REVIEW
Strengths and limitations of industry vs. academic randomized controlled trials P.-F. Laterre1 and B. François2 1) Department of Critical Care Medicine, Saint Luc University Hospital, Université Catholique de Louvain, Brussels, Belgium and 2) Medical-Surgical Intensive Care Unit, Limoges University Hospital, Limoges, France
Abstract Clinical research has evolved substantially over the last two decades, but industry-sponsored research is still substantially superior to academic research in preparing, organizing and monitoring studies. Academics have to realize that conducting clinical research has become a real job with professionalism requirements. The primary objectives of research and development clearly differ between industry and academics. In the first case, new drug development is expected to generate profit, whereas in the latter case, research is aimed at understanding mechanisms of disease, promoting evidence-based medicine, and improving public health and care. However, a large number of clinical studies do not achieve their goals, and the reasons for failure may also differ between sponsored and academic studies. Industry and academics should develop better constructive partnerships and learn from each other. Academics should guide industry in study design and in investigator site selection, and academics should benefit from industry’s expertise in improving monitoring and reporting processes. Finally, the existing database from former studies should be opened and shared with academics, to enable the exploration of additional scientific questions and the generation of new hypotheses. The two types of research should not be opposed, but should take the form of a constructive collaboration, increasing the chances of reaching each individual goal. © 2015 Clinical Microbiology and Infection published by Elsevier Ltd on behalf of European Society of Clinical Microbiology and Infectious Diseases. Keywords: Academic research, clinical research, controlled trials, data monitoring, industry, study design Article published online: 13 July 2015
Corresponding author: P.-F. Laterre, Department of CCM, St Luc University Hospital, Avenue Hippocrate 10, 1200 Brussels, Belgium E-mail: [email protected]
developments for the benefit of patients and healthcare. The aims of this article are to review the benefits and pitfalls of academic and industry research, to provide some explanations for the failure of studies, and to suggest some potential improvements for the future (Table 1).
Introduction Research driven by academics has often been considered to have conflicts with industry. Pharmaceutical companies probably have the image of developing new compounds for profit only. Therefore, research and studies carried out by industry are regularly criticized and perceived as potentially biased. However, collaboration between industry and academics has led to significant advances in drug and technical developments. Instead of opposing these two types of research, it would be more profitable to consider what can be learned from both approaches and how this can improve new study design and
Primary objectives of industry and academic research Industry is likely to develop new drugs or devices that are expected to generate profits for the company and shareholders. After a new drug registration by the authorities, industry will perform further work on marketing strategies, to increase sales and thereby increase the return on investment. Companies may also develop new drugs for orphan disease with an expected high sale price, on the assumption that the healthcare system will often accept the charges, as it cannot be
Clin Microbiol Infect 2015; 21: 906–909 © 2015 Clinical Microbiology and Infection published by Elsevier Ltd on behalf of European Society of Clinical Microbiology and Infectious Diseases http://dx.doi.org/10.1016/j.cmi.2015.07.004
CMI
Laterre and François
TABLE 1. Potential improvements for conducting academic and industry research Academic
Industry
Improve study and ethics committee submission preparation Improve data-monitoring quality and learn from the industry Safety reporting not restricted to unexpected events Knowledge of regulatory authorities’ policies and expectations Study agenda to be better established Multicentre rather than single-centre studies
Study designed in collaboration with actively involved academics Phase III trials not defined by weak phase II signals to meet marketing targets Site selection based on objective metrics Redefine a more realistic agenda and support quality recruitment vs. volume Consider the use of a clinical coordinating centre Access to study databank when completed
perceived as abandoning patients with no access to this unique care. On rare occasions, however, mainly because of public and healthcare pressures, companies may provide financial support by reducing their sales margins for difficult-to-access treatments for low-income populations or countries, in order to maintain a good public image. Benefits for academics, even though they are different from those for industry, clearly exist. The career progression of academics is driven by their scientific production and publication metrics. It is therefore not surprising that all efforts are made by scientists to conduct studies aimed at rapid publication. This attitude may lead not only to the well-known data cheating, but also to research of limited interest and benefit for the improvement of patient care [1]. It must be recognized that some research and clinical studies have little relevance for a better understanding of disease mechanisms. Also, some academics, when involved in sponsored studies, may pay more attention to the expected impact of the subsequent associated publications than to the value of the conducted research. The publication policy of industry studies should often better defined before the conduct of a clinical trial, to more adequately reward the actual investment of active investigators.
