Generating clinical utility, increasing sensitivity, and improving scalability are just a few of the goals in the ongoing pursuit to improve assay development.
Philippe Goix, PhD, is president and chief executive officer at Singulex Inc. (Hayward, CA). Prior to joining Singulex,
he founded Guava Technologies Inc.
(Hayward, CA), where he held successive positions as president and chief
technology officer. He has held positions at CNRS in France, Sandia National Laboratories, and Stanford University, where he developed sophisticated laser-based diagnostic tools to investigate complex fluidics systems. He can be reached at firstname.lastname@example.org.
The movement from research and discoveries about a biomarker to putting an assay on a commercially viable platform can be a lengthy and intricate process. The development of a quality assay requires not only well-documented clinical data but also demonstrations of the reproducibility and cost-efficiency of producing the test.
In the enduring quest to build better, faster, and more- cost-effective assays, the importance of generating clinical utility cannot be overlooked. Translating clinical data and understanding protein interactions can, at times, be limited by technologies with poor sensitivity, requirements for large sample volumes, and a lack of high-throughput capability.
To find out how early-stage IVD companies tackle this challenge, IVD Technology editor Richard Park spoke with Philippe Goix, president and chief executive officer of Singulex Inc. (Hayward, CA). In this interview, Goix discusses the importance of clinical utility and assay value in the assay development process. He also talks about the advantages for IVD companies of building relationships with the pharmaceutical industry and academic researchers.
IVD Technology: What have been the biggest technological advances in assay development during the past few years?
Philippe Goix: The completion of the human genome project (HGP) has provided a tremendous amount of genetic information, and a lot of researchers are mining this information to discover new biomarkers. There has been a lot of progress in developing assays; however, these are more along the lines of disease-screening assays. They can only provide “yes” or “no” types of answers.
Utilizing HGP information in genomic assays has been the focal point of many diagnostic companies. These types of genomic assays have necessitated the development of many innovative tools and technologies that allow for easier access to genomic components, including both DNA and RNA. This effort has generated a wealth of information on the mechanisms that lead to disease onset and progression. It is unclear, however, whether genomic tests alone can provide the information that will be most valuable for clinical development and monitoring, which are necessary as clinical candidates move into patient trials. Because a single gene can trigger many different pathways within a cell and cells produce a vast number of proteins in response to these triggers, the most important information is found at the protein level. In other words, protein biomarkers can provide a wealth of information about the cell that cannot be found at the genomic level, which increases the utility of a potential proteomic assay significantly.
There should be a shift in protein assay development, not necessarily in the discovery part but in the validation part because you need to have assays that have very tight performance in terms of the coefficient valuations. And that is the challenge. You have a lot of valuable market targets, but there is a tremendous amount of effort that needs to be applied to generate robust assays that are clinically validated.
How do IVD companies go from discoveries about biomarkers and genetic information to a product with commercial viability?
It's a big challenge. There are several emerging technologies from the discovery of genes that are enabling the discovery of proteins, and from these comes the development of highly parallel immunoassays on microarrays.
That's one way of trying to move the focus of the assay from gene to protein and to develop assays from them. Also, analyzing a protein's spectral signature using mass spectrometry enables you to visualize a large number of proteins at the same time. And even if you don't know which protein is of interest, at least the assay enables the measurement of a large number of proteins, from which you are likely to uncover clinical applications within the sample. Developing a balanced system with a lot of proteins may generate assays with clinical importance faster than developing good old immunoassays one at a time, which can be very time-consuming.
This approach, however, has compromised several important things. High-throughput assays can compromise sensitivity, where significant clinical signal may reside, and well-known signals may go undetected by this type of platform. Additionally, in order to validate an assay and show clinical importance, the assay has to have tight coefficients of variation (CV) and reproducible results. It does not matter how many proteins the assay monitors, if the results do not have a tight CV, there can be no quantification, and quantification is the number-one goal of any assay.
Number two is sensitivity. There is a lot of interesting clinical action that is happening at very low levels. So you have to have sensitivity. And the two together have to be done on a very good sample. You don't necessarily need to analyze hundreds of proteins, but you could possibly get 5 or 10 protein immunoassays per sample.
So I think the answer is to develop immunoassay platforms in small volume with very high-level reproducibility without compromising the sensitivity.
Developing for the Pharmaceutical Industry
What are the latest trends in diagnostic assay development?
