You are here Biopharmaceutical/ Genomics Glossary Homepage > Best practices > Drug discovery & Development

Biopharmaceutical Best practices, lessons learned & ongoing challenges  DRAFT 
Evolving Terminology for Emerging Technologies
Comments? Questions? Revisions? 
Mary Chitty MSLS
Last revised January 07, 2020

Challenges: Spending more and more on R&D and getting fewer and fewer new drugs is not a sustainable business model. Does pharmacogenomics hold the potential to make every drug an orphan drug and every disease a rare disease?  Will there be enough patients in the world to fill Randomized Controlled Clinical Trials? 

Bridging the silos between biologists, biophysicists, chemists, IT, toxicologists, patent attorneys, CFOs, CEOs and clinicians [and patients] present some of the biggest challenges of all. My own undergraduate anthropology major is more relevant to me now than it was when I graduated.  

I'm realizing that the variety of skills and domain expertise needed for breakthroughs in the life sciences is more than anyone discipline and/'or lifetime(s) can muster.  While new technologies might be helpful, cultural shifts, incentives for collaborating, and pre-competitive cooperation will also be important.  Incremental changes can be helpful more frequent than true paradigm shifts. Industrialization, scalability and automating and ramping up processes to move from the R&D lab into the clinic are under-appreciated challenges, as are marketing dilemmas that come about with disruptive technologies.

Best practices
Biology :
 Nature is very clever. Leveraging biomimetics [high-tech pharmacognosy] can help the odds.  

Chemistry: Fail fast
Killing compounds early has been said to be like “drowning puppies in front of small children”.  How many companies reward this vital task?   But even failed drugs may be repurposed and less interesting compounds licensed out for others to investigate. Repositioning existing drugs can leverage existing toxicity profiles.  Who would have thought thalidomide would be back on the market?  But if compounds are going to fail, killing them quickly saves time and money

Company size is not an indicator of success in terms of R&D productivity … the strongest single correlator with success (odds ratio 3.9) was having a high termination rate in preclinical/Phase I stages. This indicates that companies have an early idea of which assets are likely to succeed, and that the companies most willing to face the hard decisions about which assets to terminate do better than companies that let assets linger. Does size matter in R&D productivity? If not, what does? Michael Ringel, Peter Tollman, Greg Hersch and Ulrik Schulze Nature Reviews Drug Discovery 12:901-902, Dec 2013 

Clinical trials and personalized medicine
Selecting patients who will respond to compounds, without toxic effects [pharmacogenomics/ personalized medicine] will enable better, faster, cheaper, more definitive clinical trials, and help to meet unmet clinical needs. But biomarkers for patient stratification to identify non-responders and toxic responders are still nascent. 

Innovation & productivity I had hoped drug companies which encouraged open sharing of scientific information would prosper in the long run, without finding much evidence (even anecdotal) until I read this report, which quantified the positive correlation between companies encouraging (or requiring) peer reviewed scientific publication and productivity (patents issued to company scientists and articles published in peer- reviewed journals by company scientists). Diffusion of Science Driven Drug Discovery Organizational Change in Pharmaceutical Research, Iain M. Cockburn, Rebecca Henderson and Scott Stern, NBER, Sept. 1999   I asked a few years ago if this paper had been updated – but it hasn't  been.  Adoption was noted as being extremely slow.

Are there other best practices we can identify?

Lessons learned

Biology Genes and proteins, genomics and proteomics are just molecular biological starting points. If DNA is 2-dimensional, and protein structures are 3D, then post-translational modifications and alternative splicing are 4D temporal-spatial relationships constituting profoundly dynamic systems. DNA makes RNA makes protein[s] is still true -- but with infinitely more variations than originally envisioned.

in most contexts, talk about being "post- genomic" seems premature. "Post Mendelian" seems more accurate as we move from an era in which genetics has been rooted in monogenic diseases with high penetrance to a greater awareness (but limited understanding) of polygenic diseases (and traits) often with relatively low penetrance. How little we know about polygenic diseases with varying penetrance.  

