Advertisement

Journal of Commercial Biotechnology

, Volume 15, Issue 4, pp 309–323 | Cite as

How can pharmaceutical and biotechnology companies maintain a high profitability?

  • Klaus J Nickisch
  • Joachim M GreuelEmail author
  • Kerstin M Bode-GreuelEmail author
Original Article
  • 1.4k Downloads

Abstract

Biotechnology investors are increasingly concerned about taking the risk of investing in the development of innovative drugs, and pharmaceutical companies are worried about maintaining their high profitability in the future. The question is how to build a portfolio of research and development (R&D) projects that fulfils the financial expectations of investors and shareholders. State-of-the-art net present value algorithms are applied to different types of projects at entry into development in order to evaluate their financial attractiveness and their ability to generate adequate returns. Based on the currently applied cost of capital for pharmaceutical and biotechnology companies the attractiveness of the so-called blockbuster model is clearly supported. The increasingly favoured specialty model, however, will only provide sufficient returns to biotechnology investors if significant sales volumes are reached. Complementing a company's development portfolio with risk-reduced projects could be an attractive way to ensure sustained growth for both biotechnology and pharmaceutical companies.

Keywords

financial analysis risk management business models portfolio optimisation 

INTRODUCTION

Pharmaceutical R&D is a highly risky endeavour. Only around 10 per cent of drugs entering development finally reach the market, and only 20 per cent of marketed drugs recover their investment.1, 2 After the initial enthusiasm driven by the tremendous valuations of early-stage biotechnology companies in the late 1990s, investors are now more careful in considering biotechnology investment opportunities. Many venture capital funds are still suffering from the decline in valuations of the early century, hardly providing the returns that are expected for biotechnology investments. 3 In fact, venture capitalists are much more risk averse these days and prefer investments in companies whose products have the potential to reach the market either faster or at lower cost, 4 or they reduce investments in biotechnology altogether5, 6 and shift to other assets such as, for example, clean energy technologies. 7 The question arises whether there is an objective rationale for investors’ declining enthusiasm in supporting biotechnology companies.

Also in the pharmaceutical industry it is currently debated how high profitability can be maintained. Rising R&D cost and some setbacks caused by withdrawals of launched products, as well as expensive late development failures, raised concerns whether pharmaceutical companies can keep the comfortable profitability levels of the past. The industry has responded with numerous initiatives to enhance its efficiency, including attempts to improve the productivity of their huge R&D investments. For example, portfolio management techniques are increasingly applied in R&D in order to maximise portfolio value and to balance portfolio risk across projects.8, 9, 10, 11 It is an accepted fact that a significant number of companies currently rely on very few high volume products. Given the well-known uncertainties in the life science industry, the most important and also the most difficult question is not to determine the high priority projects but, rather, which projects should be terminated based on their limited economic potential. Companies that manage this challenge better than others have a significant advantage over their competitors because they more likely allocate R&D resources to projects that will create significant value. In other words, such companies should get a higher return on their R&D investment.

In order to investigate the question which projects would have sufficient expected value to satisfy the demands of venture capital investors and capital markets, the present paper analyses the risk and profitability of three different business models pursued by biotechnology and pharmaceutical companies, in light of their respective cost of capital. Three projects typically representing those models are analysed. Each project's net present value is determined at entry into preclinical development, taking into account the probability of success for the forthcoming development milestones. The results are discussed in the light of return expectations for different categories of companies.

BUSINESS MODELS

In the life science industry there is no single business model guaranteeing success. There are two predominant models that companies pursue in order to outperform competition:
  • Blockbuster model

  • Specialty model

Blockbuster model

The well-known blockbuster model emerged in the 1980s and is characterised by a focus on mass markets and on highly prevalent diseases such as, for example, hypertension, lipid metabolism disorders and gastric ulcers. Successful products were often not first in class molecules but fast followers. The advantages of these second generation products were not always clinically significant; instead, their commercial success could clearly be correlated with the marketing capabilities of the respective companies. Well-known examples are enalapril, ramipril, cimetidin, ranitidin and especially atorvastatin (Lipitor™), which was only the sixth HMG-CoA reductase inhibitor entering the market for the treatment of lipid metabolism disorders. The overall risk profiles of these projects were quite favourable because the scientific and clinical proof of concept had already been delivered by another company/molecule. The billion dollar annual sales were reached and later far exceeded by continued massive investments in development, life cycle management, and marketing and sales initiatives that were also used to create or expand the markets.

Most big international pharmaceutical companies currently rely on a few blockbuster products that are supported by significant life cycle management investments protecting these important assets. It is quite obvious that products like Lipitor™ with annual sales exceeding US$10 billion provided the returns that made the pharmaceutical industry the most profitable one over the last decade.

