The high attrition in drug discovery is responsible for the extremely high cost of developing a new medicine. Some suggestions aimed at reducing the attrition have been put forward such as: improving efficacy and safety profiles, reducing toxicity, improving preclinical models, better understanding of mechanism (Nat Rev Drug Discov. 2004, 3, 711-716) and shifting the attrition to earlier phases (Nat. Rev. Drug Discov. 2010, 9, 203–214).
Based on this, scientists at AstraZeneca have established a five-dimensional framework (5R framework) (Nat. Rev. Drug Discov. in 2014) aimed at improving the low success rates in the process. The framework includes five determinants: right target, right tissue, right safety, right patient and right commercial potential, identified as key features in the drug discovery process (summary depicted in Figure 1).
Figure 1. The 5R framework (from Nat. Rev. Drug Discov. 2018).
In addition to the “5R framework”, AstraZeneca scientists decided to reduce their diseases portfolio and to potentiate their capabilities for target selection and validation (better understanding of biology and mechanism of the disease, stronger target rational).
Furthermore they improved their lead generation strategy (expansion of their compounds library including library sharing, and integration with other screening approaches), and their pharmacokinetic/pharmacodynamics modelling, patient stratification and biomarkers. In a more recent paper published in Nat. Rev. Drug Discov. in 2018, they show how the application of these guidelines have led to an increase in project success rates (Fig. 2) and to a reduction of cycle time and cost (Fig. 3).
Figure 2. Project success rates for the AstraZeneca (AZ) portfolio.
The overall project success rates have increased from 4% (2005-2010) to 19% (2012-2016); the cost to reach clinical proof of concept has decreased by 31% when comparing the two time cohorts, and by 42% compared to the industry average.
Figure 3. Metrics for projects costs (a. first good laboratory practice dose; b. clinical proof of concept) and cycle times.
With regard to cycle times, they observed a considerable reduction of the length for phase II (50% shorter than industry average) although this may return in line to the average in the future.
To further support how the introduction of the 5R framework has influenced AstraZeneca pipelines, Table 1 reports the new molecular entities and new biologics that entered phase III in 2012-2016, highlighting how their progression was influenced by these guidelines. The 5R framework can be used at any stage of the process, as it has been done for Olaparib (selected as drug candidate before 2011) resulting in the initiation of novel clinical studies. (Readers are encourage to check Box 1 in the paper for an example of the 5R framework applied to Osimertinib).
Table 1. Influence of 5R framework for 15 projects new molecular entities and new biologics entering phase III.
Although these guidelines have clearly improved productivity, 81% of the projects still failed at some stage of the process, but it is clear they are moving the process in the right direction and it will be interesting to see how these guidelines may affect the R&D strategy of other companies.
Blog written by Marco Derudas