Biomarkers & molecular diagnostics Diagnostics almost always precede therapeutics, and “companion diagnostics” or theranostics are increasingly important and closely aligned.  

Bioprocessing & manufacturing Just getting enough of compounds can be the first challenge.  Biotech started as a means for bioproduction, but now biotechnologies, genomic, proteomic and metabolomic tools and insights permeate every/many? stage of pharmaceutical R&D. Scaling up is a profoundly underestimated challenge.  

Business: Some technologies are improved in a linear fashion or incrementally.  Others truly change the paradigm.   Clayton Christensen writes about these in The Innovator's Dilemma. What is interesting about Christensen's analysis (based on data from the disk drive industry) is that he found disruptive technologies tended to be much cheaper than existing technologies. Existing companies were quite capable of developing the technologies (and had). What they couldn't do was figure out how to market them and whether it made sense to devote sufficient resources to them (which in many cases would not have been the responsible thing to do.)  The pharmaceutical industry is mentioned only in passing, but the success of larger established companies either partnering with smaller less established ones (clearly happening in the pharmaceutical and biotechnology sectors) or spin- off of promising developments as separate companies makes a lot of sense. 

We conclude that a substantial sector of the pharmaceutical industry has not modified its drug design practices and is still producing compounds with suboptimal physicochemical profiles. Paul D. Leeson and Stephen A St-Gallay The influence of the "organizational factor" on compound quality in drug discovery, Nature Reviews Drug Discovery, 10:749-765, Oct 2011

Ongoing challenges
There isn’t enough matter in the universe to make all possible compounds.  High throughput screening needs careful compound prioritization to avoid being the chemical parallel of trying to get monkeys with typewriters reproducing Shakespeare, 

Drug development
 Proofs of concept enable better experiments and fewer [expensive] unpleasant surprises. 
ADME -- We know a lot about absorption and distribution, not so much about metabolism and excretion. 

Drug safety & pharmacovigilance From preclinical through Phase IV and post marketing surveillance, drug safety just never ends. Idiosyncratic toxicities don’t emerge until drugs reach larger populations.  

Drug targets Greatly increased numbers of targets mean even more need to be validated as hits and to optimize leads.  Papers per compound have dropped from > 100 to < 10 according to Lehman Brothers 2001 report The Fruits of Genomics.   

Informatics isn’t as much about old knowledge becoming obsolete, but becoming more and more granular, with elements of chaos theory thrown in. .  

"A month in the lab can save you an hour in the library." said bioinformatician Phoebe Roberts.

"The biomedical literature is even noisier than microarray data." says John Quackenbush, of the Dana Farber Cancer Center

Moravec’s Paradox:  the discovery by artificial intelligence and robotics researchers that, contrary to traditional assumptions, high-level reasoning requires very little computation, but low-level sensorimotor skills require enormous computational resources. The principle was articulated by Hans MoravecRodney BrooksMarvin Minsky and others in the 1980s. As Moravec writes, "it is comparatively easy to make computers exhibit adult level performance on intelligence tests or playing checkers, and difficult or impossible to give them the skills of a one-year-old when it comes to perception and mobility."[1]  Similarly, Marvin Minsky emphasized that the most difficult human skills to reverse engineer are those that are unconscious. "In general, we're least aware of what our minds do best," he wrote, and added "we're more aware of simple processes that don't work well than of complex ones that work flawlessly."[2]   Wikipedia accessed 2018 Jan 25

Artificial intelligence and machine learning have been developing for decades, but recent advances in computing power and scale seem to hold promising developments for pharmaceuticals and healthcare.