It is, however, believed that the blockbuster model is now at risk because of declining R&D productivity and changing market conditions.12, 13, 14, 15 Branded products with total yearly sales of $170 billion will lose their patents by 2015. 16 In spite of the expected significant changes, large primary care products will remain an integral part of major pharmaceutical and biotechnology companies and will therefore be evaluated in the context of this study.

Specialty model

The term ‘specialty pharma’ is not clearly defined and covers a wide range of different approaches. 17 The first wave of such companies focussed on niche therapeutic areas such as ophthalmology and various orphan diseases. Gradually, they have come to include a whole gamut of companies, ranging from therapeutic area concentrators (such as Lundbeck and Actelion) to drug delivery experts (such as Elan) and generic specialists (such as Teva and Barr).

According to Kambhammettu, 17 specialty pharma can be differentiated as follows:
  • Niche therapeutic area/orphan disease concentrators Companies focus on therapeutic segments such as ophthalmology, specific tumour types and other rare but serious diseases that have been neglected by big pharma. Examples for this model are Genzyme, Alcan and Actelion.

  • Portfolio adapters Companies pursuing this model acquire products that are de-prioritised by big pharma. Efforts are made to increase the sales volume of such products significantly by additional development investments and focussed intensive marketing. Companies successfully applying this business model are, for example, Shire and King Pharmaceuticals.

  • Licensing experts Such companies in-license early and late stage development candidates and complete development towards approval. One of the first companies applying this model was Marion Laboratories; today, Helsinn Pharma is a typical example.

  • Drug delivery experts Companies reformulate existing molecules in order to enhance their therapeutic application, their benefit/risk profile or their convenience. The company that first applied this model was Alza; today, companies such as Nektar and Elan are representing this strategy.

  • Specialty generic companies The increasing competitive pressure in the generic industry made some generic companies develop their own branded generics by reformulating existing products. Examples of this approach are Barr and Watson.

A further definition of the specialty pharma model is added to the classifications described above.

Indication specialists

Companies belonging to this class typically focus on severe diseases whose incidence is lower than that of the classical blockbuster indications such as hypertension and lipid metabolism disorders. However, drug prices are usually higher 18 and products are often biologics and injectables. Patients are predominantly treated by specialists rather than primary care physicians. Many biotechnology companies are indication specialists. The most broadly addressed indication specialist market is oncology. Twenty-eight per cent of drugs currently in R&D are cancer therapeutics. 19 The development of this market with products like Avastin™, Gleevec™ and Taxotere™ indicates that also specialty products can exceed $1 billion annual sales. Based on their high prices these products are very profitable. The great majority of products in this class generate, however, much lower sales.

The specialty markets are becoming increasingly relevant for both international pharmaceutical companies and biotechnology companies. The development investments in specialty products are similar to those of large primary care products up to proof of concept, but Phase III trials are usually less expensive. 20 Extensive post-approval outcomes studies are rarely required. Two subtypes of the speciality model are included in the present investigation, that is, the indication specialist model and the drug delivery model. The large inherent risk involved in drug discovery and development has created a demand for business models delivering commercially attractive products with a higher probability of launch, shorter development time and lower development cost. From the various options that have been pursued it can be concluded that strategies using well-known and well-characterised drug substances and modifying them to create incremental value to patients are especially attractive.21, 22 It is an obvious approach to reformulate already approved drugs in order to reduce some disadvantages of the original formulation. Such reformulations could simply address convenience issues, but in certain cases also improve efficacy and safety. Four strategic objectives associated with drug reformulations can be differentiated: 23

Switch and grow

Switch of patients from old to new and competitively differentiated formulation (example: J&J developed Risperdal™ consta, long acting risperidone, a once monthly depot formulation for the treatment of schizophrenia).

Expand and grow

Expanding the scope of a product, that is, gaining access to new segments of patient population or new indications (example: Wyeth developed Protonix™ i.v., a reformulation of pantoprazole, especially indicated for severe forms of gastroduodenal, esophageal inflammation and Zollinger Ellison Syndrome).

Generic defence

Switch of patients to competitively differentiated formulation that provides extended patent protection, usually late in product life cycle (example: Prozac™ weekly).

Market grab

Reformulation of a generic drug to capture market share both from the original brand and from generic manufacturers. This can be associated with a different regulatory label (example: Cephalon launched a reformulation of fentanyl (Actiq™) for the treatment of breakthrough cancer pain).