Reasons for failure in academic and industry research Study design and selected population Despite the fact that the regulatory authorities are more likely to approve a new drug if it has shown superiority to a comparator used as part of the standard of care, numerous companies have designed non-inferiority studies to access the market. By evaluating the potential benefit of their new compound in a low-risk population, these trials were often unable to detect some clinical cure failures or even inferiority as
Academic vs. industry research
907
compared with standard care. Numerous confirmatory trials or analyses of larger samples performed after market launching of a new drug have demonstrated the limitations of the initial studies that had resulted in its registration by the authorities [2]. Academics are more likely to explore the possible efficacy of an intervention or a drug in a more severe group of patients with comorbidities and a higher risk of death, without marketing objectives, but targeting a population with important unmet medical needs. Exploring antibiotic dosing regimens and associated outcomes in the critically ill provides one example of such academic studies. These studies have often supported the need to consider the original label for the most severe patients corresponding to a population for which more efforts should be made [3]. However, industry cannot always be blamed for this non-inferiority design approach. Indeed, recent examples of new antibiotic approvals by regulatory authorities are questionable, and may have been facilitated by political considerations, as the current era of bacterial multidrug resistance represents a threat to the community [4]. The tight agenda of industry research Industry has a well-defined and tight agenda for a research plan. Drug development takes years, and the patent-restricted period after launching potentially limits the profits that a company can expect. Potential launching and marketing strategies have to be established well in advance, when a promising drug is under early development. Also, industry is responsible to shareholders, and needs to complete its clinical studies within a very short period. This agenda may significantly impact on study quality, by resulting in the enrolment of a suboptimal population. Indeed, industry is often confronted by the problem of slower recruitment than expected or not meeting the predefined targets. Sponsors may therefore potentially facilitate recruitment by unintentionally opening the window to an inadequate patient population for the initial primary objectives of the study, which is discovered later when the database has been locked, and the part of the industry team responsible for the trial has changed position in the company. Continuous monitoring of the enrolled population should be systematically implemented, in order to more rapidly detect inappropriate enrolments, and investigator sites should be warned, or even closed, if a suboptimal patient population is recruited on multiple occasions. Finally, the investigator fee and financial support offered by industry may bias the enrolment. Indeed, the amount of money provided by the sponsor per patient included in a study may far exceed the actual personnel cost. The associated positive balance may help an academic group in future non-sponsored development, but may, on the other hand, in some areas, directly benefit the researcher. The high cost associated with research and development has caused industry to move in the direction of emerging markets [5]. In addition to
© 2015 Clinical Microbiology and Infection published by Elsevier Ltd on behalf of European Society of Clinical Microbiology and Infectious Diseases, CMI, 21, 906–909
908
Clinical Microbiology and Infection, Volume 21 Number 10, October 2015
some of the ethical concerns about the way in which some studies have been conducted, the investigator’s fee offered for study participation may, in some countries, exceed the physician’s monthly salary, resulting in the enrolment rate being inappropriately enhanced, with an associated bias. Investigator contracts should be systematically reviewed by an independent board from the clinical institution, to prevent excessive unjustified fees. Study site learning curve and clinical coordinating centres In large, multicentre, sponsored trials, an extensive study protocol description and a list of specific inclusion/exclusion criteria are generally provided. However, numerous potentially eligible patients may present with symptoms or conditions that were not described or considered in the initial proposed protocol. The inclusion criteria may be too wide, and confounding factors with a possible impact on the primary objective of a study may have been omitted. Unfortunately, investigators too often consider that they have sufficient expertise to recognize and select the most appropriate patients to enrol for a dedicated study. This may be the cause of suboptimal population enrolment in studies. It has also been demonstrated that the first patients enrolled within a single investigator centre have more protocol violations regarding inclusion/exclusion criteria or study drug administration than the subsequently included patients [6]. Over time, a learning curve phenomenon has been observed, and larger centres with higher recruitment, if monitored, do provide more consistent enrolled populations. Also, in order to meet the recruitment objectives, a sponsor may decide to augment the number of study sites. This may lead, based on the former observations, to a higher proportion of inappropriate patients being recruited for the study, with an associated