Clinical Therapeutics/Volume 36, Number 10, 2014
Original Research
Taking the Pulse of Strategic Outsourcing Relationships Kenneth A. Getz, MBA; Mary Jo Lamberti, PhD; and Kenneth I. Kaitin, PhD Tufts Center for the Study of Drug Development, Tufts University School of Medicine, Boston, Massachusetts. ABSTRACT Purpose: Articles in peer-reviewed journals and the trade press presuppose that strategic outsourcing relationships have been formed to replace preexisting collaborative approaches with contract research organizations. They do not consider that large, fragmented pharmaceutical and biotechnology companies may be supporting competing and conflicting relationship models simultaneously. A recent Tufts Center for the Study of Drug Development study quantifies actual strategic outsourcing practices among drug development companies and sheds new light on why these relationships may be failing. Methods: Tufts Center for the Study of Drug Development conducted an in-depth assessment of 43 Phase II and III clinical studies completed since 2012 to examine the outsourcing relationships used by 9 major pharmaceutical and biotechnology companies to support key functional areas. Descriptive statistics were assessed and t tests were performed to characterize outsourcing practices by function and to determine differences in study performance between transactional and strategic outsourcing relationships. Findings: The results indicate that sponsor companies are using a variety of outsourcing relationship models to support their studies, mixing and matching the use of internal staff, and using traditional transactional and strategic outsourcing relationships simultaneously. Specifically, despite the fact that each sponsor company had entered into several strategic outsourcing relationships, in no instance did a single contract research organization manage all functional areas supporting an individual Phase II or III study. In addition, sponsor companies vary the types of outsourcing relationship models that they use on a study-by-study basis. Accepted for publication September 17, 2014. http://dx.doi.org/10.1016/j.clinthera.2014.09.008 0149-2918/$ - see front matter & 2014 Published by Elsevier HS Journals, Inc.
October 2014
Implications: The inability of pharmaceutical and biotechnology companies to consistently embrace and coordinate sourcing strategies is creating internal friction and inefficiency. As a result, the expected impact of strategic outsourcing relationships on drug development performance, quality, and cost remains elusive. (Clin Ther. 2014;36:1349–1355) & 2014 Published by Elsevier HS Journals, Inc. Key words: contract research organizations, functional relationships, outsourcing, pharmaceutical company, research and development, strategic alliances, transactional relationships.
INTRODUCTION In 2014, half of every pharmaceutical research and development (R&D) dollar was spent on contract service providers. Moreover, the level of R&D spending apportioned to outsourcing will likely continue to rise.1 The current operating environment demands more variable infrastructure from an integrated and coordinated collective of internal and external sources to support complex, global R&D programs. During the past 20 years, the decision to hire contract service providers was driven in large part by the need to employ a variable head count to support peak periods of drug development activity. Sponsor companies also looked to their contract service providers to gain access to scientific expertise and for assistance in expanding into unfamiliar emerging global regions.2 More recently, pharmaceutical and biotechnology companies have experienced massive downsizing and consolidation, in part due to global economic conditions. According to Challenger, Gray & Christmas, Inc Scan the QR Code with your phone to obtain FREE ACCESS to the articles featured in the Clinical Therapeutics topical updates or text GS2C65 to 64842. To scan QR Codes your phone must have a QR Code reader installed.
1349
Clinical Therapeutics (Chicago, Illinois), since 2003, 4325,000 biopharmaceutical positions have been eliminated, with an estimated 35% of the total coming from R&D divisions.3 As a result of downsizing and other cost-saving measures, pharmaceutical and biotechnology companies are turning to outsourcing to supplement or replace critical functional support that is no longer available internally. At the same time that the receptivity to and use of outsourcing are increasing, a growing number of pharmaceutical and biotechnology companies have implemented more integrated and coordinated engagements with partnered contract research organizations (CROs). In contrast to traditional outsourcing relationships, where sponsor companies contract with service providers to deliver single or multiple tasks on a per-project basis, these newer strategic relationships involve CRO companies providing single functional or multifunctional support for entire programs across large portions of sponsor company portfolios. Strategic relationships hold the promise of offering higher levels of efficiency at lower cost to the sponsor organization through the use of advanced planning, dedicated staffing, shared governance, and shared data and management control systems and procedures, with fewer sponsor staff overseeing CRO execution. These relationships are typically established under preferred pricing arrangements, offering cost savings to the sponsor. All of the top 30 largest pharmaceutical and biotechnology companies have now entered into Z1 strategic outsourcing relationship.4 Major sponsor companies establish a mean of 3 strategic relationships to support their portfolios. However, although strategic outsourcing relationships have been widely adopted during the past 7 years, there is little evidence to suggest that they are contributing to faster and more efficient R&D activity on a consistent basis. Research by the Tufts Center for the Study of Drug Development (CSDD) indicates that drug development remains a highly risky activity. Only 16% of drugs entering clinical testing today are approved by the Food and Drug Administration, down from 21.5% in the 1990s.5 Tufts CSDD research has also shown that clinical phase durations are 15% longer than they were in the early 1990s and that sponsor companies must typically double the planned enrollment period to give investigative sites enough time to recruit study volunteers and complete a given clinical trial. In addition, between 2003 and 2013, the pharmaceutical industry spent a cumulative $1.3 trillion (US$) on R&D, resulting in 296
