Ultra-Sensitive Immunoassays: Origins, Methods, and Applications

An immunoassay is a biochemical test that detects a target molecule, or analyte, through the specific binding of an antibody. This binding is converted into a measurable signal, such as color change, fluorescence, light emission, or electrochemiluminescence.

However, when an analyte is present at very low concentrations, the assay may record too little signal, making positive and negative samples indistinguishable from one another. Because of this issue, researchers have long tried to increase the sensitivity of these tests, so they can detect low-abundance analytes and accurately compare the quantity of target molecules between samples.

These efforts led to the development of ultra-sensitive immunoassays.

In this article, we explain what ultra-sensitive immunoassays are, how they were discovered, the different types that exist, and what they are used for, so you know what is available and which to choose depending on your needs.

Summary:

• Ultra-sensitive immunoassays are antibody-based tests that detect target molecules at sub-picogram per milliliter (pg/mL) levels or below, improving on standard ELISA sensitivity.
• Two main approaches are used to reach ultra-sensitivity: signal amplification methods that generate more signal per binding event while keeping your existing workflow, or detection platforms that use different instruments to count single molecules or reduce background noise, but require dedicated, expensive instruments.
• Choose signal amplification when your validated ELISA/immunoassay works but needs 10–100× better sensitivity; choose an instrument platform when you need single-molecule detection, multiplexing, or are working with very small sample volumes.

The origin of immunoassays

In the 1950s, Rosalyn Yalow and Solomon Berson discovered anti-insulin antibodies in insulin-treated diabetic patients. They went on to publish a paper describing the first radioimmunoassay (RIA), one of the earliest immunoassays, which became widely adopted¹. In an RIA, radiolabeled antigen competes with unlabeled antigen in the sample for binding to a known quantity of antibodies. The total measured radioactivity is then inversely proportional to the sample’s antigen quantity. Yalow earned the Nobel Prize in Physiology or Medicine in 1977 for her discoveries.

RIAs worked well in terms of sensitivity, but, as with many other molecular techniques, radioactivity was not the ideal signal for immunoassays due to safety, costs, and isotope half-lives. In 1971, Eva Engvall and Peter Perlmann published a paper describing the first Enzyme-Linked Immunosorbent Assay, or ELISA: an immunoassay where an antibody or antigen is attached to a solid surface².

Since then, ELISA has become the most common immunoassay, due to its mix of sensitivity, usability, and reliability in laboratory settings. But there are others, such as lateral flow immunoassays, which became common during the COVID-19 pandemic. However, despite its benefits and improvements since 1971, ELISA sensitivity—and that of other standard immunoassays—is still not good enough for certain low-abundance biomarkers.

What is ultra-sensitivity?

Standard immunoassay and ELISA sensitivity can be as low as pg/mL detection limits³. Therefore, ultra-sensitive immunoassays are described as those reaching limits of detection below standard immunoassays, typically below pg/mL or lower ranges. Ultra-sensitive immunoassays rely on multiple methods to improve sensitivity, such as generating more signal, reducing background noise, or using completely different detection technologies.

In the 1980s, several approaches to reduce background noise and increase signal were published, including catalyzed reporter deposition, better known today as tyramide signal amplification, or TSA.⁴

Since then, multiple new strategies have been used to develop ultra-sensitive immunoassays. For practical purposes, methods can be classified into two groups, depending on how they reach ultra-sensitivity: signal amplification methods or signal detection technologies.

Signal amplification methods

Signal amplification methods work by generating more signal per binding event, retaining the capture and detection parts of the original immunoassay. This makes signal amplification the most straightforward way to reach ultra-sensitivity.

Signal amplification methods provide significant sensitivity gains, often around 2–200×. However, they may retain some of the limitations of the original immunoassay, such as high background noise.

Some examples are:

Tyramide signal amplification (TSA)

TSA uses horseradish peroxidase (HRP) attached to a detection antibody. When in the presence of hydrogen peroxide, HRP activates tyramide molecules labeled with fluorophores or other reporters. The activated tyramides are reactive radicals that covalently bind to nearby proteins (antibodies, antigen, others), resulting in many deposited reporter molecules around each single antibody-antigen complex. This produces a more concentrated signal than conventional immunoassays using enzymatic detection, which can make an ELISA or other immunoassay ultra-sensitive.⁴

