This action might not be possible to undo. Are you sure you want to continue?
Best practices in deal-making, valuations and strategic management
By Steven Seget
Steven Seget is Principal at Delphi Pharma, and provides independent strategic consulting services to the pharmaceutical and biotechnology industries. Steven previously managed the strategic healthcare consulting function at Datamonitor and has an MBA from the London Business School. firstname.lastname@example.org
Delphi Pharma provides strategic, financial and market–based solutions to clients, focusing primarily on the portfolio management, business development and licensing functions. Delphi Pharma combines an extensive research network, applied analytical expertise and an established track record to deliver high value results and measurable impact to its clients. www.delphipharma.com
Copyright © 2006 Business Insights Ltd This Management Report is published by Business Insights Ltd. All rights reserved. Reproduction or redistribution of this Management Report in any form for any purpose is expressly prohibited without the prior consent of Business Insights Ltd. The views expressed in this Management Report are those of the publisher, not of Reuters. Reuters accepts no liability for the accuracy or completeness of the information, advice or comment contained in this Management Report nor for any actions taken in reliance thereon. While information, advice or comment is believed to be correct at the time of publication, no responsibility can be accepted by Business Insights Ltd for its completeness or accuracy. REUTERS and dotted and sphere logos are the house trade marks of Reuters Limited in more than 25 countries world-wide.
Printed and bound in Great Britain by FPC Greenaway. Ormolu House, Crimscott Street, London SE1 5TE. www.greenaways.com
Table of Contents
Pharmaceutical Licensing Strategies
Introducing pharmaceutical licensing Licensing trends Licensing process Licensing valuations Licensing best practices
10 10 12 13 14
Introducing pharmaceutical licensing
16 17 17 19 20
The age of the partnership Definitions Report outline
Summary Introduction Headline deal trends
22 23 23 27 32 36
Licensing deal partners Licensing deal types Licensing deal subjects
Chapter 3 Summary Introduction A complex process Licensing process 42 42 43 43 45 47 49 50 52 56 57 59 60 60 61 62 64 In-licensing versus out-licensing Licensing strategy Opportunity identification In-licensing Out-licensing Licensing evaluations General portfolio management Applications for licensing evaluations Deal-making and agreement Key elements of a pharmaceutical license agreement Post-deal management and analysis Alliance management Using outside agencies Chapter 4 Summary Introduction Valuing deals Licensing valuations 70 70 71 71 73 76 77 82 88 91 Current best practices Deal-making valuation model Model inputs Evaluation modeling Model outputs Model refinements Chapter 5 Summary Introduction Licensing best practices 100 100 101 101 101 102 102 Top licensing deals of the 21st century Genentech-Roche Idenix-Novartis Millennium-Ortho Biotech iv .
AstraZeneca-AtheroGenics Preferred licensing partners Leading in-licensing companies Novartis Leading out-licensing companies Cephalon Recommendations for the future Licensing trends Licensing process Licensing valuations Licensing best practices 103 104 104 105 107 107 109 109 109 110 110 Chapter 6 Sources Index Appendix 112 112 115 116 Primary research survey v .
15: Figure 3.22: Figure 4.8: Figure 2. 2000 78 R&D lead times by phase. 2006 66 Parties involved in the valuation and negotiation of potential licensing deals.34: Figure 4.30: Figure 4.32: Figure 4.24: Figure 4.7: Figure 2. 2001-2005 39 Expected change in number of potential partners chasing each licensing deal during 2006 44 Expected change in the length of time required to complete a licensing deal during 2006 44 The pharmaceutical licensing process 46 Parties involved in identifying potential licensing opportunities.38: Figure 5.5: Figure 2.3: Figure 2.List of Figures Figure 2. 2006 74 Valuation techniques used in determining optimal licensing deal terms.37: Figure 5. phase II and phase III drugs 87 Discounted expected real values for new phase I.11: Figure 2. 2001-2005 37 Number of R&D licensing deals by development stage.1: Figure 2.35: Figure 4. 2001-2005 33 Number of top 10 biotech company licensing deals by deal type.31: Figure 4. 2006 107 Licensing trends survey respondents by company focus 112 vi .18: Figure 4. 2001-2005 35 Proportion of product-based licensing deals by therapy area. 2000 95 In-licensing partner of choice. 2001-2005 36 Number of product-based licensing deals by therapy area. 2006 104 Business development and licensing department.28: Figure 4.10: Figure 2.39: Figure 5.29: Figure 4.26: Figure 4. 2000 80 Drug market diffusion curve – product lifecycle 81 Likelihood of outcomes for new phase I. 2006 65 Parties involved in conducting due diligence for potential licensing opportunities.23: Figure 4. phase II and phase III drugs (adjusted for lower R&D cost inflation) 88 Deal outcomes for out-licensor 89 Deal outcomes for in-licensor 89 Share of expected deal outcomes by partner 90 R&D costs by phase by therapy area. phase II and phase III drugs 85 Expected real values (non-discounted) for new phase I. 2000 94 Peak sales and year of peak sales by therapy area. 2001-2005 28 Number of top 10 biotech company licensing deals by partner. 2000 79 R&D success probabilities by phase.36: Figure 5. Novartis 106 Licensing process at Novartis 106 Out-licensing partner of choice. 2006 75 R&D costs by phase.9: Figure 2. 2001-2005 30 Number of biotech out-licensing deals by partner.27: Figure 4.40: Figure 6.6: Figure 2.2: Figure 2. phase II and phase III drugs 86 Discounted expected real values for new phase I. 2006 67 Information shared between partners during licensing negotiations. 2000 94 Discounted value of sales by therapy area. 2001-2005 31 Number of top 10 pharmaceutical company licensing deals by deal type.14: Figure 3.21: Figure 4.41: Number and average value of top 10 pharmaceutical company licensing deals.33: Figure 4.17: Figure 3. 2001-2005 38 Number of biotech R&D licensing deals by development stage. 2000 93 R&D success probabilities by phase by therapy area.20: Figure 4. 20012005 24 Expected change in number of licensing deals during 2006 25 Expected change in average value of licensing deals during 2006 26 Number of top 10 pharmaceutical company licensing deals by partner.13: Figure 3.25: Figure 4.4: Figure 2.19: Figure 4.12: Figure 3.16: Figure 3. 2000 92 R&D lead times by phase by therapy area.
2006 (expressed in 2006 dollars) 83 vii .43: Licensing trends survey respondents by functional responsibility Licensing trends survey respondents by licensing responsibility 113 114 List of Tables Table 4.Figure 6.42: Figure 6.1: R&D cost by phase and peak sales.
Executive Summary 9 .
1978). value and complexity over the past 20 years or so. However. The pharmaceutical industry remains one of the most risky industries in the world. Since Eli Lilly and Genentech forged the first truly ‘strategic’ licensing agreement almost 30 years ago (Humulin. with the licensing agreement providing the best safeguard for managing. the licensing deal has been a fundamental part of every pharmaceutical and biotechnology company’s strategy. some of the inherent risks involved with pharmaceutical R&D. or at least sharing. likely to be the result of deal sizes increasing to include multiple development compounds.Executive Summary Introducing pharmaceutical licensing With falling R&D productivity and continued healthcare cost containment and generic competition pushing down the returns available for successfully launched products. 10 . It appears that the leading pharmaceutical companies have determined that the size and quality of the deal is more important than signing a greater number of deals. Given its relative infancy. there has been a consistent increase in average deal values between 2002 and 2005. only those companies able to complement internal efforts with a strong partnering strategy will be able to remain competitive over the next five-to-ten years. the strategic alliance – one that involves some level of ongoing collaboration between partners – has grown in frequency. Licensing trends A pharmaceutical company’s limited resources to manage inter-company relationships and collaborative projects places an upper limit to the number of different agreements that can formed each year.
these trends appear to confirm the leading pharmaceutical companies’ new found confidence in committing to long-term alliances. As with deal numbers and deal values. Licensing agreements in later stages. particularly in phase II. However. 11 . including monoclonal antibodies and growth factors. pharmaceutical partners. Many biotech approaches. with leading pharmaceutical companies significantly increasing licensing activity with biotech partners at the expense of intra-pharmaceutical deals. a period of rationalization in the following three years saw the leading pharmaceutical companies turn away from risky biotechnology companies and back to more traditional. but relatively less risky. before falling and steadily increasing to a new peak in 2005. However. 2005 saw a shift in licensing activity by partner. This trend is consistent with the increased maturity of the biotech sector and the emergence of biotech companies with the resources and capabilities to develop lead drugs to later stages of clinical development before seeking a pharmaceutical partner. have their most valuable applications in the treatment of cancer.2001 was at the height of the ‘golden age’ for biotechnology. where company valuations were high and every pharmaceutical company wanted to access their technologies. The greatest growth in product-based pharmaceutical licensing over the past four years has been in agreements involving cancer therapies. Again. appear to be driving growth in licensing activity over the past five years. the proportion of relationship-based alliances was also high in 2001.
Biotech companies look towards specialist agencies and key investors. Pharmaceutical companies appear to have greater levels of licensing resources and expertise inhouse. requiring greater levels of resources to be committed to the identification and evaluation of potential opportunities and partners. Deal failure is often the result of one or more partners not clearly identifying their strategic aims for a licensing deal. Pharmaceutical companies must not enter into an agreement without having determined that the licensing opportunity satisfies a real and valuable objective for the company. If you are a small biotech looking to agree a blockbuster late-stage licensing agreement with a top 10 pharmaceutical company you can either spend money on a good lawyer now to negotiate and agree a favorable contract or spend double the money later down the line having to continually defend claims from your eventual licensing partner. in order to support the licensing process. with the majority completing the entire licensing process without outside help. 12 . such as venture capitalists. but ambitious biotechnology companies have managed to do a good job with the preparation of presentation materials and the identification of target licensing partners only to make a bad job of establishing a first contact with the company and failing to make any further progress. Having a clear understanding of what you can and cannot offer potential partners is critical in order not to over promise or waste time negotiating over the wrong deal with the wrong partner.Licensing process Growth in the number of companies involved in chasing each licensing agreement will lead to increased pressure on licensing lead times. Many small.
If two companies are unable to agree upon the true value of the licensing asset and subsequently on the fair distribution of risks. with more than 80% of companies sharing at least a single point financial projection and almost 50% sharing a more detailed probabilistic evaluation. The most common evaluation technique used in determining optimal deal terms is a discounted cash flow net present value (NPV) calculation. which is used in more than 70% of companies according to surveyed licensing executives. While the approach might be similar to that used to provide recommendations for selecting the most valuable licensing opportunities to pursue the outputs of the model are quite different.Licensing valuations According to a survey of 142 licensing executives. The first test as to whether two companies are going to be successful in collaborating together as part of a strategic licensing agreement is whether or not they are able to negotiate an agreement based on significant common ground. returns and responsibilities then there is little hope that they will be able to reach a satisfactory conclusion once the real work of implementing a licensing deal begins. A deal-making valuation model is one used specifically to agree licensing terms between partners. 13 . If the inputs into the valuation model cannot be agreed upon by both licensing parties then the outputs of the exercise will not provide any common ground for negotiation. Similarly. pharmaceutical companies are more likely than biotech companies to share financial evaluations during licensing negotiations. if the modeling approach and any assumption are not clear and defendable then the effective use of the evaluation model for negotiation will be limited.
but the application of best practices must always be directed by the specifics of the deal. Novartis employs a single gateway for all opportunities allowing for improved coordination and contact management. in order to maximize the returns from its internal assets. The leading strategic pharmaceutical licensing deal. rather than reverting to a list of generalized benchmarks. particularly in the US and Japan. there are no hard-and-fast rules for successful pharmaceutical licensing. Cephalon’s licensing model of acquiring rights to marketed products in order to generate revenues cash flow to help fund the development of in-house products appears to have worked well. the company has been willing to share promotional rights for its lead products. 14 . Over the course of the 16 year relationship Roche has acquired non-US marketing rights to a range of Genentech products. The company has worked hard to position itself as a preferred licensing partner. as an out-licensor. Herceptin (trastuzumab) in 1998 and Avastin (bevacizumab) in 2003. As part of this standard review process. Novartis was considered to be the in-licensing partner of choice in 2006. Cephalon was considered to be the out-licensing partner of choice in 2006. mentioned by the greatest number of surveyed licensing executives. is the ongoing relationship between Roche and Genentech. including Rituxan (rituximab) in 1995. Lessons can be learned by looking at the leading deals and deal-makers. employing a structured process providing a quick evaluation of opportunities and the early involvement of senior management to expedite decision-making. More importantly.Licensing best practices With no two drugs or partnerships the same.
CHAPTER 1 Introducing pharmaceutical licensing 15 .
the strategic alliance – one that involves some level of ongoing collaboration between partners – has grown in frequency.Chapter 1 Introducing pharmaceutical licensing Summary With falling R&D productivity and continued healthcare cost containment and generic competition pushing down the returns available for successfully launched products. The pharmaceutical industry remains one of the most risky industries in the world. Since Eli Lilly and Genentech forged the first truly ‘strategic’ licensing agreement almost 30 years ago (Humulin. with the licensing agreement providing the best safeguard for managing. some of the inherent risks involved with pharmaceutical R&D. 16 . only those companies able to complement internal efforts with a strong partnering strategy will be able to remain competitive over the next five-to-ten years. Given its relative infancy. value and complexity over the past 20 years or so. or at least sharing. 1978). the licensing deal has been a fundamental part of every pharmaceutical and biotechnology company’s strategy.