1350
new drug and biologic approvals. This translates into $4.4 billion (US$) mean annual cost per approved molecule. In that 10-year period alone, the mean development cost per new chemical and biologic entity increased 30% from the first half of the decade to the second half.6 Initial reports in 2010 from 2 early adopters—Pfizer, Inc (New York, New York) and Eli Lilly & Co (Indianapolis, Indiana)—suggested that strategic, integrated relationships resulted in cost savings, largely through reductions in the level of oversight required and in the number of vendors bidding on project work. Eli Lilly, for example, reported 20% cost savings on data management and monitoring and a 93% improvement in monthly patient enrollment volume. Pfizer reported saving $20 million annually through consolidated management of its vendors, from 150 to 17; a 26% reduction in enrollment cycle time; and an 80% reduction in the number of contracts delayed by 4120 days.7 But recent assessments from a variety of sources, based on much larger numbers of sponsor companies, present a more mixed picture. For example, a study of 89 sponsor companies conducted by the Avoca Group (Princeton, NJ) in 2012 found that 1 of 5 sponsors had terminated a strategic relationship with a CRO. That same study noted that approximately one third of sponsors surveyed were not satisfied with the performance of their strategic relationships, and nearly half did not believe that their strategic relationships would ever achieve the intended outcome.8 A 2013 study conducted by Vantage Partners (Boston, Massachusetts) of 81 sponsor company representatives and 88 CRO company representatives found that a large percentage of strategic relationships were unable to establish a collaborative working style and failed to align expectations and capabilities. In addition, one third of the sponsor organizations reported that their strategic alliances had not delivered expected cost and time savings, and 56% said that their relationships were not delivering innovative solutions.9 In 2012, as part of a broader study, Tufts CSDD assessed the frequency of change orders in studies for which most of its budget was allocated to contract service providers. Change orders are defined as formal, written changes to the original scope of work agreement. Tufts CSDD hypothesized that strategic relationships would result in fewer change orders because the parties share planning and governance, and the CRO partner is given more autonomy to manage the study. However, the findings revealed no Volume 36 Number 10
K.A. Getz et al. difference in the frequency of change orders between traditional, transactional relationships and strategic relationships, suggesting that the latter are having limited effect on R&D efficiency (Figure 1).10 The authors of the Avoca Group and Vantage Partners studies concluded that the primary causes of relationship failure are cultural barriers, poor communication, failure to achieve buy-in from senior staff and affiliates, and unreasonable expectations. Other industry professionals have argued that strategic relationships are a fallacy and that true partnership can never be achieved as long as sponsor companies lack the ability to truly measure their operating costs and profitability.11,12 Most articles in the trade press that discuss sponsor-CRO relationships primarily focus on negotiating and establishing collaborative processes and on measuring general attitudes and satisfaction. Most of the articles presuppose that strategic outsourcing relationships have been formed to replace preexisting collaborative approaches with CROs. They do not consider that large, fragmented organizations may be supporting competing and conflicting relationship models simultaneously and that, in doing so, they may detract from the promise offered by new strategic relationship models. The recent Tufts CSDD study, described herein, sheds new light on why strategic outsourcing relationships may be failing.
STUDY BACKGROUND In early 2014, Tufts CSDD contacted each of the top 20 largest pharmaceutical companies to solicit their
interest in participating in an in-depth assessment. All the companies contacted use outsourcing extensively, and they had entered into strategic relationships with CROs at least 3 years before the start of the study. Nine companies agreed to participate. Between March and May 2014, these 9 companies evaluated specific contract services used on their individual clinical studies. Each of the companies selected representative Phase II and III studies conducted and completed since 2012. Representative studies were those with similar scope (eg, number of enrolled patients, number of countries, and number of investigative sites) and therapeutic areas as those found in each company’s development portfolio. For each study, sponsor companies met with their clinical teams to document outsourcing relationship models used to support specific functional areas. Sponsor companies also indicated whether internal staff (including temporary personnel) supported a given functional requirement. Transactional relationships are those in which the sponsor engages a contract service organization to provide a specific task for an individual project requirement. Strategic relationships are those in which the sponsor enters into a longerterm relationship with a contract service organization to provide single functional or multifunctional support across projects in a sponsor’s portfolio under shared governance and integrated systems and practices.13 In total, 11 functional areas were examined. These areas were selected to capture outsourcing practices across 3 broad clinical study requirements: (1) study
Transactional Relationships
Strategic Relationships
(Mean Number per Study)
(Mean Number per Study)
5.3
4.1
2.4
2.3
2.5
1.1
Phase I
2.2
1.1
Phase II
Phase III
Phase IV
Phase I
Phase II
Phase III
Phase IV
Figure 1. Change orders by outsourcing relationship model. Source: Tufts CSDD 2012.
