The media has been constantly reporting breakthrough discoveries in medical research, so much so that it is no longer exhilarating to read about the "next cure for cancer" or "how to live forever". Nonetheless, rarely these discoveries can be translated into useful, practical and realistic medical treatments for real patients. Experts believe that the average success rate to “translate” laboratory findings into fruitful clinical trials is a miserable 8% (1). In other words, nine out of ten “breakthrough discoveries” never make it to human trials, let alone being approved for treatment in the real world.

So why does this happen? The scientific community has provided many factors that had led to such astonishing failure rate, but these can be summarised into three main reasons:

1. Animal models do not adequately represent complex human physiology

Despite being an essential part of biomedical research, animal models suffer from the lack of complexity necessary to mimic human disease conditions in real life. There have been enormous support and endorsement from the scientific community over the value of animal models, which can be summarised in the following statement: “Virtually every medical achievement of the last century has depended directly or indirectly on research with animals” (2). However, the value and reliance on animal models cannot be exaggerated. There was an increasing awareness about the challenges to directly use results obtained from animals and apply onto human trials.

The mouse model, a popular animal model used in laboratory work, is actually a very poor model to mimic human diseases condition. For example, several crucial genetic, molecular, immunologic and cellular differences between mice and humans had rendered results from mice inappropriate for cancer research (1). Researchers in genetic studies also discovered that the binding sites for transcription factors to homologous genes were different between mice and humans, further complicating the effort for any successful translational research (3). Publication bias also contributed to the seemingly inflated efficacy of animal studies. In a report published in PLOS Biology, researchers found that publication bias was accounted for one-third of the efficacy reported, where the actual efficacy may be lower after adjustment for publication bias (4).

2. Low reproducibility of laboratory findings

Earlier this year, Nature had conducted a survey among its scientist readers about their perception of the “reproducibility crisis”. Not surprisingly, over 70% of these scientists claimed that they had tried and failed to reproduce another group’s experiment. Reproducible experiments are the pillars of good science. They allow scientists to test and re-test the theories and findings, and subsequently, more knowledge can be built upon this solid foundation.

Yet, the definition of reproducibility itself may be ambiguous, as discussed in a Nature editorial “Those who study the science of science joke that the definition of reproducibility itself is not reproducible” (5). The term "reproducibility" may cover different scopes of a study, or it may imply varying degree of repeatability, robustness, or reliability of an experiment according to different people. Nonetheless, we cannot brush over the fact that basic research that cannot be repeated fail to provide a solid foundation for translational efforts.

3. Failure of clinical trials to prove efficacies

A promising molecule may have multiple safety and efficacy hurdles to overcome before it can be tested in human, but once the molecule is approved for clinical trial, it must be tested using robust trial designs in order to satisfy strict regulatory requirements. However, most often than not, clinical trials are criticised as "outdated and impractical" where "a significant number of good drugs are being lost" (6). In other words, it is the failure of trials rather than the failure of the efficacy of the compound tested.

Rachel Sherman, the associate director for medical policy at the Center for Drug Evaluation and Research at the U.S. FDA commented that the old clinical trial paradigm was designed “based on single trials, carried out at single sites, and designed to answer single questions” (6). Current trials face much more stringent criteria, and it was much more challenging to prove efficacy, particularly when the compound confer only small benefits in comparison to established treatments.  MIMS

Read more:
Ethical controversies in today’s medical research
Cancer research body criticized for methods of evaluation
Research finds 13 people with 'superhero DNA'

1. Mak S, Evaniew N, Ghert M. Lost in translation: animal models and clinical trials in cancer treatment. Am J Transl Res. 2014;6(2):114–8.
2. Matthews RA. Medical progress depends on animal models - doesn’t it? J R Soc Med. 2008 Feb 1;101(2):95–8.
3. Odom DT, Dowell RD, Jacobsen ES, Gordon W, Danford TW, MacIsaac KD, et al. Tissue-specific transcriptional regulation has diverged significantly between human and mouse. Nat Genet. 2007 Jun 21;39(6):730–2.
4. Sena ES, van der Worp HB, Bath PMW, Howells DW, Macleod MR. Publication Bias in Reports of Animal Stroke Studies Leads to Major Overstatement of Efficacy. Roberts I, editor. PLoS Biol. 2010 Mar 30;8(3):e1000344.
5. Reality check on reproducibility. Nature. 2016 May 25;533(7604):437–437.
6. Ledford H. Translational research: 4 ways to fix the clinical trial. Nature. 2011 Sep 28;477(7366):526–8.