Poly-HRP conjugates

Standard assays using HRP rely on a few enzymes per antibody. Poly-HRP is a commercially available reagent, in which multiple HRP molecules are attached to a single streptavidin or antibody. With this conjugate replacing a single HRP molecule, each captured antigen recruits multiple enzyme molecules instead of one. This results in more substrate turnover per antibody-antigen binding, increasing the total signal.⁵

Binding Oligo Ladder Detection (BOLD)

Instead of an enzyme, detector antibodies are conjugated or coupled to oligo-dT primers. Using poly-A RNA as a template, a polymerase with reverse transcriptase activity adds BrdU to the primers, creating a DNA:RNA hybrid. Then, anti-BrdU antibodies coupled to an enzyme or reporter molecule of choice can bind the nucleic acid products, with multiple binding sites available per original detector antibody. BOLD has been successfully used to detect multiple targets at lower concentrations than standard immunoassays, and can provide ultra-sensitivity to most types of immunoassays, including ELISA and other ligand-binding formats.

While many other signal amplification methods have been published (using nanoparticles, enzyme cascades, nanozymes, and others), most are not commercially available and are thus harder to implement from scratch^6,7,8.

Signal detection technologies

Standard immunoassays use enzymes, fluorophores, or radioactive labels attached to the detector antibodies to generate the signal measured by the reader⁹. To overcome the sensitivity limitations of standard immunoassays, different technologies have been developed based on alternative detection rationales. These achieve a higher sensitivity but typically require dedicated instruments.

An important example dates back to 1992, when three researchers published the lowest LOD ever reported at the time for an immunoassay: 580 antigen molecules. This was done through immuno-PCR, an ultra-sensitive immunoassay that relies on attaching DNA markers to detector antibodies and then amplifying the DNA using PCR¹⁰. The final DNA amplification products are then detected through gel electrophoresis (alternatively, qPCR can be used for amplification and detection).

However, immuno-PCR can also introduce background signal and additional workflow steps, which limits its usefulness in some settings.

New breakthroughs in 2007 and 2010 improved the readout rather than the signal, leading to single-molecule counting and digital ELISA, methods that allow researchers to detect single immunocomplexes or detection reagents^11,12. Microfluidics and automation have also allowed scientists to reduce error, increase reproducibility, and in some cases achieve ultra-sensitivity⁷. The nucleic acid amplification problem was refined further using dual antibody binding to reduce background signal issues¹³.

Some of the best-known signal detection technologies used to achieve ultra-sensitivity in immunoassays are summarized below:

Digital ELISA

This bead-based ELISA system captures target molecules on beads, and then distributes said beads randomly across hundreds of thousands of femtoliter wells. The small volume of the wells allows the detector to identify which wells contain even a single enzymatic reaction. Using Poisson statistics and knowing which wells are positive and which are negative, users can determine the analyte concentration in a single sample.

Single-molecule counting, or SMC

Here, a bead- or plate-based sandwich ELISA is performed, but for the readout, the fluorescent reporter is eluted into a solution. The volume is passed through a capillary system where a focused laser illuminates the sample. Each fluorescent molecule or complex is counted as it passes through. Multiple systems and instruments have been developed around SMC.

Proximity DNA methods using dual antibody detection

Multiple approaches focus on solving the same fundamental problem of immuno-PCR: the non-specific background signal from PCR amplifying DNA or primers that don’t come from antibody-target binding. The most common solution is using two antibodies that bind to the same target, each with different DNA strands. Only when both antibodies bind the same target are the DNA strands close enough that ligation or polymerase extension can create an amplifiable DNA molecule. Then, qPCR or NGS can be used to amplify and read the assay. Note that multiple technologies have been commercialized around this principle, each with additional steps and different workflows. Regardless, the dual antibody target recognition works similarly to solve the background signal problem.

Microfluidics and automation

Microfluidic ELISA platforms and automation tend to use sealed, pre-loaded, single-use cartridges and automated processing to reduce variability and background signal through controlled capture antibody-analyte exposure times, and standardized incubations. Some of these platforms achieve detections of sub pg/mL of analyte, reaching ultra-sensitivity.

