October 1, 2014 (Vol. 34, No. 17)

John H. Leamon, Ph.D. vice president CyVek

A System for Automated, Highly Sensitive Analysis of Multiple Biomarkers

Traditionally, protein biomarkers or analytes have been measured individually in ELISAs, which can attain a high degree of analytical specificity by testing only a single analyte with a dedicated antibody pair. However, clinical researchers have become increasingly aware that effective detection of complex, multimodal, multivariate diseases including sepsis, rheumatoid arthritis, cancer, neurodegenerative diseases, and traumatic brain injury often requires the analysis of multiple biomarkers.

Medical research has demonstrated instances where a single analyte provided little benefit, but monitoring a suite of multiple, multiplexed biomarkers has clinical value.  Unfortunately, adoption of multiplexed analytes in clinical research has been severely limited for many reasons, including technical concerns regarding assay reproducibility, decreased sensitivity or increased variability, and reported noncorrelation with conventional ELISA data.

Another important limitation of multiplexed technologies is that the detection of multiple biomarkers in a single sample requires a mixture of multiple capture and detection antibodies in the same reaction. This often results in cross-reactivity between noncompatible antibodies, where the antibody for a given analyte could react to, and detect, any number of additional antibodies. Poor quality data is generated, elevating concentration of some biomarkers, or even incorrectly detecting biomarkers that are not even present in the sample.

Multianalyte Capabilities

To address this issue, CyVek has developed an automated immunoassay platform (Figure 1A) that enables simultaneous multianalyte quantification. The sensitivity and specificity of single-analyte ELISAs are retained without the associated negatives of high sample volume requirements, slow reaction rates, high labor requirements, and cost inefficiencies typically encountered when analyzing multiple analytes.

The system enables simultaneous quantitation of four analytes from 16 individual samples in a single disposable microfluidic cartridge (Figure 1B) in under an hour. Each sample is analyzed in a unique circuit within the cartridge. The automated microfluidic system divides 20 µL of sample into four parallel channels; each channel contains the immunoassay for a specific analyte (Figure 1C).

As each assay is isolated in a discrete channel, this methodology provides the specificity of a traditional single-analyte ELISA sandwich immunoassay, as the sample is assayed by a singular antibody pair (capture and detect) rather than a cocktail of multiple antibodies. This eliminates potential negative interactions or interference from the antibody pairs for other assays, while simultaneously providing the improved efficiencies of a multiplexed antigen analysis and rapid microfluidic reaction kinetics.

Additional advantages include: (A) low, 20 µL sample volume requirement per 4-plex analysis; (B) lack of cross-reactivity between nontarget antibodies; and (C) validated assays in the immunoassay cartridge are functionally equivalent, or superior to validated ELISA assays.

Figure 1. (A) The CyPlex LS 1000 analyzer, an automated immunoassay platform. (B) The CyPlex 16-sample, 4-analyte cartridge. (C) The microfluidic infrastructure supporting the CyPlex 16-sample cartridge. (D) A sequential diagram of the automated steps that constitute the CyPlex assay workflow, shown for 1 of the 16 circuits on the cartridge.

Capture and Detection

CyVek’s disposable, automated, microfluidic cartridge combines ease of use with rapid, high-sensitivity multianalyte analysis. The hands-on effort required to operate the CyPlex system is limited to placing the cartridge in the instrument, pipetting 16 samples into individual sample wells, adding running buffer to the cartridge, and starting the assay software.

The remainder of the assay is fully automated; the cartridge’s computer controlled microfluidic pumps and infrastructure reconstitute the assay reagents stored on the cartridge, after which each sample is split into multiple parallel assay channels (Figure 1D).

Each assay channel contains three glass nanoreactors (GNRs)—hollow, cylindrical reaction chambers composed of fused silica, each functionally equivalent to an individual ELISA well. The chemical and optical properties of the GNRs are ideally suited for both chemical attachment of biomolecules and subsequent fluorescent detection.

CyVek has exploited these properties to immobilize antibodies on the internal surface of the GNR. The attached antibodies (capture antibodies, or cAb) bind the respective analyte from the sample as it flows through the GNRs.