The range of strategic approaches indicate that virtually all companies can benefit from drug reformulations. The advantage is the overall higher success rate of such projects and the usually front-loaded risk structure, meaning that the greatest risk of failure is early in development and R&D budget erosion is minimised. Furthermore, product prices for reformulations are often not significantly different from the original and may, depending of the competitive advantage, even be higher; however, the market positioning of the reformulation requires a significant marketing budget. For companies following this business model it is very important to generate a strong intellectual property position in the technology that they apply because formulation patents can be circumvented much easier than composition of matter patents protecting a new active ingredient. In any case, the huge number of companies pursuing drug delivery strategies,24, 25 many of them backed by venture capital, is a living proof that this risk-minimising strategy is attractive to investors.

In our financial modelling we considered two different development strategies for a drug reformulation, one just requiring a bioequivalence study (for projects where the pharmacokinetic profile is not changed) plus a small clinical trial to establish the added value, and another one where additional Phase III studies are necessary in order to establish efficacy and safety.

VALUATION MODEL AND INPUT PARAMETERS

In order to compare the financial value created through the different business models (for example, blockbuster, specialty and drug reformulation), representative project examples were defined using input parameters based on industry standards 1 and Bioscience Valuation's database. The specialty model evaluated in this analysis reflects the indication specialist definition.

The following parameters were defined:
  • Development time

  • Development cost

  • Probability of development success (PoS)

  • Cost of goods sold (CoGs)

  • Marketing and sales (M&S) cost

  • Peak sales and time to reach peak sales

  • Patent duration

For our financial modelling we used the concept of augmented net present value (NPV); (other commonly applied terms are expected NPV, or risk-adjusted NPV based on the project's milestone structure), taking advantage of discrete probability distributions to model development risk, potential outcomes and respective consequences. The model is generally accepted for the valuation of R&D projects.8, 26, 27, 28, 29, 30, 31

Projects were evaluated at entry into preclinical development after successful completion of the discovery phase. We assumed that the input variables for the investigated project categories are not significantly different if undertaken in large pharmaceutical or biotechnology/specialty companies, with the only exception of the cost of capital. Arguments for this assumption are provided below.

The proportion of total R&D expenditure by functions 1 is as follows:
  • Clinical 35 per cent

  • Chemistry, Manufacturing & Controls (CMC) 15 per cent

  • Non clinical 5 per cent

  • Regulatory 4 per cent

  • Discovery 23 per cent

  • Miscellaneous plus fees 17 per cent

If only development costs are considered, clinical development is most expensive with 45 per cent, followed by CMC nearly 20 per cent, non clinical with 6 per cent and regulatory with 5 per cent.

The overall external R&D expenses are created in the following areas 1 :
  • Clinical 60 per cent

  • CMC 12 per cent

  • Non clinical 5 per cent

  • Regulatory 2 per cent

  • Miscellaneous plus fees 12 per cent

It is assumed that biotechnology companies will depend to a large extent on external contract research organisations for their preclinical, clinical and CMC development activities. Pharmaceutical companies also use contract research organisations intensively for their clinical development, whereas the preclinical and the CMC work is more often done in-house. Whether drug development can be done more cost effectively in house or through contract research organisations is a matter of intensive debate and depends on the special situation. For areas in which both types of companies intensively rely on external vendors, such as clinical development, we assume that there are no significant differences in development cost between biotechnology/specialty and pharma companies because the potentially smaller amount of overhead in biotechnology/specialty companies might be counterbalanced by better deals that big pharma is able to negotiate with clinical research organisations, based on the larger amount of work they are bringing to them.

Whereas for established large pharmaceutical companies the cost of capital can be assumed to be in the order of 10 per cent, the cost of capital (or, in other words, the return expectations) for biotechnology companies funded by venture capital is considerably higher. We applied 20 per cent in our model, being aware that some venture capitalists apply significantly higher figures.

The base case assumptions for cost, PoS, and time lines for the blockbuster, specialty and drug delivery (with/without Phase III) projects are illustrated in Figures 1, 2, 3 and 4.
Figure 1

Project structure for blockbuster and specialty model. R&D milestones, timelines, R&D costs and probability of development success for the development of a new molecular entity are illustrated. The data represent activities to obtain first approval. Indirect costs are included in activity-specific figures. The model assumes that timelines and probability of development success for first approval are not significantly different for blockbusters versus specialty products. R&D costs and sales differ for the scenarios investigated (see Tables 2 and 3). Post approval expenses, for example, for Phase IV studies, are included in marketing and sales costs represented as percent of sales.

Figure 2

Example of a cash flow profile underlying the net present value calculation. The displayed cash flows represent Scenario 1 in Table 2. These cash flows reflect the launch scenario with peak sales of EUR 500 million. The probability of launch is 11 per cent (see Figure 1). For the net present value calculation, all cash flows are inflated by 2 per cent, the tax rate is 40 per cent. The terminal value is calculated based on the period 19 net cash flow reduced by 80 per cent, assuming generic competition after patent expiry. The augmented (expected, risk-adjusted) net present value includes all failure scenarios and their respective values in a probability-weighted manner, according to the decision trees. For the failure scenarios, cash flows were included up to the respective milestone of termination.