Applications of ultra-sensitive immunoassays

Ultra-sensitivity was developed to provide sensitivity for applications where the sensitivity of standard ELISAs or other immunoassays was not enough. Because of that, most developed ultra-sensitive methods were tested in applications with low-analyte concentrations. Below are some examples of ultra-sensitive methods that have been used and their research and diagnostic applications:

Cytokines and immunology

Proximity Ligation Assay (PLA), published by Fredriksson et al. in 2002, successfully detected zeptomole levels of platelet-derived growth factor (PDGF), a signaling protein and cytokine¹³. Cytokines and other immune response molecules are often present at extremely low levels in the blood of healthy individuals, and when doing research in cell culture, organoids, or small animals, sample volumes are often too low to accurately quantify and compare cytokine levels with standard immunoassays. Ultra-sensitive immunoassays add most value in immunology when small differences close to the limit of detection or quantification matter, or when interfering factors in serum and culture medium force a dilution that would push the analyte below the detection limit of a standard ELISA.

Cardiac injury biomarkers

Single-molecule counting was originally tested using different biomarkers, including cardiac troponin I (cTnI), a marker of myocardial injury¹¹. Ultra-sensitive immunoassays can detect smaller cTnI increases earlier than conventional assays, which, in diagnostic settings, may allow faster evaluation of suspected acute coronary syndrome. BOLD was also successfully implemented in a cTnI ELISA for research purposes.

Infectious diseases

BOLD was first developed attempting to detect the SARS-CoV-2 virus. Often, infectious diseases can be transmitted before symptoms show, but protein levels in fluids may be below standard immunoassay sensitivity before symptom onset. A recent paper featured in SLAS shows how BOLD improves the sensitivity of an ELISA against SARS-CoV-2 spike protein¹⁴. Sensitivity for pathogen detection is also an issue in chronic infection, where hepatitis B is a clear example. Tyramide signal amplification combined with digital ELISA has reached ultra-sensitive levels for the surface antigen in patients with very low viral protein loads, showing that amplification and single-molecule readout can be used together rather than as alternatives¹⁵.

Cancer monitoring

The first paper behind digital ELISA measured sub-femtomolar concentrations of prostate-specific antigen (PSA)¹². PSA is a biomarker of prostate abnormalities, including cancer and other conditions. PSA is used to monitor recurrence of prostate cancer after treatment, but its concentration in a sample depends on tumor size and tissue damage, so early detection requires better sensitivity. Dual antibody proximity DNA methods have also been used to monitor cancer-related protein presence in blood¹³.

Neurological disorders

Detecting neurological biomarkers of disease is extremely hard in some cases, as only very low quantities make it to the bloodstream. Digital ELISA has been used to detect phosphorylated Tau 217 and other markers in serum, supporting Alzheimer’s disease research alongside other methods such as PET scans¹⁷. Real-time immuno-PCR has also been used to detect ultra-low levels of prions in research settings¹⁸. BOLD has also been successfully implemented in neuroscience applications.

When to go ultra-sensitive

Sometimes, your immunoassay’s sensitivity is not enough to detect or quantify what you are after. It’s important to understand when it is a good idea to go for an ultra-sensitive immunoassay, and what option you should choose depending on your needs.

Immunoassay sensitivity and standard numbers

There are three parameters to consider when moving towards an ultra-sensitive immunoassay:

1. Limit of detection, or LOD: Often calculated as the mean blank signal + 2–3× standard deviation. Readouts above this value are considered positive.
2. Lower limit of quantification, or LLOQ: Usually calculated from standard curves, LLOQ is the lowest concentration of an analyte that can be accurately quantified with acceptable accuracy and precision.
3. Dynamic range is the interval between the LLOQ and the upper limit of quantification, where the calibration curve is steep enough to resolve concentrations and a diluted sample still returns the correct result (dilution linearity).

Immunoassay sensitivity is often reported as the LOD, commonly at pg/mL of target. But ultra-sensitivity changes the LOD, LLOQ, and dynamic range, while the final sensitivity will still depend on antibody pairs, sample, and workflow.

Which ultra-sensitive method should you choose?

The straightforward answer is to choose the least complex method that can quantify your analyte at the required concentrations. To make a good decision, first establish the LOD, LLOQ, and dynamic range coverage you require. If a standard ELISA (or your standard immunoassay of choice) does not achieve the levels you need, look at ultra-sensitive options.

Signal amplification add-ons like BOLD are the easiest choice, as they improve sensitivity without the need for capital investment in new detection platforms and instruments. For many analytes, extremely low LODs are not needed. If your immunoassay is already working well, signal amplification can easily add 10–100× more sensitivity to it. TSA and Poly-HRP work well for standard ELISAs, but BOLD can be used with ELISAs as well as other immunoassays and reporter molecules and tends to achieve more sensitivity gains.

