Non-clinical drug development activity today represents a substantial proportion of total R&D spending. According to the Pharmaceutical Research and Manufacturers of America (PhRMA), in 2011 major pharmaceutical and biotechnology companies spent $10.5 billion or 22% of total annual R&D costs on non-clinical research, more than the total spent on Phase I and II activity combined.¹

Non-clinical development strategies have also been changing rapidly. As part of industry-wide efforts to consolidate infrastructure and lower fixed operating costs, pharma and biotech companies have been increasing their use of contract service providers in both clinical and non-clinical areas for capacity and expertise. A 2012 study conducted by the Tufts Center for the Study of Drug Development (Tufts CSDD) estimates that the global contract non-clinical services market is an estimated $4.1 billion. And the number of contract research organizations (CROs) operating in the U.S. between 2000 and 2011 has quadrupled — from 800 to slightly more than 3,200 — with 26% of them offering non-clinical research services.² According to Jeffries & Company, the market for contract non-clinical services is growing at an annual rate of 9.7%.³

Despite the substantial level of research spending and the growing reliance on outsourcing within the non-clinical domain, to our knowledge very little data exists on the economics of specific non-clinical activities and the comparative cost of internal vs. outsourced support. Andrews, Laurencot and Roy in 2006 reported that the direct cost to conduct specific non-clinical tests for a single compound ran from tens to hundreds of thousands of dollars.⁴ Lee-Brotherton also analyzed the costs of specific tests per compound, broken out by animal model, and derived similar estimates.⁵ Ferrandiz, Sussex and Towse in 2012 calculated that the average development costs from first toxicity dose to first human dose for a single compound was $6.5 million (2011$) with the costs ranging from as low as $100,000 to as high as $27 million.⁶ This wide range suggests many different variables affect the cost of non-clinical development.

In 2012, Tufts CSDD conducted a study to capture more detailed cost measures of non-clinical research activity and contract service support and to characterize current practices and perceptions of non-clinical outsourcing  (process chemistry to IND submission). It is our hope that the results of this study will assist R&D managers in making more informed decisions with regards to non-clinical outsourcing and budgeting practices.

Methods
Nine biopharmaceutical companies agreed to participate in intensive interviews and the collection of cost data. Tufts CSDD conducted in-depth interviews by telephone between December 2012 and February 2013. Seven of the nine participating firms represent top 20 biopharma companies. Eight of the nine firms focus on both small and large molecules across multiple therapeutic areas. Interviewees were senior managers (e.g., director to vice president-level) directly involved with non-clinical procurement, budgeting and outsourcing decision-making.

The data collection and in-depth interviews focused on:
1) Budgeting practices and pricing paradigms for non-clinical activities;
2) Perceptions of non-clinical program costs;
3) Perceptions of non-clinical CRO capabilities, service quality, and costs compared to internal costs

Economic data on non-clinical activity was gathered across five primary areas associated with an investigational new drug (IND) application along with resources required to put together the IND. These are:

• Chemistry, Manufacturing and Control (CMC)
• Pharmacology
• Metabolism
• Pharmacokinetics (PK)
• Toxicology
• FTEs involved in IND preparation

The survey consisted of 11 questions that were both qualitative and quantitative in nature. Companies were also asked to define fully loaded costs and the various inputs that are used to calculate them. The data collection instrument measured costs of specific programs or molecules starting from process chemistry or the decision to commit resources to the compound to IND submission. If the respondent company did not fill out the data collection instrument on specific program costs prior to the interview, individuals were asked to provide either cost of an average program or their perceptions of costs.

For questions involving perceptions of CRO costs relative to internal costs, respondents rated variables using a four-point Likert scale ranging from ‘1’ More Expensive to ‘4’ Less Expensive. Respondents also provided their overall perception of the efficiency of service provided by non-clinical CROs compared to in-house practices. Participants used the following scale: ‘1’ More Efficient; ‘2’ Somewhat More Efficient; ‘3’ Somewhat Less Efficient or ‘4’ Less Efficient.

Respondents were asked to rate the criteria that played a major role in the decision to select a vendor and to prioritize their importance. Selection criteria included cost, reputation, animal facility capability, CRO location, and location of the animal facility.

Results
Internal Cost Accounting and Budgeting Practices
Half of the executives interviewed noted that while non-clinical cost data is accessible in their organizations, they lacked the tools to analyze and manage this information.

Four out of eight respondents reported that their non-clinical cost data are not available at a granular level. Non-clinical costs are typically aggregated within the broad category of preclinical R&D expenditure. As a result, interviewees felt that separating out detailed cost data was not necessary as it would be too resource intensive. One executive from a midsize company commented that it would be cost prohibitive: “[O]ne would need to have a full-time group to collect and maintain such a database.”

Similarly, respondents report that they do not monitor and analyze CRO costs for providing non-clinical services. Several interviewees noted that doing so is costly and offers limited value since demand for CRO services is expected to increase.