October 2014
1351
Clinical Therapeutics selection. Most studies also outsourced data management (80%), medical monitoring (73%), patient recruitment and retention (68%), and statistical analysis (63%). The results have been organized into 3 groups of functions supported by outsourcing—study design and regulatory affairs, site management and monitoring, and data management and dossier creation—to summarize relationship use. Despite the fact that each sponsor company had entered into several strategic outsourcing relationships, in no instance did a single CRO manage all functional areas supporting an individual Phase II or III study. Internal staff typically managed protocol design and regulatory affairs functions, with less than 1 in 5 supported by a transactional or strategic outsourcing relationship (Figure 3). Site management and study monitoring functions were the most heavily outsourced. Functional requirements in this area were supported by transactional relationships for 40% to 50% of studies. Strategic relationships were used to meet functional requirements for 25% to 40% of all studies (Figure 4). Outsourcing relationship models used for data management and dossier creation and submission were the most varied. For most studies, internal staff was used to support medical writing and regulatory document submission preparation. Data management and statistical analysis functions, however, used outsourcing relationships extensively. As shown in Figure 5, approximately 40% of all studies used
design and regulatory affairs, (2) site management and study monitoring, and (3) data management and dossier creation and submission. Detailed data on 43 Phase II and III studies were collected. The studies were representative of multiple disease conditions: 20% in central nervous system diseases and disorders; 18% in oncology; 10% each in respiratory, endocrinology, immunology, and infectious diseases; and 5% each in cardiovascular and hematologic disorders. Participating companies also provided performance and quality data for each individual study to determine whether the independent variables (outsourcing relationship model used) had a measurable impact. Studyspecific performance and quality data collected included planned to actual study startup cycle time, planned to actual study conduct cycle time, planned to actual study closeout cycle time, screen failure rate, study volunteer dropout rate, number of protocol amendments, number of protocol violations; number of change orders, and queries per thousand clinical data points collected. Twotailed t tests were performed to evaluate differences between dependent variables by transactional versus strategic outsourcing relationship used per study.
RESULTS Figure 2 describes the incidence of services that were outsourced for each Phase II and III study evaluated. Nearly all of the studies (95%) outsourced site monitoring, and 86% outsourced site identification and
Percentage of Total Studies Site Monitoring
95% 86%
Site Identification and Selection
80%
Data Management 73%
Medical Monitoring
68%
Patient Recruitment and Retention
63%
Statistical Analysis 38%
Central Site Payment Administration
34%
Medical Writing
30%
Submission Preparation Regulatory Protocol Design
18% 15%
Figure 2. Incidence of services outsourced for 43 Phase II and III studies.
1352
Volume 36 Number 10
K.A. Getz et al. 12% 2%
14% 5%
86%
81%
Protocol Design
Regulatory Affairs
Strategic Relationships (FSP and Integrated Alliances) Transactional (Full and Niche Service Models) In-house
Figure 3. Relationship models used for study design and regulatory affairs for 43 Phase II and III studies. FSP ¼ full-service provider.
transactional outsourcing relationships, and 25% to 40% used strategic outsourcing relationships. Clinical research studies conducted by the same sponsor company used different outsourcing relationship models without a clear or systematic pattern. One company, for example, varied its use of in-house staff and used transactional (full-service provider) as well as strategic relationships on each study to manage the site monitoring function. Another company alternated between transactional service providers and a strategic partner to provide statistical analysis. The same variation in outsourcing relationship use was observed
40%
26%
in site identification and management and in data management functions. With respect to performance and quality, no statistically significant differences were observed between studies managed by transactional versus strategic relationship partners with 2 exceptions: the number of screen failures was higher in studies managed by strategic outsourcing relationships (mean [SD] difference, 0.133 [0.22]; P o 0.03) and the frequency of protocol amendments was lower in studies managed by strategic outsourcing relationships (mean [SD] difference, 1.28 [2.01]; P o 0.02).
26%
40%
19% 28%
42%
51% 51%
51%
54%
32% 9% Site Identification Patient and Selection Recruitment and Retention
9% Site Monitoring
23% Medical Monitoring
Central Site Payment Administration
Strategic Relationships (FSP and Integrated Alliances) Transactional (Full and Niche Service Models) In-house
Figure 4. Relationship models used for site management and monitoring for 43 Phase II and III studies. FSP ¼ full-service provider.
October 2014
1353
Clinical Therapeutics
37%
26%
21% 12%
12% 19%
37% 42% 67%
69%
Medical Writing
Submission Preparation
37% 21% Data Management
Statistical Analysis
Strategic Relationships (FSP and Integrated Alliances) Transactional (Full and Niche Service Models) In-house
Figure 5. Relationship models used for data management and dossier creation for 43 Phase II and III studies. FSP ¼ full-service provider.
DISCUSSION Based on the Tufts CSDD assessment of sponsor company use of different outsourcing relationship models, it is apparent that companies mix and match a variety of outsourcing approaches in clinical studies to meet specific functional needs. Despite the investment and effort required to identify, select, and integrate strategic outsourcing relationships, sponsor companies clearly choose to keep all options on the table. The outsourcing practices observed in this study indicate that sponsor companies are not fully invested in strategic relationships, ie, they are not ready to hand over the keys—and operating risk—to their strategic partners. Although the mix-and-match approach offers pharmaceutical and biotechnology companies some flexibility, this study suggests that it compromises the conceptual benefits of strategic relationships. Specifically, sponsor company staff must simultaneously support processes and procedures that accommodate multiple relationship models. In addition, collaborative practice experience and efficiency are isolated to individual studies and do not carry across the development portfolio. Numerous dynamics play a role in the use of multiple sourcing models by sponsor companies. The high level of fragmentation within major pharmaceutical companies and the difficulty in implementing and enforcing new practices no doubt contribute to the difficulties in integrating and coordinating the use of select providers. Moreover, aversion to risk entices sponsor company
1354
staff to hedge their bets by maintaining multiple contract service providers under a variety of sourcing models. Interestingly, the labor- and time-intensive nature of sponsor company participation in this Tufts CSDD study provided insight into the level of fragmentation within participating companies. Clinical teams had considerable difficulty determining which outsourcing relationships were used across multiple functional areas for their recently completed studies, suggesting that cross-functional awareness, communication, and coordination are limited. Perceptions and attitudes about contract service providers—as noted by the Avoca Group and Vantage Partners—are deeply ingrained in company culture and also are a major factor in the types of relationships companies establish with CROs. The long-held view that CROs are commodity service providers that must be carefully policed by sponsor staff to ensure quality is hard to change and contributes to distrust in strategic relationship partners. Given the way that sponsor companies are using their strategic relationships, it is not surprising that study performance and quality are not improving. Significantly higher screen failure rates in studies supported by strategic relationships are likely associated with the use of less experienced investigative sites in emerging global regions. The lower frequency of protocol amendments in studies managed by strategic partners is a positive outcome that may reflect greater collaborative input into study design feasibility before protocol approval.
