify in this category and there are good examples of how detailed knowledge of the disease can produce new medicines.

Pompe disease is an inherited, rare (approximately 1 in 40,000 births) and often a fatal disorder that occurs following mutation in an enzyme acid alpha-glucocidase (GAA). This enzyme is used to break down lysosomal glycogen. In Pompe disease, mutations in the GAA gene reduce or eliminate the enzyme resulting in heart and muscle malfunction. A number of different mutations in the GAA disease can cause the malfunction and age of onset and severity of disease are related to the degree of enzyme insufficiency. This under – standing led to the development of an enzyme replacement therapy with alglucocidase alfa (Myozyme) for early onset form of the disease. This is an example of where our understanding of the biology of the disease highlighted a potential pharmacological intervention. However, the challenge is to find the underlying cause of disease in individuals. This strategy ultimately leads to better and more targeted therapies but with a challenge of more defined and smaller patient groups that might be suitable for therapy.

A number of companies are now actively researching orphan diseases as they have a more detailed understanding of the pathogenesis of disease. As our knowledge progresses further the better understanding of differences in individual clinical phenotypes coupled with knowledge of the aetiology of disease will allow the industry to focus on these well validated targets.

Even with much better understanding of the pathogenesis of disease, there may still be a challenge to the industry. How to identify diseases that are tractable to drug modalities? Competition is likely to be fierce in these areas and excellence in discovery and development will be a key feature for companies that succeed.

Eckhard von Keutz: At present, the pharma – ceutical industry is facing significant challenges. It appears that at the end of the day only medical breakthroughs, i.e. real innovations, will have the chance to get full cost compensation. Such innovations require risk-taking approaches and may only be achievable by relying on nonvalidated targets. The risk of such a strategy is obvious. It is thus necessar y to also work on well validated targets. While such projects may only lead to incremental advantages, they do contribute to de -risking the pipeline. A balanced portfolio ensuring innovations but also having drug candidates with a comparably high likelihood of technical success is key to a sustainable drug discovery process.

It should also be noted that there are recent publications indicating that 50 per cent of published targets from academic laboratories cannot be repeated in an industrial setting. The reason for this is not always known. Against the backdrop of falling research and development budgets, increased financial risk-aversion and high attrition rates, broader strategies to improve data robustness are needed. It appears that the researchers in the pharmaceutical industry have become more cautious when work ing with published targets. Whereas in the past they sometimes used to set up highthroughput screens against published targets prior to internal validation, they now tend to check out at least some of the data themselves at the bench first. While data robustness is an issue, it is equally important to mention that external sources will become even more important to provide the pharmaceutical industry with new targets implicated in diseases.

Most approved drugs are small molecules. Although considerable effort has been expended in discovering biological based therapies, have the challenges associated with discovering and developing these therapies been underestimated? What do you envisage as the optimal balance between small molecule and biological based therapies within the pharmaceutical industry pipelines?

Andrew Parsons: Small molecules and biological based therapies cover many different approaches which all have different issues and risks associated with them. For example, synthetic molecules include a wide range of different modalities including the traditional low molecular weight (<400) molecules, oligo – nucelotide based drugs and even peptides. Biological based therapies can include monoclonal antibodies, soluble protein and perhaps even cell based therapies.

There are some common issues irrespective of the approach. These include the familiar issue of selecting a suitable target for the chosen modality and finding the right experiments to optimise the agent for the clinical situation.

A number of trade-offs will need to be reviewed by an individual company to see where the optimal balance is within the portfolio. Some of these key questions would include the following. Can I make the product reproducibly? Do I have the skills and capabilities or do I need to acquire them to enter the synthetic / biological space? Will the product have the right pharmaceutical properties to be useful to patients in the general practice? Are there any special considerations need to distribute the product e.g. cold or warm distribution networks. These are key considerations for any investor into a drug discovery portfolio.

There may have been a better success rate for biological agents in the past, but does this mean it will continue in the future? Without increasing our understanding of the pathobiology of disease, it might be difficult to select targets suitable for a ‘bio’ approach. A key decision point for any potential therapy is at the stage of proof of concept in patients. At this point data can be reviewed in terms of the potential benefits of the agent in patients and the risk associated with adverse finding and potential toxicity. Therefore, for any approach that a company decides upon, they need to be confident to find the right balance of tractable targets for the approach they select.

I believe there are now many technologies in development that will aid the identification of more tractable small molecules for diseases. These include approaches to identify small molecules with high binding efficiency and appropriate physical chemical properties for development. Several approaches are now also in investigation to make peptides more stable and even delivering drugs to the target tissue. Other modalities are also under investigation such as oligonucleotide based drugs including antisense approaches with Isis (USA) and exon skipping with Prosensa NV (The Netherlands).

These approaches have developed in parallel to the advent of biological therapies which have been incredibly successful so perhaps companies should also consider new modalities to synthetic and biological modalities as an alternative approach. Ideally, a variety of drug modalities could be explored using both internal and external resources via outsourcing and collaboration to optimise small molecule novel modalities and biologicals. In a recent review of recent FDA approvals, of 259 agents between 1999 and 2008, 75 of these approvals were first in class medicines of which 67 per cent were synthetic molecules and 33 per cent biologics2. It will be interesting to review how this profile changes in the next 10 years.