Figure 3

Project structure for reformulation (Phase III not required). R&D milestones, timelines, R&D costs and probability of development success are shown for a reformulation project. The model assumes a limited clinical development programme with a Phase I and an additional Phase II-like trial to establish the added benefit of the new product. Time to market is only 7 years compared to 10 years for new chemical entities/new biological entities, and the probability of development success is significantly higher with 65 per cent compared to 11 per cent.

Figure 4

Project structure for reformulation (Phase III is required). The development of a reformulation may require a Phase III programme depending on the application and targeted label. In such a case, development time lines and costs approach ranges typical for new chemical entities. However, probability of development success will usually be higher based on the knowledge and experience accumulated with the original product. The case example assumes a probability of development success of 43 per cent compared to 11 per cent for a new chemical entities.

Table 1 summarises the input parameters for the three different business models. R&D related parameters are based on Bioscience Valuation's database and on publications by Parexel and CMR.24, 25 Marketing, sales and G&A costs (SG&A) reflect data published in representative annual reports.
Table 1

Overview of input parameters for the three business models; costs are only referring to the development of one project and do not include the costs of failures to achieve one launched product

 

Development time (years)

R&D cost (US$million)

PoR (%)

Peak sales (US$million)

CoGs (% of sales)

SG&A (% of overall sales)

Blockbuster

10

108–1260

11

1000–2000

10–30

20–30

Specialty product

10

108

11

140–500

10–30

20–30

Reformulation

7–10

36–108

43–62

55–500

10–30

30

Assumptions for reformulation products are based on 22 .

The NPV model includes all project related cash flows from the start of preclinical development up to patent expiration. Cash flows are inflated by 2 per cent per year. The tax rate is 40 per cent. Patent application was assumed to be in the year before start of preclinical development, leading to patent protected sales for about 10–12 years. Peak sales are achieved in 5 year on the market and maintained for 5 years. A sales decline of 5 per cent per year thereafter was assumed because of emerging new treatment options. Our models do not take into account licensing deals that may be closed at certain stages of development. It is assumed that the value of the deal would correlate with the value of the project at the respective development stage, following commonly observed rules of value split between licensors and licensees.

The influence of the different input parameters was investigated to understand the value drivers and to address the overall question which projects create sufficient value for pharmaceutical and biotechnology companies.

RESULTS

Pharmaceutical companies

The high profitability of the pharmaceutical industry was clearly correlated with blockbuster products that were able to overcompensate the failures of other projects. Our model clearly supports this view, showing a positive risk adjusted NPV of $229 million for a project entering preclinical development with forecasted annual sales of $2 billion, development cost of $216 million and the standard assumptions for CoGs of 10 per cent and M&S costs of 20 per cent of total sales (see Table 2, example 13). The risk-adjusted internal rate of return (IRR) for such a project is 26 per cent, further underlining its attractiveness.
Table 2

Scenario analysis for projects undertaken in pharmaceutical companies

Big pharma

Scenario

Development time (years)

Development cost (US$million)

PoR (%)

Peak sales (US$million)

Discount rate

CoGs (% of sales)

SG&A as % of overall sales

Risk-adj. NPV (US$million)

Non Risk-adj. NPV (US$million)

RA IRR (%)

 1

10

108

11

500

10

10

30

38

520

18

 2

10

108

11

250

10

10

30

9

236

13

 3

10

108

11

100

10

10

30

−9

65

8

 4

10

108

11

100

10

10

20

−7

85

7

 5

10

108

11

175

10

10

30

0

149

10

 6

10

108

11

500

10

15

30

33

471

17

 7

10

108

11

500

10

30

30

17

323

14

 8

10

216

11

500

10

10

30

19

474

13

 9

10

108

11

1000

10

10

30

96

1079

24

10

10

108

11

1000

10

10

20

117

1277

26

11

10

425

11

1000

10

10

30

38

950

13

12

10

625

11

1000

10

10

30

0

855

10

13

10

216

11

2000

10

10

20

229

2500

26

14

10

1260

11

2000

10

10

20

38

2083

11

15

10

108

24

500

10

10

30

109

520

24

The analysis represents NMEs at entry into preclinical development. A discount rate of 10 per cent is applied.

The lack of sufficient projects of this kind forces the industry to look for alternatives. Almost all pharmaceutical companies decided to engage in specialty markets, and many moved into oncology. For this study three projects were considered where peak sales ranged between $100 and 1000 million. With the input parameters described above, our model shows that projects with a sales potential of $500 million or higher are very attractive targets, provided that they could be developed in the average time frame and with the average development cost (examples 1, 6 and 9). Risk adjusted NPVs range from $33 to 96 million.