impact on the final primary objectives and study results. Industry and academics should limit the number of active sites for a dedicated study. Finally, in a recent publication, it has been shown that some countries may be selected more readily for participation in a study, because the study approval process by local authorities or ethics committees may be facilitated [7]. The centres should be selected according to their size, infrastructure, capacity to enrol on a 24/7 base, and the demonstration of their previous expertise in the field with a dedicated investigator team. In the vast majority of sponsored studies, it has been observed that 10% of the sites recruit >50% of the patients. Industry should collaborate with academics to select centres for quality and consistency. Industry studies: More favourable results and fewer adverse events? There is a general concern that industry research is biased, that findings are false, and that study results are more often positive
CMI
when the research is sponsored, even for non-inferiority trials [8,9]. This raises the question of whether industry studies are truly biased or better designed. Industry has, in general, strong preclinical data, and may design trials that are more likely to match with their primary objectives and marketing targets. Academic single-centre phase II trials are less likely to lead to positive phase III studies [10]. This may be explained by the variability in the standard of care when multiple centres are involved, but also by a different patient profile for recruitment in later developments or a centre effect in the initial phase II trial. A review of the literature supports the idea that industry phase III trials are more likely to be positive than nonsponsored studies. However, on review of some cancer research conducted by industry as compared with academics, the more pronounced benefits observed initially in the former also tended to occur in the latter [11]. This may be explained by improvements achieved by academics in the way in which research is structured and organized, and the way in which the studied population is selected. In two recent publications, the authors compared industry with academic studies in the field of statins, and demonstrated that the benefits of the studied drugs were observed in both types of study. The magnitude of favourable effects was mostly driven by the compounds’ intrinsic properties, rather than the origin of the sponsorship [12,13]. Data monitoring and reporting after drug market launching Industry has clearly demonstrated substantial superiority over academics in the way that study protocols are submitted to the regulatory authorities and how monitoring needs to be organized. In addition, for the vast majority of trials conducted by industry, data collection is complete, and the entire process is facilitated by the use of a contract research organization (CRO). This process has, by definition, an associated cost that academics are less likely to meet in order to achieve the same objectives. However, because of the scope of data collection combined with multiple associated queries, often suggested for their own profits by the CRO, the database closure is potentially delayed, as is the final study report. Academics, on their side, have to learn from industry regarding the way to conduct study preparation, ethics committee submission, and the quality of data monitoring [14]. Despite a limited number of publications supporting the idea that academic studies are performing adequately in the reporting process, it is generally admitted that academic studies too often have much missing information, together with restricted adverse event data collection [15]. This is partially explained by the lack of sufficient financial resources, but is probably too often secondary to limited experience of
© 2015 Clinical Microbiology and Infection published by Elsevier Ltd on behalf of European Society of Clinical Microbiology and Infectious Diseases, CMI, 21, 906–909
CMI
Laterre and François
the way in which to conduct adequate clinical research. However, sponsored research has been sometimes criticized for minimizing some side effects described later when the drug has been launched [16].
Conclusions The restricted investments currently made by both industry and public authorities in the field of healthcare threaten future research developments and possible improvements in the care of patients for numerous diseases that are either not profitable for the former or have limited political impact for the latter. Research conducted by industry should not be opposed to the academic world developments, but these should be combined for the benefit of all. The financial and organizational strengths of industry should be shared with academics to enable the better design, organization and performance of studies for the benefit of scientific knowledge and healthcare. Industry should collaborate more with academics in the design of studies, the definition of endpoints, and the selection of centres to perform the research. The industry database should be accessible to and shared with academics, to further explore additional questions and raise new hypotheses. A recent report on the open access provided by some companies is promising in this regard [17]. Also, the new initiative from the European Commission to provide financial support to academics and industry organized in collaborative consortia to conduct studies for the development of new diagnostic tools and drugs is promising. If this is successful, both partners should achieve their respective goals for the benefit of patients and scientific developments.