The latest trends are driven by the pharmaceutical industry because they are under huge pressure to develop assays and benchmark the efficacy or toxicity of their drug. There is a need to be able to develop technology, benchmark an assay, and assess normal levels.
Another trend is sample utilization. You want to monitor well-documented samples in clinical trials over a long period of time. You want to analyze as many markers as possible, and be able to reanalyze some samples in the future.
Also, when we talk about assay development for preclinical and clinical trials, an important factor is having a very reproducible platform with tight performance. More important than the limit of detection is the limit of quantification, and that requires very tight assay performance.
How has assay development been driven by the pharmaceutical industry, and how will pharmacogenomics continue to affect assay development?
Molecular diagnostics has affected assay development because of the cost value in the marketplace for new molecular tests. Historically, diagnostics has been a commodity business with assay valuation at a fixed and predictable price. With the advent of better, more informative assays, the paradigm is shifting and diagnostic assays are starting to be priced on the value of the assay. This shift is evidenced in the variation of price for molecular diagnostic tests, which can range into the thousands of dollars. This valuation increase for molecular diagnostic tests has allowed diagnostic companies to invest more expertise, effort, and money into developing assays with enhanced clinical utility. The drive to develop assays with greater utility and higher value is driving assay development now, as opposed to the development of better, faster, and cheaper assays.
What is the future of pharmacogenomics, and what challenges lie ahead for pharmacogenomics to become an integral part of healthcare?
The challenges lie in the areas of clinical utility, adding value, reducing costs, improving drug development decision making, and improving patient outcomes. All of these factors are going to require significant focus for every assay development company to demonstrate improvements in these areas. If there are more companies, especially in the academic space, that focus on developing a few assays that generate a lot of clinical utility instead of focusing on higher assay volume, that will secure the future for the field.
I'm very excited about this, because the way we are going to be monitored in 5–10 years will be very different from the way we are monitored now. Currently, we go to physical exams, we have the regular battery of tests, and we are treated today mostly by surgery or chemotherapy. Ten years from now, things are going to be very different.
Partnering with Technology Developers
What are the primary challenges that IVD manufacturers encounter in designing and developing a diagnostic assay product?
IVD manufacturers have been extremely good over the years at developing efficient diagnostic platforms that are completely automated, are very robust, can handle a large number of samples, and can deliver diagnostic answers with the lowest possible cost.
IVD manufacturers develop extensive platforms, and the problem is these platforms are not necessarily equipped to handle this new class of molecular diagnostics. Assays with higher added value would be useful; however, they would be applied on a smaller number of clinical samples or a similar population, and they are not yet validated on the market. They are going to take some time to become full scale, and they require a very fast development timeline.
So IVD manufacturers are well prepared to deliver a low-cost commodity. But that doesn't necessarily prepare them to handle the new wave of molecular assays that will generate the greatest value for the future.
How do IVD manufacturers overcome these challenges?
Manufacturers will have to partner with new, emerging technology companies that are more tuned in to what a pharmaceutical partner is doing. At Singulex, we do not consider ourselves IVD manufacturers. We are a technology developer and provider to the IVD market; not only the old market, but also the new, molecular diagnostic market.
Being closer to the drug development industry puts the IVD manufacturer in a better position to develop diagnostics with higher added value and a faster development timeline. In order to do this, the manufacturer should select an emerging company that will be able to scale up assay development and stay very close to the value chain by working with a pharmaceutical partner. Then the manufacturer can deploy the assay into the conventional diagnostic channels.
Can you describe Singulex's role as a provider to the IVD market and tell us a little bit about your latest technology, the Erenna bioassay system?
Singulex is expanding immunoassays to the next level. With the Erenna platform, Singulex provides a means to increase sensitivity, increase clinical sample utilization, and provide a scalable platform to validate immunoassays in a way that will generate information that could not be provided before. This information allows the measurement of drug effect or toxicity from normal levels. These measurements help monitor the departure of diseases and provide an early quantification of the departure of the disease in a very reliable way. Singulex is providing this capability to pharmaceutical companies and introducing it for preclinical and clinical trial validation.