A sensitivity analysis regarding the influence of higher CoGs and higher development costs revealed that both factors have the expected modest influence. Doubling the development cost to $216 million reduces the risk adjusted NPV from $38 to 19 million (example 1 and 8). For projects with peak sales of $1000 million, the overall development cost could go up to $625 million in order to reduce the risk adjusted NPV to zero (example 12). Similarly, a 50 per cent increase of the CoGs from 10 to 15 per cent has only a marginal effect on the attractiveness of a project leading to a reduction of the risk adjusted NPV from $38 to 33 million (example 6). The risk adjusted NPV still remains positive ($17 million) when CoGs are tripled from 10 per cent to 30 per cent (example 7).

The analysis of the influence of the peak sales potential showed a somewhat different picture. For projects with a peak sales potential of $250 million the risk adjusted NPV falls from $38 to 9 million and for projects with a peak sales potential of just $100 million the risk adjusted NPV turns negative to $−9 million (examples 1,2 and 3). A reduction of the M&S expenditures from 30 per cent to 20 per cent only improves the NPV marginally (example 4).

In summary, our modelling supports the attractiveness of the blockbuster and specialty models most pharmaceutical companies adopt these days.

The risk-adjusted IRR for most of the projects with peak sales up to $500 million is in the range of 10–20 per cent. Rates above 20 per cent can only be reached with projects having peak sales of $1000 or higher, emphasising the importance of blockbuster products for the overall profitability of the industry. To provide shareholders with a minimum annual return of 10 per cent, projects undertaken in the pharmaceutical industry require a peak sales potential of at least $175 million (it should be noted, however, that pharmaceutical companies sometimes use higher hurdle rates for internal decision purposes than the true cost of capital, in order to focus the R&D organisation on more ambitious corporate goals).

Biotechnology companies

As mentioned above, we assumed equal effectiveness and efficiency for biotechnology and pharmaceutical companies in our model. Arguments can be given for changing these assumptions in both directions, for example, smaller companies with a smaller overhead should have lower development cost or larger companies could be faster in recruiting for pivotal trials. However, the goal of the present paper is not to investigate the impact of effectiveness and efficiency on value. Therefore, these variables are assumed to be similar for both, big pharma and biotechnology.

There is, however, one highly significant difference between pharmaceutical and biotechnology companies: the cost of capital. Whereas for established pharmaceutical companies in the Western world a discount rate of 8–10 per cent is commonly applied (we use 10 per cent for our modelling), the return expectations of venture capitalists and other investors in the biotechnology industry are significantly higher given the higher risk of such investments. There are no generally accepted numbers available. Although the cost of capital may differ among biotechnology companies, we chose 20 per cent in this study, which is a reasonable portfolio return for most venture capital companies. Start-up companies often choose the same drug development targets as pharmaceutical companies. We therefore kept all parameters constant for the NPV model with the exception of the discount rate that was changed from 10 to 20 per cent (see Figure 5).
Figure 5

Sensitivity analysis with respect to discount rate and peak sales, using base case assumptions for development risk, cost and time lines. New molecular entity entering development for which peak sales of EUR 500 million are forecasted, have a negative risk-adjusted net present value at a discount rate of 20 per cent, whereas the net present value is clearly positive at a discount rate of 10 per cent. In view of their return expectations, biotechnology investors prefer projects that have blockbuster potential (for example, based on ‘breakthrough innovation’) or a lower risk of failing in development (‘incremental innovation’, for example, new drug delivery technologies).

Even for projects that have a peak sales potential of $500 million the risk adjusted NPV is negative ($−3 million, Table 3, example 1), meaning that with such projects biotechnology companies and their investors cannot expect to realise the return they are looking for. The actual risk adjusted IRR is 18 per cent. Expectedly, the results are even more bleak when the peak sales potential is reduced to $250 million (NPV: $−10 million) with an IRR of 13 per cent and to $150 million (NPV: $−13 million) with an IRR of just 9 per cent (examples 3 and 5). Again, lowering the marketing and sales expenses from 30 to 20 per cent doesn’t have a significant influence on the NPVs (examples 2 and 4). Reducing development costs by 50 per cent to $54 million leads to a positive NPV for a $500 million peak sales level but is still negative for the $150 million level (examples 6 and 7).
Table 3

Scenario analysis for projects undertaken in biotechnology companies

Biotech

Scenario

Development time (years)

Development cost (US$million)

PoR (%)

Peak Sales (US$million)

Discount rate

CoGs (% of sales)

