Probabilities of success in drug development
Translation: Original comment published in Finnish on 9/26/2024 at 7:27 am EEST
The second part of our series of articles focusing on the basics of investing in Life Science companies deals with the likelihood of successful drug development. Drug development companies typically examine the safety, tolerability and efficacy of a drug candidate in three clinical phases (see the first article of this series on clinical phases here). If the safety or efficacy of a candidate is not proven sufficient in the trials, its development is typically ended and the candidate may prove worthless. It is therefore crucial for an investor to understand the probability of drug development being successful based on each clinical phase (Table 1) and the probability at which it moves to the next clinical phase (Table 2). Probabilities can be estimated based on statistics and studies on the successes and failures of previous projects.
The road from the early development phase to the market is very long
In Clinical Phase I of drug development, information on the safety, tolerability, efficacy and behavior of the candidate is still highly limited. This is reflected as high uncertainty about market entry. The probability of obtaining a marketing authorization is historically below 10% (Table 1). However, drug development usually progresses to the next clinical phase and the average probability of reaching Phase II is approximately 60% (Table 2). In Phase I, drug development is typically interrupted due to lack of tolerability. Reasons related to pharmacokinetics can also interrupt development. For example, poor absorption of the drug may prove an obstacle for an otherwise good candidate being a working drug. In the early development phase, extensive efficacy data are not yet available, so lack of efficacy is rarely the cause of failure.
The middle phase is the valley of death in drug development
In Clinical Phase II, the focus moves a step away from safety toward initial demonstration of efficacy. The number of patients and consequently the costs of the trial also increase. Based on the results, decisions are made on the implementation of the costly third phase. Indeed, companies are typically most critical in this phase, and the development is most likely to end in Phase II, from which the probability of reaching marketing authorization is still low (around 15%). The probability of progressing to Phase III is roughly 30%. Development can typically end if efficacy results that would justify significant Phase III investments are not obtained. Data on tolerability and safety from a larger number of patients can also tip the scales to ending the development. Here, just like in other phases, development may also stop due to commercial reasons, such as changes in the competitive situation or a reassessment of the business.
Phase III – the goal is looming
Implementation of the critical Clinical Phase III is very costly and time-consuming, so drug development companies will carefully consider the rationale for this investment. In the last phase, the probability of success has been clearly higher than for Phase II. This is probably due to the tight screening in the previous phase, which means candidates with high success potential progress to the final phase to limit the risks. Historically the probability of candidates in Phase III reaching marketing authorization has been around 50% and the probability for submitting an application for marketing authorization has been some 60%. In this phase, development is most likely to fail due to a lack of efficacy. An even more extensive group of patients may also highlight less common side effects that result in abandoning the development.
Marketing authorization assessment and follow-up
If the drug development company believes that the results of clinical trials support the commercialization of the drug, it submits a marketing authorization application for assessment by the authority. In Europe, applications are processed by the European Medicines Agency (EMA) and in the United States by the Food and Drug Administration (FDA). When the marketing authorization has been submitted, the probability of obtaining the authorization is already around 90% but some projects still stumble in the final stages of the authority assessment. After approval, the safety and efficacy of the drug continue to be monitored throughout its time on the market. Sometimes a granted marketing authorization may be withdrawn if unexpected issues emerge about the drug that have not been detected in the clinical trials. Withdrawal of authorization is more likely when based on an accelerated procedure (such as the FDA's Accelerated Approval) and therefore on more limited clinical evidence.
Probability of success depends on the discipline
The above figures are average probabilities of success of drug development and provide a rough framework based on a large historical sample size. However, on closer examination, there are significant differences between different medical disciplines. Table 3 below shows the probabilities by discipline, from the highest to the lowest. Historically, drug development has been the most successful in hematology (e.g. anticoagulants). Urology, in turn, has historically been the most challenging discipline for drug development.
Additional information is available for those interested
The probabilities of drug development success have been studied in much more detail relative to different variables than we can discuss here. For example, the probability of success has been higher for biologics and vaccines compared to small molecules. The use of biomarkers has yielded good results and in rare diseases, development has been more successful than in common chronic diseases. If you are interested we recommend that you take a look at the openly available reports by Thomas, 2021 and Zhou & Johnson, 2018.
Sources used in this article:
Dimasi et al, 2010: Trends in risks associated with new drug development: success rates for investigational drugs.
Hay, et al, 2014. Clinical development success rates for investigational drugs.
Thomas et al, 2016: Clinical Development Success Rates 2006-2015.
Thomas et al, 2021: Clinical Development Success Rates and Contributing Factors 2011–2020.
Zhou & Johnson, 2018: Pharmaceutical Probability of Success.