We foresee tremendous opportunity with the Erenna platform, especially in the field of cardiotoxicity. Singulex is not changing the biology of the immunoassays. Singulex technology builds on existing biology, and our immunoassays rely on good antibodies and strong development protocols. Singulex has demonstrated over the last year and a half that any immunoassay can be enhanced in terms of sensitivity and dynamic range by the Erenna platform. We do not reinvent immunoassays. We build on them. We make immunoassays better and more reliable.
We also have a cardiac troponin I (cTnI) flagship assay that will be made available for cardiotoxicity studies. With this assay, Singulex can change the way people are using the cTnI information for acute myocardial infarction diagnosis. We make our assay more sensitive and we expand the dynamic range to measure and monitor the cTnI activity from normal to disease. We have been able to demonstrate that with our capability, we can indeed provide better dynamic range with our assay.
How has Singulex approached developing a cTnI assay that distinguishes it from all the other cTnI assays on the market?
Troponin I is a half-a-billion-dollar industry, and the major players have been working on this assay for a long period of time. Yet Singulex is taking this assay to the next stage; an unprecedented level of sensitivity. We are the first company to measure normal levels for this assay.
We started this assay basically from scratch about a year ago. We partnered with researchers at the University of California, San Francisco (UCSF), and we sourced the best antibody pair and applied our well-proven immunoassay development. We integrated this immunoassay development into our platform, the Erenna system, which caught every single antibody in the immunoassay. That's what makes it so sensitive, reproducible, and statistically sound. It also provides a level of quantification at the low range that's completely unprecedented.
Partnering with Academia
How do IVD companies view the relationship between academic researchers and private industry when developing diagnostic assays?
You have to stay close to where the expertise is. If you work with an academic medical school, laboratory, or hospital, you're very close to documented sample patients and outcomes and to clinical action.
You're also close to experts in the field that have worked with many different technologies and understand the challenges. Establishing and maintaining relationships with experts in the field helps not only to guide assay development but also to guide clinical trial monitoring and evaluation. Academia provides critical samples with comprehensive documentation and, working together, we can design the best possible assay to generate clinical utility. We crystallize the clinical work from academia and get it to a level where a diagnostic partner can look at it.
The validation platform and assay development capability that Singulex has and the scalability of this assay development can be a good catalytic piece to generate clinical utility and attract diagnostic partners for the long run and get them excited about the production of this assay or another assay.
So Singulex continues to work with key academic institutions on the clinical front to provide a means to better develop assays and validate assays on existing platforms or markers that have been developed in academia.
How did the relationship between Singulex and UCSF first emerge and evolve, and how has it contributed to your assay development efforts?
I met Alan Wu, PhD, professor of laboratory medicine, at a technical conference where he presented his work. I asked him about the challenges he was facing in assay development, and he mentioned that the main challenge to his troponin I work was sensitivity. One of Singulex's strengths is improving the sensitivity of conventional or unconventional immunoassays, and we decided to take the challenge, even without knowing yet what the market would be.
So we entered into collaboration with Dr. Wu and his research team at UCSF for the validation of the Singulex technology using clinical samples to demonstrate clinical utility. We looked at the well-documented clinical samples that he had, and we convinced him that we had a very robust technology and that our assay worked.
When he was convinced, we started exchanging samples, analyzing them, and producing results, and we developed a very productive collaboration. It was very dynamic, flexible, and rapid. It was also pretty demanding technical work. But we are excited we did it, we want to do more, and we are continuing to work with Dr. Wu's group.
What future challenges will emerge in diagnostic assay development?
There is a lot of pressure now from FDA with the critical path initiative to develop more discipline in biomarker assay development. In particular, there is difficulty developing clinical diagnostic guidelines into drug discovery and assay development.
It is also very difficult to predict the direction the regulatory environment is going to go. And so some flexibility is required by the key players in the IVD industry. The best way to get prepared is to follow the best product development practices, which are ISO driven.
Additionally, there are challenges with building up intellectual property (IP) landscapes. IVD companies have to be more diligent with these issues, especially with gene research and genomics. With this type of research, the IP is more complex and crowded, and this lends an advantage for the proteomics space, where there are fewer IP issues.
What new trends can we expect to see this year and in the future in the area of diagnostic assay development?
I see a move toward increased development on protein assays and more advances with sample utilization. Diagnostic product development requires a high-quality, well-documented clinical sample, and I think sample utilization is going to become increasingly important. Instead of focusing on discovery and pure analytical capability, I see an intensified focus on validation and clinical utility.