SG&A as % of overall sales

Risk-adj. NPV (US$million)

Non Risk-adj. NPV (US$million)

RA IRR (%)

 1

10

108

11

500

20

10

30

−3

105

18

 2

10

108

11

500

20

10

20

0

131

20

 3

10

108

11

250

20

10

30

−10

35

13

 4

10

108

11

250

20

10

20

−8

48

14

 5

10

108

11

150

20

10

30

−13

7

9

 6

10

54

11

150

20

10

30

−4

24

14

 7

10

54

11

500

20

10

30

6

122

24

 8

10

108

11

700

20

10

30

3

159

21

 9

10

108

11

1000

20

10

30

11

244

24

10

10

216

11

2000

20

10

20

32

576

26

11

10

108

11

700

20

30

30

−5

84

17

12

10

108

11

1000

20

30

30

0

142

20

13

10

216

11

2000

20

30

20

9

364

22

The analysis represents NMEs at entry into preclinical development. A discount rate of 20 per cent is applied.

This result has potentially far reaching consequences for the question which projects and which companies will likely be funded by venture capitalists. If the well accepted specialty model applied to biotechnology companies does not provide investors with the expected returns the question arises with which projects or business models would be able to deliver it. Two different options were considered in this context: first, an increase in peak sales going from a specialty product towards a blockbuster drug or, second, a strategy that is based on minimising overall development risk. For the second case reformulation approaches were considered.

By increasing the peak sales potential from $500 million to 1000 and 2000 million the risk adjusted NPV increases from $−3 to +11 and +32 million, respectively (examples 1, 9 and 10), meaning that the required return could be delivered. An IRR of 20 per cent would only be accomplished with a peak sales level of $500 million if M&S expenses were 20 per cent (instead of 30 per cent, example 2). The overall conclusion is that the expected return for biotechnology investments can be accomplished if the project has blockbuster potential (‘breakthrough innovation’).

Finally, we investigated risk-mitigating strategies that some companies apply successfully, such as reformulation strategies (‘incremental innovation’, see Table 4). For such projects the overall PoS was increased from 11 per cent to 43 per cent or 62 per cent, assuming that risk is significantly reduced based on the experience with the original product (see assumptions in Figures 2 and 3). Development costs were reduced from $108 to 36 million for cases in which clinical proof of efficacy and safety is not needed (for example, 505 B2 regulatory route in the United States) or to $86 million for projects where safety and efficacy data are required, respectively. Examples for the latter are transdermal delivery approaches in the female health and pain area.
Table 4

The analysis represents reformulation projects

Reformulation

Scenario

Development time (years)

Development cost (US$million)

PoR (%)

Peak Sales (US$million)

Discount rate

CoGs (% of sales)

SG&A as % of overall sales

Risk-adj. NPV (US$million)

Non Risk-adj. NPV (US$million)

RA IRR (%)

 1

7

36

62

500

20

10

30

110

187

42

 2

7

36

62

250

20

10

30

48

86

35

 3

7

36

62

100

20

10

30

11

25

25

 4

7

36

62

100

20

30

30

1

9

21

 5

7

36

62

55

20

10

30

0

7

20

 6

7

36

62

55

10

10

30

32

59

20

 7

10

86

43

500

20

10

30

37

111

30

 8

10

86

43

250

20

10

30

8

42

23

 9

10

86

43

250

20

30

30

−3

15

19

10

10

86

43

100

20

10

30

−9

0

15

11

10

86

43

180

20

10

30

0

23

20

12

10

108

43

500

20

10

30

31

104

28

The profile of existing drugs is improved through reformulations or alternative drug delivery routes. This may result in improved convenience, safety or efficacy, and may lead to attractive sales at a lower development risk and cost. The returns provided to investors can be substantial.

The results explain why this business model is so popular among investors these days. For the investigated cases that do not require a Phase III study and that, therefore, have lower overall development cost and shorter development times, the risk adjusted NPVs were solid positive at peak sales between $100 and 500 million (at CoGs of 10 per cent, NPV: $11–110 million); IRRs were between 25 and 42 per cent (Table 4, examples 1, 2 and 3). Even at peak sales as low as $55 million a 20 per cent IRR could be accomplished. If a 10 per cent discount rate were applied the risk adjusted NPV would increase to $32 million (example 6), making such projects extremely interesting also for established specialty companies. For products with CoGs at 30 per cent, however, risk adjusted NPVs would approach zero, IRR would be 21 per cent, already at peak sales of $100 million (example 4).

