Despite 150 years of history and an aggregated actual R&D expenditure of $150 billion per year, innovation in the modern pharma industry is essentially still based on trial and error. Only 1 out of 5,000 to 10,000 new molecules investigated becomes a commercialized product approved by regulatory authorities for specific therapeutic uses. On average about 250 of the investigated molecules will show sufficient promise for further evaluation by in vitro and animal laboratory tests. Many such projects are never transferred from the small-volume laboratory synthesis to large-scale commercial manufacturing.
Based on past results, only about 10 will qualify for clinical tests on humans and one will ultimately be approved. In total, less than 30 new medicines based on new molecular entities have been launched per year during the past 10 years. Moreover, during this period, one third of the new drug substances were high molecular weight molecules made by mammalian cell technology, which is only accessible to a small number of fine chemical companies. The share of biopharmaceuticals is expected to increase further in the years to come.
The relationship between the pharma and the fine chemical industries is challenging. On the one hand, pharma is the major customer segment, accounting for almost two-thirds of the $85 billion fine chemicals market.1 On the other hand, the reservoir of attractive new product opportunities is small. An antagonism exists with the need of the fine chemical industry, plagued with overcapacity, to rejuvenate and expand its product portfolios. The high attrition rate of pharma R&D and the challenges associated with a drug selection process are the main reasons why the fine chemical industry is affected by a low success rate for new drug projects. As fine chemical companies/contract manufacturers attempt to improve the decision process for “picking the winners”— identifying, and realizing profitable new revenues — they have to cope with three challenges (see Table 1), namely, finding new drug substances with:
- a promising market and profit potential
- a high probability of the pharma company obtaining regulatory approval for commercializing its new drug
- a good chance of being selected as supplier
In Table 2, we develop three scenarios quantifying the overall probability of success for a CMO to win a new project for a new pharmaceutical active ingredient. Assuming, for example, a pharma company issues a request for proposal (RFP) in clinical Phase IIb/III with a 66% chance of FDA approval, and the attractiveness of the CMO’s offering is average (35% chance), the overall probability of securing the business is 0.66 x 0.35 = 0.2.
In order to reinforce the CMO’s new business development efforts, unsolicited offers will have to be prepared proactively. Key parameters for selecting developmental drugs with high revenue potential are listed in Table 3.
1. Screening the list of the most promising developmental drugs published by equity research or industry research firms monitoring the pharmaceutical markets
Examples include Biopharm Insight, Datamonitor, Decision Resources, EvaluatePharma, FiercePharma, Jefferies & Cie., Piper Jaffray & Cie., Sanford C. Bernstein, Zacks.
In Table 4, a list of most valuable pharma projects from Evaluate Pharma2 is reproduced.
While a lot of valuable quantitative information is available from such sources on drug products, there is a lack of data on drug substances (i.e. APIs). A semi-quantitative approach can be made by the following equations:
2. For determining the market potential of the drug product (X), sales of the top-selling drug in the therapeutic class (see Table 5) can be considered as the upper limit unless the mechanism of action is substantially different. For example, one can assume a newly introduced anti-hypertensive could achieve an 8.6% (b) market share, equal to Diovan, the top selling drug with sales of $4.42 billion (a) in 2012 (size of the therapeutic class $51.6 billion).
3. As a first estimate, one can assume a sales volume of the drug substance (Y) equal to 3-5% share (c) of the final drug price. Taking again the anti-hypertensive Diovan (Valsartan) as an example, it would be in the range of $133-220 million (3-5% of $4.42 billion). Since pharma calculates the cost of a drug substance on the base of a comparison with the cost of a daily dosage of an existing drug in the same therapeutic class, substantial discrepancies can result.
An example is Biogen Idec’s Tecfidera, approved early 2013 for treating multiple sclerosis. The drug substance is dimethyl fumarate, a commodity. The daily maintenance dosage of 480 mg translates in a yearly consumption is 175 grams, which costs less than $1. This represents 0.002% of the yearly treatment cost of $50,000!
4. A literature/patent search for the synthesis of the pertinent drug substance. It will enable assessing the fit with the in-house technical capabilities and an approximate cost calculation on the basis of raw material and standard operating cost.3
5. An assessment of the two types of probabilities of success as described in chapters 2 and 3.
When drug companies issue requests for proposals (RFP) for their new drug candidates, they typically also provide a forecast for the quantities of the pharmaceutical fine chemical they intend to purchase. These forecasts, however, are not always reliable, especially if the drug belongs to a therapeutic category that is outside the core therapies of that pharma company. An example is Roche’s anti-obesity pill Xenical. In anticipation of a huge demand, the company invested $750 million in a plant in Florence, SC.
