HCS roundtable; Achieving the potential of HCS

Introducing the Panel

Name: John Dunlop, PhD
Job title: Director of Neuropharmacology and Neurophysiology
Company: Wyeth Research
Biography: Dr John Dunlop is Director of Neuropharmacology and Neurophysiology in the Neuroscience Discovery Research division of Wyeth Research. Dr. Dunlop oversees a multidisciplinary technology group interfacing with the psychiatric and neurological disease therapeutic areas within Neuroscience. Dr. Dunlop is a member of the Society for Neuroscience, American Society of Pharmacology and Experimental Therapeutics and Society for Biomolecular Sciences.

Name: Dr Edward Ainscow
Job title: Research Scientist
Company: AstraZeneca R&D Charnwood
Biography: Dr Ainscow has been working in the Advanced Science and Technology Laboratory within AstraZeneca for the last five years, developing High Content Analysis screens – particularly in the field of predictive toxicology – and approaches for data analysis and interpretation.

Name: Stuart Gibb
Job title: European Sales Manager / BD Bioimaging & BD Gentest
Company: BD Biosciences
Biography: Mr Gibb’s career in the Life Science Research market began 15 years ago in Scotland as a Bio-Analytical Chemist. He has been working for the last 10 years at BD Biosciences focused on the Life Science Market. In the last six years he has focused on the Pharmaceutical/Biotechnology/Academic arena, with the last two years based in BD Biosciences in Erembodegem, Belgium – the European Headquarters building of the Bioimaging business.

Name: Len Pagliaro
Job title: Vice President, Business Development
Company: Fisher BioImage ApS, Thermo Fisher Scientific Inc., Denmark
Biography: From 2000-2003 Mr Pagliaro served in line and project management roles at BioImage, most recently as VP, Discovery Projects. Prior to joining BioImage, he was Principal Scientist at Cerep, Inc., where he led development of pharmaceutical profiling assays and served as study director.

Name: Dr Gabriel Gradl
Job title: Vice President Cell Handling and Analysis
Company: Evotec Technologies, a PerkinElmer company
Biography: Gabriele Gradl is responsible for cell handling and analysis instrumentation. She is a pharmacist who received her PhD for isolation of a novel human receptor mediating density dependent inhibition of cell growth in Mainz, Germany in 1992. In her Post Doc period she worked on signaling events in growth control and on apoptosis at Genesis R&D in Auckland, New Zealand.

Name: Mark A Collins, PhD
Job title: Marketing Manager, HCS Software
Company: Thermo Fisher Scientific, Integrative Technologies Division, Cellular Imaging and Analysis
Biography: Mr Collins has a Ph.D in microbiology from the University of Surrey in the UK and a Post doc in modelling microbial stress. He has worked with the Cellomics product line since 2001, initially as a software developer and then as a product manager and now as a marketing manager for the CIA business. He is responsible for the commercialisation of Cellomics’ range of High Content Informatics (HCi(tm)), the development of new approaches to cellular (bio)informatics as well as Cellomics’ professional services business.

High Content Screening has become an invaluable tool for pharmaceutical research laboratories the world over. In order to chart the progress of this important technology over recent years, as well as its applications, we spoke to a selection of industry experts. Read what the technology users and vendors have to say about its value and the future of HCS.

End-user view

Q1. Traditionally, what has HCS been used for in the pharmaceutical industry?

Dunlop: In our case with the focus on Neuroscience we have made heavy use of neurite outgrowth and have witnessed the available algorithms for doing this improve significantly over the years. Most recently, we have also begun to combine this approach with measurements of synaptic spine density and multiple neurite parameters to provide a more information-rich approach to neuronal morphology. Other workhorse applications for us have been receptor internalisation, cytoplasm to nucleus translocation and signalling pathway activation. Of course there are other approaches to these types of assays – the value of HCS is the ability to multiplex and also get a sense of cellular toxicity simply from total neuron count or neuron morphology.

Ainscow: We have used HCS in all phases of early drug discovery, including primary screening (e.g. Norak for GPCRs and other in-house assays), target validation, secondary screening, selectivity assessment, pathway de-convolution and early safety assessment.

Q2. Currently, what are the greatest benefits of using HCS in R&D?

Dunlop: In the case of neurite outgrowth being able to take a very labor intensive, manual assay and to perform it in an automated format in 96- and 384-well formats such that it can now be implementing in screening paradigms. In other cases, the ability to use information rich cell-based assays is important, and as discussed above the ability to multiplex assays. That said, there is still a great opportunity for more widespread implementation of multiplexed assays, the true power of high content; in practice many users are still running fairly simple assays.

Ainscow: The greatest benefits can be outlined as the following:

• Access to biology that would be intractable for screening through non-HCS methods, or is best represented through using targets in the appropriate cellular environment. This includes looking at target pathways and sub-cellular, organelle related responses.
• Giving a detailed yet broad indication of the effects of cells through multiplexing of assays and multiple readouts of the phenotypic profile induced by a compound.

Q3. HCS is being applied to several different areas and is combined with different techniques (such as RNAi). In the future, what do you consider will be the most exciting applications of the technology?