With falling R&D productivity and continued healthcare cost containment and generic competition pushing down the returns available for successfully launched products. primary research survey results and a profile of best practices in pharmaceutical licensing in order to present a set of actionable recommendations for optimizing dealmaking.Introduction Pharmaceutical licensing strategies: Best practices in deal-making. The age of the partnership The purpose of this report is to provide a strategic perspective on the increasingly important area of pharmaceutical licensing. valuations and strategic management provides a detailed analysis of licensing strategies in the pharmaceutical and biotechnology industries. only those companies able to complement internal efforts with a strong partnering strategy will be able to remain competitive over the next five-to-ten years. Questions the report attempts to answer include: Why is licensing important in the pharmaceutical and biotechnology industries? What are the recent trends in pharmaceutical licensing activity? What are best practices in developing a robust licensing process? What can be done to optimize licensing deal values through collaborative evaluations? What does it take to become a partner of choice? What lessons can be drawn from the key successful deals formed over the past 10 years? 17 . The report draws upon deal-making trend data.
some of the inherent risks involved with pharmaceutical R&D. The pharmaceutical industry remains one of the most risky industries in the world. and the pharmaceutical industry would be in a sorry state without the breakthrough innovations licensed from biotech companies. Leading deal-tracking databases including MedTRACK and Recombinant Capital provide licensing trend data. The strategic alliance – one that involves some level of ongoing collaboration between partners – has grown in frequency. Licensing has been used to provide a more flexible mechanism through which R&D. The biotechnology industry could not exist on venture capital alone. practices and expectations regarding future deal-making trends. or at least sharing. Much of the growth in pharmaceutical licensing is linked to the emergence and development of the biotechnology industry. with deal-making activity broadly tracking funding levels and valuations for biotechnology companies. Ever since Eli Lilly and Genentech forged the first truly ‘strategic’ licensing agreement almost 30 years ago for recombinant insulin (Humulin. most important of all. sales and marketing and. A survey of 142 senior licensing executives drawn from across the pharmaceutical and biotechnology industries provides a real-time assessment of current licensing processes. Finally. 1978).In order to answer these questions the report brings together research and analysis from multiple sources. company profiles and licensing case studies provide a detailed insight into successful licensing strategies and best practices. value and complexity of pharmaceutical licensing better than a story that starts with a headline value of US$2 billion and concludes with an initial nonapprovable letter from the Food and Drug Administration (FDA). a rewrite of the licensing contract and. with the licensing agreement providing the best safeguard for managing. as a sub-plot. jail time for one the key protagonists. Nothing illustrates the importance. 18 . revenue and income streams can be balanced over time. value and complexity over the past 20 years or so. An illustration of the rise in value and complexity of strategic licensing is provided by the 2001 agreement for Erbitux between Bristol-Myers Squibb and ImClone Systems. the licensing deal has been a fundamental part of every pharmaceutical and biotechnology company’s strategy.
Licensing – the act of forming a license. In order to avoid confusion the following definitions of key terms can be applied throughout the remainder of the report: License – an agreement between two or more parties relating to the ownership of intellectual property (IP) rights. such as a co-development or co-commercialization agreement. Pharmaceutical license – an agreement as above that involves IP rights relating to a pharmaceutical technology.Definitions The term ‘strategic pharmaceutical licensing’ presents many questions regarding scope and interpretation. For the purposes of this report a pharmaceutical license is used as an umbrella term to include IP relating to both pharmaceutical and biotechnology products and technologies. development compound or marketed product. Out-licensor – the company divesting rights to a technology or product. In-licensor – the company acquiring rights to a technology or product. and in particular those rights pertaining to a technology. 19 . In-licensing – the act of acquiring the rights to a technology or product. Strategic license – an agreement as above that also involves an ongoing collaboration between partners. relating either to process or ongoing activity in forming licenses. development compound or marketed product. Deal-making – see licensing above. Out-licensing – the act of divesting the rights to a technology or product.
Report outline The report has been divided into four key sections. 20 . The licensing valuations chapter outlines a simple licensing valuation model based on independent sources to support negotiation and deal-making efforts between prospective partners. the licensing best practices chapter presents a set of detailed cases studies for successful licensing deals and leading licensing partners in order to provide key lessons for optimizing strategic pharmaceutical licensing. The licensing trends chapter presents an overview of recent trends in pharmaceutical licensing and details industry expectations for future deal-making. The licensing process chapter provides a practical guide to pharmaceutical licensing and sets out the different phases involved in optimal deal-making for both in.and out-licensors. Finally.
CHAPTER 2 Licensing trends 21 .
particularly in phase II. Many biotech approaches. This trend is consistent with the increased maturity of the biotech sector and the emergence of biotech companies with the resources and capabilities to develop lead drugs to later stages of clinical development before seeking a pharmaceutical partner. It appears that the leading pharmaceutical companies have determined that the size and quality of the deal is more important than signing a greater number of deals. However. the proportion of relationship-based alliances was also high in 2001.Chapter 2 Summary Licensing trends A pharmaceutical company’s limited resources to manage inter-company relationships and collaborative projects places an upper limit to the number of different agreements that can formed each year. but relatively less risky. before falling and steadily increasing to a new peak in 2005. 2001 was at the height of the ‘golden age’ for biotechnology. appear to be driving growth in licensing activity over the past five years. these trends appear to confirm the leading pharmaceutical companies’ new found confidence in committing to long-term alliances. Licensing agreements in later stages. 2005 saw a shift in licensing activity by partner. a period of rationalization in the following three years saw the leading pharmaceutical companies turn away from risky biotechnology companies and back to more traditional. including monoclonal antibodies and growth factors. likely to be the result of deal sizes increasing to include multiple development compounds. However. 22 . Again. The greatest growth in product-based pharmaceutical licensing over the past four years has been in agreements involving cancer therapies. where company valuations were high and every pharmaceutical company wanted to access their technologies. However. As with deal numbers and deal values. pharmaceutical partners. with leading pharmaceutical companies significantly increasing licensing activity with biotech partners at the expense of intra-pharmaceutical deals. there has been a consistent increase in average deal values between 2002 and 2005. have their most valuable applications in the treatment of cancer.
The number of licensing deals involving top 10 pharmaceutical companies has both increased and decreased at different periods between 2001 and 2005. AstraZeneca. The top 10 companies were considered to be those with the highest pharmaceutical product sales in 2005 (Pfizer. GlaxoSmithKline. Sanofi-Aventis. numerous and complex than ever before. As a result these deal values often refer to the maximum potential deal value and usually exclude any royalty payments to be paid once a drug is brought to market. Wyeth and Bristol-Myers Squibb). Merck & Co. Today’s strategic licensing deals are more valuable. Average deal values are based on the headline deal values released by partnering companies at the time of signing an agreement. and as a consequence companies must build competences in the licensing field in order to support these trends. with all licensing activity for companies acquired by the top 10 companies between 2001 and 2005 consolidated over the period.. as shown in Figure 2. While licensing activity in 2005 is the highest it has been over the five years. while the types of company involved. Headline deal trends The number and average value of licensing deals involving the top 10 pharmaceutical companies both increased between 2001 and 2005. Average deal values for the top 10 pharmaceutical companies have risen consistently between 2002 and 2005. Novartis.1. Trends in deal-making activity have shown an increase in the value and number of licensing agreements.Introduction Pharmaceutical licensing has undergone significant changes over the past 20-30 years. similar peaks were reached in 2001 and 2003. 23 . Roche. Johnson & Johnson. stage of development of the deal subject and complexity of the agreement have also changed significantly.
licensing activity in the leading companies appears to have reached a natural limit.1: Number and average value of top 10 pharmaceutical company licensing deals.Figure 2. the same 24 . It appears that the leading pharmaceutical companies have determined that the size and quality of the deal is more important than driving increases in the number of deals. As a result. This is likely to be the result of deal sizes increasing to include multiple development compounds. 2001-2005 350 300 350 285 297 273 266 304 300 250 Number of deals 200 300 250 200 Average deal value ($ million) 187 150 100 50 0 172 149 170 150 100 50 0 2001 2002 2003 2004 2005 Number of deals Average deal value ($ million) Source: MedTRACK Deals & Alliances database Business Insights Ltd By looking at an analysis of the top 10 pharmaceutical companies only. While an average of 30 deals per company is significantly more than would have been seen 20 or even 10 years ago. A company’s limited resources and abilities to manage inter-company relationships and collaborative projects places an upper limit on the number of different agreements that can formed each year. it is noticeable that there has been a consistent increase in average deal values between 2002 and 2005. we can see that the number of licensing deals that have been signed by any one company appears to have reached a plateau between 2001 and 2005. However.
Figure 2. but growth will be primarily driven by smaller companies building up their licensing capabilities and increasing their deal-making activity to the levels found in established pharmaceutical and biotech companies. It appears that there is significant room for growth in licensing activity in the pharmaceutical and biotechnology markets. as shown in Figure 2.2: Expected change in number of licensing deals during 2006 100% 90% Proportion of survey respondents 80% 70% 60% 50% 40% 30% 20% 10% 0% Pharma Biotech Other Overall Decrease significantly Decrease somewhat Increase significantly Increase somewhat Stay roughly the same Source: Business Insights Licensing Trends Survey.number of licensing partners are providing a greater level of licensing value over a greater number of licensed compounds.2. Around two thirds of the 142 licensing executives surveyed for this report considered the number of pharmaceutical licensing deals likely to increase during 2006. This increase in licensing agreements was considered likely to be higher in the biotechnology industry than in the pharmaceutical industry. 2006 Business Insights Ltd 25 .
such as generics and diagnostic products. drug delivery. such as specialty and drug delivery companies. generics and diagnostic companies. Again. as shown in Figure 2. the likely increase is considered to be higher for biotech companies than for pharmaceutical companies. Figure 2. or are in industry sectors where licensing plays a less important role.The slowest level of growth in licensing activity is expected in specialty.3. Only a very small percentage of survey respondents (just over 10%) expect deal values to decrease in 2006. Around 60% of surveyed licensing executives expect average deal values to increase in 2006.3: Expected change in average value of licensing deals during 2006 100% 90% Proportion of survey respondents 80% 70% 60% 50% 40% 30% 20% 10% 0% Pharma Biotech Other Overall Decrease significantly Decrease somewhat Increase significantly Increase somewhat Stay roughly the same Source: Licensing trends survey. 2006 Business Insights Ltd 26 . These companies are either well-established licensors. Trends in increased average deal values found over the past four years appear set to be continued in the near future.
In 2004.Alongside increases in industry deal numbers and headline values. 2005 saw a shift in licensing activity by partner. 2001 was at the height of the ‘golden age’ for biotechnology. Those companies not able to negotiate their way through these sophisticated deal terms will quickly find themselves on the wrong end of a bad deal. Licensing deal partners Between 2001 and 2005 it appears that the leading pharmaceutical companies have returned back to the biotech industry as a source for licensing. it also appears that pharmaceutical licensing deals are becoming more complex. a period of rationalization in the following three years saw the leading pharmaceutical companies turn away from the more risky biotechnology companies and back to more traditional. However. GlaxoSmithKline and Theravance signed a multi-compound. 27 . contingent equity valuations and complex territorial and market splits for co-developed and co-promoted compounds are fast becoming the norm. with leading pharmaceutical companies significantly increasing licensing activity with biotech partners at the expense of intra-pharmaceutical company deals. but relatively less risky. Sliding royalty rates. where company valuations were high and every pharmaceutical company wanted to access their technologies. cross-portfolio compounds include the 2003 agreement between Novartis and Idenix and the landmark agreement between Genentech and Roche. However. multi-therapy area deal that involved sophisticated financial put and call options on Theravance shares in support of the biotech company’s imminent initial public offering (IPO). first signed in 1990 and revised in 1995. pharmaceutical partners.4. as shown in Figure 2. Other deals involving major equity stakes and multiple.
3% 35.2% 18.Figure 2.2% 38.1% 21.0% 40.7% 18.1% Pharma 44.4: Number of top 10 pharmaceutical company licensing deals by partner. 2001-2005 350 Number of top 10 pharma licensing deals 300 Pharma 250 135 116 128 122 90 200 150 Biotech 157 106 106 100 100 50 115 Other 50 63 63 51 57 0 2001 2002 2003 2004 2005 100% Proportion of top 10 pharma licensing deals 90% 29.8% 2002 2003 2004 2005 Source: MedTRACK Deals & Alliances database Business Insights Ltd 28 .7% Other 22.6% 40% 30% 20% 10% 0% 2001 16.7% 36.7% 43.6% 80% 70% 60% 50% 45.7% Biotech 51.6% 37.
licensing activity for the top 10 biotech companies in 2005 saw an increase in the number of deals signed with other biotechnology companies. as shown in Figure 2.Over the last five years the biotechnology industry has matured to become a tried and tested source for good science and collaborative development. MedImmune and Cephalon) appear to closely mimic those found in leading pharmaceutical companies. Chiron. Genentech. Interestingly. It is clear that leading pharmaceutical and biotechnology companies are much less likely to partner amongst themselves than they are to partner with smaller companies that represent less of a strategic. As was the case for the top 10 pharmaceutical companies. Gilead. Biogen Idec. Taking a look from the other side of the industry.5. Both pharmaceutical and biotech companies stand to benefit as a result of increased licensing activity between the two – with pharmaceutical companies gaining access to innovative and valuable compounds and biotech companies receiving improved deal values and longterm collaborative partners. CSL. Serono. 29 . Genzyme. it appears that the leading biotechnology companies are not behind the increase in pharmaceutical-biotech licensing found for the top 10 pharmaceutical companies. competitive threat. with the number of deals between leading biotech companies and pharmaceutical companies decreasing consistently between 2003 and 2005. This has been helped along by a period of consolidation in the industry to allow stronger biotech companies to emerge with full pipelines and robust development capabilities. the trends for the top 10 biotech companies based on total 2005 revenues (Amgen.
5% 39.2% 28.6% Other 10.8% 12.2% 38. 2001-2005 100 Number of top 10 biotech licensing deals 90 80 70 30 26 Pharma 60 50 40 30 20 41 35 42 32 Biotech 48 34 32 33 Other 10 8 11 8 9 18 0 2001 2002 2003 2004 2005 100% Proportion of top 10 biotech licensing deals 90% 80% 70% 60% 50% 52.5: Number of top 10 biotech company licensing deals by partner.6% 0% 2001 2002 2003 2004 2005 Source: MedTRACK Deals & Alliances database Business Insights Ltd 30 .0% 40% 30% 20% 10% 44.3% Pharma Biotech 51.1% 13.Figure 2.8% 9.2% 43.9% 42.0% 43.8% 51.2% 19.
6% Biotech Pharma Source: Recombinant Capital.6.7% 41. Figure 2. with trends in licensing activity for the top 10 pharmaceutical companies showing an increase in the number of deals signed between leading pharmaceutical companies and biotech companies it appears that the growth in biotech licensing is greater for intra-biotech deals than for pharmabiotech deals.5.4% 44. other biotech partners continue to provide just over 60% of all licensing partners. It is also evident that smaller pharmaceutical companies are not yet fully embracing the renewed interest in biotech licensing found in their larger contempories.6: Number of biotech out-licensing deals by partner.9% 60. However.1% 39.4% 58. and is consistent with the figures for leading biotech companies in Figure 2.3% 56. Analyst’s Notebook Business Insights Ltd 31 . This proportion has been on the increase over the previous two years.3% 58.7% 43. as shown in Figure 2. 2001-2005 100% Proportion of biotech out-licensing partners 90% 80% 70% 60% 50% 40% 30% 20% 10% 0% 2001 2002 2003 2004 2005 41.As a proportion of all biotech out-licensing deals.6% 55.
