We've updated our Privacy Policy to make it clearer how we use your personal data.

We use cookies to provide you with a better experience. You can read our Cookie Policy here.

Advertisement
Rectangle Image
Industry Insight

Will the Potential of Proteomics Be Realized in 2020?

Rectangle Image
Industry Insight

Will the Potential of Proteomics Be Realized in 2020?

Credit: National Cancer Institute on Unsplash.

Proteomics, the study of the proteome, has been anticipated to majorly advance research areas such as personalized medicine and drug discovery. But has it delivered on such promise yet? Is 2020 the year in which the potential of proteomics will be "realized"?

Technology Networks
recently spoke with Stephen Williams, Chief Medical Officer SomaLogic, to learn more about research progress in the proteomics space and the technology that has enabled it.

Molly Campbell (MC): In your opinion, will proteomics "realize" its potential in 2020? If so, why? What factors have prevented this from happening thus far?

Stephen Williams (SW):
There are two ways to look at the “potential” of proteomics.

In terms of the potential to discover new biology and drug targets in association with genetic changes, drug pharmacology and environmental impacts in disease, the application of highly multiplexed proteomics (measuring thousands of proteins and their concentration levels simultaneously) at scale is already extremely impactful.

In terms of its potential to deliver medically actionable information to patients and care-teams, we’ve only just started; the first 10 tests based on protein patterns associated with specific health states and disease risks identified using the SomaScan platform technology are commercially available only at limited scale today, and the full pipeline of hundreds of tests is still under development.

It is important to note that proteomics has been completely overshadowed by genomics. Until recently, we didn't have the technology to measure proteins at scale, whereas DNA sequencing has become both fast and relatively cheap. Plus, genomics was backed by big international initiatives with lots of funding. I do think that proteomics is gaining traction as people are beginning to acknowledge the limitations of genomics in delivering on the promise of precision medicine, and recognizing the need for more “real-time” measurements of health and disease.

MC: For our readers who may be unfamiliar with SomaLogic, please can you provide a bit of background information about the company and your aims?

SW:
The company was founded in Boulder, Colorado 20 years ago by Larry Gold, who had co-invented the method for developing aptamers (single strands of DNA or RNA) which bind specifically to individual proteins. While the rest of the scientific world was pursuing the genome to drive better health, Larry’s vision was to use these types of reagents to measure thousands of proteins at once in order to characterize health and disease. Over that period Larry doggedly persisted in advancing the science and continued to raise funding to pursue his vision.

It took 20 years and 600 million dollars, but SomaLogic now has the technology to accurately measure thousands of proteins in a single blood sample and translate those measurements into useful health information across a diverse set of diseases and conditions from each sample. Last year we started offering tests based on our SomaScan Platform technology through select practices in Colorado. Our goal has always been to provide actionable health information to help a person optimize their health and stay healthy for as long as possible.

MC: Please tell us about the development of your proprietary technology, the SomaScan® Platform?

SW:
It has taken 20 years to make a scalable, precise and relatively low cost assay that could measure thousands of proteins simultaneously in a high-throughput manner. After inventing the reagents, key developments that enabled multiplexing were:

  • Modifying the aptamers chemically to enhance their binding characteristics

  • Creating a second element of specificity in the assay. While the first element – high affinity binding for individual proteins – is important, it is not enough to overcome the challenge of measuring low abundance proteins in the presence of highly abundant proteins (eight logs of dynamic range in plasma). The second element of specificity is delivered through kinetics – the aptamers are selected for slow off-rates as well as high affinity. The timing difference between non-specific binding (fast off-rates) and specific binding is used to create the extra specificity

  • Creating an assay format that translates the protein molecule counting problem (which is difficult to do ) into a DNA counting problem (which is easy to do); the format of the assay is unique in having two “catches.” In the first catch, the immobilized aptamers catch the proteins of interest in complexes and unbound proteins are washed away. After the release of these complexes and the aptamers that bound no protein (and the application of the kinetic challenge to overcome non-specific binding a second time), the complexes are caught again and unbound aptamers are washed away. Thus, at the end of the assay, a “DNA soup” of aptamers, each present in proportion to the original individual protein concentration, is quantified using DNA measurement tools


We now have the largest human proteomics database in the world, and it was built by working with leading industry and academic partners to apply the SomaScan Platform to hundreds of thousands of samples coupled with clinical data.