Question from Nature column Lifelines put to Michel Brunet, palaeontologist "What is the one thing about science you wish the public understood better?" Answer "That the 'truth' is always an asymptotic ideal." Dreams of the past, Nature 423 (6939): 121, 8 May 2003

Innovation, productivity and diversity
Novelty is an essential feature of creative ideas, yet the building blocks of new ideas are often embodied in existing knowledge. From this perspective, balancing atypical knowledge with conventional knowledge may be critical to the link between innovativeness and impact. Our analysis of 17.9 million papers spanning all scientific fields suggests that science follows a nearly universal pattern: The highest-impact science is primarily grounded in exceptionally conventional combinations of prior work yet simultaneously features an intrusion of unusual combinations. Papers of this type were twice as likely to be highly cited works. ... the production and consumption of boundary-spanning ideas can also raise well-known challenges. If, as Einstein believed, individual scientists inevitably become narrower in their expertise as the body of scientific knowledge expands, then reaching effectively across boundaries may be increasingly challenging, especially given the difficulty of searching unfamiliar domains. Moreover, novel ideas can be difficult to absorb and communicate, leading scientists to intentionally display conventionality.   Atypical combinations and scientific impact.  Uzzi B, Mukherjee S., Stringer M. Jones B.  Science 2013 Oct 25;342(6157):468-72. doi: 10.1126/science.1240474.  

This article talks about how Newton presented his laws of gravitation in Principia using accepted geometry instead of his newly developed calculus, and Darwin used well accepted knowledge of selective breeding of animals in the first part of Origin of the Species.

Expertise does matter: Teams publishing in high-impact journals have a high fraction of incumbents. But diversity matters too: Teams with many former collaborative links offer inferior performance. …When forming a “dream team” make an effort to include the most experienced people, whether or not you have worked with them before. The temptation to work mainly with friends will eventually hurt performance …  A mathematical theory of human dynamics may not be the solitary achievement of a genius scientist but will likely emerge from the combined efforts of an expert team with just the right combination of expertise, collaborative experience, and fresh ideas. Network Theory--the Emergence of the Creative Enterprise Albert-László Barabási* Science 29 April 2005: Vol. 308 no.5722 p. 639-641 
DOI: 10.1126/science.1112554  [abstract only]

Here, we investigate how the mechanisms by which creative teams self-assemble determine the structure of these collaboration networks. We propose a model for the self-assembly of creative teams that has its basis in three parameters: team size, the fraction of newcomers in new productions, and the tendency of incumbents to repeat previous collaborations. ...research shows that the right balance of diversity on a team is elusive. Although diversity may potentially spur creativity, it typically promotes conflict and miscommunication  … We analyzed data from both artistic and scientific fields where collaboration needs have experienced pressures such as differentiation and specialization, internationalization, and commercialization (41011): (i) the Broadway musical industry (BMI) and (ii) the scientific disciplines of social psychology, economics, ecology, and astronomy (Table 1). For the BMI, we considered all 2258 productions in the period from 1877 to 1990 (1213).  …. For each of the scientific disciplines, we considered all collaborations that resulted in publications in recognized journals within the fields studied (14): seven social psychology journals, nine economics journals, 10 ecology journals, and six astronomy journals   Guimerà R, Uzzi B, Spiro J, Amaral LA. Team assembly mechanisms determine collaboration network structure and team performance. Science. 2005 Apr 29;308(5722):697-702. doi: 10.1126/science.1106340. PMID: 15860629; PMCID: PMC2128751.  [free full text]

The most productive teams were composed of experienced members and novices. These outperformed teams composed only of experienced people – a great argument for diversity.

Women and Biotech: Defining and analyzing the problem; Solutions and Strategies, Radcliffe Symposium

Molecular Medicine
 Medicine is being reorganized at the molecular and biochemical levels. Traditional pharmaceutical franchises  and clinical medicine are being supplemented by molecular networks, pathways and insights. How will scientists and practitioners adapt to this?

Regulatory: GxP -- from GLP Good Laboratory Practice to GCP Good Clinical Practice and GMP Good Manufacturing Practice to INDs Investigational New Drugs and NDAs New Drug Approvals -- and truly NCEs New Chemical Entities and NBEs New Biological Entities -- is a long, expensive, and often unpredictable journey. Regulatory authorities are struggling with trying to figure out how to regulate new and emerging technologies.

Research: Progress from basic to applied research and commercial R&D is distinctly non-linear.  Human lifetimes aren't long enough to learn everything needed to get drugs to market safety -- and keep them there. Collaboration and partnerships seem essential

Learning to live with uncertainty and tradeoffs is also essential and unavoidable. Uncertainty became a PubMed MeSH heading in 2003. 