The smaller number of simple license/acquisitions in 2005 is also representative of a down-turn in pharmaceutical mergers and acquisitions. As with deal numbers and deal values.Licensing deal types In order to better understand trends in pharmaceutical licensing agreements signed over the past five years three key categorizations were used: License/acquisition – including all agreements that involve a simple transaction involving intellectual property (IP) rights.7. which are often followed by portfolio consolidation and the divestment of overlapping products/compounds. 32 . ongoing relationship. before falling and steadily increasing to a new peak in 2005. these trends appear to confirm the leading pharmaceutical companies’ new found confidence in committing to long-term alliances. Collaboration – including all agreements involving a significant. R&D/S&M – including all agreements involving a transactional exchange of IP rights and some exchange of expertise and know-how. the proportion of relationship-based alliances was also high in 2001. either in R&D or marketing and sales. For the top 10 pharmaceutical companies. Again. there has been a move away from simple licensing/acquisition deals towards more relationship-based alliances involving collaborative R&D and sales and marketing. as shown in Figure 2.
0% 31.9% 28.3% 24.3% 23.7% 31.2% 24.2% Source: MedTRACK Deals & Alliances database Business Insights Ltd 33 .7% 11.9% 32.9% 32.0% 26.8% 7.Figure 2.3% 28.9% 14. 2001-2005 350 Number of top 10 pharma licensing deals 300 250 200 79 95 68 87 86 91 92 75 78 105 Collaboration R&D/ S&M 150 100 50 56 License/ acquisition 74 91 83 88 44 73 Others 40 34 0 2001 2002 2003 20 2004 2005 100% Proportion of top 10 pharma licensing deals 90% 30.3% 32.3% 25.3% License/ acquisition Others 13.5% 80% 70% R&D/ S&M 60% 50% 40% 30% 20% 10% 0% 2001 2002 2003 2004 2005 18.7: Number of top 10 pharmaceutical company licensing deals by deal type.6% Collaboration 34.9% 27.
the leading biotech companies have increased R&D/S&M licensing activity. As shown in Figure 2. At the same time. particularly with other smaller biotech companies.For the top 10 biotech companies. Big biotech have grown to reach a critical mass and can now avoid having to sign collaborative out-licensing agreements in order to bring their drugs to market.8. the five-year trend towards relationship licenses is similar to that found for the leading pharmaceutical companies. Interestingly. the number of collaborative agreements has remained relatively unchanged between 2001 and 2005 – likely to be a function of big biotech turning to smaller biotech to drive growth in their licensing activity over the past five years. 34 . in order to fill their pipelines and maintain a throughput in R&D. the leading biotech companies have increased the number of R&D/S&M agreements at the expense of simple license/acquisition agreements.
8% 9.3% 22.3% 20.4% 21.8% 28.3% 24. 2001-2005 100 Number of top 10 biotech licensing deals 90 80 70 60 50 40 30 20 29 24 20 23 24 15 15 15 21 17 30 20 26 25 28 30 26 Collaboration R&D/ S&M License/ acquisition Others 10 0 2001 8 11 2002 2003 2004 2005 100% Proportion of top 10 biotech licensing deals 90% 32.8% 30.3% 36.4% 37.0% 32.8% 14.6% 27.3% 28.6% License/ acquisition Others Source: MedTRACK Deals & Alliances database Business Insights Ltd 35 .9% 16.8% 29.Figure 2.7% 18.9% 31.8: Number of top 10 biotech company licensing deals by deal type.3% Collaboration 80% 70% R&D/ S&M 60% 50% 40% 30% 20% 10% 0% 2001 2002 2003 2004 2005 36.
Agreements involving infections and central nervous system (CNS) indications accounted for approximately 14% each.5% Source: MedTRACK Deals & Alliances database Business Insights Ltd The greatest level of growth in product-based pharmaceutical licensing over the past four years has been in agreements involving cancer therapies. Figure 2. 2001-2005 Cancer Infections Central Nervous System Cardiovascular and Circulatory System Autoimmune and Inflammation Metabolic/ Endocrinology Dermatology Respiratory and Pulmonary System Digestive System Blood and Lymphatic System 8.2% 4.9% 13.7% 6.9. closely matching trends found for biotech licensing in general. followed by cardiovascular and circulatory system indications and autoimmune and inflammation indications.10. cancer indications have grown to the levels found in 2001. as shown in Figure 2.6% 6. Many biotech approaches in monoclonal antibodies and growth factors have their most 36 . as shown in Figure 2.Licensing deal subjects More than 20% of all product-based pharmaceutical licensing agreements signed between 2001 and 2005 involved cancer indications.8% 20.9: Proportion of product-based licensing deals by therapy area.4% 7.0% 3.2% 5. As a share of all product-based licensing.7% 13. 2001-2005 Proportion of product-based licensing agreements.
With pharmaceutical licensing often involving agreements for the most successful R&D projects available. particularly in 2005. Figure 2. with more disappointing growth rates anticipated for CNS and cardiovascular and circulatory system therapies. Other areas of licensing growth can be found in infections and autoimmune and inflammation indications. Leading therapy areas with more disappointing licensing growth rates include CNS and cardiovascular and circulatory system. deal trends by therapy area provide a good indication of current and future R&D trends. 20012005 Proportion of product-based licensing deals 24% 22% 20% 18% 16% 14% 12% 10% 8% 6% 4% 2% 0% 2001 Cancer Infections Central Nervous System Autoimmune and Inflammation Cardiovascular and Circulatory System 2002 2003 2004 2005 Source: MedTRACK Deals & Alliances database Business Insights Ltd 37 . It appears clear that we are likely to see some significant growth in the number of late stage cancer therapies coming through the pipeline.valuable applications in the treatment of cancer.10: Number of product-based licensing deals by therapy area.
5% 37.8% 23. Figure 2.3% 20.8% 20.3% 42.2% 19.0% 38.5% 22.5% 18.7% 28. 2001-2005 100% 90% Proportion of R&D licensing deals 80% 70% 60% 50% 40% 30% 20% 10% 0% 2001 2002 2003 2004 2005 19. Licensing agreements in later stages. there has been trend in licensing activity from phase II to phase I licensing over the period 2001 and 2005. The general share of R&D licensing for phase III compounds has remained relatively unchanged over the five-year period at around 20%.2% 18.12. appear to be driving growth in licensing activity over the past five years. there has been some recent movement towards pre-clinical licensing in 2005.2% 17.0% 19.5% 16.11.3% Phase III Phase II Phase I Pre Clinical Source: MedTRACK Deals & Alliances database Business Insights Ltd Looking at biotech R&D licensing deals specifically. This trend is consistent with the maturing of the biotech sector and the emergence of biotech companies with the resources and capabilities to develop lead drugs to later stages of clinical development before seeking 38 . as shown in Figure 2. More generally.4% 34. particularly in phase II. as shown in Figure 2.By stage of development. For biotech R&D licensing deals.6% 39. there appears to have been a significant trend towards later stage deals between 2001 and 2005. the proportion of agreements signed in phase II or later has grown from less than 20% in 2001 to almost 30% in 2005.2% 21.11: Number of R&D licensing deals by development stage.7% 23.4% 19.
0% Research Source: Recombinant Capital. the value of licensing deals will inevitably increase as the risk involved with compounds found at later stages of development is reduced.3% 7.1% 11. The increased number of licensing candidates available at later stages of clinical development also helps to explain the stagnant growth in the number of agreements been signed by major companies.4% 10.2% 19.12: Number of biotech R&D licensing deals by development stage. Figure 2.9% 13. It is now possible for pharmaceutical companies to limit risk by in-licensing a single phase II compound rather than licensing two phase I compounds or a handful of preclinical opportunities.4% 23. It appears that the biotech industry is better prepared to bear the costs of unsuccessful early stage compounds alone in order to share higher rewards with its partners for successful compounds in later stages of development. As a consequence of this trend.7% 56.0% 29.a pharmaceutical partner. 2001-2005 100% 90% Proportion of R&D licensing deals 80% 70% 60% 50% 40% 30% 20% 10% 0% 2001 2002 2003 2004 2005 73.0% 63. Analyst’s Notebook Business Insights Ltd 39 .7% Phase II and above Preclinical/ Phase I 66.5% 9.4% 30.9% 59.5% 26.
CHAPTER 3 Licensing process 41 .
If you are a small biotech looking to agree a blockbuster late-stage licensing agreement with a top 10 pharmaceutical company you can either spend money on a good lawyer now to negotiate and agree a favorable contract or spend double the money later down the line having to continually defend claims from your eventual licensing partner. in order to support the licensing process. 42 . Biotech companies look towards specialist agencies and key investors. with the majority completing the entire licensing process without outside help. requiring greater levels of resources to be committed to the identification and evaluation of potential opportunities and partners. Having a clear understanding of what you can and cannot offer potential partners is critical in order not to over promise or waste time negotiating over the wrong deal with the wrong partner.Chapter 3 Summary Licensing process Growth in the number of companies involved in chasing each licensing agreement will lead to increased pressure on licensing lead times. Many small. Pharmaceutical companies appear to have greater levels of licensing resources and expertise inhouse. but ambitious biotechnology companies have managed to do a good job with the preparation of presentation materials and the identification of target licensing partners only to make a bad job of establishing a first contact with the company and failing to make any further progress. such as venture capitalists. Pharmaceutical companies must not enter into an agreement without having determined that the licensing opportunity satisfies a real and valuable objective for the company. Deal failure is often the result of one or more partners not clearly identifying their strategic aims for a licensing deal.
in-licensing companies will inevitably have to expand their initial opportunity screening numbers in order to maintain the same levels of successful deals at the end of the process.14. A complex process Pharmaceutical licensing is a complex process involving the identification and valuation of multiple potential opportunities and partners. Pharmaceutical companies must determine a clear licensing strategy. there are five key steps that must be completed on the way to agreeing a successful licensing deal. more than 25% of surveyed licensing executives expect the average time taken to complete the licensing process to increase in 2006.13. identify appropriate opportunities. evaluate those opportunities. Whether approaching licensing in order to acquire or divest intellectual property. as shown in Figure 3. Growth in the number of companies involved in chasing each licensing agreement will lead to increased pressure on licensing lead times. negotiate favorable terms and then ensure agreements are successfully implemented and managed. More than 75% of surveyed licensing executives expect the number of potential partners chasing each licensing deal to increase in 2006.Introduction The pharmaceutical licensing process is both lengthy and complex. As shown in Figure 3. 43 . With more companies chasing each deal. with greater levels of resources required for the identification and evaluation of potential opportunities and partners. with less than 20% expecting the time to decrease.
2006 Increase significantly Increase somewhat Stay roughly the same Decrease somewhat Decrease significantly Biotech Other Overall Business Insights Ltd Figure 3.13: Expected change in number of potential partners chasing each licensing deal during 2006 100% 90% Proportion of survey respondents 80% 70% 60% 50% 40% 30% 20% 10% 0% Pharma Source: Licensing trends survey.14: Expected change in the length of time required to complete a licensing deal during 2006 100% 90% Proportion of survey respondents 80% 70% 60% 50% 40% 30% 20% 10% 0% Pharma Source: Licensing trends survey. 2006 Increase significantly Increase somewhat Stay roughly the same Decrease somewhat Decrease significantly Biotech Other Overall Business Insights Ltd 44 .Figure 3.
As shown in Figure 3. 45 . some degree of evaluation must be undertaken in order to begin prioritizing those of greatest value in order to determine which opportunities should be pursued further or require more detailed scrutiny. Licensing valuations provide both a selection and deal structuring tool.and outlicensor. whereby each deal provides lessons to inform better deal-making in the future. but involves a similar process of identifying and reviewing potential compounds/ technologies (for the in-licensor) or potential partners (for the outlicensor) in order to begin approaching and evaluating those opportunities. the key stages through which companies must progress as part of the licensing process are similar. Finally. Good licensing valuations also provide an essential input into the deal-making and agreement stage of the licensing process. Identifying opportunities means slightly different things to the in. it is important that deal outcomes are regularly fed-back to those charged with the responsibility for earlier phases of the licensing process. Setting out the strategic aims for licensing is essential in determining the direction and value benchmarks for subsequent phases of the process. Licensing is an iterative process. helping to structure terms and values in order to reach a satisfactory agreement between licensing partners.15.In-licensing versus out-licensing While in-licensing or inward licensing is a very different proposition to out-licensing or outward licensing. Once a list of possible opportunities has been identified. it is important not to forget that the licensing process does not stop once signatures are added to a contract. In order to make sure all previous phases of the licensing process help to endorse good post-deal relations between licensing partners. the licensing process always begins with a review of strategy. The real hard work begins when agreements are being implemented and when problems arise that require solutions.
Figure 3. returns and responsibilities? Deal-making and agreement What are my target/ acceptable risks. returns and responsibilities? What could be done better next time? Post-deal management and analysis What could be done better next time? Source: Author’s research and analysis Business Insights Ltd 46 .15: The pharmaceutical licensing process In-licensing Where are the gaps? What can we offer? Licensing strategy Out-licensing What are our limitations? What are our needs? What do we need? Where will we find it? Opportunity identification What have I got? Who will value it? What is the value? What is the value to us? Licensing evaluations What is the value? What is the value to us? What are my target/ acceptable risks.
where investors require returns in both the long and short term. rather than rigorous strategic evaluations. Drug development is a long-term game that involves significant rewards. licensing provides companies. However. This does not mean that all deals have to be led by top-down strategic management. which from a portfolio perspective represents a sound investment when compared with the strategic alternatives. 47 . the in-licensing of a similar late stage project or the out-licensing of the failed project for development in other indications or by a company with lower ‘success’ thresholds provide real compensatory actions that serve to balance the strategic aims of the company over a short time period. Portfolio management will continue to be the catalyst for building a balanced portfolio and maintaining strategic alignment across therapy areas. companies must not enter into an agreement without having determined that the licensing opportunity satisfies a real and valuable set of objectives for the company. However.Licensing strategy Deal failure is often the fault of one or more partners not clearly identifying their strategic aims from a licensing deal. Licensing presents an integral part of today’s corporate strategy for pharmaceutical companies. but where much of the ‘low hanging fruit’ for drug discovery and development has been exploited. licensing has emerged as an effective tool for establishing alignment. particularly where changes and adjustments are required over a short time period. geographies and time-scales. large and small. In the current pharmaceutical environment. The unforeseen failure of a key product in phase III trials is difficult to compensate for with an injection of R&D investment into a given therapy area. with ways to limit costs and risk and bring about new products or revenue streams more quickly. but only after undertaking costly investments over a long time period with a great risk of failure. However. nor does it mean that there is no room for opportunistic deal-making that arises out of serendipity.