The capacity exceeded demand by far. In the worst-case scenario, the firm or regulatory authorities withdraw the drug from the market, i.e. if adverse side effects occur. This was the case with other weight loss pills, Fen-phen (Fenfluramine/Phentermine) from Wyeth (now Pfizer) and Reductil/Meridia from Knoll/Abbott.
Other well-known examples of withdrawals are the anti-diabetic Avandia, (rosiglitazone) from GlaxoSmithkline, the anti-inflammatory drugs Vioxx (rofecoxib) from Merck, Bextra (valdecoxib) from Pfizer, Iressa (gefitinib) from AstraZeneca and Xigris (drotrecogin alfa) from Eli Lilly.
An estimation of the market potential for the API for a developmental drug is a demanding task, particularly for an outsider. With regard to the drug product, the therapeutic class, the market penetration and the number of patients must be taken into consideration. The corresponding data for the best-selling drug with the same indication can be used as a guideline. The market for the drug substance, the yearly dosage is relevant for the volume requirement, and the unit price of the API for the achievable revenues, respectively. A compilation of the relevant data for Gilead Science’s hepatitis C drug Sofosbuvir is shown in table 6.
Probability of Success in Development
With a total size of 10,4525 R&D projects in the pharma pipeline, the reservoir of opportunities for new business appears enormous — at first glance. Reality is different, when taking into account the high attrition rate for new drugs. In preclinical development, more than 99% of the projects fail. Once in clinical development, the probability of success increases from 16% in Phase I, to 26% in Phase II and 66% in Phase III, to close to 100% after submission to the competent health authority authorizing the commercialization (see Figure 1).7
The high drop-out rate in drug development during the clinical phases has significant negative consequences for both the pharma and the fine chemical industry. For the former it means exorbitant R&D expenditures of about $1.5 billion per new drug launched. In addition to the high attrition rate, the staggering cost of R&D is also driven by a lengthy and demanding new product development process involving clinical tests with a large number of volunteers and patients, technical and regulatory challenges. For the fine chemical industry, it is a major source of uncertainty for CMOs looking to rejuvenate their product portfolio. It is also the reason why the “CRAM” strategy (Con-tract Research & Manufacturing) pushed by many Indian companies was not a particular success. It is based on the hypothesis that a backward integration to contract research generates business opportunities for custom manufacturing. In reality this is rarely the case; it is more a stroke of luck than a repeatable strategy.
If fact, when eliminating the drug projects with an unacceptably low probability of success, the adjusted reservoir of opportunities amounts closer to 1,000 developmental drugs than 10,000.
Probability of Securing the Business
Securing the business for a CMO is a major success factor. The main risk elements are the timing of the negotiations with the pharma company, the CMO’s reputation with the customer, and the attractiveness of the CMO offering. In the best case, if the timing is appropriate, the CMO is esteemed and the service offering is attractive, the success factor will be high. In contrast, the CMO is unlikely to win business if its offering is unattractive for the targeted drug, or if the pharma company decides in favor of in-house production.
There are four points of entry along the life cycle of a drug (Figure 38) for a CMO to step in. At gate 1, samples of the new drug molecule are needed for in vitro tests. This business is not attractive because of the small chance of success and small quantities of APIs being produced. Therefore, it should be considered primarily as a promotional tool for acquiring a new customer. At gate 3, the FDA has approved the new drug for commercialization. However, at this point, suppliers have often been selected and supply contracts have already been signed. Therefore gate 3 can only be considered if the price pressure on a product is increasing and there is a chance to submit an attractive offer, e.g., on the basis of cost reductions achieved after development of a substantially more economical second generation manufacturing process. At gate 4 the proprietary drug is approaching patent expiration and generic companies are preparing to enter the field. This gate is suitable for a fine chemical company involved in the manufacturing of API for generics.