Volume 36 Number 10
K.A. Getz et al. This study has 2 limitations of note, both associated with selection bias. The 9 major pharmaceutical companies participating in this study represent a highly active drug development subgroup, but their outsourcing practices may not reflect those of smaller companies. In addition, bias may have been introduced when participating companies were asked to self-select studies for evaluation. Future studies will benefit by a larger and more representative sample of companies and projects. Although a growing number of sponsor companies expect their use of strategic outsourcing relationships to continue to increase, it seems that the anticipated positive impact of these relationships on performance, quality, and cost may be elusive. To date, the implementation and integration of strategic outsourcing relationships has fallen short, and some sponsor companies may lose patience with this approach. Indeed, at recent industry conferences, several sponsors have indicated that they are evaluating whether to increase the number of internal staff to support their studies.
CONCLUSIONS If the conceptual benefits of strategic outsourcing relationships are to be realized, it is critical that sponsor companies recognize that their internal sourcing approaches lead to friction and inefficiency. New dedicated internal mechanisms may assist sponsor companies in identifying and managing this friction more effectively. And this mechanism may ultimately serve as a catalyst in assisting organizations in transitioning to a more consistent and leveraged approach to strategic outsourcing relationships. Internal metrics to assess the level of interaction and coordination between different outsourcing relationship models will also assist sponsor companies in understanding ways to adjust their use of sourcing approaches. Sponsor organizations, arguably facing their most difficult operating environment in history, must improve the effectiveness of their contract service provider management practices to best leverage this integral drug development collaborative partner.
ACKNOWLEDGMENTS Mr. Getz was the principal investigator on this study involved in all aspects of the research including study design, execution, analysis and reporting on the
results and co-authoring the manuscript. Dr. Lamberti was the senior project manager on this research involved in study design, execution, analysis, reporting on the results and co-authoring the manuscript. Dr. Kaitin was a co-author of this manuscript.
CONFLICTS OF INTEREST The authors have indicated that they have no conflicts of interest regarding the content of this article.
REFERENCES 1. Getz K, Lamberti MJ, Mathias A, Stergiopoulos S. Resizing the global contract R&D services market. Contract Pharma. 2012;14:54–62. 2. Scott N, Binkowitz B, Ibia EO, et al. Results of a survey of PhRMA member companies on practices associated with multiregional clinical trials. TIRS. 2011;45:609–617. 3. Chemjobber. 2013 not shaping up to be a great year for pharma: Challenger reports 19,507 jobs lost. http:// chemjobber.blogspot.com/2013/11/2013-not-shaping-upto-be-great-year.html. Accessed September 30, 2014. 4. Getz K. Profound changes in the outsourcing landscape. Pharmaceutical Executive. Published February 29, 2012. http://www.pharmexec.com/pharmexec/Outsourcing/Pro found-Changes-in-the-Outsourcing-Landscape/ArticleStan dard/Article/detail/762383. Accessed September 30, 2014. 5. DiMasi J, Feldman L, Seckler A, Wilson A. Trends in risks associated with new drug development: success rates for investigational drugs. Clin Pharmacol Ther. 2010;87:272–277. 6. Kaitin K, DiMasi J. Pharmaceutical innovation in the 21st century: new drug approvals in the first decade 2000-2009. Clin Pharmacol Ther. 2011;89:183–188. 7. Getz K. FIPNet: pharma’s new, sexy, but not yet ready for print-time model. Applied Clinical Trials. 2009(Suppl): 10– (12,14,16). 8. Henderson L. The state of CRO and sponsor relationships. Applied Clinical Trials. 2013;10:26–28. 9. Hughes J, Rayer E, Wadd J. Sponsor-CRO Collaboration Study. Boston MA. Vantage Partners LLC. May 2013. 10. Kaitin KI, ed. Outsourcing to maximize clinical trial operating efficiency. Tufts Center for the Study of Drug Development R&D Management Report. 2013 Oct;8(4). 11. Parrett A. Challenging the value of strategic partnerships in clinical trial outsourcing. Applied Clinical Trials. 2013;4:30–33. 12. Waife R. Partnership heresy. Journal of Clinical Research Best Practice. 2012;8:1–4. 13. Getz KA. Perils and promise of risk imbalance. Inside Outsourcing. 2012(December):10–16.