Transparency declaration PF L is a consultant at Ferring, Tigenix, Lascco and Versantis. B F is a consultant at LASCCO, Aridis, Tigenix, Inotrem and GSK.
References [1] Elia N, Wager E, Tramèr MR. Fate of articles that warranted retraction due to ethical concerns: a descriptive cross-sectional study. PLoS One 2014;9(1):e85846.
Academic vs. industry research
909
[2] Prasad P, Sun J, Danner RL, Nathanson C. Excess deaths associated with tigecycline after approval based on noninferiority trials. Clin Infect Dis 2012;54:1699–709. [3] Blot S, Koulenti D, Akova M, Bassetti M, De Waele JJ, Dimopoulos G, et al. Does contemporary vancomycin dosing achieve therapeutic targets in a heterogeneous clinical cohort of critically ill patients? Data from the multinational DALI study. Crit Care 2014;18(3):R99. [4] Doshi P. Speeding new antibiotics to market: a fake fix? BMJ 2015;350: h1453. [5] MacMahon S, Perkovic V, Patel A. Industry-sponsored clinical trials in emerging markets: Time to review the terms of engagement. JAMA 2013;310(9):907–8. [6] Macias WL, Vallet B, Bernard GR, Vincent JL, Laterre PF, Nelson DR, et al. Sources of variability on the estimate of treatment effect in the PROWESS trial: implications for the design and conduct of future studies in severe sepsis. Crit Care Med 2004;32(12):2385–91. [7] Gehring M, Taylor RS, Mellody M, Casteels B, Piazzi A, Gensini G, et al. Factors influencing clinical trial site selection in Europe: the Survey of Attitudes towards Trial sites in Europe (the SAT-EU Study). BMJ Open 2013;3(11):e002957. [8] Flacco M, Manzoli L, Boccia S, Capasso L, Aleksovska K, Rosso A, et al. Head-to-head randomized trials are mostly industry sponsored and almost always favor the industry sponsor. J Clin Epidemiol 2015;68(7): 811–20. [9] Lundh A, Sismondo S, Lexchin J, Busuioc OA, Bero L. Industry sponsorship and research outcome. Cochrane Database Syst Rev 2012;12: MR000033. [10] Chan JK, Ueda SM, Sugiyama VE, Stave CD, Shin JY, Monk BJ, et al. Analysis of phase II studies on targeted agents and subsequent phase III trials: what are the predictors for success? J Clin Oncol 2008;26(9): 1511–8. [11] Djulbegovic B, Kumar A, Miladinovic B, Reljic T, Galeb S, Mhaskar A, et al. Treatment success in cancer: industry compared to publicly sponsored randomized controlled trials. PLoS One 2013;8(3):e58711. [12] Naci H, Dias S, Ades AE. Industry sponsorship bias in research findings: a network meta-analysis of LDL cholesterol reduction in randomised trials of statins. BMJ 2014;349:g5741. [13] Krauth D, Anglemyer A, Philipps R, Bero L. Nonindustry-sponsored preclinical studies on statins yield greater efficacy estimates than industry-sponsored studies: a meta-analysis. PLoS Biol 2014 Jan;12(1): e1001770. [14] van Lent M, Rongen GA, Out HJ. Shortcomings of protocols of drug trials in relation to sponsorship as identified by Research Ethics Committees: analysis of comments raised during ethical review. BMC Med Ethics 2014;15:83. [15] Patwardhan S, Gogtay N, Thatte U, Pramesh CS. Quality and completeness of data documentation in an investigator-initiated trial versus an industry-sponsored trial. Indian J Med Ethics 2014;11:19–24. [16] Hughes S, Cohen D, Jaggi R. Differences in reporting serious adverse events in industry sponsored clinical trial registries and journal articles on antidepressant and antipsychotic drugs: a cross-sectional study. BMJ Open 2014;4(7):e005535. [17] Strom BL, Buyse M, Hughes J, Knoppers BM. Data sharing, year 1—access to data from industry-sponsored clinical trials. N Engl J Med 2014;371(22):2052–4.