48 . The difference between the two serves as a set of objectives for future licensing activity. Once a strategic review has been completed. the resulting licensing strategy needs to be presented to and agreed upon by senior management.Formulating a good licensing strategy is really a process of determining where you are as a company today and where you would like to be in the future. Resources and financing – determining the excess or shortfall in the amount of money and resource capacity available compared with the amount required. Key areas of analysis involve: Therapy area and physician profiles – understanding in which areas a company has strengths and deficiencies based on the current portfolio and pipeline of new drugs. R&D pipeline – entering new indication or supporting new brands with follow-up drugs by accessing either additional marketing expertise or additional products in specific markets. Drug delivery/ line extensions – partnering with drug delivery companies in order to enhance the formulation of a product in order to extend its life-cycle. The licensing strategy serves two purposes: it provides internal clarity on what needs to be done. Geographical coverage – maximizing the sales of key products across geographies. such as proteomics or nanotechnology. R&D technology platforms – understanding current R&D capabilities and the potential productivity benefit of novel platforms. but also provides potential partners with the reassurance that the company is committed to a clear strategy and is unlikely to under-resource either the licensing process or post-deal implementation. particularly those in which companies have no established in-house sales force.
Saying you are young. As such companies looking to present their compounds as good licensing opportunities or themselves as good licensing partners must find new ways to credibly elevate themselves above the mass of potential opportunities available for scrutiny. Companies need to demonstrate their strengths rather than simply listing them as a proof of suitability. There are two key functions to perform – the presentation of yourself as a potential partner and the selection of appropriate profiles of potential partners. attractive and blonde but not attaching a photo will not work in attracting a partner today. the process for identifying licensing opportunities is not an exact science and is therefore fraught with inflated expectations and considerable disappointment. As in the ‘market for lemons’ – a microeconomics paper written by Nobel laureate George Akerlof about the market for second hand cars – if you cannot easily determine whether an opportunity or partner is a good one you inevitably assume it is a bad one. No matter how much effort is put into these two functions. Opportunity identification Pharmaceutical licensing is best understood using the analogy of a dating agency. 49 . which is that a highly skeptical opportunity screening process has emerged. As with personal dating. The process of opportunity identification is somewhat different for in-licensing and out-licensing. pharmaceutical licensing suffers from an unfortunate externality resulting from the inflation of expectations and disappointment.Having a clear understanding of what you can and cannot offer potential partners is critical in order not to over promise or waste time negotiating the wrong deal with the wrong partner.
However. 50 . geographies and development stages will already provide the focus for the opportunity search. Development stage and expected launch dates. geographies and development stages – are used to eliminate specific opportunities. refinement of this list will take into account a more targeted set of criteria including: Do we have the required R&D capabilities for this product? Do we have the required sales and marketing capabilities for this product? Is the product likely to be made available for licensing? Are we likely to be considered to be an attractive partner for this product? A full list of search criteria – once an initial selection has been limited to the required therapy area. Key performance objectives (once-a-day dosing.In-licensing Companies looking to identify potential compounds/technologies for in-licensing must first determine an appropriate set of search criteria with which to produce a manageable list of potential opportunities. These would include the following: Specific therapeutic classes and indications. broad criteria regarding specific market sectors. low side effects etc). Maximum and minimum projected sales performance. Sales and marketing data. Geographic availability. Identifying potential opportunities can involve significant research. but should be carried out in a logical way using the criteria identified above. reformulation etc). Having determined an agreed licensing strategy. Risk profile (first in class. available from IMS Health or through other aggregators of product sales information provides a useful tool for identifying potential opportunities.
there are significant roles to play for R&D managers and regional marketing teams. Informa’s PharmaProjects. There are a number of well-organized licensing network events and a number of websites. In order to reduce the broad list of potential licensing opportunities to a more manageable size. In this way. some of the best research sources for licensing opportunities are R&D databases such as LifeScience Analytics’ MedTRACK. such as Pharmalicensing. These standardized profiles then form the basis for a formal discussion regarding potential strategic fit. These profiles present the relevant product information alongside status details for each of the key selection criteria. a short list of suitable targets that meet these criteria can be formed. are exposed to many new developments and opportunities in the various therapy area-focused conferences and industry journals. It is still the major responsibility of any licensing or business development manager to form networks through which information about potential opportunities can quickly be shared. in particular. Aside from the desk research involving sales. Above and beyond the direct networks between licensing and business development managers. R&D managers. More detailed profiles for key drugs are also found in a number of the main databases which can further refine choices around sales projections and the likely risk involved with further development. value and prioritization. Each has the ability to refine searches by indication.Given that a high proportion of licensing occurs for development compounds. A formal review stage is undertaken using a series of licensing opportunity profiles.and R&D-based database querying. ADIS R&D Insight and IMS Health’s R&D Focus. These can often provide excellent sources for identifying new opportunities or for fleshing-out an early evaluation of a potential opportunity using the most up-to-date information. that are now providing networking short cuts in order to pass-on details of available opportunities to potentially interested parties. As mentioned earlier. any determination of opportunity value must include an introspective look at whether you as a company can offer sufficient value to the partner to make it a valuable opportunity 51 . development stage and company. the main source for identifying potential opportunities is the old-fashioned route of networking. technology. a process of prioritization must be applied.
Intellectual property status. clinical targets etc). Other types of ‘transactional’ 52 . Originator/owner.to them too. a late stage compound in phase III development is likely to yield significant returns for the out-licensor and include both co-development and co-commercialization rights. Abstract (history. companies must look to both identify and present their own opportunities (compounds/technologies) as well as identify partnership opportunities with other companies. Determining when to license a compound involves the consideration of many factors such as how much risk and financial burden is the company able to bear and how well equipped is that company to continue further development of a compound alone. Market profile (size. Deciding how to out-license a compound is often subject to the decision over when to out-license. Development status (global/ regional). Clinical profile (patient size. but is unlikely to include the co-commercialization of the compound once it reaches the market. mechanism of action. key players etc). A licensing opportunity profile for a development compound should include the following: Drug name (brand/ chemical). Before preparing presentational material for a licensing opportunity. a company looking to out-license must determine when and how to go about it. However. Licensing at an early stage of clinical development may involve some element of co-development. Out-licensing For out-licensing. Planned indication(s). potential competitive benefits etc). The prioritization of opportunities must include a realistic determination as to the likely success of subsequent deal negotiations.
a team of qualified staff need to be brought together to draft the presentational licensing material or licensing prospectus. The main sections to be included in the prospectus include: Overview/summary – a one-to-two page summary presenting the key product features and potential value. in vitro and in vivo studies and any comparative data. Intellectual process – a review of the current patent position. Experimental studies – a review of relevant data and observations from experimental work. expiry dates and any other intellectual property such as confidential information etc. For a development project this will focus on the expected clinical profile. and a more detailed confidential prospectus that is only shared after initial contact has been formed and the appropriate confidentiality agreements have been signed. pharmacology and study results including a detailed summary of completed pivotal studies. Therapeutic rationale – an outline of the mechanism of action and how and why this differs from other products.licensing with limited collaboration often occur in early stage development or involve products been divested as part of post-merger anti-trust actions etc. Having loosely determined the optimal window for licensing and the type of deal been sought. Chemistry and pharmacy – an outline of the product’s chemistry including likely formulations and manufacturing processes. Usually this involves a non-confidential dossier that can be provided to any and all interested parties. Clinical development – an outline of clinical objectives. Therapeutic potential – a review of the product’s commercial potential and how this links with a significant market opportunity. The confidential prospectus will include all relevant data available to support the product’s potential. 53 . the associated regulatory risks and the status of patent claims.
The non-confidential brochure provides a more concise product summary (five-to-six pages) aimed to encourage potentially suitable partners to make contact and find out more information. Aside from brevity. When preparing presentational materials for marketed products the key difference is the addition of more significant therapeutic and competitive potential sections. A references and bibliography section is often included to help support some of the claims made in the experimental data and clinical development sections and to provide further references to papers not referenced within the core text of the prospectus. replaced simply with ‘…more information is available under confidentiality’. The first is that some sensitive data may be removed from the latter document. Who is currently marketing. A good non-confidential brochure would also provide enough information to deter unsuitable partners from making contact and wasting time in providing further information. Market data will already be available and this will be reflected in a more robust commercial evaluation for the product in current or future potential markets. developing or in-licensing products 54 . The first filter for identifying potential partners is to look for activity in the product’s target market sectors. The brochure should include a company section that outlines scientific and commercial competences as well as key contact details. tolerance and efficacy for each indication. non-confidential brochure will require a short introduction to the company at large and not simply the product available for outlicensing.Competitive potential – a detailed profile of the product’s expected dosing. as well as likely launch dates. The second licensing opportunity process involves the active identification of potential partners. The second difference is that the wider audience exposed to the brief. there are two further differences between the confidential prospectus and the non-confidential brochure. The licensing brochure and prospectus form the primary mechanism for presenting a potential out-licensing opportunity to potential partners.
is determining the closeness of fit between the licensing opportunity and the potential licensing partner’s portfolio and pipeline strategy. The final. biotechnology companies have managed to do a good job with the preparation of presentation materials and the identification of target licensing partners only to make a bad job of establishing a first contact with the company and as a result fail to make any further progress. These potential partners must be contacted very carefully. Europe and Japan. but often the most important screening. as the wrong sort of introduction with the wrong message to the wrong person may prove disastrous to future communications and potential negotiations. such as the US. but ambitious. Identifying opportunities includes identifying the right approach to the right person within the short list of target companies. 55 . Many small. Would the product add something of value? Would the out-licensing partner be able to add something of value? Ideally. this process should provide around 20 companies to actively target as preferred partners.for the target therapy sector or physician group? Secondary considerations involve understanding which companies are strong in targeted regional markets.
However. Understanding the relative value of an underlying licensed asset is not enough to make choices among different licensing opportunities. For an in-licensor this is the most attractive product. The first involves a selection of the most appropriate and valuable opportunities. This deal-making evaluation is not covered in this section of the report. Licensing evaluations for selecting appropriate opportunities to advance to a more detailed stage of negotiation and deal-making are similar to those found as part of an internal portfolio management process. which is the contractual relationship with a third party. 56 . Valuations need to be consistent and transparent in order to provide the relative ranking and prioritization required. potential licensing agreements involve an extra level of uncertainty. the key trade-offs are to be found between sophisticated financial models and more simple rating models that include strategic elements such as portfolio fit and balance. particularly for complex licenses which might include multiple compounds. more detailed evaluation supports negotiations and deal-making. while for an out-licensor this is the most attractive partner. First of all. Once the selection process is completed a second. but is presented in detail in the licensing valuations chapter. there are two levels of valuation. compound or technology. Some expectation as to the likely deal terms available for a licensing deal is required in order to compare opportunities. Unlike portfolio management evaluations. Like portfolio management evaluations.Licensing evaluations Evaluating licensing opportunities can fast become an overly detailed and sophisticated pursuit if the right parameters are not agreed beforehand. understanding a deal’s potential terms and financial metrics at an early stage of opportunity assessment can be difficult.
General portfolio management Licensing evaluations need to provide relative values and rankings across all external opportunities, but need not necessarily provide consistent measures across both inhouse and external opportunities. However, it is desirable in a robust portfolio management process to include both internal and external potential projects in order to arrive at optimal resource allocation decisions. Parameters set-up to aid decisionmaking for internal projects and resources can easily be adapted for use in evaluating external projects.
Portfolio evaluation approaches allow decision-makers to establish preferences across projects by measuring them against the explicit objectives identified by senior management. As a result, project evaluations must include measurable criteria to assess the extent to which key objectives are likely to be satisfied. Identifying these value criteria is a key part of the project evaluation framework. Once criteria have been established, project evaluations can begin.
Project scoring approaches accommodate the fact that multiple criteria are often required in order to select and prioritize pharmaceutical projects effectively. Different scoring systems use different criteria, and include financial measures, strategic fit, competitive advantage, market attractiveness, the degree to which projects leverage core competencies, technical feasibility and risk versus return.
After a project has been scored against relevant criteria, a weighted score is calculated. This relative superiority or attractiveness score is used as a proxy for the value of the project. Once all projects have been evaluated, minimum threshold scores can be set in order to select and prioritize the most attractive projects and eliminate those of insufficient value.
Although project scoring is a key approach to project evaluation, and enables projects to be evaluated with respect to a number of different dimensions, the project rankings
derived from scoring can be misleading, which as a result can lead to sub-optimal project prioritization and decision-making. Scoring systems provide a means of capturing and quantifying project information in a consistent manner for future decision-making, but as a basis for discussion rather than a decision in itself.
Pharmaceutical companies develop approaches to portfolio management in order to allocate resources optimally with respect to both the minimization of risk and the maximization of value across the portfolio. Different projects are associated with different values and risk profiles. Although additional parameters can be incorporated, the key role for project evaluations is to provide a critical assessment of value and risk across the portfolio.
A broad range of different financial tools and evaluation techniques are available for use in project valuation. Financial analysis is able to combine an evaluation of a project’s value, risks and associated costs. However, financial approaches do not represent a viable standalone approach to project evaluation. Portfolio decision-making is characterized by the range of trade-offs made across different criteria, including meeting unmet medical need while maximizing cash-flows, balancing short-term and long-term performance, and protecting existing franchises alongside investments in new technologies. It is clear that these trade-offs cannot be adequately captured by project evaluations based solely on financial analysis. Instead, financial analysis must be employed alongside a broader scoring methodology enabling non-financial dimensions to be assessed and included in decision-making.
Although it is beyond the scope of this report, a 2005 Business Insights report, Pharmaceutical Portfolio and Project Management, discusses a range of available financial analysis tools and techniques, their application in project evaluation and their relative strengths and weaknesses. Key financial approaches include: Net present value (NPV); Decision tree analysis and expected NPV (ENPV);
Monte Carlo simulations; Real options.
Applications for licensing evaluations As mentioned previously, the key difference between an internal and external opportunity is that a licensed compound’s agreement terms will determine how risks, returns and responsibilities will be shared between parties. It is only by including an evaluation of this package of contractual terms that a potential licensing opportunity can be evaluated. Other complications for evaluating licensing opportunities include the partnering risks brought about by third-party decision-makers and the limited availability of data for generating evaluations for external opportunities.
As a check list, some of the key intellectual property (IP) licensing evaluation questions include: Who owns the IP rights (sole versus joint ownership, originator versus sub-license, exclusive versus non-exclusive etc.)? Is the patent in force and valid (have maintenance fees been paid, is claim enforceable, are there any blocking patents etc.)? What is the scope and length of the patent (how useful in preventing competitors, how does regulatory progress affect legal and economic lifespan of patent etc.)?