Address correspondence to: Kenneth Getz, MBA, Tufts CSDD, 75 Kneeland St, Suite 1100, Boston, MA 02111. E-mail: [email protected]
October 2014
1355
Original Research
Taking the Pulse of Strategic Outsourcing Relationships Kenneth A. Getz, MBA; Mary Jo Lamberti, PhD; and Kenneth I. Kaitin, PhD Tufts Center for the Study of Drug Development, Tufts University School of Medicine, Boston, Massachusetts. ABSTRACT Purpose: Articles in peer-reviewed journals and the trade press presuppose that strategic outsourcing relationships have been formed to replace preexisting collaborative approaches with contract research organizations. They do not consider that large, fragmented pharmaceutical and biotechnology companies may be supporting competing and conflicting relationship models simultaneously. A recent Tufts Center for the Study of Drug Development study quantifies actual strategic outsourcing practices among drug development companies and sheds new light on why these relationships may be failing. Methods: Tufts Center for the Study of Drug Development conducted an in-depth assessment of 43 Phase II and III clinical studies completed since 2012 to examine the outsourcing relationships used by 9 major pharmaceutical and biotechnology companies to support key functional areas. Descriptive statistics were assessed and t tests were performed to characterize outsourcing practices by function and to determine differences in study performance between transactional and strategic outsourcing relationships. Findings: The results indicate that sponsor companies are using a variety of outsourcing relationship models to support their studies, mixing and matching the use of internal staff, and using traditional transactional and strategic outsourcing relationships simultaneously. Specifically, despite the fact that each sponsor company had entered into several strategic outsourcing relationships, in no instance did a single contract research organization manage all functional areas supporting an individual Phase II or III study. In addition, sponsor companies vary the types of outsourcing relationship models that they use on a study-by-study basis. Accepted for publication September 17, 2014. http://dx.doi.org/10.1016/j.clinthera.2014.09.008 0149-2918/$ - see front matter & 2014 Published by Elsevier HS Journals, Inc.
October 2014
Implications: The inability of pharmaceutical and biotechnology companies to consistently embrace and coordinate sourcing strategies is creating internal friction and inefficiency. As a result, the expected impact of strategic outsourcing relationships on drug development performance, quality, and cost remains elusive. (Clin Ther. 2014;36:1349–1355) & 2014 Published by Elsevier HS Journals, Inc. Key words: contract research organizations, functional relationships, outsourcing, pharmaceutical company, research and development, strategic alliances, transactional relationships.
INTRODUCTION In 2014, half of every pharmaceutical research and development (R&D) dollar was spent on contract service providers. Moreover, the level of R&D spending apportioned to outsourcing will likely continue to rise.1 The current operating environment demands more variable infrastructure from an integrated and coordinated collective of internal and external sources to support complex, global R&D programs. During the past 20 years, the decision to hire contract service providers was driven in large part by the need to employ a variable head count to support peak periods of drug development activity. Sponsor companies also looked to their contract service providers to gain access to scientific expertise and for assistance in expanding into unfamiliar emerging global regions.2 More recently, pharmaceutical and biotechnology companies have experienced massive downsizing and consolidation, in part due to global economic conditions. According to Challenger, Gray & Christmas, Inc Scan the QR Code with your phone to obtain FREE ACCESS to the articles featured in the Clinical Therapeutics topical updates or text GS2C65 to 64842. To scan QR Codes your phone must have a QR Code reader installed.
1349
Clinical Therapeutics (Chicago, Illinois), since 2003, 4325,000 biopharmaceutical positions have been eliminated, with an estimated 35% of the total coming from R&D divisions.3 As a result of downsizing and other cost-saving measures, pharmaceutical and biotechnology companies are turning to outsourcing to supplement or replace critical functional support that is no longer available internally. At the same time that the receptivity to and use of outsourcing are increasing, a growing number of pharmaceutical and biotechnology companies have implemented more integrated and coordinated engagements with partnered contract research organizations (CROs). In contrast to traditional outsourcing relationships, where sponsor companies contract with service providers to deliver single or multiple tasks on a per-project basis, these newer strategic relationships involve CRO companies providing single functional or multifunctional support for entire programs across large portions of sponsor company portfolios. Strategic relationships hold the promise of offering higher levels of efficiency at lower cost to the sponsor organization through the use of advanced planning, dedicated staffing, shared governance, and shared data and management control systems and procedures, with fewer sponsor staff overseeing CRO execution. These relationships are typically established under preferred pricing arrangements, offering cost savings to the sponsor. All of the top 30 largest pharmaceutical and biotechnology companies have now entered into Z1 strategic outsourcing relationship.4 Major sponsor companies establish a mean of 3 strategic relationships to support their portfolios. However, although strategic outsourcing relationships have been widely adopted during the past 7 years, there is little evidence to suggest that they are contributing to faster and more efficient R&D activity on a consistent basis. Research by the Tufts Center for the Study of Drug Development (CSDD) indicates that drug development remains a highly risky activity. Only 16% of drugs entering clinical testing today are approved by the Food and Drug Administration, down from 21.5% in the 1990s.5 Tufts CSDD research has also shown that clinical phase durations are 15% longer than they were in the early 1990s and that sponsor companies must typically double the planned enrollment period to give investigative sites enough time to recruit study volunteers and complete a given clinical trial. In addition, between 2003 and 2013, the pharmaceutical industry spent a cumulative $1.3 trillion (US$) on R&D, resulting in 296