© 2015 Clinical Microbiology and Infection published by Elsevier Ltd on behalf of European Society of Clinical Microbiology and Infectious Diseases, CMI, 21, 906–909
Strengths and limitations of industry vs. academic randomized controlled trials P.-F. Laterre1 and B. François2 1) Department of Critical Care Medicine, Saint Luc University Hospital, Université Catholique de Louvain, Brussels, Belgium and 2) Medical-Surgical Intensive Care Unit, Limoges University Hospital, Limoges, France
Abstract Clinical research has evolved substantially over the last two decades, but industry-sponsored research is still substantially superior to academic research in preparing, organizing and monitoring studies. Academics have to realize that conducting clinical research has become a real job with professionalism requirements. The primary objectives of research and development clearly differ between industry and academics. In the first case, new drug development is expected to generate profit, whereas in the latter case, research is aimed at understanding mechanisms of disease, promoting evidence-based medicine, and improving public health and care. However, a large number of clinical studies do not achieve their goals, and the reasons for failure may also differ between sponsored and academic studies. Industry and academics should develop better constructive partnerships and learn from each other. Academics should guide industry in study design and in investigator site selection, and academics should benefit from industry’s expertise in improving monitoring and reporting processes. Finally, the existing database from former studies should be opened and shared with academics, to enable the exploration of additional scientific questions and the generation of new hypotheses. The two types of research should not be opposed, but should take the form of a constructive collaboration, increasing the chances of reaching each individual goal. © 2015 Clinical Microbiology and Infection published by Elsevier Ltd on behalf of European Society of Clinical Microbiology and Infectious Diseases. Keywords: Academic research, clinical research, controlled trials, data monitoring, industry, study design Article published online: 13 July 2015
Corresponding author: P.-F. Laterre, Department of CCM, St Luc University Hospital, Avenue Hippocrate 10, 1200 Brussels, Belgium E-mail: [email protected]
developments for the benefit of patients and healthcare. The aims of this article are to review the benefits and pitfalls of academic and industry research, to provide some explanations for the failure of studies, and to suggest some potential improvements for the future (Table 1).
Introduction Research driven by academics has often been considered to have conflicts with industry. Pharmaceutical companies probably have the image of developing new compounds for profit only. Therefore, research and studies carried out by industry are regularly criticized and perceived as potentially biased. However, collaboration between industry and academics has led to significant advances in drug and technical developments. Instead of opposing these two types of research, it would be more profitable to consider what can be learned from both approaches and how this can improve new study design and
Primary objectives of industry and academic research Industry is likely to develop new drugs or devices that are expected to generate profits for the company and shareholders. After a new drug registration by the authorities, industry will perform further work on marketing strategies, to increase sales and thereby increase the return on investment. Companies may also develop new drugs for orphan disease with an expected high sale price, on the assumption that the healthcare system will often accept the charges, as it cannot be
Clin Microbiol Infect 2015; 21: 906–909 © 2015 Clinical Microbiology and Infection published by Elsevier Ltd on behalf of European Society of Clinical Microbiology and Infectious Diseases http://dx.doi.org/10.1016/j.cmi.2015.07.004
CMI
Laterre and François
TABLE 1. Potential improvements for conducting academic and industry research Academic
Industry
Improve study and ethics committee submission preparation Improve data-monitoring quality and learn from the industry Safety reporting not restricted to unexpected events Knowledge of regulatory authorities’ policies and expectations Study agenda to be better established Multicentre rather than single-centre studies
Study designed in collaboration with actively involved academics Phase III trials not defined by weak phase II signals to meet marketing targets Site selection based on objective metrics Redefine a more realistic agenda and support quality recruitment vs. volume Consider the use of a clinical coordinating centre Access to study databank when completed
perceived as abandoning patients with no access to this unique care. On rare occasions, however, mainly because of public and healthcare pressures, companies may provide financial support by reducing their sales margins for difficult-to-access treatments for low-income populations or countries, in order to maintain a good public image. Benefits for academics, even though they are different from those for industry, clearly exist. The career progression of academics is driven by their scientific production and publication metrics. It is therefore not surprising that all efforts are made by scientists to conduct studies aimed at rapid publication. This attitude may lead not only to the well-known data cheating, but also to research of limited interest and benefit for the improvement of patient care [1]. It must be recognized that some research and clinical studies have little relevance for a better understanding of disease mechanisms. Also, some academics, when involved in sponsored studies, may pay more attention to the expected impact of the subsequent associated publications than to the value of the conducted research. The publication policy of industry studies should often better defined before the conduct of a clinical trial, to more adequately reward the actual investment of active investigators.
