Attrition in the Pharmaceutical Industry

"Innovative drug discovery can only be partially guided by knowledge from known chemical and pharmacological space, so a level of attrition is therefore inevitable. This refreshingly readable book provides an engaging combination of background historical and current reasons for attrition, combined with a panorama of some of the possible ways to covert “attrition” into the informed risk-taking necessary for innovative drug discovery. The book is very logically presented across the phases of drug discovery out across the technologies an across the phases of drug discovery and thus builds in depth reference source especially for those entering the challenging environment of drug discovery. As the authors point out, converting molecules into drugs remains difficult and engaging in more projects as a way to ensure a minimal level of success is not sustainable, so challenging and understanding better the reasons for attrition are of fundamental importance. The reasons for attrition change over time as some factors, notably AMDE and PK are better understood. But an inability to accurately predict still hampers the industry. This book makes a very useful reference source by highlighting where progress is being made, for example, the AstraZeneca 5Rs approach or the translational data analysis by Pfizer showing three parameters which correlate combined confidence in pharmacology an exposure with confidence in Phase II success. The book then nicely moves the reader on from the improvements in Phase II attrition by asking the critical question, “Why do drugs fail in Phase III if efficacy failures in Phase II are being better managed?” One of the advantages of this book is that it not only provides a well written background review, but that it combines this with examples of where progress it still needed, something of particular importance as regulators increasingly place emphasis on safety profiles. Although the book does make some comments to other modalities, it focuses on small molecule drug discovery, for which the authors take the reader through all stages of drug discovery form target identification to post-marketing attrition with extensive use of informative case studies. These case studies are used to highlight where attrition has been reduced, where improvements are still needed, and for preclinical research in particular where attrition isn’t necessarily bad, but rather a consequence of innovative drug discovery, that is best managed in a structured approach where knowledge can be transferred between projects. As the pharmaceutical industry moves to a more fragmented but networked environment changes in the ways in which knowledge is acquired and transferred between companies will significantly change the ways in which attrition is confronted. This book is therefore an excellent source material that will be of great value to all those embarking in drug discovery in smaller more agile companies. As evidenced in Chapter 2, preclinical research has made significant inroads in managing attrition with structured approaches to ADME profiling and PK/PD modelling. This is picked up and integrated into more detailed discussion later in Chapters 7–10, covering reasons for attrition associated with the various technologies employed in preclinical research. Whilst attrition in preclinical research can be mitigated and to varying extents managed, attrition in clinical studies represent failure of a project or mechanism. Clinical and post-marketing failures continue to limit the overall efficiency of the drug discovery industry. The reasons are many and starting in chapter 3 with Phase I this book systematically reviews the factors influencing attrition in each phase, combined with examples of how some may be reduced. Attrition due to PK and tolerability issues remain the main causes of Phase I attrition, although PK attrition can be attenuated by preclinical in vitro CYP profiling combined with in vivo PK studies. Phase I oncology studies are more susceptible to tolerability problems and in general tolerability issues are a common reason for termination of dose escalation studies across disease areas. The chapter finishes with an interesting discussion on the addition of Target Occupancy readouts in Phase I studies for a range of different target classes. Determining the Target Occupancy required for efficacy significantly improves the probability of a success in subsequent Phase II studies. In the following Chapter, the discussion moves to attrition in Phase II/III studies with a detailed series of well-chosen case studies that highlight that despite improving Phase II success rates, lack of efficacy in some cases compounded by addition toxicological issues remains the main reason for Phase II/III failures. Failure is evenly spread across small molecules, antibodies, and biologics, though some disease areas such as Alzheimer’s disease are more difficult. Well-selected case histories are also used to highlight post-marketing attrition arising from both on- and off-target pharmacologies, where unacceptable benefit– risk scenarios have led to drugs being removed from the market or subject to restricting label restrictions. And as highlighted in Chapter 5 some off-target side effects are only revealed in large (postmarketing) populations which show highlight second target or other side effects. Whilst not perhaps directly as source of attrition, changes in the regulatory environment are presented in Chapter 6 as they have a significant retroactive effect on drug discovery. For example, no project today would be progressed without extensive studies on liability for drug-induced QTc prolongation. Drug withdrawals due to safety are thankfully relatively rare, nevertheless FA and EMA guidance impact significantly on introducing additional parameters in pre-clinical and clinical research that need to be effectively controlled to avoid attrition. Chapter 9 contrasts the different attrition scenarios contained in phenotypic screening and target-based drug discovery projects, using key case studies from anti-infective and CNS projects. The values of each approach and the associated potential attrition factors, such as; complex SAR in phenotypic screening or the disconnection from pharmacological relevance in target-based approaches are compared. Attrition is also affected in each approach by technological factors and implications for data-driven compound optimization and translation into clinical studies. As screening technologies advance the distinction between the two approaches becomes less clear, and for now the combination of both approaches coupled with in silico modelling appears to best method for project progression and mitigation of downstream attrition. As discussed in Chapter 10 of the book, data integration and interpretation via in silico modelling has significantly helped reduce pre-clinical attrition. ADME and toxicity profiling in particular have benefitted from ability to predict compound properties via iterative cycles of in silico modelling. Indeed, knowledge sharing of compound data and properties either by public databases or industry-academic collaborations has proven an effective route to help further reduce attrition. The book closes with two chapters looking to the future and to emerging new approaches to tackle attrition rates that are emerging from pre-competitive collaborative research and new business models and above all the continued need for highly motivated knowledge seeking researchers that make the drug discovery business successful. As highlighted in the conclusion, “if there were no attrition, it would not be research”. Attrition in drug discovery will always be a factor as new targets and new mechanisms are investigated. Attrition is a necessary element of innovation, and through constant improvements in our understanding of the causes of attrition continued reduction of failures due to lack of efficacy and in particular safety issues can be expected."
Prof. Roberto Pellicciari, TES Pharma, Perugia (ChemMedChem, July 2017)