Dunlop: There is still a lot of potential in the combination of HCS and RNAi. Also there will likely be HCS studies moving to more complex biology, e.g, imaging in brain slices and other complex tissue.

Ainscow:

• Improved bioinformatics utilities will better enable access, visualisation and analysis of HCS data. This will also then tie in non-HCS data.
• Also use of primary or stem cells in 3D culture models should enable more relevant and predictive biological systems to be used in automated culture.

Q4. Looking specifically at your operations, what techniques are you using in your laboratory?

Ainscow: We routinely use ICC and fluorescent labelling of cells. Our current drive is to improve throughput and assay robustness through development of automated platforms for the preparation of plates for imaging. We are also developing new methods of image analysis (e.g. for live cell imaging) and informatics approaches for data handling and decision making. Cell models used include clonal cell lines, primary cells and 3D culture models.

Q5. What do you consider to be the most ground-breaking technologies in the market place?

Dunlop: Each of the hardware platforms now has a good level of validation and each has its own strengths and weaknesses so new users really need to assess their needs in light of what is our there. We have been heavily focused on the Cellomics platform and have had very positive experiences with it; the Opera also appears as a good option for true HTS application.

Ainscow:

• Improved image analysis, giving the end user the ability to quantify and characterise what is being seen by eye.
• Extension of the HCS into areas that it has not traditionally been applied to – for example, pathology and in vivo studies.

Q6. What improvements/developments need to be made to existing products in order to further drive the use of HCS in the marketplace?

Dunlop: Something I see coming up often – better validation of reagents for HCS and better ways to use images across different platforms – some progress here but could more faster.

Ainscow: As described above, the key limitation is development of the informatics infrastructure, which is archiving and visualisation of the images. Development of analysis of multiparametric compound responses across looking at time series, compound dose responses etc. are needed.

Vendor view

Q1. From your point of view, what do you see as the most exciting application of HCS?

Gibb: Cancer research; endothelial cell tube formation as a model for angiogenesis. The application provides the ability to monitor the state of angiogenic outgrowths. These allow the flow of oxygenated blood throughout the body and, importantly, to newly formed cancers. Imaging is the tool of choice for making quantitative measurements in the amount of vascularisation with matrigel-like substrates.

Pagliaro: Rather than a specific application, the most exciting development is the increasing availability of a range of vendor products, services and expertise that did not exist even a few years ago. At Thermo Fisher we have assembled a breadth of ‘complete solutions’ that are focused on addressing customers’ needs in the best and most complete way possible.

Gradl: The most interesting applications are the ones that cannot be addressed with other methods than imaging. One of the most exciting here is to use stem cell differentiation assays for development of drugs for treatment of, for example, stroke or cardiac infarction.

Collins: Almost all of drug discovery can benefit from HCS, the drive to understand disease requires more knowledge about targets and their potential mediators and HCS’ benefits are providing this today. A recent analysis of Phase II to Phase III failures showed that up to 60% of compounds failed because they were not effective. The efficacy shortfall was shown to be a lack of understanding of how the novel mechanism worked in the disease process. HCS, we believe, can have a significant impact in driving that understanding.

“Exciting” really equates to the areas where HCS’ unique capabilities can be best used – for example: Neuroscience. Much of neuroscience involves screening morphological phenotypes – this is a great match for HCS and scientists have been using Cellomics products for several years to measure neuronal phenotypes in cultured cells, primary cells as well as tissues. The breadth of information that can be obtained we believe is set to accelerate the pace of neuroscience research and offer the hope of new therapies for many debilitating diseases.

Q2. How do you see HCS evolving within the industry?

Gibb: There has been a big shift over the last few years in this type of Analysis, moving from the traditional large Pharma/Biotech to the large academic institutes all over the world. With the introduction of the newer, faster and less expensive HCS systems, this advanced technology capability is available to almost anyone.

Pagliaro: It is (finally) becoming more mainstream as tools/technologies become more commoditised (I see this as a good thing, not a negative!) and as the operational tractability (ability to do lead optimisation with HCA, progression to clinic of HCA-discovered compound classes, etc.) of the high content discovery process becomes clear to end users.

Gradl: All major players in the pharmaceutical industry use HCS in their research today. The days of evaluating feasibility for finding hits in cellular HCS screens are over and it is used heavily in target identification, secondary screening and early ADME/T today. Instrumentation has matured to address primary screening. The hope is to find better drugs earlier, especially for cancer, inflammation and neurodegenerative disorders.

Collins: We consider that it will become a standard technique. Every scientist will have access to ‘cellomic’ data in much the same way as genomic data is today. HCS, which started out in secondary screening, will proliferate throughout the drug discovery process as well as into early clinical trials (through Biomarkers).

Q3. What do you consider to be the most ground-breaking technologies in the market place and how do you see the market place changing over the next few years?

Gibb: The most ground breaking is not one single technology but how companies, like BD, are bringing together plates/reagents/hardware/software and providing comprehensive solutions. This is critical, as the list of ever-expanding applications is being developed on HCS platforms.