Signal detection platforms can be good options when sample volume is scarce, you require multiplexing, or the analyte is present at extremely low quantities. These often require changing the existing immunoassay, with different antibodies, readout systems, and workflows. Your choice of platform depends on sensitivity, multiplexing, sample volume, precision, dynamic range, throughput, validation requirements, and total cost. If you have one-time needs, you can often find laboratory service companies such as CROs that will run your tests on these systems, but if your need is long-term, owning the instruments may give you more assay flexibility and lower costs.

About BOLD and Cavidi

BOLD is the signal amplification system behind Exazym®, Cavidi’s commercial RUO reagent line that provides ultra-sensitivity to immunoassays as a simple add-on. Exazym^® was commercialized in 2024 as an easy-to-use system that can be used with ELISAs and other immunoassays such as Luminex and Gyrolab immunoassays.

Exazym® provides reagents to add oligo-dT primer to existing detection antibodies, the modified nucleotide, BrdUTP, and the polymerase needed to amplify the primers, and multiple tertiary antibodies with different reporter systems (biotin, HRP, APC, and PE) that recognize or bind to the polymerized DNA:RNA hybrid strand.

This makes BOLD the most flexible and simplest commercial signal amplification system to use for making existing immunoassays ultra-sensitive and achieving a 10–100× lower LOD.

Conclusions

The existence of multiple ultra-sensitive immunoassay approaches offers you different possibilities when considering your sensitivity needs. While ultra-sensitive immunoassays add complexity and costs to assays, for many research and diagnostic applications, the benefits outweigh the costs.

Whether you look for signal amplification add-ons like BOLD or TSA for their flexibility or settle for signal detection systems, it is clear that new advances will push the field forward so researchers can detect and quantify what was not possible before.

Frequently Asked Questions (FAQs)

1. What is an ultra-sensitive immunoassay?

An ultra-sensitive immunoassay is an immunoassay capable of detecting target molecules below standard sensitivity, usually at sub pg/mL concentrations or lower. This is beyond standard ELISA sensitivity, which typically is in the pg/mL or ng/mL range. The category includes signal amplification add-ons such as BOLD, Poly-HRP, or TSA, and instrument-based platforms with dedicated detection technologies such as digital ELISA and single-molecule counting systems.

2. What is the typical sensitivity of an ELISA?

A typical standard sandwich ELISA’s sensitivity is between 1 pg/mL and 100 ng/mL. The exact sensitivity varies depending on the antibody pair, sample, and workflow. Some variants can reach around 0.1 pg/mL. Ultra-sensitive methods push detection lower, into fg/mL or below.

3. How can I improve my ELISA/immunoassay sensitivity?

Optimize your assay when possible or use ultra-sensitive methods. Most commercial ELISA kits have already been optimized for sensitivity by the manufacturer. However, if you are designing the assay on your own, start by optimizing the antibody pair, blocking, buffers, and incubation conditions. In either case, if the sensitivity is not enough, add a signal amplification step. BOLD works as an add-on to ELISAs and other immunoassays and lowers the limit of detection by 10–100× without requiring new equipment.

4. What is the difference between LOD and LLOQ?

The limit of detection (LOD) is the lowest signal distinguishable from a blank, while the lower limit of quantification (LLOQ) is the lowest concentration the assay can quantify accurately with precision. An analyte can be detected by showing values above the LOD while the readout may still be below the LLOQ, meaning the target is present in the sample but not accurately quantifiable.

5. When should I use an ultra-sensitive immunoassay instead of a standard ELISA?

Try an ultra-sensitive immunoassay when you need to detect or quantify an analyte in samples you currently can’t. Some cases include working with low-volume samples, earlier detection of biomarkers, or when comparing sick and healthy individuals. Typical analytes for which ultra-sensitivity is beneficial include low-abundance cytokines and immune system markers, injury or disease biomarkers, and neurological proteins in blood.

