Research Model Trends: How will CRISPR technology affect the pharmaceutical industry?

May 27, 2015 in Blog by ws-lssg

By Dr. Wade Sunada, Managing Consultant

With the recent introduction of CRISPR/Cas9 gene editing technology by companies who develop research models for the pharmaceutical industry, the possibilities for custom generating research models for pharmaceutical research and development have increased.   Interest in CRISPR/Cas9 from pharma is high, but what do pharma researchers need to know about CRISPR?  What is possible?  What is changing?  How will this affect research programs going forward?  The OutsourcePharma Advisory Board spoke with representatives from some of the leading companies using CRISPR technology to custom generate research models about their views on how CRISPR technology will affect the pharmaceutical industry.

For Alexandre Fraîchard, CEO of genOway, conversations with clients often start with CRISPR.  “What I can tell you is, clearly people hear about the Cas9 story and we have more people coming to us about okay, then how long would it take and how much would it cost to create a new model using the Cas9 technology.  So clearly there is an interest.  For me it clearly means that there’s a lot of scientists in the industry that are still believing that creating an animal model and using it is a two year program, risky and with maybe an average outcome.  So, this Cas9 story clearly opens up a new window to these guys that can say, maybe it is different now.

While nuclease-based gene editing technologies have been around for several years (TALEN, zinc finger, etc.), it is really the advent of CRISPR/Cas9 technology that is starting to change how pharma looks at research models.  Personalized medicine is a catch phrase that entered the lexicon since the turn of the century, but is really now becoming a process instead of a far-off goal.  From genetically modified research models to biomarkers to companion diagnostics to adaptive trial designs, it is difficult to bring a therapeutic to market without at least considering a targeted approach.

The ability to quickly modify genes with CRISPR gives unprecedented power and flexibility in the creation of genetically modified research models.  This translates into 3 tangible benefits to pharma researchers: 1) Reduced timelines for custom generation of research models; 2) Lower cost of custom generating research models; and 3) feasibility of more complex custom generated research models.  Clearly the ability to custom generate models cheaper and faster has obvious benefits, but there are a few hidden benefits as well.  For example, pharma companies can now custom generate a simple knockout model fast enough that it may be a better choice to custom generate a model that is available from repositories strictly for IP reasons, as the time and cost tradeoffs get smaller and smaller.

What is the impact of reduced timelines for CRISPR custom generated research models?  According to James Vitale, Director, Product Management at Taconic Biosciences, “CRISPR will allow investigators at pharmaceutical companies to generate GEMS very quickly.  This speed will provide quick answers to critical questions.  The CRISPR models may be simple, for proof of concept in model development, or may serve as the model used in drug development discovery research.

How much faster is CRISPR?  Lara Reid, Product Manager at Sage Labs, project times for custom generating models have decreased by months.  “I would say at least a couple months.  We used to quote our knockout rat models using the ZFNs around six to eight months and now we can do them in as little as five, and we see the same trend with all of our knockin projects.  Those could take almost a year sometimes and now we can get conditional knockins between six and eight months.”  Having access to custom generated models faster can allow for the use of custom models in situations where previously researchers may have not had the time to wait for custom model development, or did not have the ability to plan far enough ahead to develop a custom model.  Fraîchard believes that the strength of CRISPR technology is the speed that simpler models can be created.  “With CRISPR/Cas9, you benefit if you are really in a hurry and if you want low sophistication model.  Then you will benefit from this technology and you will save a few weeks or few months in your development time.

While the lower cost of CRISPR models is certainly a benefit, Fraîchard cautions that the design of the model is more important than the cost. “If you do it quick and dirty, you will save a little bit of time, a few months and maybe a few tens of thousands of U.S. dollars but then when you’re going to use it, you’re going to spend much more time and much more resources to use it.  One very simple example is if it’s not properly designed, after a year you will see the limit of the model.  You will say okay, I got the first result but if I go more in detail, I can see that it’s not working so you need a new model.”

Creating more complex models furthers the possibilities of developing personalized medicines.  By more closely mimicking the human response, animal models can go beyond target validation and better predict results in humans.  Vitale notes that custom generated rat models are a key benefit of CRISPR technology.  “Developing GEM rats has been difficult; CRISPR technology offers a way to generate GEM rats in a targeted manner which is not possible through traditional transgenic technology.  Reid notes that Sage thinks the future of animal models is in more complex species than mice.  “I know that’s always been the hope of Sage is that the rat models would really pick up and become as popular model as the mouse.  I think that that trend is increasing and I know we just hope to see it continue to increase, especially over the next five years.”

Humanized models offer a stern test for CRISPR technology.  While clearly a fantastic technology, CRISPR is still a young technology with potential pitfalls.  The possibility of off-target effects is greater with CRISPR than with ES cell technology, but that can be mitigated by screening for off-target effects and with the quality of design of the model.  CRISPR may be less suited for humanized models until there is more experience creating those models using CRISPR because of the potential difficulties inserting the large fragments required for humanization.   Fraîchard notes: “If you do need a full humunization of the target or if you need something sophisticated like chimeric proteins – partly human, partly mouse – so that you get signal transduction, if you should need that today, we don’t use Cas9 for that because it’s not robust enough.  So, that’s exactly the issue with the Cas9 today is simply the efficacy level for insertion of large fragments plus the off target risk.”  Reid notes that Sage only uses CRISPR technology, and believes the technology is validated based on their successful completion of complex models such as humanized models, conditional knockins and conditional knockouts.  “We’ve proven that the technology is useful across the board.”


So is CRISPR the right technology for your next research model?  Vitale notes that “We are finding that scientists come to Taconic with a CRISPR model in mind, but after discussions with our Scientific Project Managers realize that a different technology will be better suited to deliver the model they had in mind.  I would offer this piece of advice; seek out a vendor who has capabilities beyond just CRISPR, otherwise the model may be developed using CRISPR, but it may not be the best possible model to meet the goals of your study.  Seek out a vendor who has a variety of model generation tools and who can also provide the downstream breeding and analysis.  This will maximize the chances of success.”

As exciting as CRISPR technology is today, in the end the models that can be created using the technology are going to have the greatest effect on the pharma industry.  Fraîchard states: “Will it change anything in the way we do business in the industry?  No, because it’s just one more tool in the toolbox.  It’s a nice tool, a great tool but that’s not what is important.  What is important for pharma or for the scientist or the PI is the intrinsic value that will be growth to the project by the animal model.  The way the animal model was created does not matter.