Managing change and expectations are also crucial. Homeostasis is a very powerful barrier to change. More research is pre-competitive than anyone likes to admit.

A useful metaphor is sailing and tacking” When I started to sail I wanted to go straight ahead – which gets you “in irons” and unable to steer.  Only by tacking from side to side do you get anywhere – but that also required ongoing adjustments.  Another goal – as hockey great Wayne Gretzky is known for saying “I just skate to where the puck is going to be.:

I've taken heart from Steven Weinberg's Four Golden Lessons, Nature 426: 389, 27 Nov 2003 "How could I do anything without knowing everything that had already been done? ... [graduate school] was sink or swim.... I did learn one big thing: that no one knows everything, and you don't have to. Another lesson to be learned ... is that while you are swimming and not sinking you should aim for rough waters.   My advice is to go for the messes -- that's where the action is...  My third piece of advice is probably the hardest to take. IT is to forgive yourself for wasting time... If you want to be creative, then you will have to get used to spending most of your time not being creative, to being becalmed on the ocean of scientific knowledge.  Finally, learn something about the history of science, or at a minimum the history of your own branch of science. The least important reason for this is that the history may actually be of some use to your ... More importantly, the history of science can make your work seem more worthwhile to you. 

Robert Weinberg's Racing to the Beginning of the Road : The Search for the Origin of Cancer   is a very readable account of top rate biomedical research, a good reminder that these "races" are marathons and not 100 yard dashes. The title is one of my favorite metaphors for the complexity of biology. His explanation of how nonlinear progress from lab to clinic can be is highly recommended. 

failure, learning from: Dr. [Stephen] Friend criticized the current scientific process as one that catalyzes failure because there is no way to learn from others. “Our understanding of disease is based on narrow or generalized knowledge… He added that there needs to be a cultural change in science where scientific results are completely shared. “Good research shouldn’t have to always be a positive result. Negative results should also be reported.”  FasterCures, Learning to Love Failure panel video

The biotechnological innovations of the 1970’s took until the 1990’s to integrate.  "The Pharmaceutical Industry and the Revolution in Molecular Biology: Exploring the Interactions between Scientific, Institutional and Organizational Change, Iain M. Cockburn, Rebecca Henderson, Scott Stern, 1999.  No longer on the web?

Integration of genomics (and proteomics and transcriptomics) into drug discovery and development seems likely to be an ongoing process as well. 

New technologies such as microarrays, next generation sequencing and RNAinterference coexist with improvements in century old mass spectrometry and Nuclear Magnetic Resonance.  DNA, genes, RNA and proteins are intimately related, but the technologies for learning about nucleic acids and proteins can constitute almost unbridgeable chasms

Amara’s law:  "We tend to overestimate the effect of a technology in the short run and underestimate the effect in the long run.”  

Roger Brent compared microarrays to the microscope and telescope because they "enable observation of the previous unobservable" [transcripts expressed under different conditions in cells, tissues, and organisms. But “Making new technology work may be easier than using it to discover truth” Roger Brent, "Functional genomics: learning to think about gene expression data" Current Biology 9: R338- R341, 1999 

 Industrialization, scalability and automating and ramping up processes to move from the R&D lab into the clinic are under-appreciated challenges. too

What are your biggest hurdles and challenges?  I welcome your comments.  Mary Chitty 

Heath, Chip and Dan, Switch: How to change things when change is Hard, Broadway Books, 2010
Sutton, Robert L. and Huggy Rao, Scaling Up Excellence, Crown Business, 2014
Scaling up Excellence video

Tett, Gillian, The Silo Effect: The Peril of Expertise and the Promise of Breaking Down Barriers,  Gillian Tett

CHI’s Drug Discovery and Development Map  text   We're trying to update this popular graphic


How to look for other unfamiliar  terms      Tips/FAQ

Contact | Privacy Statement | Alphabetical Glossary List | Tips & glossary FAQs | Site Map