In evaluating a licensing opportunity it is optimal to first value the asset and then apply expected deal terms to bring about a company-specific value. As is shown in the licensing valuations chapter, a number of different parameters affect compound valuations. The costs, risks and returns associated with a compound all differ depending on the stage of development and therapy area of the specific compound. Further adjustments can be made if a specific peak sales figure can be forecast, while company specific characteristics such as size and experience can also impact on values. Consistent discount factors, applied for internal products, will allow for relative
There are many pitfalls when agreeing terms for a pharmaceutical licensing agreement. product. guarantors and active date of the agreement. Grant clauses relate specifically to what is being agreed. can be applied in order to determine valuations for each licensing opportunity. can save significant time and unnecessary confusion during negotiations. expected deal terms. reservations and improvements. Key elements of a pharmaceutical license agreement The key sections of a pharmaceutical license agreement. particularly where local. What is the technology. along with other more strategic considerations.rankings for licensing opportunities. with respect to milestones. to use. provide the basis for selecting the opportunities of greatest potential value to the company. there are a number of key considerations expressed in these legal terms that require due consideration. recitals. releases. These values. such as preambles. to import? 60 . regional and corporate representatives are involved in early discussions. royalties. a number of which are detailed below. Determining an early understanding as to which parties are signatures to the agreements. Finally. what is the specific right being granted – to sell. Recitals memorialize the background of a relationship and illustrate the importance of existing relationships and a reputation for successful licensing. The preambles identify the respective parties. read more like a legal text book than a guide to pharmaceutical licensing. grant clauses. field of use being licensed? What is the term of the agreement and over which territories? Finally. definitions. shared costs and responsibilities. However. However. Deal-making and agreement The valuations used as part of negotiations are outlined in more detail in the licensing valuations chapter. agreeing a successful deal involves more than a robust valuation and some persuasive negotiations.
grant clauses have been interrogated as part of an exchange of term sheets. pharmaceutical licensing continues to be one of the most complex and dynamic of legal pursuits.Usually. such as royalty payments etc? In light of all the potential pitfalls outlined above. but it is important that negotiations over deal terms are founded on explicit grant clauses. If IP rights are challenged what happens to other terms of the agreement. Similarly. but in the quality of the relationship between licensing partners. As such. It is particularly important that there is agreement over the assignment of obligations relating to the enforcement and protection of the licensed IP. but in the successful implementation of the deal terms over time. therefore. It is important that the ownership of any IP resulting from collaboration is made clear and that each party understands their obligations relating to protecting and promoting the licensed IP. that much of the success in the licensing process actually occurs in the period after making an agreement. often the company with the best lawyer wins. Improvements in licensing contracts play an important role. the sign of a successful deal is not in the quality of the eventual product. Post-deal management and analysis The sign of a successful strategic licensing agreement is not in the agreement and completion of the deal. like much adversarial law. particularly where ongoing relationships are anticipated. pharmaceutical companies must work hard to 61 . The good news is that there are a number of well-prepared legal firms available to help pharmaceutical companies negotiate their way through the complexities of the licensing agreement. The bad news is that. It is clear. If you are a small biotech looking to agree a blockbuster late-stage licensing agreement with a top 10 pharmaceutical company you can either spend money on a good lawyer now to negotiate and agree a favorable contract or spend double the money later down the line having to continually defend claims from your eventual licensing partner. over and above that which could have realistically been evaluated through due diligence.
AstraZeneca launched a collaboration management program for alliances in 2003. Eli Lilly were among the first to publicly promote their alliance management capabilities. Operational alliance leader – responsible for day-to-day management of alliance resources. liaising between Lilly and its alliance partner. Lilly established their Office of Alliance Management in 1998 as a corporate executive function looking to develop Lilly’s alliance strategy and provide a central point for alliance managers to learn and share experiences.develop capabilities to help manage and support their alliances and collaborative licensing relationships. To date. From an organizational perspective each Lilly alliance was assigned the following: High-level alliance champion – vice president level or above responsible for meeting the partner counterpart to discuss strategic and governance issues. the results show very different approaches to this critical issue. The collaboration management function is assigned with the responsibility for balancing AstraZeneca’s perspective with the perspective of external partners. Alliance management Alliance management is currently one of the hottest topics in pharmaceutical licensing. The function provides a sense of continuity from deal-making through to implementation and is positioned within the corporate licensing department. Companies across the pharmaceutical and biotech industries have spent the last ten years trying to determine whether it is a necessary capability and if so how it should be organized. Alliance manager – responsible for providing a supportive and catalyst function. 62 .
Alliance management responsibilities at Roche have been elevated to the senior management level and are integrated into every stage of decision-making. Roche’s global alliance management department combines key research, technical and commercial functions into a single team to create and manage alliances more effectively. Each partnership is assigned an alliance director who is charged with providing a consistent company perspective across the due diligence, deal signing and project execution phases.
The execution of alliances involves the management of two critical relationships. The collaborative relationship between partnering companies helps to provide effective alliance leadership, coordination and communication. However, the internal relationship between the licensing team and all other relevant functional departments is also a critical element in ensuring the successful implementation of each partner’s licensing objectives.
Pharmaceutical alliances fail when significant confusion or disconnection exists between groups within or between partnering companies. Internal groups must reach agreement on alliance objectives and on each group’s respective roles in achieving their objectives. Relevant functional teams must commit an appropriate share of their resources to the partnership and internal efforts must be coordinated and integrated both with the organization as a whole and with the efforts of the licensing partner.
The big test of alliance management capabilities is that when alliances fail for technical reasons do partners come back for more?
Using outside agencies
The pharmaceutical licensing process requires significant resources in order to be completed successfully. Aside from the essential services provided by lawyers mentioned earlier, a number of other outside agencies provide services that can help complete the resource intensive or complex parts of the licensing process. Outside agencies offering appropriate services in support of pharmaceutical licensing include: Management consultants; Other specialist consultancies, including licensing or drug/marketing forecasting agencies; Key investors, including venture capitalists and private equity funds for biotech; Investment banks; Accounting firms.
More than 50% of all surveyed licensing executives revealed that their companies identify potential licensing opportunities without the help of a third party agency, as shown in Figure 3.16. However, 70% of the biotech licensing executives relied on outside help to support the identification of licensing opportunities, turning to management and other consultancies, key investors and investment banks. The main recourse for pharmaceutical companies, where less than 40% of companies rely on outside help, are other specialist consultancies. These trends reveal the difference in available information for potential in-licensors and out-licensors. It appears that pharmaceutical companies, primarily comprised of potential in-licensors, are able to identify potential opportunities without significant outside help, due in large part to the promotional efforts of companies looking to out-license their compounds. For biotech companies, who are predominantly involved in out-licensing, it appears they require greater help in identifying potential partners for their technologies.
Figure 3.16: Parties involved in identifying potential licensing opportunities, 2006
70% No outside parties Proportion of survey respondents 60% 50% 40% 30% 20% 10% Accounting firms 0% Pharma Biotech Overall Management consultancies Other consultancies
Key investors (venture capitalists etc) Investment banks
Source: Licensing trends survey, 2006
Business Insights Ltd
More than 40% of the licensing executives surveyed for this report turned to third party providers when conducting due diligence exercises as part the ongoing evaluation efforts surrounding potential licensing opportunities. As shown in Figure 3.17, 60% of biotech companies used outside agencies to support licensing due diligence, primarily using specialist consultancies and accounting firms. As was the case for identifying licensing opportunities, only a small proportion of pharmaceutical companies turn to outside service providers to support their licensing due diligence.
around two-thirds of pharmaceutical companies were able to manage the valuation and negotiation of potential licensing agreements internally without outside help. as shown in Figure 3.17: Parties involved in conducting due diligence for potential licensing opportunities.18. Again. Outside agencies were used by more than 50% of biotech companies to support licensing valuations and negotiations.Figure 3. 2006 Business Insights Ltd More than 40% of surveyed licensing executives turned to third party service providers to support the valuation and negotiation of potential licensing deals. such as venture capitalists. The main agencies used include specialist consultancies and key investors. 2006 70% No outside parties Proportion of survey respondents 60% 50% 40% 30% 20% 10% Accounting firms 0% Pharma Biotech Overall Management consultancies Other consultancies Key investors (venture capitalists etc) Investment banks Source: Licensing trends survey. 66 .
Pharmaceutical companies appear to have greater levels of in-house licensing resource and expertise. 67 . 2006 70% No outside parties Proportion of survey respondents 60% 50% 40% 30% 20% 10% Accounting firms 0% Pharma Biotech Overall Management consultancies Other consultancies Key investors (venture capitalists etc) Investment banks Source: Licensing trends survey.18: Parties involved in the valuation and negotiation of potential licensing deals. subsequent due diligence and the valuation and negotiation of specific deals. such as venture capitalists.Figure 3. in order to support the identification of licensing opportunities. 2006 Business Insights Ltd Based on the results of a survey of 142 licensing executives it appears that around 60% of biotech companies and 40% of pharmaceutical companies make use of outside agencies in supporting the licensing process. Biotech companies look towards specialist agencies and key investors. with the majority completing the entire licensing process without outside help.
CHAPTER 4 Licensing valuations 69 .
Similarly. returns and responsibilities then there is little hope that they will be able to reach a satisfactory conclusion once the real work of implementing a licensing deal begins. which is used in more than 70% of companies according to surveyed licensing executives. The most common evaluation technique used in determining optimal deal terms is a discounted cash flow net present value (NPV) calculation. 70 . A deal-making valuation model is one used specifically to agree licensing terms between partners. If the inputs into the valuation model cannot be agreed upon by both licensing parties then the outputs of the exercise will not provide any common ground for negotiation. with more than 80% of companies sharing at least a single point financial projection and almost 50% sharing a more detailed probabilistic evaluation. pharmaceutical companies are more likely than biotech companies to share financial evaluations during licensing negotiations.Chapter 4 Summary Licensing valuations According to a survey of 142 licensing executives. The first test as to whether two companies are going to be successful in collaborating together as part of a strategic licensing agreement is whether or not they are able to negotiate an agreement based on significant common ground. While the approach might be similar to that used to provide recommendations for selecting the most valuable licensing opportunities to pursue the outputs of the model are quite different. if the modeling approach and any assumption are not clear and defendable then the effective use of the evaluation model for negotiation will be limited. If two companies are unable to agree upon the true value of the licensing asset and subsequently on the fair distribution of risks.
The majority of this chapter will deal with the latter valuation. much of the approach taken can be applied to the financial element of valuations made for selecting appropriate licensing opportunities to pursue. The problem with tasks that are both important and difficult is that they usually get handed over to specialists who neither understand the true importance of the task nor are able to fully deliver against the objectives set for the task.Introduction As mentioned in the licensing process chapter. However. there are two main reasons for evaluating licensing opportunities. 71 . Explaining in simple terms some of the interactions between risks and returns inherent in the deal. involving financial projections for deal-making. Valuing deals The valuation of a licensing opportunity is both an important and extremely difficult task that usually falls under the responsibility of a licensing manager more suited to making contacts and negotiations than to sitting in a dark room churning out financial projections. Presenting an estimation of potential costs and risks associated with a deal. One is to support the optimal selection amongst numerous opportunities and involves more simplistic and top-line valuations. The second reason is to support negotiations around reaching a licensing agreement and involves a more detailed financial evaluation around which specific deal terms can be agreed. Those important objectives for a licensing valuation include: Presenting an estimation of potential value for a deal. often referencing a range of strategic as well as financial measures.
there is much common ground between partners that does not require significant negotiation based on upper targets set by each party. Target and minimum acceptable deal terms are the established form of preparing negotiation parameters for pharmaceutical licenses. Usually. The key to having more productive licensing negotiations is to try to come about an agreement over the total deal value prior to 72 . Secondly. any discussion regarding common value becomes obsolete. Firstly the negotiations become needlessly adversarial. More and more pharmaceutical and biotech companies are beginning to move away from the “sit around a table and argue over numbers approach” to try tackling negotiations in slightly different way. Valuations are not shared explicitly. both potential partners produce their own separate valuation models. Producing a financial estimate that can be explained and defended by all key stakeholders. calculate target and minimum acceptable terms and then begin negotiating. It is the last two objectives mentioned above that represent the current challenge for licensing executives. Often. The problems with this approach are two-fold. who often have a financial model produced for them by some centralized financial or forecasting department. and once negotiations begin there is limited recourse to check new numbers against the valuation model. by taking separate approaches to valuing the deal.Enabling the testing of different contract terms in order to see the impact for both partners. However. It is important to make the distinction here between a valuation model used as an active negotiation tool and one that is used to set target and minimum acceptable deal terms. The ‘win-win deal’ approach is very difficult to negotiate if one company values the deal at a distinctly different level than the other. if a model cannot be shown to be both fair and clear in representing values used in negotiations with a potential licensing partner it does not provide a useful negotiation tool. Building an independent and unbiased model for representing deal values.
A little over 10% of biotech companies divulge simulated decision-model outcomes during licensing discussions with a potential partner. The key to this approach is to use independent and unbiased estimates for the major valuation parameters and to keep all financial modeling at the simplest. is a much easier exercise to embark upon.20. risks and responsibilities. pharmaceutical companies are more likely than biotech companies to share financial evaluations during licensing negotiations. only around 15% of biotech companies are engaged in simulation modeling in determining optimal licensing deal terms. However.negotiations. 73 . According to a survey of 142 licensing executives. with more than 30% of all companies failing to share any evaluation data at all during negotiations. most digestible level. As can be seen in Figure 4. along with costs. only a few leading biotech companies are likely to be engaging in such sophisticated modeling techniques and therefore most do not have this information to share.19. with more than 80% of companies sharing at least a single point financial projection and almost 50% sharing a more detailed probabilistic evaluation. Current best practices Current practices for the sharing of licensing valuations during negotiations differ by the type of company involved. the picture is very different. as shown in Figure 4. If this can be agreed upfront. The remainder of this chapter looks to set-out an approach to collaborative licensing valuations that provides both parties with a clear way of expressing how deal terms affect their share of deal values. costs and risks. then determining how that value is shared between partners. Around a third of all pharmaceutical companies share detailed decision-models during negotiations with simulated outcomes based on probabilistic parameters. For biotech companies.