1350
new drug and biologic approvals. This translates into $4.4 billion (US$) mean annual cost per approved molecule. In that 10-year period alone, the mean development cost per new chemical and biologic entity increased 30% from the first half of the decade to the second half.6 Initial reports in 2010 from 2 early adopters—Pfizer, Inc (New York, New York) and Eli Lilly & Co (Indianapolis, Indiana)—suggested that strategic, integrated relationships resulted in cost savings, largely through reductions in the level of oversight required and in the number of vendors bidding on project work. Eli Lilly, for example, reported 20% cost savings on data management and monitoring and a 93% improvement in monthly patient enrollment volume. Pfizer reported saving $20 million annually through consolidated management of its vendors, from 150 to 17; a 26% reduction in enrollment cycle time; and an 80% reduction in the number of contracts delayed by 4120 days.7 But recent assessments from a variety of sources, based on much larger numbers of sponsor companies, present a more mixed picture. For example, a study of 89 sponsor companies conducted by the Avoca Group (Princeton, NJ) in 2012 found that 1 of 5 sponsors had terminated a strategic relationship with a CRO. That same study noted that approximately one third of sponsors surveyed were not satisfied with the performance of their strategic relationships, and nearly half did not believe that their strategic relationships would ever achieve the intended outcome.8 A 2013 study conducted by Vantage Partners (Boston, Massachusetts) of 81 sponsor company representatives and 88 CRO company representatives found that a large percentage of strategic relationships were unable to establish a collaborative working style and failed to align expectations and capabilities. In addition, one third of the sponsor organizations reported that their strategic alliances had not delivered expected cost and time savings, and 56% said that their relationships were not delivering innovative solutions.9 In 2012, as part of a broader study, Tufts CSDD assessed the frequency of change orders in studies for which most of its budget was allocated to contract service providers. Change orders are defined as formal, written changes to the original scope of work agreement. Tufts CSDD hypothesized that strategic relationships would result in fewer change orders because the parties share planning and governance, and the CRO partner is given more autonomy to manage the study. However, the findings revealed no Volume 36 Number 10
K.A. Getz et al. difference in the frequency of change orders between traditional, transactional relationships and strategic relationships, suggesting that the latter are having limited effect on R&D efficiency (Figure 1).10 The authors of the Avoca Group and Vantage Partners studies concluded that the primary causes of relationship failure are cultural barriers, poor communication, failure to achieve buy-in from senior staff and affiliates, and unreasonable expectations. Other industry professionals have argued that strategic relationships are a fallacy and that true partnership can never be achieved as long as sponsor companies lack the ability to truly measure their operating costs and profitability.11,12 Most articles in the trade press that discuss sponsor-CRO relationships primarily focus on negotiating and establishing collaborative processes and on measuring general attitudes and satisfaction. Most of the articles presuppose that strategic outsourcing relationships have been formed to replace preexisting collaborative approaches with CROs. They do not consider that large, fragmented organizations may be supporting competing and conflicting relationship models simultaneously and that, in doing so, they may detract from the promise offered by new strategic relationship models. The recent Tufts CSDD study, described herein, sheds new light on why strategic outsourcing relationships may be failing.
STUDY BACKGROUND In early 2014, Tufts CSDD contacted each of the top 20 largest pharmaceutical companies to solicit their
interest in participating in an in-depth assessment. All the companies contacted use outsourcing extensively, and they had entered into strategic relationships with CROs at least 3 years before the start of the study. Nine companies agreed to participate. Between March and May 2014, these 9 companies evaluated specific contract services used on their individual clinical studies. Each of the companies selected representative Phase II and III studies conducted and completed since 2012. Representative studies were those with similar scope (eg, number of enrolled patients, number of countries, and number of investigative sites) and therapeutic areas as those found in each company’s development portfolio. For each study, sponsor companies met with their clinical teams to document outsourcing relationship models used to support specific functional areas. Sponsor companies also indicated whether internal staff (including temporary personnel) supported a given functional requirement. Transactional relationships are those in which the sponsor engages a contract service organization to provide a specific task for an individual project requirement. Strategic relationships are those in which the sponsor enters into a longerterm relationship with a contract service organization to provide single functional or multifunctional support across projects in a sponsor’s portfolio under shared governance and integrated systems and practices.13 In total, 11 functional areas were examined. These areas were selected to capture outsourcing practices across 3 broad clinical study requirements: (1) study
Transactional Relationships
Strategic Relationships
(Mean Number per Study)
(Mean Number per Study)
5.3
4.1
2.4
2.3
2.5
1.1
Phase I
2.2
1.1
Phase II
Phase III
Phase IV
Phase I
Phase II
Phase III
Phase IV
Figure 1. Change orders by outsourcing relationship model. Source: Tufts CSDD 2012.