Reasons for failure in academic and industry research Study design and selected population Despite the fact that the regulatory authorities are more likely to approve a new drug if it has shown superiority to a comparator used as part of the standard of care, numerous companies have designed non-inferiority studies to access the market. By evaluating the potential benefit of their new compound in a low-risk population, these trials were often unable to detect some clinical cure failures or even inferiority as
Academic vs. industry research
907
compared with standard care. Numerous confirmatory trials or analyses of larger samples performed after market launching of a new drug have demonstrated the limitations of the initial studies that had resulted in its registration by the authorities [2]. Academics are more likely to explore the possible efficacy of an intervention or a drug in a more severe group of patients with comorbidities and a higher risk of death, without marketing objectives, but targeting a population with important unmet medical needs. Exploring antibiotic dosing regimens and associated outcomes in the critically ill provides one example of such academic studies. These studies have often supported the need to consider the original label for the most severe patients corresponding to a population for which more efforts should be made [3]. However, industry cannot always be blamed for this non-inferiority design approach. Indeed, recent examples of new antibiotic approvals by regulatory authorities are questionable, and may have been facilitated by political considerations, as the current era of bacterial multidrug resistance represents a threat to the community [4]. The tight agenda of industry research Industry has a well-defined and tight agenda for a research plan. Drug development takes years, and the patent-restricted period after launching potentially limits the profits that a company can expect. Potential launching and marketing strategies have to be established well in advance, when a promising drug is under early development. Also, industry is responsible to shareholders, and needs to complete its clinical studies within a very short period. This agenda may significantly impact on study quality, by resulting in the enrolment of a suboptimal population. Indeed, industry is often confronted by the problem of slower recruitment than expected or not meeting the predefined targets. Sponsors may therefore potentially facilitate recruitment by unintentionally opening the window to an inadequate patient population for the initial primary objectives of the study, which is discovered later when the database has been locked, and the part of the industry team responsible for the trial has changed position in the company. Continuous monitoring of the enrolled population should be systematically implemented, in order to more rapidly detect inappropriate enrolments, and investigator sites should be warned, or even closed, if a suboptimal patient population is recruited on multiple occasions. Finally, the investigator fee and financial support offered by industry may bias the enrolment. Indeed, the amount of money provided by the sponsor per patient included in a study may far exceed the actual personnel cost. The associated positive balance may help an academic group in future non-sponsored development, but may, on the other hand, in some areas, directly benefit the researcher. The high cost associated with research and development has caused industry to move in the direction of emerging markets [5]. In addition to
© 2015 Clinical Microbiology and Infection published by Elsevier Ltd on behalf of European Society of Clinical Microbiology and Infectious Diseases, CMI, 21, 906–909
908
Clinical Microbiology and Infection, Volume 21 Number 10, October 2015
some of the ethical concerns about the way in which some studies have been conducted, the investigator’s fee offered for study participation may, in some countries, exceed the physician’s monthly salary, resulting in the enrolment rate being inappropriately enhanced, with an associated bias. Investigator contracts should be systematically reviewed by an independent board from the clinical institution, to prevent excessive unjustified fees. Study site learning curve and clinical coordinating centres In large, multicentre, sponsored trials, an extensive study protocol description and a list of specific inclusion/exclusion criteria are generally provided. However, numerous potentially eligible patients may present with symptoms or conditions that were not described or considered in the initial proposed protocol. The inclusion criteria may be too wide, and confounding factors with a possible impact on the primary objective of a study may have been omitted. Unfortunately, investigators too often consider that they have sufficient expertise to recognize and select the most appropriate patients to enrol for a dedicated study. This may be the cause of suboptimal population enrolment in studies. It has also been demonstrated that the first patients enrolled within a single investigator centre have more protocol violations regarding inclusion/exclusion criteria or study drug administration than the subsequently included patients [6]. Over time, a learning curve phenomenon has been observed, and larger centres with higher recruitment, if monitored, do provide more consistent enrolled populations. Also, in order to meet the recruitment objectives, a sponsor may decide to augment the number of study sites. This may lead, based on the former observations, to a higher proportion of inappropriate patients being recruited for the study, with an associated