20 provide the proportions of respondents from companies utilizing the various evaluation approaches for licensing. The uptake of ‘Monte Carlo’ simulations and real options is around 50% higher in pharmaceutical companies than it is in biotech companies. with each separate technique able to be applied alongside another.Figure 4.19: Information shared between partners during licensing negotiations. which is used in more than 70% of companies according to surveyed licensing executives. The results shown in Figure 4. 2006 Business Insights Ltd The most common evaluation technique used in determining optimal deal terms is a discounted cash flow net present value (NPV) calculation. 74 . there are some small differences when it comes to the use of more sophisticated approaches. While the use of NPV calculations are applied in similar proportions of companies in both the pharmaceutical and biotech segments. 2006 100% 90% Proportion of survey respondents 80% 70% 60% 50% 40% 30% 20% 10% 0% Pharma Biotech Other Overall No hard financial forecasts A single point financial forecast A decision-model simulating outcomes for both partners A detailed probabilistic forecast model Source: Licensing trends survey.
the sharing of this information with a potential partner is far greater for pharmaceutical companies than it is for biotech companies. 2006 Business Insights Ltd Interestingly.20: Valuation techniques used in determining optimal licensing deal terms. the company able to hire the best lawyer usually does disproportionately well during negotiations. As was mentioned earlier. A cynic might also add that it may be possible that the quality of these biotech evaluations would not stand up to scrutiny from a more experienced licensing 75 . One way to place a check on this imbalance for the biotech company is undertake extra evaluation work and keep it undisclosed in order to place a control on deal values as they evolve during negotiations. aside from the small difference in uptake of sophisticated modeling found between biotech and pharmaceutical companies.Figure 4. However. This means that a greater proportion of biotech companies are withholding valuation data from their partners than is the case for pharmaceutical companies. the utilization of modeling techniques to support licensing deal-making is similar. The reason may be distrust between a biotech company and the often larger pharmaceutical partner. 2006 80% Proportion of survey respondents 70% 60% 50% 40% 30% 20% 10% 0% Pharma Biotech Overall `Monte carlo' simulation models Real option valuations Discounted cash flow NPV calculations Decision-tree expected NPV calculations Simple product sales forecasts Source: Licensing trends survey.
If the inputs into the model cannot be agreed upon by both licensing parties then the outputs of the exercise will not provide common ground for negotiation. but a reputation for leaving nothing on the table for their partners will have disastrous consequences on the company’s ability to make deals in the future. The first test as to whether two 76 . the reason to invest time in building and using a deal-making evaluation model is to help bring about a set of negotiated deal terms that maximize deal values for each partner. A deal-making valuation model must demonstrate two important qualities: Independent and unbiased inputs. but the aim of every strategic licensing deal must be to find a way of maximizing deal values for both partners. Clear and defendable outputs. Similarly. in preparing useful licensing evaluations to support the licensing negotiation process. A ‘win-win licensing deal’ is an overused phrase in today’s pharmaceutical licensing lexicon. A tough negotiator armed with an even tougher lawyer might be able to squeeze the last cent of value to the benefit of their company. Either way.protagonist and are therefore shielded from any unwelcome criticism. Deal-making valuation model A deal-making valuation model is one used specifically to agree licensing terms between partners. As mentioned earlier in this chapter. there is some work to be done for most companies. particularly those in the biotech segment. if the modeling approach and any assumptions are not clear and defendable then the use of the evaluation for negotiation will be limited. the outputs of the model are quite different. While the approach might be similar to that used to provide recommendations for selecting the most valuable licensing opportunities to pursue.
companies are likely to be successful in collaborating together as part of a strategic licensing agreement is whether or not they are able to negotiate around common ground during the deal-making process.21. 2003). as shown in Figure 4. This cost estimate has become widely established as the benchmark for R&D development costs in the pharmaceutical industry and the article’s R&D data are therefore considered to be a good estimate of cost. Ronald W. If two companies are unable to agree upon the true value of the licensing asset and on the fair distribution of risks. R&D out-of-pocket costs increase significantly as a product progresses through the various phases of clinical development. Long term animal 77 . DiMasi. The article gave a total pre-approved cost estimate of US$802 million in 2000 dollars. The data was calculated using a random selection of 68 new drugs drawn from a survey of 10 pharmaceutical firms. which takes into account the cost of all the failed projects required in earlier stages to get one successful product. Figure 4. Grabowski and published in 2003 (DiMasi. Given that most strategic pharmaceutical licensing agreements are formed during R&D.21. returns and responsibilities then there is little hope that they will be able to reach a satisfactory conclusion once the real work of implementing a deal begins. Model inputs The best starting point for any dealing-making valuation model is to begin with a generic model based on independent industry inputs. Hansen and Grabowski. lead time and risk associated with pharmaceutical R&D must be determined. these adjustments need to be both agreed by each party and make a relative reference to the generic inputs put together as part of the base model.23 are taken from the Journal of Health Economics article ‘The price of innovation: new estimates of drug development costs’ written by Joseph A. Once this model has been completed a number of additional adjustments to the model can agreed upon by both licensing partners in order to bring about a more robust valuation of the licensed asset. specific inputs relating to the cost.22 and Figure 4. lead time and risk. The industry numbers presented in Figure 4. Hansen and Henry G. However.
5 86. 78 .22.studies are considered to occur in parallel with phase I and phase II trials. 2000 160 140 Cost (US$ million.0 60 40 20 15.21: R&D costs by phase. or 7. being completed before a drug begins phase III trials. Post approval trials relate to lifecycle drug development including the development of a launched drug in new indications or new formulations. 2000) 120 100 80 140. The greatest amount of development is spent in late stage phase III trials.3 5.2 0 Phase I Phase II Phase III Long-term animal Post approval Source: DiMasi. These figures are based primarily on receiving approval from the Food and Drug Administration (FDA) in the US. but an additional 18 months are added once a marketing application is made before a drug receives final approval. Hansen and Grabowski (2003) Business Insights Ltd As shown in Figure 4. as well as any necessary post approval regulatory trial requirements. the total lead time for a drug from entering human trials to being approved for launch is approximately 90 months.2 23.5 years. Figure 4.
79 . At a late stage of clinical development only those drugs with the greatest chance of progressing through the stage and eventually being approved for marketing are entered into phase III trials. 2000 40 Mean time to next phase (months) 35 30 25 20 33.23 shows the average attrition rates for a typical pharmaceutical product as it progresses through human testing and the approval stage. Hansen and Grabowski (2003) Business Insights Ltd Figure 4.Figure 4. the greatest level of attrition occurs before a drug enters expensive phase III trials.2 Approval phase Source: DiMasi.8 15 26.22: R&D lead times by phase.3 18. As would be expected.0 10 5 0 Phase I Phase II Phase III 12.
John Vernon and Joseph A. Mean average peak sales are skewed by the disproportional impact of the small number of blockbuster drugs generating significant 80 . lead times and risks. The drug market diffusion curve shown in Figure 4. Vernon and DiMasi.Figure 4.4% 10% 0% Phase II Phase III Approval Source: DiMasi. 2002).0% 68. It is assumed that market sales erosion following loss of patent occurs at a rate of 50% per annum. There are a number of different approaches to including sales forecasts for pharmaceutical drugs. The diffusion curve assumes a 14 year post-launch patent life and peak sales occurring in year 10. The article estimated mean average peaks sales for a pharmaceutical drug of US$458 million in 2000 dollars. 2000 Probability of entering phase (from previous) 80% 70% 60% 50% 40% 71. DiMasi published in 2002 (Grabowski.24 has been adapted from the Pharmacoeconomics article ‘Returns on research and development for 1990s new drug introductions’ written by Henry Grabowski. but perhaps the simplest is to use a peak sales number and then apply a diffusion curve representing the likely uptake and eventual erosion of sales across the product lifecycle.5% 30% 20% 31.23: R&D success probabilities by phase. a drug valuation also requires a determination of likely sales returns for a pharmaceutical product. Hansen and Grabowski (2003) Business Insights Ltd Alongside an analysis of R&D costs.
24: Drug market diffusion curve – product lifecycle 100% 97% 97% 94% 94% 88% 88% 75% 75% 100% 90% Proportion of peak sales 80% 70% 60% 50% 40% 30% 20% 10% 3% 6% 0% 1 2 3 4 5 6 25% 50% 38% 19% 13% 9% 5% 2% 1% 7 8 9 10 11 12 13 14 15 16 17 18 19 20 Year (post-launch) Source: Author’s research and analysis Business Insights Ltd 81 . the direct costs for sales were calculated as a proportion of sales expected in the following two years. particularly around the launch period for a new drug. a contribution margin of 45% is used in order to reflect both the cost of sales and direct selling costs (amounting to 55% of total sales). Figure 4. the very real scenario of investing significant marketing spend on a potential new product only to receive a final ‘non approvable’ letter from the FDA can captured by the model. In order to reflect the lead time involved in marketing and sales. and as a result the mode (50th percentile) average peak sales are much lower than the mean. As a result. When including the direct costs associated with sales.peak sales revenues.
In order to arise at 2006 cost and revenue levels a general US dollar inflation rate must also be included alongside the price increases in constant dollar amount. It is clear that clinical trials have gotten larger over the past 10 years or so and this increase in costs seems broadly reasonable. For example. Vernon and DiMasi. 2002.Evaluation modeling The first stage of evaluation modeling is to transpose all input values into current values.9% per annum in constant 2000 dollar values. key trends in costs and values were extrapolated from the two key source papers – DiMasi. This will allow evaluation results to be expressed in a common value measure (namely 2006 US dollars). The resulting sum is as follows: 2000 phase I trial cost (2000 dollars) x (1 + 11.8% + 3%) ^ 6 = 2006 phase I trial cost (2006 dollars) 82 . and was estimated to be a further 3% per annum.8%) as well as accounting for the change in dollar value between 2000 to 2006 dollars (3% per annum). 2003 and Grabowski. In order to do this we first have to arise at 2006 values and then account for any changes in the real cost and value of real inputs. The smaller rise in sales compared with drug development costs seems reasonable given the continued rise of healthcare cost containment measures across major markets and the impact of aggressive generic competition on prices. Hansen and Grabowski. average peak drug sales have increased at a rate of 2. the 2006 cost of a phase I trial expressed in 2006 dollars will involve calculating 6 years of price increases as a result of increased trial costs (11. In order to account for the changes in R&D costs and likely drug values between 2000 and 2006. Over the ten year period leading up to 2000 it was shown that clinical development costs have increased at a rate of 11. Over the same ten-year period.8% per annum in constant 2000 dollar values.
0% 3. 83 .5 11.0 458.0% 3.0 Year-on-year increase in cost/ revenue 11.0% 2006 (expressed in 2006 dollars) 34.5 645. It is important to point out that the base case model must involve cost and revenue projections that account for real increases over time and are expressed in a common value measure (e.9% Year-on-year increase in dollar value 3. all costs and revenues are expressed in a common. all figures can be added together to represent true total value.g.2 23.0% 3. constant value measure and reflect the real levels of increase in costs and values that will occur over time.8% 11. In this way.9 320.8% 11.2 140. 2006 (expressed in 2006 dollars) 2000 (expressed in 2000 dollars) 15.8 53.8% 2. The baseline valuation for a licensing deal should have a real discount rate of 0%. without needing to discount values in order to adjust for inflation.8% 11.1 R&D costs Phase I Phase II Phase III Long term animal Post approval Peak sales All figures in US$ millions Source: Author’s research and analysis Business Insights Ltd Once the 2006 levels for all key costs and revenues have been established the real increase in these values over time must also be added to the model. By having a simple model accounting for real price increases only there is no need to discount values to begin calculating value.Table 4.9% per annum respectively) must also be extrapolated. 2006 dollars) in order to allow a real valuation aggregation. Most approaches fail to separate inflation from a general level of nominal discount rate and the resulting model is needlessly complicated.5 86. Costs and revenues remain in 2006 dollar values but the real increases in costs and revenues (11.8 197.1: R&D cost by phase and peak sales.3 5.8% and 2.8% 11.0% 3.0% 3. In this way.
adjusted for risk. given that most approaches to pharmaceutical drug evaluations make use of a decisiontree or expected value approach it appears better that specific development phases are actually discounted based on their technical risk rather than using a common projectwide discount rate across all phases.g. It is important to recall that net present value (NPV) measures are comparator measures and not actual values. success in phase I but failure in phase II etc) can be weighted by their 84 . failure in phase I. potential licensing partners need to relieve themselves from the need to discount value and instead look at the real value being created. the greater the discount rate threshold. It is better to allow a pharmaceutical company to compare different internal rates of return for projects than to simply reply on a comparison of value with a mythical alternative investment project (opportunity cost). It would be difficult to be more provocative than stating that net present value is the wrong tool for valuing licensing agreements. This separation between valuing a project rather than ranking it amongst its comparators is particularly important when dealing with licensing projects. the main reason for discounting values is to account for risk – the greater the risk. If for a pharmaceutical investment I normal expect a rate of return equivalent to 10% per annum. but rather the project is likely to return US$10 million less than the opportunity cost of capital (an average investment in a similar project). If the resulting NPV for a project is a loss of US$10 million. different outcomes (e. given that the discount rate used by one company can be quite different the one used by another. this does not mean the project has no value. particularly given that we have established that more than 70% of companies are using NPV to support licensing evaluations. then discounting a pharmaceutical investment by the 10% factor in order to arrive at an NPV only represents the value of the project over and above that expected on average from similar investments. and then determine how this might be shared between parties. As well as adjusting for price inflation. In this way. However. NPV is a decision-making tool rather than a valuation tool.When calculating net present values. However. future values are discounted in order to represent the opportunity cost of investing in an alternative but similar investment.
The likelihood of specific outcomes for a new phase I. phase II and phase III drugs 80% 70% Likelihood of outcome 60% 50% 40% 69% 69% 30% 49% 20% 29% 32% 7% 10% 15% 22% 10% 0% Phase I only Phase 2 only New phase I Phase 3 only Launch New phase II New phase III Source: Author’s research and analysis Business Insights Ltd The non-discounted real values for each potential drug outcome are shown in Figure 4. phase II and phase III drugs can be calculated. only those outcomes that eventually result in a successful drug launch have a positive real value in constant 2006 US dollars. Obviously. Figure 4. phase II and phase III drug are shown in Figure 4.likelihood and eventual outcomes can be used to calculate a weighted average of all possible outcomes based on the risk of project failure.25.25. the expected real values for new phase I. Based on these expected 85 . while a new phase III drug has a 69% chance of reaching the market. When weighted against the likelihood of different outcomes. It is clear that an NPV projection with a common discount rate would do a bad job in capturing the differences between these two investments. A new phase I drug has only a 15% chance of reaching the market.25: Likelihood of outcomes for new phase I.26. as shown in Figure 4.