Eckhard von Keutz: The answer to the question is pretty much dependant on the capabilities / history of the individual companies. Companies who have already made large investments in biological development, in particular into the production facilities, may wish to maximally utilise their capacities by moving as many biopharmaceuticals as possible into development. The situation may be different for companies newly entering the field. While the challenges for biological based therapies are pretty much known, they should not prevent the pharmaceutical industry from investing in such opportunities. It is clear that the low hanging fruits are all gone and that targets which cannot be addressed by small molecules may become druggable only by bio – pharmaceuticals. In my view, a ratio of 2/3 (small molecules):1/3 (biological based therapies) may be considered a good balance.

Steve Street: Biological based therapies have already shown themselves to be of immense medical and commercial benefit. I am certain that the pharmaceutical industry and patients will continue to benefit from both small molecule and biological based therapies. That having been said, I do believe that the early success of monoclonal antibodies in certain diseases lead many to believe that all targets and all biological modalities would be similarly successful. This was always an unrealistic expectation. As different biotherapeutic modalities have been advanced into different disease areas then attrition due to toleration, off target activity, differentiation and commercial pressures have all become more common … and in many ways mirror the same challenges faced in small molecule therapy.

William Strohl: While this question is absolutely accurate in a literal sense, it misses a key difference between small molecule and large molecule approaches, i.e., time. Small molecule approaches to treat diseases have been with us for more than a century now, with significant sophistication of the small molecule approaches taking place as far back as the 1960s and 1970s. The first biologic to be approved for marketing was Humulin® (recombinant human insulin) which was approved in 1982. After that, the approvals of major MAbs and FcFPs, which have driven the significant recent growth in the biologics industry, started in the 1997 – 1998 timeframe. Currently, only about 165 thera – peutic proteins of all classes have been approved for marketing, of which 39 are MAbs or FcFPs. Another 36 MAbs and FcFPs are in post- Phase IIb clinical trials, of which about 27 should be eventually approved for marketing based on current probabilities of success of about 75 per cent (for initiation of Phase III to the market). Thus, within the next four to five years, we should see more than 60 MAbs and FcFPs on the market, treating a variety of significant disease indications – all within about 20 years of the first approval of a recombinant MAb (ReoPro®, 1994). Additionally, it has been predicted that by 2016, eight of the top 10 pharmaceuticals by sales will be biologics. Thus, biologics are already becoming significant players in the pharmaceutical arena. Moreover, some of the current and future MAbs and FcFPs will represent what we sometimes call ‘pipeline in a product’. This terminology is used to describe medicines that have broad medical applications such as Remicade®, which is currently approved for at least seven different disease indications (e.g., Crohn’s disease, pediatric Crohn’s disease, ulcerative colitis, rheumatoid arthritis (in combination with methotrexate), ankylosing spondylitis, psoriatic arthritis, plaque psoriasis). These data suggest to me that the challenges of developing biologics are pretty well understood and are being managed, and that biologics will continue to grow as a significant part of the pharmaceutical armamentarium.

The absolute most significant challenge for discovery and development of biologics, especially MAbs and FcFPs, is the depth of knowledge about the target biology. By and large, MAbs and FcFPs do not fail in development for many of the problems that plague small molecules, e.g., off-target activity, pharmacokinetics, chemical class-related toxicities, P450-based drug metabolism and drug-drug interactions, and oral bioavailability (there is none with these proteins). Thus, most MAbs and FcFPs that fail in development do so at the Phase IIa proof-of-concept stage (53 per cent POS), while a smaller but significant percentage fail for toxicities associated with their mechanisms of action. The reason for both of these, as mentioned above, is a lack of knowledge surrounding the target biology. Thus, as I mentioned in my answer to a previous question, if we can learn more about MAb and FcFP targets earlier, and understand the downstream biology better, we not only mitigate potential toxicological risks, but also increase our chances of getting the indications right. Most pharma and biopharma companies working on MAbs and FcFPs have made significant investments in translation medicine and biomarker research to try to link preclinical biology with human disease states. I think this approach, where applied in depth, will pay off to help more biologics achieve success in Phase IIa/Phase IIb clinical trials.

The decision of balance between small and large molecule development in any given company is really company-specific and should be made in respect to the expertise, resources and strategy that the company has in each space. Several large pharma companies are driving toward models that approach 60/40 or 50/50 small molecule to large molecule development candidate ratios, which will have two major consequences in the future. First, it will flood the clinical trials programs with new biologics against a variety of known and new mechanisms of action, and two, ultimately it will put even greater pressure on the cost of biologics to the patients, and what third-party payers are willing to cover for treatment of those diseases for which the biologics are indicated. Biologics, and particularly MAbs and FcFPs, will play an increasingly significant role in the pharma industry and in treatment of diseases in the future, and they will likely play a significant role in reshaping the healthcare industry over the next few decades.

References

1. Gamo et al., 2010. Nature 465; 305-310
2. Swinney and Anthony, 2011, Nature Reviews Drug Discovery. 10: 507-519