How to speed up the development of a drug?

João Vítor Perez de Souza
3 min readDec 16, 2020

Clinical trials, vaccine development, drug development…

Hearing or reading these terms became common in our daily lives as the COVID-19 pandemic hit the world.

As of today, in an unprecedented time, we have at least a few vaccines with proven efficacy against COVID-19, and several countries already started vaccination schemes.

In “common situations”, the development of a new drug or vaccine takes from some years to more than a decade from initial tests to the delivery of the final product. Estimates indicate that this process costs at least U$ 2 billion for completion.

This enormous amount in expenses comes from the thousands (often millions) of failures in early research due to lack of activity, presence of side effects, or characteristics that are not interesting for commercial use (eg. difficulty of synthesis).

Standard drug development workflow. Most molecules end up being discarded in the initial steps of research and only a select number will proceed to clinical trials. Source: PhRMA research.

The figure above shows us how the majority of drug development efforts fail to go through the initial steps of research (i.e. in vitro tests). These failures come as important wastes of research resources, time, investment, and personnel effort.

As the biggest bottleneck is in the early stages of research, we can apply strategies to avoid or at least minimize the chances of occurrence of the most common issues found.

But what can we do to avoid these issues?

In my Ph.D. thesis, I search for possible new antibiotic drugs against tuberculosis. As a way to speed up this process, I use bioinformatic tools that allow me to accomplish things that would be impossible in the same proportion or time if done otherwise:

  1. I can test hundreds of thousands of molecules and select only the ones with the most promising activity for future tests;
  2. To discard molecules with predicted toxicity, making the final drug safer;
  3. Evaluate how the metabolism of my candidate drug could be, which aids in the selection of optimal dosage for eventual treatment.

All this saves much time and research resources!

PdbId 2XB8, Mycobacterium tuberculosis 3-Dehydroshikimate dehydratase

The figure above exemplifies an in silico assay called molecular docking, where an algorithm tries to evaluate the possibility of binding of a candidate drug (blue) in an enzyme (green), thus leading to its inhibition and the death of the pathogen.

It is important to point out that although in silico (computational) assays are of great aid, in vitro tests are indispensable to confirm findings!

This research is developed at the State University of Maringá, Brazil, with a partnership of researchers Dr. Rosilene Cardoso, Dr. Erika Kioshima, Dr. Flávio Seixas, and at the LORIA Institute, France with Dr. Bernard Maigret

Published By João Souza

Originally published in Portuguese at https://www.linkedin.com.

--

--

João Vítor Perez de Souza

I am a PhD student at State University of Maringá, Brazil, Data Science enthusiast and eager to learn about Health and Data!