26: Expected real values (non-discounted) for new phase I.outcomes a new phase III drug is around four times as valuable as a new phase II drug and around seven times as valuable as a new phase I drug. 2006) 2. the resulting valuation represents the expected additional value added by the licensing opportunity (e.27.000 464 1.500 2.000 Drug value (US$ million. phase II and phase III drugs 3. US$271 million for a new phase I drug) in constant 2006 US dollars. These results show that at a real discount rate of 6% the NPV of a new phase I drug falls below zero.500 1.g. Given that pharmaceutical companies often base R&D investment decisions on a real NPV discount rate of around 8-10% these numbers clearly do not 86 . the expected discounted real values for drugs at different development stages are shown in Figure 4.000 1. while at a discount rate of 8% the NPV of a new phase II drug falls below zero. Given that the risks associated with drug development are accounted for in the expected value calculations and that any market risks have been averaged out into a single point peak sales forecast.797 500 0 -500 Phase I only Phase 2 only Phase 3 only Launch 271 Expected value New phase I New phase II New phase III Source: Author’s research and analysis Business Insights Ltd In order to illustrate the impact of different discount rates on values. Figure 4.
600 1. phase II and phase III drugs Expected drug value (US$ million. Figure 4. However.000 800 600 400 200 0 -200 0% 1% 2% 3% 4% 5% 6% 7% 8% 9% 10% Real discount rate (annual) phase 3 phase 2 phase 1 Source: Author’s research and analysis Business Insights Ltd For example.800 1. a similar rate to the increase in likely revenues. the real annual increase in clinical development costs could be reduced. the impact on NPV calculations is significant. Either the NPV model fails to accurately capture the risk associated with pharmaceutical R&D or some of the base model inputs would need adjusting to ensure the results appear reasonable.8% per annum.28 shows the impact on the discounted expected real drug values of changing the rate at which costs escalate over time.8% to just 3. Both new phase I and phase II compounds show positive NPVs with real discount rates of up to 10%. for the purpose of this illustrative example the rate of clinical cost increase has been kept at 11. in order to ensure the model results appear more reasonable.400 1.0%. This check for reasonableness can help to inform modeling and bring about agreement between licensing partners. 87 . 2006) 1. If the rate of increase in the real cost of clinical trials is reduced from 11.27: Discounted expected real values for new phase I.stack up. Figure 4.200 1.
The deal terms modeled involve an upfront payment to the out-licensor of US$20 million.500 2.29 and Figure 4.000 750 500 250 0 0% 1% 2% 3% 4% 5% 6% 7% 8% 9% 10% Real discount rate (annual) phase 3 phase 2 phase 1 Source: Author’s research and analysis Business Insights Ltd Model outputs While an understanding of total deal values are generated by the base case model.250 1. royalties based on gross sales of 10% are payable to the out-licensor.750 1. 88 . negotiations around licensing agreements are informed by looking at how that value is shared between potential partners with respect to risks. As an example. a new phase II product has been used to illustrate the potential returns for each partner.Figure 4.500 1. 2006) 2.000 1.750 2. Upon the drug reaching the market. phase II and phase III drugs (adjusted for lower R&D cost inflation) Expected drug value (US$ million. The first model outputs are shown in Figure 4. a milestone payment of US$40 million for successfully completing phase II trials.28: Discounted expected real values for new phase I. a further US$60 million for completing phase III trials and US$100 million for gaining regulatory approval.250 2. returns and responsibilities.30 and outline all potential cash flows for each partner over the lifecycle of the licensing agreement.
2006) 80 60 40 20 0 1 2 3 4 5 6 7 8 9 1011 12 13 1415 16 17 1819 20 21 2223 24 25 2627 28 -20 -40 -60 Source: Author’s research and analysis Business Insights Ltd Out-licensor license payments/ royalties Out-licensor costs/ sales Year Figure 4.29: Deal outcomes for out-licensor 120 100 Cost/ sales (US$ million. 2006) 400 300 200 100 0 1 2 3 4 5 6 7 8 9 10 1112 13 1415 16 1718 1920 21 2223 2425 26 2728 -100 Year -200 -300 Business Insights Ltd In-licensor costs/ sales In-licensor license payments/ royalties royalties Source: Author’s research and analysis 89 .Figure 4.30: Deal outcomes for in-licensor 600 500 Cost/ sales (US$ million.
200 1.Understanding potential cash flows over the lifecycle of the licensed compound does not inform each party as to its expected share of potential value.600 Drug value (US$ million. However.400 1.31: Share of expected deal outcomes by partner 1. In the illustrative example the maximum out-of-pocket cash flow for the in-licensor is more than double that of the out-licensor. Figure 4. as shown in Figure 4. Using the same licensing terms as outlined above the respective shares of expected licensing value can illustrated. 2006) 1. As with the total licensing opportunity evaluation. expected values are calculated using a weighted average of all potential project outcomes.31. These expected values allow for a negotiated agreement over the sharing of risk and expected values.000 800 600 400 200 0 -200 -400 Phase 2 only Phase 3 only Out-licensor Launch In-licensor Expected value 200 264 Source: Author’s research and analysis Business Insights Ltd 90 . in order to reward bearing a greater share of development risk the out-licensor is able to secure a greater share of the sales returns for a successful product and as a result a greater share of the expected overall licensing agreement value.
as shown in Figure 4. by in-licensors.Having an understanding of both potential licensing cash flows and expected. Henry G. with 91 .32. largely unsubstantiated commercial claims by out-licensors are always going to be ignored. Grabowski and John Vernon and published in 2004 (DiMasi. R&D costs vary significantly. As a final word on deal-making licensing valuations. independent value parameters and using a flexible but clear approach to modeling licensing opportunities helps potential partners to facilitate dealmaking negotiations around common estimates of value and not around pre-determined target deal terms prepared in isolation by each party. However. Therapy area specific data for analgesic/anesthetic. for companies to exhaust all industry-wide valuation parameters before beginning to add drug-specific data relating to the potential claims of the drug.33 and Figure 4. an independent valuation should be provided by the out-licensor as part of any detailed licensing prospectus and any challenges to the model’s validity should be dealt with as a priority. DiMasi. A collaborative valuation will only prove to be a useful negotiation tool where both licensing parties have confidence in the valuation and are fully committed to the process. Model refinements A base case valuation model is used as a starting point to arrive at mutually agreeable licensing deal valuations based on unbiased sources. Grabowski and Vernon. therefore. the likely valuation of a drug varies significantly by development stage. R&D costs. cardiovascular and central nervous system (CNS) drugs were presented in the Drug Information Journal article ‘R&D costs and returns by therapeutic category’ written by Joseph A. As has already been shown in the base model. it is important to generate ‘buyin’ for the collaborative valuation and negotiation process as early as possible. antiinfective. Agreeing on broad. riskadjusted licensing values provides a benchmark for constructive licensing negotiations. Figure 4. other key value parameters include therapy area and company profile.34. lead times and risks all vary by therapy area. or at best challenged. 2004). Ideally. Single point. It is important.
Grabowski and Vernon (2004) Business Insights Ltd Drug development lead times are greatest for CNS drugs. 2000) 160 140 120 100 80 60 40 20 0 Phase I All Analgesic/ Anesthetic Phase II Anti-infective Phase III Cardiovascular CNS Source: DiMasi. 2000 Development costs (US$ million. which are almost double those found for analgesic/anesthetic and anti-infective drugs.33. Both CNS and cardiovascular drugs take significantly more time for regulatory review.phase III trials for anti-infective drugs (which include large scale HIV and hepatitis drug trials) more than twice as expensive as average analgesic/anesthetic and CNS drug trials.32: R&D costs by phase by therapy area. as shown in Figure 4. Figure 4. 92 . largely as a result of the increased complexity of many of the indications involved in these therapy areas.
0 15.4 12.0 22.1 46. phase II anti-infective drugs are slightly more likely to progress through into phase III trials than drugs from other therapy areas. as shown in Figure 4.34. However.1 Clinical phase Approval phase Source: DiMasi. 2000 Mean time to complete phase (months) 140 120 100 80 60 40 20 0 All Analgesic/ Anesthetic Anti-infective Cardiovascular CNS 72.5 61. while phase III analgesic/anesthetic drugs are more likely to receive final approval than other drugs. 93 .33: R&D lead times by phase by therapy area.2 21.Figure 4.5 18.4 50.5 92. Grabowski and Vernon (2004) Business Insights Ltd The probabilities of progressing through clinical phases do not vary significantly between therapy areas.
2000 90% Probability of entering phase (2000) 80% 70% 60% 50% 40% 30% 20% 10% 0% Phase II All Analgesic/ Anesthetic Phase III Anti-infective Approval Cardiovascular CNS Source: DiMasi.34: R&D success probabilities by phase by therapy area.35: Peak sales and year of peak sales by therapy area. Grabowski and Vernon (2004) Business Insights Ltd Figure 4.Figure 4. 2000 900 Peak year sales (US$ million. Grabowski and Vernon (2004) Business Insights Ltd 94 . 2000) 800 700 600 Year 10 Year 12 Year 9 500 Year 9 400 300 Year 6 200 100 0 All Analgesic/ Anesthetic Anti-infective Cardiovascular CNS Source: DiMasi.
000 5. The greatest peak sales revenues are generated by CNS drugs. Peak sales for cardiovascular drugs are almost twice the sales for an average anti-infective drug. 2000) 7.36. Grabowski and Vernon. as shown in Figure 4.000 2.000 0 Mean All 25th percentile Median 75th percentile Cardiovascular CNS Analgesic/ Anesthetic Anti-infective Source: DiMasi. CNS drug sales are highly skewed by a small proportion of CNS drugs generating significant revenues. 2000 8. including multiple blockbuster products for the treatment of depression and psychosis.000 6.35.000 1. Cardiovascular drugs appear to promise the greatest minimum value. Grabowski and Vernon (2004) Business Insights Ltd 95 . occurring in year 12. generating the highest NPV at the 25th percentile level.A range of peak sales by therapy area are also presented in DiMasi. 2004. Cardiovascular peak sales occur slightly later in the product lifecycle to other drugs.000 NPV sales (US$ million.000 3. which generated sales almost twice that of pharmaceutical drugs on average. As shown in Figure 4.36: Discounted value of sales by therapy area. Figure 4.000 4.
the likelihood of an alliance drug progressing through to phase III is 9 percentage points higher than when an 96 . 2005). with hormonal preparations having the highest probability of success (78%). which require perfecting dosing and generating statistical evidence for efficacy across large patient samples. In phase II. However. The results showed that indications developed in an alliance are no more likely to progress through phase I trials than those developed independently. 2005 presented further trends relating to the overall effect on development success for drugs developed within an alliance rather than by one single company. respiratory indications have the lowest predicted probability of being approved (30%). The likelihood that an indication will complete phase III trials successfully for a company that has previously participated in only one phase III trial is just 51% compared with 81% for a firm that has developed 30 phase III drugs. However. For drugs in phase I. Danzon. Nicholson and Pereira. Nicholson and Pereira. alliance drugs are significantly more likely to complete phase II and phase III than those developed by a single company. Danzon. The likelihood of phase II success for a company that has previously participated in only one phase II trial is 69%.Further differences by therapy area were presented in the Journal of Health Economics article ‘Productivity in pharmaceutical-biotechnology R&D: the role of experience and alliances’ written by Patricia M. There were no discernable trends between experience and success for phase I trials. which is perhaps consistent with the idea that experience only really impacts on large. Data taken from over 900 pharmaceutical firms showed that the predicted probability that an indication in phase III trials will be approved by the FDA is 22 percentage points below the sample average for CNS drugs and 21 percentage points above the average for hormonal preparations. complex trials in phase II and phase III. 2005 also presented some key trends relating to the experience of the developing company. Danzon. Sean Nicholson and Nuno Sousa Pereira and published in 2005 (Danzon. this likelihood of success increases to 81% for a firm with experience in 25 phase II compounds. Nicholson and Pereira.
97 .indication is developed independently. In phase III. co-developed indications have a 14 percentage point higher likelihood of success.
CHAPTER 5 Licensing best practices 99 .
Novartis employs a single gateway for all opportunities allowing for improved coordination and contact management. as an out-licensor. employing a structured process providing a quick evaluation of opportunities and the early involvement of senior management to expedite decision-making. the company has been willing to share promotional rights for its lead products. Herceptin (trastuzumab) in 1998 and Avastin (bevacizumab) in 2003. Cephalon’s licensing model of acquiring rights to marketed products in order to generate revenues cash flow to help fund the development of in-house products appears to have worked well. Cephalon was considered to be the out-licensing partner of choice in 2006. More importantly. Over the course of the 16 year relationship Roche has acquired non-US marketing rights to a range of Genentech products. As part of this standard review process. 100 . particularly in the US and Japan. including Rituxan (rituximab) in 1995. mentioned by the greatest number of surveyed licensing executives. is the ongoing relationship between Roche and Genentech. there are no hard-and-fast rules for successful pharmaceutical licensing. in order to maximize the returns from its internal assets.Chapter 5 Summary Licensing best practices With no two drugs or partnerships the same. The company has worked hard to position itself as a preferred licensing partner. Lessons can be learned by looking at the leading deals and deal-makers. The leading strategic pharmaceutical licensing deal. rather than reverting to a list of generalized benchmarks. Novartis was considered to be the in-licensing partner of choice in 2006. but the application of best practices must always be directed by the specifics of the deal.