Determination of Total Non-clinical and Fully Loaded FTE Costs
Six out of nine organizations included staff salaries and benefits, overhead, supplies, infrastructure, hardware, and operating costs in their definition of total (fully loaded) non-clinical costs.

Very few respondent companies have determined the fully loaded cost (e.g., salaries, benefits, overhead) of their FTE personnel. Of those who have, the cost estimates varied widely from company to company. Several respondents mentioned that they are comfortable using a benchmark value of $300,000 as the fully loaded cost for one FTE in non-clinical development, based on a case example that has been shared at conferences and in the trade press.⁷

Breakdown of Non-clinical Costs
Non-clinical costs represent those from process chemistry to IND submission and encompass staff salaries and benefits, overhead, supplies, infrastructure, hardware, and operating costs, in addition to outsourcing costs. Data are not corrected for differences in the cost of living among all organizations.

The average total non-clinical cost supporting a program is $6.2 million, ranging from a minimum of $698,000 to a maximum of $20 million. CMC costs represent the largest cost area. On average, $3.1 million (coefficient of variation (CoV) 1.0) is spent on CMC activity, accounting for 50% of the total non-clinical budget. Toxicology studies are nearly a quarter of the total non-clinical budget, or $1.5 million (coefficient of variation 0.9). Pharmacology studies represent the third largest area, at 10% of the total non-clinical budget, or $600,000 (coefficient of variation 1.0). Table 1 contains a breakdown of total non-clinical costs across six primary areas supporting an IND application.

Outsourcing Practices in Non-clinical Research
Seven out of nine participating companies reported that they regularly outsource select non-clinical study tasks. Toxicology and PK studies are top activities that are outsourced regularly. Interviewees indicate that they frequently outsource CMC, pharmacology and metabolism studies. IND preparations are reportedly rarely outsourced. Figure 1 presents the frequency of outsourcing across the six primary non-clinical areas.

Two companies reported that they are not outsourcing non-clinical activity, as they have the infrastructure, capacity and capability to handle their needs in-house.


Respondents believe that the decision to outsource depends on a number of factors, including resource availability and in-house manufacturing capabilities. Many companies reported that they tend to keep early stage preparation work, such as dosing studies, in-house, while more routine screenings and tasks (e.g. toxico-kinetic reports) are outsourced to CROs.

Other tasks typically kept in-house are those requiring specialized knowledge, including strategy, planning and IND preparations. These are considered core capabilities.

Criteria for Selecting a Non-Clinical CRO Partner
Eight participating companies reported that they ‘regularly’ consider CRO reputation when selecting a service provider. Seven companies reported ‘regularly’ considering a non-clinical service provider’s animal facility capabilities as a selection criterion. CRO animal facility location, CRO location and cost were criteria used far less frequently according to participating companies. Five participating companies reported that they rarely, if ever, consider a CRO’s animal facility location, and six report that they rarely consider a CRO’s location and cost when choosing a non-clinical service provider. Figure 2 has a breakdown of selection criteria and the frequency of their use.

All participating companies agreed that quality, experience, and reputation were the three most important factors that influenced a company’s decision to select a non-clinical CRO partner. Six companies mentioned that they prefer working with CROs in the U.S. or Europe, as these countries adhere to strong animal welfare standards. Respondents were hesitant to partner with CROs in countries where standards for animal welfare are behind those established in the U.S. and Europe.

Perceptions of Non-Clinical CRO Costs and Efficiency
All respondents agreed that CROs are more cost-effective and time-efficient than internal teams. Participating companies believe that CRO business models are very different than their biopharma company counterparts in that CROs are profit motivated. As such, CROs are constantly looking for ways to improve profitability and continuously improve their operating performance and efficiency. Moreover, participating companies believe that CROs have the ability to spread overhead costs over a larger number of active non-clinical studies.

Respondents also believe that CROs have access to larger databases from past and present studies from which they can draw relevant data points for modeling and simulation analysis.

Eight participating companies noted that CROs have established scale efficiencies as they have specialized in select non-clinical areas, thus giving the CRO a competitive advantage over internal staff and in-house activity. As one company put it, “CRO non-clinical units are more efficient because that is their focus, it’s what they do. [The CRO] deals with a large number of different sponsors in areas that they specialize in.”

Participating companies also reported that they perceive non-clinical CROs lack staff that is highly specialized. Respondent companies also perceive that non-clinical CROs are spread too thin and unable to provide individualized focus and attention in their efforts to service many different clients — a problem that an in-house development team within a pharma company does not have to face.

Another area where CROs are perceived to fall short is service quality. Five participating companies said that the quality of work performed by external vendors tends to be lower than if the procedures were performed in-house. One major pharma company stated, for example, “Although CROs tend to do things faster, the output quality is lower. They have assembly lines ready but don’t have the specialized staff.”

