Why Do Some Drug Trials Kill?

[DapostropheS]

The Modern Pharmaceutical Machine

Modern pharmaceuticals have a huge impact on the health and well-being of patients and their caregivers by slowing or preventing disease, alleviating symptoms, and increasing length or quality of life. In recent years, biopharmaceutical and pharmaceutical innovation has led to accelerated progress in the fight against many diseases. Drug approvals are increasing, with data from U.S. FDA showing a recent upward trend in novel drug approvals.



This comes despite the truly mindboggling costs, currently estimated at $2.6 billion, to develop and win regulatory approval for a new drug. With 43% of cost attributed to the pre-human development stage and 57% attributed to the three human clinical phases. That’s $1.4 billion on clinical development alone for each approved drug. [For more information of the phase process try this great infographic.] Phases are structured to ensure patient safety and drug efficacy, in that order. The first phase is the riskiest; it is the first time the drug enters a human. Phase 1 focuses on safety, typically with a small cohort of volunteers. Phases 2 and 3 tend to be larger trials examining the drug’s effectiveness, as well as safety. The whole development process, including clinical, takes 7-10 years.

However, with blockbuster drugs like adalimumab (marketed as Humira by Abbvie) able to take $10 billion revenue in a single year, it is clear why many companies invest both the time and money; the gamble can certainly pay off quickly.

Do We Know All We Can About Candidate Drugs?

Testing new drugs requires techniques at the cutting edge of science. New chemical or biological entities can now be characterized exhaustively using the very latest analytical technologies in a laboratory environment; from cell-based toxicological studies to chromatographic separation of highly immunogenic therapeutic protein aggregates [Want to find out how to do this? Download this new free eBook]. Using the latest analytical tools, pharmaceutical development scientists can determine almost everything there is to know about the physical and chemical properties of even the most complex candidate drug. I’m not saying this is trivial. Modern therapeutic proteins are enormously complex, but it can be done.

On the other hand, do we know everything about the patient? Sadly not. Humans are biologically complex. Socially complex too, but that’s another conversation. There remains a lack of understanding of many biological processes and their combinatorial interplay. We can know everything about the drug in the lab. But, when inside a human, its effects (both on- and off-target) become a component of our personal physiology and genetic variation.

Of course, the safety and effectiveness of these drugs can be, and is, rigorously tested in animals under highly regulated conditions. This testing, like it or loathe it, furnishes the clinician with a wealth of toxicology and metabolism data. Society in general, has deemed it reasonable to test and sacrifice a wide variety of animals in the pursuit of longer human life; rats, monkeys, beagles – all acceptable collateral damage. Sadly, a dog isn’t human. Toxicological effects measured in laboratory animals can be modeled but not perfectly extrapolated to humans. After a drug is confirmed as safe and efficacious in pre-clinical studies, it must be tested in healthy human volunteers. This can be lucrative for some. A volunteer can be paid up to ten thousand dollars for simply lying in a bed and taking a drug.  Marginally less effort than grinding out a blog to pay the rent. But, perhaps somewhat riskier.

Why Does It Sometimes Go Wrong?

Sadly, the news is periodically filled with stories of human drug trials ‘that went wrong.’ At the start of 2016, sadly one man died and several were hospitalized during a Phase 1 trial, using 128 healthy volunteers in France. The candidate drug, from Portuguese company Bial, has been reported as an inhibitor designed to act on a pain signaling pathway that had previously not been explored by other drugs on the market.  In 2006, another Phase 1 clinical study was conducted of an antibody TGN1412 in 6 human volunteers in the U.K. After the very first infusion of a dose 500 times smaller than that found safe in animal studies, all six human volunteers faced life-threatening conditions involving multi-organ failure. Subsequent investigations showed that pre-clinical data pointed to possible risks in humans, but this was overlooked. Possibly time will show human errors also lead to this years’ disaster in France?

Hindsight often gives perfect clarity where foresight is myopic and filled with error. Predictive pre-clinical models are extremely advanced. But, the complexity of human biology will continue to make predictive models a best-guess. Thankfully, that same beautiful complexity of genetic and physiological diversity lead to only 1 death in 128 volunteers, rather than all 128.

It is not just biology - we must also consider the human decision factor. Models and trials are operated and reviewed by humans. Humans are fallible. Humans can miss data. Humans can feel pressure to make decisions, get clinical data fast, and get drugs to market quickly. We also must consider that sometimes societies are willing to accept human collateral damage in the pursuit of greater health. After all, if society demanded zero risk in clinical trials, the 45 new drugs approved last year would still be stuck in the pre-clinical phase, rather than saving lives today. Maybe the blame lies with all of us? Maybe we are happy to accept one death to save ten thousand more?