Pagliaro: On the reagent side, the key developments are in the rapidly growing breadth and depth of reagent offerings. Here again, Thermo Fisher has lead the way by assembling a range of business units that cover the spectrum of needs high content users have. There are antibody-based HitKits, dyes and antibody reagents (Cellomics and Pierce); RNAi reagents (Dharmacon); stable cell line-based redistribution assays (BioImage); microarray technology (Searchlight); growth media and sera (HyClone); nucleic acid amplification and storage (ABgene); plastic ware and multiwell plates (Nunc) and much more. Rather than a single technology, it is the commercial availability of broad, high quality reagent offering such as this that will enable users to focus on biology rather than technology.

Gradl: Automation of imaging for HCS was driven by, for example, Nipkow spinning disc technology for confocal imaging, automated fast focussing and software for automated image analysis. On the reagent side the fluorescent proteins are most important for many HCS applications today, but more generic assay systems and reagents will be required in the future to broaden the horizon of HCS.

The market place will also need instrumentation at lower price points in order to make HCS accessible to a much larger number of labs.

Collins: We don’t consider there to be any ground–breaking technologies, per se, in the HCS arena today. Instrumentation, for example, is evolving rather than exhibiting ground breaking shifts. Image analysis has evolved, becoming more efficient and easier to use, but the so-called ground-breaking ‘black-box’ pattern recognition based software has not gained acceptance, probably because it is hard to validate and ignores the biological domain knowledge. Reagents and assay technologies have some rising stars, such as RNAi, new fluorophores as well as novel assay formats (cell patterning, assay substrates that mimic tissues for example). We see that HCS platforms that raise the value-performance ratio will become more widely accepted in the mainstream market. The future paradigm shifts in HCS may well come from systems that generate a high content readout, but don’t perhaps use traditional imaging to do that and nanotechnology may hold promise here.

Q4. What improvements/developments need to be made to existing products in order to further drive the use of HCS in the market place?

Gibb: I think mutli-level systems which can be tailored to meet the specific needs of the researcher/research focus will be the most successful. There needs to be a good compliment between technology/complexity and ease-of-use. And, of course, better flexibility in the software for the non-expert imaging user.

Pagliaro: On the reagent side, the key developments are in the rapidly growing breadth and depth of reagent offerings. This will allow customers to have a ‘one stop shop’ for a broad range of reagents for high content pathway biology. Importantly, we are starting to cross validate these integrated offerings and these joint offerings will increasingly be ready to run.

Gradl: HCS analysis instrumentation and software generally needs to make further improvement in ‘user-friendliness’ in order to assist application development. Significant developments still also have to be made in data analysis. High Content data require multi-level analysis and efficient management of data flow and storage. A major hurdle for widespread use of HCS is also data handling with respect to comparing and using HCS data, regardless of the hardware used. Scalable and seamless solutions are required on an enterprise level.

Collins: For HCS to be able to penetrate ubiquitously, ease of use and complete solutions that cover the workflow are very important. Ease-of-use really means making HCS approachable for the average biologist, rather than requiring them to have image analysis expertise. The challenge is to provide a software environment that is powerful and flexible enough for the expert, yet easy enough for the novice. Such software will have a lower learning curve and hence systems that have such software will be much more productive. The workflow of discovery and the workflow of HCS need to match up better and that is the reason to drive towards complete solutions that realise that.

Q5. In your opinion, what are the driving factors moving the industry forward and where do you think the major industry growth is going to be in the next few years?

Gibb: Driving factors moving this industry are saving lives, finding the most effective drug against certain diseases, like cancer. The most growth in this market area will more than likely be in the academic research centres.

Pagliaro: The most important single driving factor is (and will continue to be) rapid progression of projects to the clinic as a result of contributions made by HCA.

Gradl: In general, increasing demand of all life science research labs for automation in cell and tissue experimentation will drive further growth of HCS technologies over the next years. Major contribution is expected by, for example, authority approval of HCS technologies in safety assessment.

Collins: Attrition rate is clearly Pharma’s number one issue and this is a problem of productivity and better disease understanding. HCS is key to this better understanding and we believe that this will be a driver for more widespread adoption of HCS throughout the discovery process. R&D costs will certainly be monitored closely and so we see the need for platforms that are easier to use (to lower training costs and boost productivity) and with a good value-performance ratio, so more researchers can take advantage of them.

Q6. In your opinion, how has the uptake of HCS techniques in Europe compared to the USA?

Gibb: There is strong growth in the US but the European market is growing even faster. BD has been expanding the products, services and the technical support all over Europe to meet the needs of these users and ensure they will have success.

Pagliaro: From our perspective, the US market has emphasised services to a greater extent, while the EU market has emphasised products. I suspect these trends will even out over the next year or two.

Gradl: We have the impression that, with a few exceptions, the pharmaceutical industry in Europe was more careful in adopting HCS techniques. Today we see no major difference in the application of HCS in Europe and the US.

Collins: In the beginning, a little slower, but we have seen significant growth in the adoption of HCS in Europe and don’t expect that to change. As more and more academics use HCS, these are a source of growth too.