Discussion
In this study, the average total cost to support a non-clinical program is $6.2 million. CMC and toxicology activities are the most expensive elements of non-clinical programs. The results from this Tufts CSDD study corroborate the findings from the Ferrandiz, Sussex and Towse 2012 study. Minor differences are due to variations in study methodology. Both studies suggest that non-clinical costs are not consistent across organizations, suggesting that variables other than staff salaries and benefits, overhead, supplies, infrastructure, hardware, and operating costs affect actual costs. One such element could be the variations in the cost of living across different organizations, as well as differences in outsourcing levels.

The results also indicate that major pharma and biotech companies are not routinely collecting non-clinical cost data. Companies are finding it difficult to justify the expense of doing so given that the data is dispersed and not easily accessed or aggregated. Additionally, sponsor companies are not routinely and methodically tracking non-clinical outsourcing costs. Most major pharma company executives perceive that CROs are more cost-effective and more time-efficient than are internal teams. However, companies have doubts about the quality of work performed by CROs. The study findings also show that companies are regularly outsourcing toxicology and PK studies while retaining IND Document preparation in-house.

CRO company reputation was considered by participating companies as the most important selection factor. A close second was the company’s adherence to European/American animal welfare standards. Cost was not considered as important a factor due in part to sponsor company perceptions that CROs are less expensive, more efficient and faster than their internal counterparts.

Currently, sponsor companies are primarily outsourcing the more routine, later stage and less IP-generating tasks such as toxicology studies. As sponsor companies continue to consolidate and look for operating efficiencies, earlier stage non-clinical areas may become more commonly outsourced activities. Biopharma companies are not yet outsourcing activities that are further down in the non-clinical process.

As senior management increases its pressure on non-clinical functions to more effectively manage non-clinical costs, we expect sponsor companies to become more adept at routinely and systematically planning, collecting, evaluating, and forecasting detailed non-clinical economic data. Technology solutions will likely play a growing role in assisting sponsor companies in coordinating and accessing this data.

There are a number of limitations to this study that should be noted. The companies that participated in this study represent a convenience sample of primarily top biopharma companies. The results should be viewed with some caution as they may not be indicative of the industry as a whole — particularly mid-sized and smaller organizations.

Some of the specific cost data analyzed was incomplete, as companies have great difficulty accessing and compiling this data. As such, sample sizes in some cases are small, suggesting again that the results should be viewed with some caution. Still, this study represents an initial effort to gather data that will no doubt become more important for sponsor companies to collect and monitor in the future. Subsequent studies will strive to gather more extensive and more granular cost data.

The results of this Tufts CSDD study provide insight into sponsors’ non-clinical budgeting and outsourcing practices and perceptions. The results also assist CROs in understanding the current state of non-clinical outsourcing and in anticipating future opportunities as CRO usage expands and evolves. 

References

1. Pharmaceutical Research and Manufacturers of America: 2013 Biopharmaceutical Research Industry Profile.  http://www.sciencedirect.com/science/book/9780080466170. Published April 2013.  Accessed May 20, 2013.
2. Getz KA, Lamberti MJ, Mathias AB, Stergiopoulos S.  Resizing the Global Contract R&D Services Market.  Contract Pharma. May 30, 2012.  http://www.contractpharma.com/issues/2012-06/view_features/resizing-the-global-contract-rd-services-market/.  Accessed May 20, 2013.
3. Jeffries & Company.  CRO Industry Size Model, 2011-2015.  In: Mathieu M, ed.  Parexel Biopharmaceutical R&D Statistical Sourcebook 2012/2013.  Waltham, MA: Parexel International Corporation, 2012:47-48.
4. Andrews, Laurencot, Roy.  Estimates of Nonclinical Study Costs and Durations.  In: Mathieu M, ed.  Parexel Biopharmaceutical R&D Statistical Sourcebook 2012/2013.  Waltham, MA: Parexel International Corporation, 2012:239.
5. Lee-Brotherton V.  Preclinical Development Planning for Emerging Pharma and Biotech Firms.  Presented at: MaRS Workshop: From Benchtop to IND; May 13, 2008.  http://www.slideshare.net/webgoddesscathy/preclinical-development-planning-for-emerging-pharma-and-biotech-firms.  Accessed May 20, 2013.
6. Mestre-Ferrandiz J, Sussex J, Towse A; Office of Health Economics.  The R&D Cost of a New Medicine.  http://www.ohe.org/object/display.cfm?serv=2&id=124#124.  Published December 2012.  Accessed May 20, 2013.
7. Rydzewski, R. Chapter 1 – The Drug Discovery Business to Date.  In: Rydzewski, R.  Real World Drug Discovery: A Chemist’s Guide to Biotech and Pharmaceutical Research.  Oxford, UK: Elsevier; 2008:33.  http://www.sciencedirect.com/science/book/9780080466170.  Accessed May 20, 2013.

---

Stella Stergiopoulos is a project manager at the Tufts Center for the Study of Drug Development (Tufts CSDD). She can be reached at stella.stergiopoulos@tufts.edu. Jennifer Kim is a research analyst at Tufts CSDD. Kenneth Getz, MBA, is the director of Sponsored Research and associate professor at Tufts CSDD. This project was funded by an unrestricted grant from Aptuit.