First signed in 1990. Lessons can be learned by looking at the leading deals and deal-makers. Genentech-Roche The leading strategic pharmaceutical licensing deal. An understanding of key licensing trends. understanding the most important objectives for each deal-making process will always be the value added by a capable and experienced licensing manager. outline trends towards long term. with no two drugs or partnerships the same. the major objectives underpinning licensing processes and the success stories to be found in recent licensing deals will help to inform and support strategic licensing. rather than a list of generalized benchmarks. all mentioned as leading strategic licensing agreements by surveyed licensing executives. multi-product alliances and a shift in power towards the biotech out-licensor. and renewed in 1995. Top licensing deals of the 21st century The following four deals. The Genentech-Roche and Idenix-Novartis agreements both illustrate the profound effect a long term agreement can have on the pipeline of the in-licensor and the financial security of the out-licensor. The Millennium-Ortho Biotech and AstraZeneca-AtheroGenics deals illustrate the new found confidence of biotech companies with strong late stage products to hold out for the best possible deal. but the application of best practices must always be directed by the specifics of the deal.Introduction Licensing best practices provide guidance and benchmarks for successful deal-making. mentioned by the greatest number of surveyed licensing executives. a licensing agreement gave Roche a 10-year option to acquire the ex-US rights to Genentech’s developmental 101 . However. However. there can be no hard-and-fast rules for pharmaceutical licensing. is the ongoing relationship between Roche and Genentech.
similar to the Roche-Genentech deal. Millennium managed to retain all US commercialization rights. Idenix-Novartis Another licensing deal of note. Independence in a biotech company. as well as double digit royalties for 102 . the decision to keep Genentech an independent company has remained the critical success factor for the 16-year agreement. Novartis was able to use this deal to rapidly accelerate the development of its antiviral therapeutic franchise. However.drugs once they have completed phase II trials. the deal also promised significant milestone payments to Idenix in order to fund the continued development of the two hepatitis B drugs along with further development activity in other areas. Aside from turning a profit of around US$2 billion through the acquisition and divestment. Successful products arising out of the deal will be co-promoted by both partners in the US and Europe. As a result of this relationship Roche has acquired Rituxan (rituximab) in 1995. this position was soon reversed and by the beginning of 2000 Roche had sold 40% of Genentech’s shares back to the public. Primarily formed around two clinical stage hepatitis B products. mentioned by several surveyed licensing executives. is critical for maintaining a creative and innovative partner for new drug development. is the 2003 agreement between Novartis and Idenix. Millennium-Ortho Biotech A recent deal that truly illustrates the new-found bargaining power of biotech companies with late stage compounds is the 2003 deal between Millennium and Ortho Biotech (Johnson & Johnson) for cancer therapy Velcade (bortezomib). the deal also included the right for Novartis to access all suitable Idenix drugs reaching phase II trials. Significantly. In 1999. Roche exercised its option to buy Genentech outright. As well as providing Idenix’s private shareholders with significant funds (51% of the company was acquired for US$255 million). Herceptin (trastuzumab) in 1998 and Avastin (bevacizumab) in 2003. both in management and the freedom to form partnerships with other companies.
centering on partnering for the ex-US rights only. was finally selected in an attempt to retain medium to long term upside as well as balance the strategic goals of both partners. In late 2002. A third deal structure. following an earlier agreement for AGI-1067 with Schering-Plough in 1999. Millennium was then able to find a sophisticated deal structure to mirror the specific business objectives of each partner. the first of which was a copromotion agreement in the US alongside a profit-sharing arrangement on Velcade and one of the partner’s products. It was against this significant level of competition that Ortho Biotech was able to secure the deal. The anti-inflammatory cardiovascular drug was already in phase III trials when the agreement was made. Millennium collected five term sheets and three deal structures. Millennium maximized its returns from Velcade by advancing development and bringing multiple potential partners to the table. but the agreement also made a clear distinction between future development responsibilities which were divided by tumor rather than by geography. conducting CEO-level talks with more than ten different companies. The agreement includes potential royalty rates in the mid-30s depending on cumulative 103 . which could be worth up to US$1 billion plus royalties. Millennium considered three different deal structures. The second deal involved co-promotion in the US with Millennium retaining over 50% of profits. Not only did the deal result in a clear split between marketing rights. Once detailed negotiations with Ortho Biotech were underway. AstraZeneca-AtheroGenics Another recent example of a biotech company holding out for a good deal is AtheroGenics’ recent 2005 agreement with AstraZeneca for atherosclerosis treatment AGI-1067. Having narrowed down the list of potential partners in early 2003. The deal comes after AtheroGenics reacquired full rights for the drug back in 2001.European sales and up to US$500 million in upfront and milestone payments. another two parties re-established their interest. Millennium was speaking to more than twenty different interested parties about Velcade. Following Velcade’s approval in May 2003.
Other leading in-licensing partners include Pfizer. Johnson & Johnson (J&J) and Roche. Interestingly. 2006 Number of survey respondents Novartis Pfizer J&J Amgen Roche GSK Eli Lilly AstraZeneca Merck Sanofi-Aventis 1 2 3 4 5 6 6 7 9 14 Source: Licensing trends survey. illustrating the biotech company’s affinity with smaller biotech companies as a preferred alternative partner to big pharma. 2006 Business Insights Ltd 104 . as shown in Figure 5. Figure 5.37. Amgen was the fourth ranked in-licensing company.37: In-licensing partner of choice. Preferred licensing partners Leading in-licensing companies A survey of 142 licensing executives revealed Novartis to be the in-licensing partner of choice in 2006.sales levels as well as a 125-person dedicated sales force funded by AstraZeneca to copromote the drug.
Novartis Aside from the Idenix agreement detailed above. http://www.39 shows the licensing process as described in the company’s ‘Novartis: Your partner of choice’ presentation1. Novartis has worked hard to position itself as a licensing partner of choice. Anadys Pharmaceuticals for an early stage hepatitis C treatment. Arakis and Ventura for a phase II chronic obstructive pulmonary disease (COPD).ppt 105 . Arrow Therapeutics for an early stage respiratory syncytial virus (RSV) treatment. Its website includes a downloadable presentation explaining the company’s key qualities as a licensing partner. SeBo for an early stage renal disease treatment.com/downloads_new/bizdevelopment/Novartis_Presentation. Figure 5. Recent deals signed in 2006 include: An agreement with Human Genome Sciences for hepatitis C interferon drug Albuferon. Otsuka for a phase III treatment for dry eye. highlighting a structured process providing a quick evaluation of opportunities and an 1 ‘Novartis: Your partner of choice’ presentation. Novartis signed deals with: Senju Pharmaceutical for an early stage glaucoma treatment.38 shows the structure of the business development and licensing team at Novartis and Figure 5. An agreement with Servier for a novel phase III antidepressant. set to enter phase III trials in late 2006. In 2005.novartis. Novartis has stepped up its licensing activity over the past five years. The website also includes a clear explanation of the process for licensing employed at Novartis.
Csendes Search/Evaluation G. Weston Mature Business A. Strub Japan I. Ceulemans Research Alliances S. Novartis employs a single gateway for all opportunities allowing for improved coordination and contact management.M. Cupit Partnering H. Ohhashi Project Management J. Saxena Ophthalmology S. Hartmann USA. Séquier Development Alliances/ Out-Licensing I. Grannatt Oncology G. Novartis Global BD&L V. Asia Finance M.early involvement of senior management to expedite decision-making. As part of this standard review process. Grannatt Source: ‘Novartis: Your partner of choice’ presentation Business Insights Ltd Figure 5.39: Licensing process at Novartis Initial Technical Evaluation Preselection Full Evaluation Negotiation S&E Negotiation Alliance Management Source: Novartis: Your partner of choice’ presentation Business Insights Ltd 106 . Golumbeski Anti-Infectous M.38: Business development and licensing department. Girsault Alliance Management D. EU Latam. Hörning Primary Care S. Figure 5. De Vries Transplantation M.
2006 Business Insights Ltd Cephalon Cephalon has an interesting business model. 2006 Number of survey respondents Cephalon Ablynx Amylin Genentech UCB 2 2 2 2 3 Source: Licensing trends survey. as shown in Figure 5.40: Out-licensing partner of choice. with no single company receiving more than 3 votes. Amylin. Cephalon has acquired rights to Trisonex (arsenic trioxide) from Cell Therapeutics in 2005.Leading out-licensing companies A survey of licensing executives revealed that the leading out-licensing partner in 2006 is Cephalon. Experience from being on the in-licensing side of multiple licensing agreements has led to a number of recent examples of successful out-licensing deals with development and marketing partners. However. Cephalon signed an agreement with Takeda 107 . Figure 5. The company has both in-licensed and out-licensed the rights to pharmaceutical products in order to facilitate its business model. From an inlicensing perspective. Other leading out-licensing partners include Ablynx. Genentech and UCB.40. the rights to Gabitril (tiagabine) from Sanofi-Synthelabo in 2002 and the rights to Actiq (fentanyl) from Elan in 2002 after acquiring Anesta in 2000. based on limiting risk through licensing agreements and acquisitions. In 2006. the survey selections of leading outlicensing partner were dispersed across a wide range of companies.
Cephalon’s licensing model of acquiring rights to marketed products in order to generate revenues to fund the internal development of in-house products appears to work well. The agreement included an option to extend the relationship to include new treatment Nuvigil (armodafinil). A 2003 agreement with Tanabe Seiyaku provided the company with a license to promote Actiq (fentanyl) in Japan. 108 .Pharmaceuticals North America in order to add 500 Takeda sales reps to co-promote narcolepsy treatment Provigil (modafinil) in the US. the company has been willing to share promotional rights for its products. in order to maximize the returns from its internal assets. particularly in the US and Japan. the agreements with Takeda and McNeil illustrate Cephalon’s willingness to partner with market leaders (and potential competitors) in order to maximize total licensing revenues. A similar agreement signed with McNeil Consumer and Specialty Pharmaceuticals in 2005 provides co-promotion support for Attenace (modafanil) in preparation for the approval of a pediatric label for attention deficit/hyperactivity disorder (ADHD). as an out-licensor. By out-licensing the US promotional rights for key products without having to share any co-development costs and risks. the resulting licensing agreements have been kept simple and focused. More importantly. Moreover.
When considering licensing opportunities. as a developer. corporate-level objectives. companies must also be introspective and evaluate what they can add to the deal. not just successfully agreed and signed. pharmaceutical and biotech companies must look to extend their deal-making to form long lasting partnerships involving multiple products. 109 . marketer or collaborative partner. companies must work even harder to find the right collaborator with the right deal in order to protect a reputation for being a reliable and productive licensing partner. Licensing trends As licensing deals continue to increase in value and complexity. A distinction between a licensing valuation used to select appropriate opportunities and a valuation used to inform negotiations and deal-making must be made in order to ensure the different objectives for each exercise are satisfied.Recommendations for the future The key recommendations from this report can be summarized under four main categories. ‘The proof of the pudding is in the eating’ and all licensing efforts should be focused on creating deals that can ultimately be successfully implemented. As more and more pharmaceutical licensing moves towards long term relationship building. Licensing process Pharmaceutical companies must align licensing activities around a determined strategy that is consistent with broader.
long lasting agreements and creative deal terms – in order to capture the unique characteristics of each deal and deal-making partner. risks and revenues are fully understood. Discount rates are not necessary to understand value and should not be used to create a base case deal-making valuation. Licensing best practices Best practices are specific to a particular deal and should not be applied as a more general set of licensing rules. 2006 US dollars) a common language for discussing deal terms between licensing partners is provided. 110 . By creating licensing valuations based on a common valuation measure (e.g.Licensing valuations Licensing valuations should ideally be shared and form the basis for open discussions over total project values and appropriate deal terms for sharing risks. Licensing valuations for deal-making should be based largely on independent and unbiased model inputs in order to limit any areas of subjective disagreement. Lessons can be learned from the leading deals and deal-makers – which tend to involve clear licensing objectives. Valuations should be simple and clear to understand in order that all interactions between costs. returns and responsibilities.
CHAPTER 6 Appendix 111 .
2006 Business Insights Ltd As shown in Figure 6. survey respondents were drawn from a range of different functional responsibilities. biotechnology. with a 112 . Figure 6. though each respondent had some level of decision-making responsibilities for pharmaceutical licensing.42. including pharmaceuticals. generics and medical equipment and diagnostics. As shown in Figure 6.41.Chapter 6 Appendix Primary research survey A survey of 142 licensing executives was completed in early 2006 in order to identify trends and best practices for this report. The survey was dominated by executives from pharmaceutical companies (55%) and biotech companies (20%). Executives from the licensing and business development function made up the majority of respondents (56%).41: Licensing trends survey respondents by company focus Medical equip/ diagnostics 3% Generics 4% Drug delivery 9% Pharma 55% Other 9% Biotech 20% Source: Licensing trends survey. drug delivery. survey respondents were drawn from across all major healthcare industry segments.
22% with responsibilities for inlicensing alone and 15% with responsibilities for out-licensing alone. 113 . Figure 6. survey respondents were comprised of 63% that had responsibilities relating to both in.42: Licensing trends survey respondents by functional responsibility Industry analyst/ consultant 5% R&D 6% Corporate & Senior Management 13% Licensing & Business Development 56% Sales & Marketing 20% Source: Licensing trends survey.43.and out-licensing.further 20% involved in sales and marketing and 13% involved in corporate functions or senior management. 2006 Business Insights Ltd As shown in Figure 6.
2006 Business Insights Ltd 114 .43: Licensing trends survey respondents by licensing responsibility Out-licensing 15% In-licensing 22% Both 63% Source: Licensing trends survey.Figure 6.
Pharmacoeconomics 20 Suppl 3 (2002) 11-29 115 . Nuno Sousa Pereira. Journal of Health Economics 22 (2003) 151-185 Grabowski. Cockburn and Rebecca M. 2001 Scale and scope in drug development: unpacking the advantages of size in pharmaceutical research. Iain M. Henry G. John Vernon and Joseph A. Journal of Health Economics 20 (2001) 10331057 Danzon. 2003 The price of innovation: new estimates of drug development costs. Drug Information Journal 38 (2004) 211223 DiMasi. Journal of Health Economics 24 (2005) 317-339 DiMasi. Hansen and Henry G. Ronald H. Grabowski. Hansen and Grabowski. Henderson. Henry Grabowski. 2005 Productivity in pharmaceutical-biotechnology R&D: the role of experience and alliances. Patricia M. Nicholson and Pereira.Sources Cockburn and Henderson. Sean Nicholson. Joseph A. Grabowski and John Vernon. Vernon and DiMasi. Grabowski and Vernon. 2004 R&D costs and returns by therapeutic category. Danzon. DiMasi. DiMasi. Joseph A. 2002 Returns on research and development for 1990s new drug introductions. DiMasi.
62 Takeda. 105 AstraZeneca. 102. 27. 101. 62. 102. 109 116 . 49. 17. 108 Sanofi-Aventis. 105 Arakis. 104 McNeil. 102. 14. 87. 104 Amylin. 105. 103 Otsuka. 105 Valuation. 14. 67. 67. 43. 108 Genentech. 63. 101. 105 Eli Lilly. 14. ii. 107. 19. 95 Ortho Biotech. 102. 1. 92. 101. 101. 23. 102. 101. 37. 105 Drug delivery. 101. 47 Cardiovascular. 103 AtheroGenics.Index Ablynx. 58. 102. 48 Servier. 107 Human Genome Sciences. 63 Amgen. 91. 23. 108 GlaxoSmithKline. 19. 23. 84. 108 Merck & Co. 27 UCB. 95 Roche. 27. 18. 103 Biotechnology. 106 NPV. 62. 101. 60. 103 Portfolio management. 85. 13. 104 Cephalon. 18. 104 Pharmaceutical. 105 Ventura. 101. iii. 86. 23. 10. 74. 12. 10. 23. 27. 107. 12. 23. 103 Novartis. 105 Pfizer. 14. 23 Millennium. 107 Alliance management. 107 Tanabe Seiyaku. 75 Idenix. 95. 102. 23 CNS. 105 Deal making. 104. 105 Arrow Therapeutics. 58. 36. 63. 96 SeBo. 76 Senju Pharmaceutical. 107 Anadys Pharmaceuticals. 27. 105 Johnson & Johnson.