impact on the final primary objectives and study results. Industry and academics should limit the number of active sites for a dedicated study. Finally, in a recent publication, it has been shown that some countries may be selected more readily for participation in a study, because the study approval process by local authorities or ethics committees may be facilitated [7]. The centres should be selected according to their size, infrastructure, capacity to enrol on a 24/7 base, and the demonstration of their previous expertise in the field with a dedicated investigator team. In the vast majority of sponsored studies, it has been observed that 10% of the sites recruit >50% of the patients. Industry should collaborate with academics to select centres for quality and consistency. Industry studies: More favourable results and fewer adverse events? There is a general concern that industry research is biased, that findings are false, and that study results are more often positive
CMI
when the research is sponsored, even for non-inferiority trials [8,9]. This raises the question of whether industry studies are truly biased or better designed. Industry has, in general, strong preclinical data, and may design trials that are more likely to match with their primary objectives and marketing targets. Academic single-centre phase II trials are less likely to lead to positive phase III studies [10]. This may be explained by the variability in the standard of care when multiple centres are involved, but also by a different patient profile for recruitment in later developments or a centre effect in the initial phase II trial. A review of the literature supports the idea that industry phase III trials are more likely to be positive than nonsponsored studies. However, on review of some cancer research conducted by industry as compared with academics, the more pronounced benefits observed initially in the former also tended to occur in the latter [11]. This may be explained by improvements achieved by academics in the way in which research is structured and organized, and the way in which the studied population is selected. In two recent publications, the authors compared industry with academic studies in the field of statins, and demonstrated that the benefits of the studied drugs were observed in both types of study. The magnitude of favourable effects was mostly driven by the compounds’ intrinsic properties, rather than the origin of the sponsorship [12,13]. Data monitoring and reporting after drug market launching Industry has clearly demonstrated substantial superiority over academics in the way that study protocols are submitted to the regulatory authorities and how monitoring needs to be organized. In addition, for the vast majority of trials conducted by industry, data collection is complete, and the entire process is facilitated by the use of a contract research organization (CRO). This process has, by definition, an associated cost that academics are less likely to meet in order to achieve the same objectives. However, because of the scope of data collection combined with multiple associated queries, often suggested for their own profits by the CRO, the database closure is potentially delayed, as is the final study report. Academics, on their side, have to learn from industry regarding the way to conduct study preparation, ethics committee submission, and the quality of data monitoring [14]. Despite a limited number of publications supporting the idea that academic studies are performing adequately in the reporting process, it is generally admitted that academic studies too often have much missing information, together with restricted adverse event data collection [15]. This is partially explained by the lack of sufficient financial resources, but is probably too often secondary to limited experience of
© 2015 Clinical Microbiology and Infection published by Elsevier Ltd on behalf of European Society of Clinical Microbiology and Infectious Diseases, CMI, 21, 906–909
CMI
Laterre and François
the way in which to conduct adequate clinical research. However, sponsored research has been sometimes criticized for minimizing some side effects described later when the drug has been launched [16].
Conclusions The restricted investments currently made by both industry and public authorities in the field of healthcare threaten future research developments and possible improvements in the care of patients for numerous diseases that are either not profitable for the former or have limited political impact for the latter. Research conducted by industry should not be opposed to the academic world developments, but these should be combined for the benefit of all. The financial and organizational strengths of industry should be shared with academics to enable the better design, organization and performance of studies for the benefit of scientific knowledge and healthcare. Industry should collaborate more with academics in the design of studies, the definition of endpoints, and the selection of centres to perform the research. The industry database should be accessible to and shared with academics, to further explore additional questions and raise new hypotheses. A recent report on the open access provided by some companies is promising in this regard [17]. Also, the new initiative from the European Commission to provide financial support to academics and industry organized in collaborative consortia to conduct studies for the development of new diagnostic tools and drugs is promising. If this is successful, both partners should achieve their respective goals for the benefit of patients and scientific developments.