For custom manufacturing projects, gate 2 is the pivotal decision point and requires close scrutiny. As drugs approach the completion of Phase IIb clinical trials, several decisive events have taken place: the probability of getting approval for the drug now exceeds 50% while the volume requirements both for the drug product and drug substance are likely to grow rapidly. The procurement office of the pharma company gets involved. It determines the procurement policy, i.e., at which stage of the manufacturing process external suppliers will be chosen for both generics and exclusive products. Requests for proposals have been dispatched to preferred suppliers as well as “niche” CMOs with the specific technical capacity and capability, while unsolicited offers are also received from interested fine chemical companies.
For a pharma company, the selection of the contract manufacturer for a developmental drug is a multi-faceted process, in which both ponderable and imponderable criteria must be taken into account. The key parameters are the depth of the relationship with the CMO, the financial stability of the CMO and the attractiveness of the offering (Figure 4). For important projects — drugs with blockbuster potential — a situation where a launch has to be postponed because of delayed delivery is of great concern to pharma companies. Therefore a trusted “preferred supplier” will be the first choice. It has gained this status due to a track record of successfully completed projects and is most likely to convey a piece of mind for the execution of such valuable projects. Price plays an import role in the valuation of the offering for commercial supplies, yet less so for quotations involving smaller supplies from pilot plants. Additional criteria taken into account in the selection of the CMO include positive audit results, a solid financial situation and the total cost of ownership.3,8 Figure 4 illustrates the probability of securing a specific order as a function of the relationship with the customer and the attractiveness of the offering. Offers from newcomers will only be considered for smaller, early stage projects, or if they have a unique technology indispensable for manufacturing a particular molecule.
Identifying, realizing and maximizing business opportunities from new drug candidates is a question of survival for the fine chemical industry. Improving the project selection process means primarily getting a better grip on the intrinsic characteristic of the CMO’s business model, namely the double exposure to the pharma company’s risk of drug failure on the one hand, and the risk of not being selected for large-scale manufacturing should the drug be approved, on the other hand. Additional challenges include:
- the limited amount of new drugs approved every year: While more than 10,000 R&D projects are active, less than 30 new drugs a year have been launched successfully during the past decade on average.
- the constraint to initiate R&D work when the overall chance of success is still low in order to comply with the project schedule of potential customers.
- Management has to define a strategy focused around the pharma and GMP requirements, articulating a compelling vision and building an appropriately skilled organization, must set performance benchmarks and tracking using a balanced scorecard, and needs to provision for financial resources appropriately.
- Business development has to aggressively identify new opportunities, conduct thorough evaluation (as described in this article), and deliver quickly comprehensive project proposals that highlight the unique competitive strengths of the CMO. Additional measures can be taken to improve the success rate in new project acquisitions: appropriate training, specific promotions and pricing flexibility, the preparation of the offering and contract negotiation are detailed in Table 7.
- R&D has to be ready to develop lab processes and assist with product technology transfers to the pilot plant while manufacturing and QA have to also participate in delivering demonstration batches and ultimately establishing industrial scale production.
- Peter Pollak, Fine Chemicals – The Industry and the Business, 2nd edition (2011). John Wiley & Sons, Hoboken NJ, USA; 280 pages
- World Preview 2013, Outlook to 2018 – Returning to Growth, Evaluate Pharma, London (2013), p.14 www.evaluategroup.com/wp13
- Peter Pollak, Andrew Badrot & Rolf Dach, API Manufacturing – Facts & Fiction, Contract Pharma, Jan./Feb. 2012, pp. 62 and 63
- IMS Health / 2012 Top-Line Market Data / Top 20 Therapeutic Classes
- Ian Lloyd, Pharma R&D Annual Review 2012, Citeline Intelligence (Informa), London (2012)
- Andrew Hill, Saye Khoo, Bryony Simmons and Nathan Ford, What is the minimum cost per person to cure HCV?, 7th IAS Conference on HIV Pathogenesis and Prevention, Kuala Lumpur, Malaysia, July 2013 [TULBPE16]
- J.A. DiMasi & al., Trends in Risks Associated with New Drug Development: Success Rates for Investigational Drugs. Clinical Pharmacology & Therapies (March 2010) 87, pp 272.277
- The Division of Science Resources (SRS) of the National Science Foundation; U.S. Institute of Applied Manpower Research, India (2011) 8. World Intellectual Property Report.
Peter Pollak is a fine chemicals business consultant. He can be reached at firstname.lastname@example.org. Andrew Badrot is partner and chief executive officer of CMS Pharma. He can be reached at email@example.com. An excerpt of this article will be co-published by CHE Manager.