The modern design of clinical trials, with the need for approval by a board of ethics, written consent from participants, and multiple other safeguards, is an underrated triumph in the history of medicine. 

For centuries, clinical trials have been unpredictable, questionable, and dangerous. In antiquity, anatomy lecturers would routinely perform live dissections on prisoners. In the 1800s, the smallpox vaccine was discovered when the English physician Edward Jenner rubbed a small amount of cowpox matter into the skin of an 8-year-old boy. And who can forget the medical atrocities committed during World War II?

We should be grateful that we now live in a world in which clinical trials are highly regulated, and the process from drug experimentation to approval is long, with a very high bar set. Medical innovation is today governed strictly by reproducible, high-quality data, and not the whims, intuition, or imaginations of any one party. 

Continue Reading

However, even modern clinical trials have room for improvement. This is brilliantly illustrated in a paper on discontinued clinical developments in Duchenne muscular dystrophy (DMD) written by a group of esteemed European researchers. This paper explores therapies that have been under investigation for DMD that eventually did not make the cut. 

Trial and Error 

Because patients with DMD eventually lose their ability to ambulate and have a shortened life expectancy, there is a sense of urgency when it comes to DMD clinical research. And progress has been made: the development of antisense oligonucleotides and gene therapies that increase dystrophin levels in DMD patients is a testament to that.

Markati et al reported that out of the 40 compounds that have been in clinical development for DMD, only 5 have received approval from the US Food and Drug Administration. As previously mentioned, the bar is set very high, and rightly so. 

Read more about DMD therapies

Markati and colleagues sought to identify the compounds for which clinical development for DMD has been terminated to determine if we can learn anything from them. They performed a comprehensive review of 4 academic databases and identified 16 compounds whose development has been discontinued. 

They identified 5 key reasons for termination. The first is the discovery of discrepancies in results between preclinical and clinical research. This could possibly be due to pressure from stakeholders and investors for positive results to be published, leading to an inflation of confirmatory trials that are not replicable in later studies. 

The second reason is insufficient exposure time or sample sizes of clinical studies. Because DMD is a rare disease, recruiting participants can be a challenge; recruiting participants willing to be studied for a long period of time is even more challenging. Unfortunately, data from clinical trials that are judged to be too small or limited in scope are less likely to be accepted.

Third is variability in patient response or the nonlinear progression of the disease. One suggestion to overcome this problem is to adopt disease modeling and the use of Bayesian statistical design in clinical trials. Patients can also be clearly stratified according to disease trajectory. 

Read more about DMD treatment

The fourth key reason for discontinuation is the insufficient characterization of surrogate endpoints. For clinical studies to be taken seriously, the relationship between surrogate endpoints and functional outcomes needs to be clear. Sometimes, a period of time is to be expected between observed differences in surrogate endpoints and observed differences in functional outcomes; this, too, must be made clear. 

Fifth is variability between investigation sites. This could mean that results from a clinical trial conducted in one part of the world differ from those reached in another. This is a problem in many clinical studies and could be due to epidemiological differences; however, strict protocols for procedures can partly overcome this problem. 

Designing Future Trials 

Scientists are already thinking about innovative new ways to change the game when it comes to clinical trial design. Kurzrock et al published a paper suggesting ways in which clinical trials can be improved in the future. 

For example, they offered the idea of recruiting more patients during early-phase trials so that therapeutic signals (and clinically relevant toxicities) could be observed early on. Generally, phase 1 trials are focused on understanding the biological mechanisms of proposed drugs, and therapeutic implications are left for later.

“In the modern era, one of the most important objectives of phase 1 trials involve finding therapeutic signals; finding response signals is more likely if more patients are treated,” they wrote.

Another idea proposed by Kurzrock and colleagues is expediting promising drug trials (theoretically going from phase 1 to phase 3). The FDA has recently established new pathways to speed the review process for emerging therapies that fill an unmet medical need.

“Drugs designated as breakthrough therapies, for example, may be approved without a control group using interim results of ongoing studies or with enrichment strategies devised to select patients who are most likely to benefit,” Kurzrock et al explained. 

Medicine is always evolving; we shouldn’t be surprised that clinical trials are too. The speed at which safe and effective COVID-19 vaccines were developed and mass-produced offer us a glimpse of what well-designed clinical trials can do. We can expect future trials to be better-designed, safer, and more effective.


Markati T, De Waele L, Schara-Schmidt U, Servais L. Lessons learned from discontinued clinical developments in Duchenne muscular dystrophy. Front Pharmacol. Published online November 1, 2021. doi:10.3389/fphar.2021.735912

Kurzrock R, Lin CC, Wu TC, Hobbs BP, Pestana RC, Hong DS. Moving beyond 3+3: the future of clinical trial design. Am Soc Clin Oncol Educ Book. Published online May 27, 2021. doi:10.1200/EDBK_319783