Some reformulations require a Phase III programme. The most significant difference to new molecular entities (NMEs) is the higher overall probability of success, whereas development time and development cost are close to those of NMEs. Again, the significant influence of peak sales on the financial attractiveness of the projects can be seen. At $500 million peak sales, the projects are highly attractive with a risk adjusted NPV of $37 million (example 7). The threshold of a 20 per cent risk adjusted IRR is reached at peak sales of $180 million (example 11). At CoGs of 30 per cent, peak sales of $250 million are to be trespassed to achieve positive risk adjusted NPVs and an IRR of more than 19 per cent (example 9).

SUMMARY AND CONCLUSIONS

Our investigation reveals that for established pharmaceutical companies the most important parameter for the decision whether to move a research project into development is the expected peak sales potential, provided that it does not differ significantly from average projects regarding time, cost and overall probability of success. Under such circumstances projects with a peak sales potential around and above $500 million are clearly attractive. Projects with peak sales around $250 million could still deliver a return above 10 per cent but projects with lower peak sales potential should be questioned. Established pharmaceutical companies should also consider complementing highly innovative and therefore risky projects with low risk alternatives because such projects may offer a very attractive return on investment and help to create a risk-balanced portfolio.

Start-up biotechnology companies can only comply with the high financial expectations of their investors by either concentrating on projects that have a high peak sales potential (blockbuster) or by focussing on alternative approaches and business models that have a significantly higher overall probability of success. Our study therefore explains the current trend in the venture capital community to prefer high value and low risk opportunities when making funding decisions. It also highlights the importance of early market research to identify a project's sales potential even if the drug is still in preclinical development.

Investigating the sales potential early may be particularly useful to substantiate the valuation if the project is to be out-licensed. In fact, young biotechnology companies rarely develop their first clinical candidates towards approval. In many cases they look for a partner after clinical proof-of-concept or earlier in order to obtain cash for building a sustainable portfolio. Shortage of cash and lack of late stage development and marketing expertise are the predominant reasons for out-licensing. The value of licensing deals is driven by a drug's risk-adjusted NPV at the time of deal closure. If the respective licensor's and licensee's cash flows resulting from published deals are analysed and risk-adjusted, it can be concluded that a project's value is commonly split at a ratio of around 50 per cent:50 per cent between licensor and licensee for projects in Phases II and III. 32 There may, however, be variations depending on market demand for licensing opportunities, and on preferences for particular therapeutic categories. Assuming that companies build their deal negotiations on the value proposition of the underlying asset, the conclusions drawn from this investigation also apply to licensed products.

One may ask how some companies managed to generate satisfactory returns based on niche products. Niche products were not included in our evaluation. The development strategy and history of such drugs varies and has to be evaluated individually. Special rules apply if, for example, they have been granted orphan drug status, which may result in lower development cost, shorter time to market and market exclusivity. In many cases there is a scientific rationale for developing the drug in additional, potentially less ‘niche’, indications, which drives their valuation upwards.

Niche products are usually commercialised after having completed major development milestones. At that stage, time to market is short, risk is low and the risk-adjusted NPV is clearly positive. However, the decision to invest in a project is made at an early development stage when the outcome is still highly uncertain, that is, the decision is not made on basis of a success scenario. Our analysis compares funding decisions for projects entering preclinical development. At that stage, NPVs are significantly impacted by development risk, and many niche projects will be rejected. If it were possible to relate the number of cases with positive outcomes to the number of companies that failed while developing true niche products, one would probably conclude that rejecting such projects would have been a rational decision. Common reasons for failure are: (a) the company did not succeed in securing continued venture capital financing, (b) clinical trials failed and could not be compensated by addressing other therapeutic indications for the development candidate, driving the company back to preclinical stage, or (c) an exit was achieved, but with economic terms that did not create value for investors. As there appears to be no systematic analysis of the economic success rates of biotechnology companies pursuing niche indications, there is a danger to fall for a perception bias driven by the publicity of the successful examples. In any case, the present analysis suggests conducting careful financial evaluations that take into account potential failure scenarios, in order to enable informed decisions.