October 2014
1351
Clinical Therapeutics selection. Most studies also outsourced data management (80%), medical monitoring (73%), patient recruitment and retention (68%), and statistical analysis (63%). The results have been organized into 3 groups of functions supported by outsourcing—study design and regulatory affairs, site management and monitoring, and data management and dossier creation—to summarize relationship use. Despite the fact that each sponsor company had entered into several strategic outsourcing relationships, in no instance did a single CRO manage all functional areas supporting an individual Phase II or III study. Internal staff typically managed protocol design and regulatory affairs functions, with less than 1 in 5 supported by a transactional or strategic outsourcing relationship (Figure 3). Site management and study monitoring functions were the most heavily outsourced. Functional requirements in this area were supported by transactional relationships for 40% to 50% of studies. Strategic relationships were used to meet functional requirements for 25% to 40% of all studies (Figure 4). Outsourcing relationship models used for data management and dossier creation and submission were the most varied. For most studies, internal staff was used to support medical writing and regulatory document submission preparation. Data management and statistical analysis functions, however, used outsourcing relationships extensively. As shown in Figure 5, approximately 40% of all studies used
design and regulatory affairs, (2) site management and study monitoring, and (3) data management and dossier creation and submission. Detailed data on 43 Phase II and III studies were collected. The studies were representative of multiple disease conditions: 20% in central nervous system diseases and disorders; 18% in oncology; 10% each in respiratory, endocrinology, immunology, and infectious diseases; and 5% each in cardiovascular and hematologic disorders. Participating companies also provided performance and quality data for each individual study to determine whether the independent variables (outsourcing relationship model used) had a measurable impact. Studyspecific performance and quality data collected included planned to actual study startup cycle time, planned to actual study conduct cycle time, planned to actual study closeout cycle time, screen failure rate, study volunteer dropout rate, number of protocol amendments, number of protocol violations; number of change orders, and queries per thousand clinical data points collected. Twotailed t tests were performed to evaluate differences between dependent variables by transactional versus strategic outsourcing relationship used per study.
RESULTS Figure 2 describes the incidence of services that were outsourced for each Phase II and III study evaluated. Nearly all of the studies (95%) outsourced site monitoring, and 86% outsourced site identification and
Percentage of Total Studies Site Monitoring
95% 86%
Site Identification and Selection
80%
Data Management 73%
Medical Monitoring
68%
Patient Recruitment and Retention
63%
Statistical Analysis 38%
Central Site Payment Administration
34%
Medical Writing
30%
Submission Preparation Regulatory Protocol Design
18% 15%
Figure 2. Incidence of services outsourced for 43 Phase II and III studies.
1352
Volume 36 Number 10
K.A. Getz et al. 12% 2%
14% 5%
86%
81%
Protocol Design
Regulatory Affairs
Strategic Relationships (FSP and Integrated Alliances) Transactional (Full and Niche Service Models) In-house
Figure 3. Relationship models used for study design and regulatory affairs for 43 Phase II and III studies. FSP ¼ full-service provider.
transactional outsourcing relationships, and 25% to 40% used strategic outsourcing relationships. Clinical research studies conducted by the same sponsor company used different outsourcing relationship models without a clear or systematic pattern. One company, for example, varied its use of in-house staff and used transactional (full-service provider) as well as strategic relationships on each study to manage the site monitoring function. Another company alternated between transactional service providers and a strategic partner to provide statistical analysis. The same variation in outsourcing relationship use was observed
40%
26%
in site identification and management and in data management functions. With respect to performance and quality, no statistically significant differences were observed between studies managed by transactional versus strategic relationship partners with 2 exceptions: the number of screen failures was higher in studies managed by strategic outsourcing relationships (mean [SD] difference, 0.133 [0.22]; P o 0.03) and the frequency of protocol amendments was lower in studies managed by strategic outsourcing relationships (mean [SD] difference, 1.28 [2.01]; P o 0.02).
26%
40%
19% 28%
42%
51% 51%
51%
54%
32% 9% Site Identification Patient and Selection Recruitment and Retention
9% Site Monitoring
23% Medical Monitoring
Central Site Payment Administration
Strategic Relationships (FSP and Integrated Alliances) Transactional (Full and Niche Service Models) In-house
Figure 4. Relationship models used for site management and monitoring for 43 Phase II and III studies. FSP ¼ full-service provider.
October 2014
1353
Clinical Therapeutics
37%
26%
21% 12%
12% 19%
37% 42% 67%
69%
Medical Writing
Submission Preparation
37% 21% Data Management
Statistical Analysis
Strategic Relationships (FSP and Integrated Alliances) Transactional (Full and Niche Service Models) In-house
Figure 5. Relationship models used for data management and dossier creation for 43 Phase II and III studies. FSP ¼ full-service provider.
DISCUSSION Based on the Tufts CSDD assessment of sponsor company use of different outsourcing relationship models, it is apparent that companies mix and match a variety of outsourcing approaches in clinical studies to meet specific functional needs. Despite the investment and effort required to identify, select, and integrate strategic outsourcing relationships, sponsor companies clearly choose to keep all options on the table. The outsourcing practices observed in this study indicate that sponsor companies are not fully invested in strategic relationships, ie, they are not ready to hand over the keys—and operating risk—to their strategic partners. Although the mix-and-match approach offers pharmaceutical and biotechnology companies some flexibility, this study suggests that it compromises the conceptual benefits of strategic relationships. Specifically, sponsor company staff must simultaneously support processes and procedures that accommodate multiple relationship models. In addition, collaborative practice experience and efficiency are isolated to individual studies and do not carry across the development portfolio. Numerous dynamics play a role in the use of multiple sourcing models by sponsor companies. The high level of fragmentation within major pharmaceutical companies and the difficulty in implementing and enforcing new practices no doubt contribute to the difficulties in integrating and coordinating the use of select providers. Moreover, aversion to risk entices sponsor company
1354
staff to hedge their bets by maintaining multiple contract service providers under a variety of sourcing models. Interestingly, the labor- and time-intensive nature of sponsor company participation in this Tufts CSDD study provided insight into the level of fragmentation within participating companies. Clinical teams had considerable difficulty determining which outsourcing relationships were used across multiple functional areas for their recently completed studies, suggesting that cross-functional awareness, communication, and coordination are limited. Perceptions and attitudes about contract service providers—as noted by the Avoca Group and Vantage Partners—are deeply ingrained in company culture and also are a major factor in the types of relationships companies establish with CROs. The long-held view that CROs are commodity service providers that must be carefully policed by sponsor staff to ensure quality is hard to change and contributes to distrust in strategic relationship partners. Given the way that sponsor companies are using their strategic relationships, it is not surprising that study performance and quality are not improving. Significantly higher screen failure rates in studies supported by strategic relationships are likely associated with the use of less experienced investigative sites in emerging global regions. The lower frequency of protocol amendments in studies managed by strategic partners is a positive outcome that may reflect greater collaborative input into study design feasibility before protocol approval.