As illustrated in Figure 1D, the sequence of automated steps that comprise the CyPlex assay is as follows: each individual circuit is primed with sample from the respective sample well. Twenty microliters of sample is then pumped through the circuit, microfluidically split into each of the four individual channels containing GNRs specific for a given analyte. The GNRs are briefly incubated in the sample before the remaining sample is washed from the circuit. In each channel, a biotinylated detect antibody (dAb) specific to the analyte in the channel is then pumped in to bind to the analyte captured on that channel’s GNRs.

Thus, if analyte A was captured by cAbs on GNRs located in channel 1, the dAb for analyte A would only flow into channel 1, not into channels 2, 3, or 4. Similarly, the dAbs specific for analytes B–D would flow only into those specific channels, not channels containing other proteins.

Segregation of both the cAbs and dAbs into analyte-specific channels prevents cross-reactivity between different assays. Unbound dAb is then washed from the circuit with a buffer wash, after which fluorescently labeled streptavidin is then flowed into all of the channels of the circuit. The fluorescent tag binds to the biotins on the dAbs, and the excess is removed by a subsequent wash.

Upon laser excitation, the bound dye in the GNRs produces a fluorescent signal proportional to the number of dAbs that, in turn, are proportional the number of bound analyte molecules. The CyPlex analysis software then uses a factory-generated, assay-specific, standard curve to convert the fluorescence intensity into protein concentration.

Enhanced Sensitivity

In addition to improving ease of use with rapid, automated assays, the CyPlex system provides superior assay performance relative to traditional ELISAs. The internal GNR volume is less than 1 nL, reducing diffusional distances compared to traditional 100 µL microplate-based assays, and greatly expediting reaction kinetics, resulting in highly sensitive assays. The kinetic nature of the CyPlex assay, where reagents are actively flowed through the capture substrate, as opposed to the static incubation used in microplate assays, can reduce nonspecific binding, lowering background levels and enabling improved lower limits of detection compared to traditional ELISAs.

This effect is demonstrated in Figure 2, where CyPlex cartridges containing multianalyte assays for IL-1β, IL-5, IL-10, and IL-12 were evaluated for sensitivity and dynamic range relative to commercially available singleplex ELISA assays (Figure 2). CyPlex assays were conducted in quadruplicate (four cartridges per data point), and the ELISAs were run in triplicate as per manufacturer’s instructions. Each of the four assays that constituted the CyPlex panel (IL-1β, IL-5, IL-10, and IL-12) outperformed commercially available ELISA kits with respect to sensitivity (more linear, low concentration points) and dynamic range, even though the commercial ELISA kits were run as a singleplex assay.

Figure 2. Comparison between CyPlex multianalyte assay and commercial singleplex ELISAs.

Next, the impact on assay sensitivity of pooled antibodies in a multiplexed assay as compared to the CyPlex assay format was explored (Figure 3). Standard curves on a multianalyte sample were run using a traditional CyPlex four analyte assay. The assay was then repeated with increasing numbers of detect antibodies pooled in the detect wells.

In this manner, the traditional CyPlex assay, which quantified IL-5 concentrations in the sample, used a homogeneous IL-5 detect antibody; the next assay used a mix of IL-5 and IL-1β; the third used a mixture of IL-5, IL-1β, and IL-10; and the last used a heterogeneous cocktail of IL-5, IL-1β, IL-10, and IL-1α.

Maximum sensitivity was achieved with a traditional CyPlex assay, where only a single detect species was employed. With each additional detect species added to the cocktail, the assay sensitivity decreased and the background signal from nonspecific binding increased. With four detect antibodies constituting the cocktail, the assay sensitivity had decreased approximately 10-fold relative to the results obtained with a single detect antibody.

CyVek’s technology provides high speed and accurate analysis of multiple analytes per sample while improving throughput and cost efficiencies by supporting analysis of 16 distinct samples on a single cartridge. With over 50 validated assays and highly customizable assay panels, the CyPlex system is a flexible solution for immunoassay users who require sensitive, reproducible analysis of multiple markers in low-volume samples. 

Figure 3. The impact on assay sensitivity of pooled antibodies in a multiplexed assay.

John H. Leamon, Ph.D. ([email protected]), is vice president of biology at CyVek

Previous articleFDA Issues Draft Guidance on LDTs
Next articleEnanta Regains Rights to Phase I HCV Candidate from Novartis