In the early days, the focus was on developing an assay that could measure thousands of proteins simultaneously. This required creating reagents that could bind a given protein tightly with high specificity in complex biological mixtures, that could be produced synthetically and that could be multiplexed—a tall order that couldn’t be accomplished with traditional antibody reagents. While we were developing the assay, computing power and data analytics advanced sufficiently to enable probing "big data" sets. The combination of a robust, high-throughput assay and advanced computational methods is what allows us to discern protein patterns that are indicative of different health states and outcomes, all from a single sample of blood.

MC: How is the protein scanning technology optimized for clinical proteomics?

SW:
Our SomaScan Platform is well suited for redefining clinical proteomics because it can measure both broadly (thousands of proteins simultaneously) and deeply (high and low abundance proteins), in a small sample of a complex mixture such as blood, with high throughput. Our Platform also appears to be disease agnostic--we have tests for 10 different health conditions currently, but have more than a hundred tests in our pipeline. The tests we’re developing can be delivered in any combination or altogether from a single blood sample—the idea is to gain a holistic view of a person’s current health status and trajectory at any given point in time to guide health management throughout life.

Specific technical strengths include:

  • The assay is almost entirely automated and runs at scale on 96-well plates, and our capacity is around 200,000 samples a year and is easily expansible

  • The assay is almost entirely automated and runs at scale on 96-well plates, and our capacity is around 200,000 samples a year and is easily expansible

  • The precision of the assay is monitored and managed carefully; CVs (coefficient of variation, a measurement of assay variability from batch to batch) are ~5% median across all 5000 measurements, and calibration and quality control samples are included on every plate such that any long-term variance is quality controlled. The assay is run in our College of American Pathologists (CAP)/Clinical Laboratory Improvement Amendments of 1988 (CLIA) certified and inspected laboratory

  • Each test for medical use is subjected to a rigorous regulatory process prior to release under CAP/CLIA rules


MC: A recent
study demonstrated that the test is as good/ better than measurements currently used for determining diabetes and CVD risk. Can you tell us more about this study?

SW:
The research published in Nature Medicine was led by a team of scientists from the University of California, SF, Cambridge University in the UK and SomaLogic. It’s the largest proteomics study that’s ever been done, involving 85 million protein measurements of 5,000 different proteins from 17,000 participants.

This proof of concept study is the first to demonstrate that a person’s current health status and future risk of cardiometabolic disease including diabetes, coronary heart disease, stroke and heart failure, as well as the impact of modifiable behaviors on the development of those diseases, can be determined soley from patterns of proteins in the blood.

To identify protein patterns that predict future outcomes, we used clinical studies with stored samples from people where those outcomes are known (who got diabetes, who died, who was hospitalized for a heart attack, etc).

To ensure that the results could be reproduced, that the population studied was robust (including people from different geographies and ethnicities and are a match for those in which the test will be used), and that the sample size was large, we used five different cohort studies from around the world.

The results were then compared to the best available measurements currently used. For the CV primary test we used the most accepted comparator, the ACC/ASCVD risk score; for the diabetes predictor, there wasn’t an accepted score so we created the best combination of lab results in our study for comparison.

MC: In the aforementioned paper, the authors say "We anticipate that, with further validation and the addition of more protein-phenotype models, this approach could enable a single-source, individualized so-called liquid health check." What are the next steps in terms of developing the technology?

SW:
Since the paper was written, we have subjected these tests to more “hardening” of their performance and now 10 tests are commercially available in a limited release to doctors in Colorado, including:

  • Risk of primary CV event within four years

  • Risk of secondary CV event within four years

  • Liver fat

  • Body fat

  • Visceral fat

  • Lean body mass

  • Alcohol impact

  • Resting energy expenditure

  • Cardio-pulmonary fitness (VO2 max)

  • Impaired glucose tolerance


MC: How feasible is the delivery of a "liquid health check" into the clinical environment in the foreseeable future?

SW:
The limited release in Colorado will be expanded to additional US states in 2020 and beyond, fully compliant with the regulatory requirements of new locations.

We have seen initial enthusiastic responses from providers and users.

More tests (in particular, relevant to disease management in health systems) will be released in 2020 and beyond. There are more than 100 in our current pipeline, each one of which can be delivered from the same blood sample.

Stephen Williams, Chief Medical Officer, Somalogic, was speaking to Molly Campbell, Science Writer, Technology Networks.

Meet The Author
Molly Campbell
Molly Campbell
Science Writer
Advertisement