Transparency declaration PF L is a consultant at Ferring, Tigenix, Lascco and Versantis. B F is a consultant at LASCCO, Aridis, Tigenix, Inotrem and GSK.
References [1] Elia N, Wager E, Tramèr MR. Fate of articles that warranted retraction due to ethical concerns: a descriptive cross-sectional study. PLoS One 2014;9(1):e85846.
Academic vs. industry research
909
[2] Prasad P, Sun J, Danner RL, Nathanson C. Excess deaths associated with tigecycline after approval based on noninferiority trials. Clin Infect Dis 2012;54:1699–709. [3] Blot S, Koulenti D, Akova M, Bassetti M, De Waele JJ, Dimopoulos G, et al. Does contemporary vancomycin dosing achieve therapeutic targets in a heterogeneous clinical cohort of critically ill patients? Data from the multinational DALI study. Crit Care 2014;18(3):R99. [4] Doshi P. Speeding new antibiotics to market: a fake fix? BMJ 2015;350: h1453. [5] MacMahon S, Perkovic V, Patel A. Industry-sponsored clinical trials in emerging markets: Time to review the terms of engagement. JAMA 2013;310(9):907–8. [6] Macias WL, Vallet B, Bernard GR, Vincent JL, Laterre PF, Nelson DR, et al. Sources of variability on the estimate of treatment effect in the PROWESS trial: implications for the design and conduct of future studies in severe sepsis. Crit Care Med 2004;32(12):2385–91. [7] Gehring M, Taylor RS, Mellody M, Casteels B, Piazzi A, Gensini G, et al. Factors influencing clinical trial site selection in Europe: the Survey of Attitudes towards Trial sites in Europe (the SAT-EU Study). BMJ Open 2013;3(11):e002957. [8] Flacco M, Manzoli L, Boccia S, Capasso L, Aleksovska K, Rosso A, et al. Head-to-head randomized trials are mostly industry sponsored and almost always favor the industry sponsor. J Clin Epidemiol 2015;68(7): 811–20. [9] Lundh A, Sismondo S, Lexchin J, Busuioc OA, Bero L. Industry sponsorship and research outcome. Cochrane Database Syst Rev 2012;12: MR000033. [10] Chan JK, Ueda SM, Sugiyama VE, Stave CD, Shin JY, Monk BJ, et al. Analysis of phase II studies on targeted agents and subsequent phase III trials: what are the predictors for success? J Clin Oncol 2008;26(9): 1511–8. [11] Djulbegovic B, Kumar A, Miladinovic B, Reljic T, Galeb S, Mhaskar A, et al. Treatment success in cancer: industry compared to publicly sponsored randomized controlled trials. PLoS One 2013;8(3):e58711. [12] Naci H, Dias S, Ades AE. Industry sponsorship bias in research findings: a network meta-analysis of LDL cholesterol reduction in randomised trials of statins. BMJ 2014;349:g5741. [13] Krauth D, Anglemyer A, Philipps R, Bero L. Nonindustry-sponsored preclinical studies on statins yield greater efficacy estimates than industry-sponsored studies: a meta-analysis. PLoS Biol 2014 Jan;12(1): e1001770. [14] van Lent M, Rongen GA, Out HJ. Shortcomings of protocols of drug trials in relation to sponsorship as identified by Research Ethics Committees: analysis of comments raised during ethical review. BMC Med Ethics 2014;15:83. [15] Patwardhan S, Gogtay N, Thatte U, Pramesh CS. Quality and completeness of data documentation in an investigator-initiated trial versus an industry-sponsored trial. Indian J Med Ethics 2014;11:19–24. [16] Hughes S, Cohen D, Jaggi R. Differences in reporting serious adverse events in industry sponsored clinical trial registries and journal articles on antidepressant and antipsychotic drugs: a cross-sectional study. BMJ Open 2014;4(7):e005535. [17] Strom BL, Buyse M, Hughes J, Knoppers BM. Data sharing, year 1—access to data from industry-sponsored clinical trials. N Engl J Med 2014;371(22):2052–4.
© 2015 Clinical Microbiology and Infection published by Elsevier Ltd on behalf of European Society of Clinical Microbiology and Infectious Diseases, CMI, 21, 906–909