References and Notes

  1. CMR International (2006/2007). Pharmaceutical R&D Factbook; PAREXEL's Pharmaceutical R&D Sourcebook 2007–2008.Google Scholar
  2. Datamonitor. (2008) Trends in Clinical Trials, Strategies to Enhance Clinical Trial Design. DMHC2445, September.Google Scholar
  3. Hargreaves, B. (2007) Investing in Drug Development – Trends and Outlook., At: R&D Leaders’ Forum, October 2007, Munich.Google Scholar
  4. Pollack, A. (2005) A new wariness as the risks shift in biotech investing. The New York Times, 30 December.Google Scholar
  5. Hamilton, D. (2007) Biotech's dirty little secret: VC's are heading for the hills – fast, http://www.venturebeat.com, 20 October.
  6. Dellenbach, R. (2008) VC funding for biotech companies withering. Genetic Engineering & Biotechnology News, 1 September.Google Scholar
  7. Johnson Hambley, C. (2008) Why Cleantech is taking biotech's shine. TechTrends, 10 August.Google Scholar
  8. Sharpe, P. and Keelin, T. (1998) How Smith Kline and Beecham makes better research allocation decisions. Harvard Business Review 76 (2): 45–57.Google Scholar
  9. Slater, S. and Zwirlein, T. (1992) Shareholder value and investment strategy using the general portfolio model. Journal of Management 16: 717–732.CrossRefGoogle Scholar
  10. Seget, S. (2005) Pharmaceutical Project and Portfolio Management – Effective Organization and Decision Making, Business Insights, Business Insights Ltd.Google Scholar
  11. Bode-Greuel, K. and Nickisch, K. (2008) Value-driven project and portfolio management in the pharmaceutical industry: Drug discovery versus drug development – commonalities and differences in portfolio management practice. Journal of Commercial Biotechnology 14 (4): 307–325.CrossRefGoogle Scholar
  12. Bain & Company. (2003) Has the pharmaceutical blockbuster model gone bust. Press release, 12 August.Google Scholar
  13. Guilbert, J., Henske, P. and Singh, A. (2003) Rebuilding big pharma's business model. In Vivo 21 (10): 73–80.Google Scholar
  14. Cutler, D. (2007) The demise of the blockbuster? New England Journal of Medicine 356: 1292–1293.CrossRefGoogle Scholar
  15. Kalin, K. I. and Franco, R. J. (2002) Beyond the blockbuster. Pharmaceutical Executive, 1 November.Google Scholar
  16. SCRIP World Pharmaceutical News. (2008) US Pharma/biotech collaborations will be crucial in 2008. issue 3322, p. 15.Google Scholar
  17. Kambhammettu, S. (2007) Evolving Paradigms in the Speciality Industry. Frost & Sullivan Market Insight, 8 August.Google Scholar
  18. Pinsault, T. (2005) The world of specialty pharma. Pharmaceutical Representative, 1 June.Google Scholar
  19. PharmaProjects Update Analysis, May 2008, Volume 29, Issue 1.Google Scholar
  20. Wyman, O. (2007) Beyond the Blockbuster. Executive Briefing October, http://www.oliverwyman.com.
  21. Henry, C. M. (2005) Morphing the Model. Chemical & Engineering News 83: 25–26, 28.CrossRefGoogle Scholar
  22. Renwick, S. (2007) Specialty Pharma as a Model for Ambitious Drug Delivery Companies. Drug Delivery Report Spring/Summer 2007.Google Scholar
  23. Datamonitor DMHC. (1994 2004) Maximizing returns through drug reformulations, August.Google Scholar
  24. Biocentury. (2008) 3 October.Google Scholar
  25. MedTrack. (2008) November.Google Scholar
  26. R&D managers sometimes argue that NPV-based models are not suitable to adequately capture the option character of discovery research and early development projects. It has been concluded from financial option theory that discounting in option pricing models would better reflect the optionality of such projects. However, there is currently no consensus among scientists and practitioners about the validity of option pricing methods and risk-neutral valuation for the evaluation of projects with a high impact of private risk. Furthermore, there is evidence that the results of both approaches converge and would lead to similar decisions.29–31 We therefore conclude that augmented NPV is a suitable approach for our investigation.Google Scholar
  27. Bode-Greuel, K. and Greuel, J. (2005) Determining the value of drug development candidates and technology platforms. Journal of Commercial Biotechnology 11 (2): 155–170.CrossRefGoogle Scholar
  28. Loch, C. H. and Bode-Greuel, K. M. (2001) Evaluating growth options as sources of value for pharmaceutical research projects. R&D Management 31 (2): 231–248.CrossRefGoogle Scholar
  29. Kellogg, D. and Charnes, J. M. (2000) Real options valuation for a biotechnology company. Association of Investment Management & Research (May/June): 76–84.Google Scholar
  30. Smith, J. E. and Nau, R. F. (1995) Valuing risky projects: Option pricing theory and decision analysis. Management Science 41 (5): 795–816.CrossRefGoogle Scholar
  31. Brach, M. (2003) Real Options in Practice. New Jersey: Wiley Finance, pp. 95–103.Google Scholar
  32. Bogdan, B. and Villiger, R. (2006) Valuation in life sciences, Springer, ISBN 978-3-540-45565-3, p. 105 ff.Google Scholar

Copyright information

© Palgrave Macmillan 2009

Authors and Affiliations

  1. 1.Bioscience Valuation BSV GmbH, Am Zigeunerbergl 3GrainauGermany

Personalised recommendations