References

1. Yalow, R.S., Berson, S.A. Immunoassay of Endogenous Plasma Insulin in Man. 1960. Journal of Clinical Investigation, 39(7), 1157–1175. https://doi.org/10.1172/JCI104130
2. Engvall, E., Perlmann, P. Enzyme-Linked Immunosorbent Assay (ELISA). Quantitative Assay of Immunoglobulin G. 1971. Immunochemistry, 8(9), 871–874. https://doi.org/10.1016/0019-2791(71)90454-X
3. Krzysica, P., Verhoog, L., et al. Optimization of Capture ELISAs for Chicken Cytokines Using Commercially Available Antibodies. 2022. Animals, 12(21), 3040. https://www.mdpi.com/2076-2615/12/21/3040
4. Bobrow, M.N., Harris, T.D., et al. Catalyzed Reporter Deposition, a Novel Method of Signal Amplification: Application to Immunoassays. 1989. Journal of Immunological Methods, 125(1–2), 279–285. https://doi.org/10.1016/0022-1759(89)90104-X
5. Li, D., Cui, Y., et al. Nanobody Based Immunoassay for Human Soluble Epoxide Hydrolase Detection Using Polymeric Horseradish Peroxidase (PolyHRP) for Signal Enhancement: The Rediscovery of PolyHRP? 2017. Analytical Chemistry, 89(11), 6248–6256. https://doi.org/10.1021/acs.analchem.7b01247. PMCID: PMC5611449.
6. Yang, H., Xu, W., et al. Signal Amplification in Immunoassays by Using Noble Metal Nanoparticles: A Review. 2019. Microchimica Acta, 186(12), Article 859. https://doi.org/10.1007/s00604-019-3904-9
7. Ai, J., Yan, L., et al. Ultrasensitive Cascade Enzyme Immunoassay Electrochemical Microfluidic Chip for Early Screening of Prostate Cancer. 2026. ACS Sensors, 11(4), 2967–2978. https://doi.org/10.1021/acssensors.5c02259
8. Hu, X., Hu, R., et al. Nanozyme-Based Cascade SPR Signal Amplification for Immunosensing of Nitrated Alpha-Synuclein. 2022. Microchimica Acta, 189, Article 367. https://doi.org/10.1007/s00604-022-05465-5
9. Darwish, I.A. Immunoassay Methods and Their Applications in Pharmaceutical Analysis: Basic Methodology and Recent Advances. 2006. International Journal of Biomedical Science, 2(3), 217–235. https://pmc.ncbi.nlm.nih.gov/articles/PMC3614608
10. Sano, T., Smith, C.L., et al. Immuno-PCR: Very Sensitive Antigen Detection by Means of Specific Antibody-DNA Conjugates. 1992. Science, 258(5079), 120–122. https://www.science.org/doi/10.1126/science.1439758
11. Todd, J., Freese, B., et al. Ultrasensitive Flow-Based Immunoassays Using Single-Molecule Counting. 2007. Clinical Chemistry, 53(11), 1990–1995. https://academic.oup.com/clinchem/article-abstract/53/11/1990/5627296?
12. Rissin, D.M., Kan, C.W., et al. Single-Molecule Enzyme-Linked Immunosorbent Assay Detects Serum Proteins at Subfemtomolar Concentrations. 2010. Nature Biotechnology, 28(6), 595–599. https://www.nature.com/articles/nbt.1641
13. Fredriksson, S., Gullberg, M., et al. Protein Detection Using Proximity-Dependent DNA Ligation Assays. 2002. Nature Biotechnology, 20(5), 473–477. https://www.nature.com/articles/nbt0502-473
14. Nordal, V., Hedin, J., et al. Reverse Transcriptase Mediated Signal Amplification in Ligand Binding Assays. 2026. SLAS Technology, 39, Article 100423. https://doi.org/10.1016/j.slast.2026.100423
15. Akama, K., Shirai, K., et al. Droplet-Free Digital Enzyme-Linked Immunosorbent Assay Based on a Tyramide Signal Amplification System. 2016. Analytical Chemistry, 88(14), 7123–7129. https://doi.org/10.1021/acs.analchem.6b01148
16. Papier, K., Atkins, J.R., et al. Identifying Proteomic Risk Factors for Cancer Using Prospective and Exome Analyses of 1463 Circulating Proteins and Risk of 19 Cancers in the UK Biobank. 2024. Nature Communications, 15, 4010. https://www.nature.com/articles/s41467-024-48017-6
17. Kubota, M., Bun, S., et al. Plasma Biomarkers for Early Detection of Alzheimer’s Disease: A Cross-Sectional Study in a Japanese Cohort. 2025. Alzheimer’s Research & Therapy, 17, 131. https://link.springer.com/article/10.1186/s13195-025-01778-8
18. Barletta, J.M., Edelman, D.C., et al. Detection of Ultra-Low Levels of Pathologic Prion Protein in Scrapie-Infected Hamster Brain Homogenates Using Real-Time Immuno-PCR. 2005. Journal of Virological Methods, 127(2), 154–164. https://doi.org/10.1016/j.jviromet.2005.04.007