Volume 36 Number 10
K.A. Getz et al. This study has 2 limitations of note, both associated with selection bias. The 9 major pharmaceutical companies participating in this study represent a highly active drug development subgroup, but their outsourcing practices may not reflect those of smaller companies. In addition, bias may have been introduced when participating companies were asked to self-select studies for evaluation. Future studies will benefit by a larger and more representative sample of companies and projects. Although a growing number of sponsor companies expect their use of strategic outsourcing relationships to continue to increase, it seems that the anticipated positive impact of these relationships on performance, quality, and cost may be elusive. To date, the implementation and integration of strategic outsourcing relationships has fallen short, and some sponsor companies may lose patience with this approach. Indeed, at recent industry conferences, several sponsors have indicated that they are evaluating whether to increase the number of internal staff to support their studies.
CONCLUSIONS If the conceptual benefits of strategic outsourcing relationships are to be realized, it is critical that sponsor companies recognize that their internal sourcing approaches lead to friction and inefficiency. New dedicated internal mechanisms may assist sponsor companies in identifying and managing this friction more effectively. And this mechanism may ultimately serve as a catalyst in assisting organizations in transitioning to a more consistent and leveraged approach to strategic outsourcing relationships. Internal metrics to assess the level of interaction and coordination between different outsourcing relationship models will also assist sponsor companies in understanding ways to adjust their use of sourcing approaches. Sponsor organizations, arguably facing their most difficult operating environment in history, must improve the effectiveness of their contract service provider management practices to best leverage this integral drug development collaborative partner.
ACKNOWLEDGMENTS Mr. Getz was the principal investigator on this study involved in all aspects of the research including study design, execution, analysis and reporting on the
results and co-authoring the manuscript. Dr. Lamberti was the senior project manager on this research involved in study design, execution, analysis, reporting on the results and co-authoring the manuscript. Dr. Kaitin was a co-author of this manuscript.
CONFLICTS OF INTEREST The authors have indicated that they have no conflicts of interest regarding the content of this article.
REFERENCES 1. Getz K, Lamberti MJ, Mathias A, Stergiopoulos S. Resizing the global contract R&D services market. Contract Pharma. 2012;14:54–62. 2. Scott N, Binkowitz B, Ibia EO, et al. Results of a survey of PhRMA member companies on practices associated with multiregional clinical trials. TIRS. 2011;45:609–617. 3. Chemjobber. 2013 not shaping up to be a great year for pharma: Challenger reports 19,507 jobs lost. http:// chemjobber.blogspot.com/2013/11/2013-not-shaping-upto-be-great-year.html. Accessed September 30, 2014. 4. Getz K. Profound changes in the outsourcing landscape. Pharmaceutical Executive. Published February 29, 2012. http://www.pharmexec.com/pharmexec/Outsourcing/Pro found-Changes-in-the-Outsourcing-Landscape/ArticleStan dard/Article/detail/762383. Accessed September 30, 2014. 5. DiMasi J, Feldman L, Seckler A, Wilson A. Trends in risks associated with new drug development: success rates for investigational drugs. Clin Pharmacol Ther. 2010;87:272–277. 6. Kaitin K, DiMasi J. Pharmaceutical innovation in the 21st century: new drug approvals in the first decade 2000-2009. Clin Pharmacol Ther. 2011;89:183–188. 7. Getz K. FIPNet: pharma’s new, sexy, but not yet ready for print-time model. Applied Clinical Trials. 2009(Suppl): 10– (12,14,16). 8. Henderson L. The state of CRO and sponsor relationships. Applied Clinical Trials. 2013;10:26–28. 9. Hughes J, Rayer E, Wadd J. Sponsor-CRO Collaboration Study. Boston MA. Vantage Partners LLC. May 2013. 10. Kaitin KI, ed. Outsourcing to maximize clinical trial operating efficiency. Tufts Center for the Study of Drug Development R&D Management Report. 2013 Oct;8(4). 11. Parrett A. Challenging the value of strategic partnerships in clinical trial outsourcing. Applied Clinical Trials. 2013;4:30–33. 12. Waife R. Partnership heresy. Journal of Clinical Research Best Practice. 2012;8:1–4. 13. Getz KA. Perils and promise of risk imbalance. Inside Outsourcing. 2012(December):10–16.
Address correspondence to: Kenneth Getz, MBA, Tufts CSDD, 75 Kneeland St, Suite 1100, Boston, MA 02111. E-mail: [email protected]
October 2014
1355
