Developments in digital PCR: finding the needle in a haystack

Digital PCR (dPCR) is the third generation of PCR technology, after conventional PCR and real-time quantitative PCR. It offers significant benefits over these other methods, making it ideal for a wide range of applications, including nucleic acid quantification and rare mutation detection.

Here, we speak to Magali Droniou and Valérie Taly, both of METHYS Dx (Paris, France), to explore dPCR’s potential and find out how the technology has transformed life science research in recent years, as well as the challenges it faces and what the future has in store.

Valérie is CEO, Scientific Director and co-founder of METHYS Dx, which develops and commercializes products for monitoring cancer patients with liquid biopsies based on measurements of circulating tumor DNA with dPCR assays for patient-agnostic methylation markers. She is also a research director at the Centre National de la Recherche Scientifique and group leader at the Cordeliers Research Center (both Paris, France). Her research focuses on applying droplet-based microfluidics and dPCR in cancer research, particularly for exploring new biomarkers and enabling liquid biopsies.

Magali is Chief Operating Officer of METHYS Dx. She has a background in microbiology and virology and, in 2013, co-founded STILLA Technologies (Villejuif, France), a company developing and marketing dPCR solutions, which was acquired by Bio-Rad Laboratories (Watford, UK) in 2025.

How does dPCR work – what sets it apart from traditional PCR techniques?

Magali: dPCR is based on partitioning of the sample prior to amplification. This partitioning can be achieved through droplets, or solid partitions, for example on microwell devices. Typically, some partitions will have one copy of the DNA target of interest and many partitions will have no DNA molecules. After the sample is segregated into individual partitions, then each of the DNA fragments are amplified independently of each other.

After PCR amplification with reagents similar to those used for quantitative PCR (qPCR), the fluorescence signals of all individual partitions are analyzed and the partitions that are deemed positive, i.e. above a signal threshold, are counted, as well as the total number of partitions that were present in a reaction. The absolute quantity of each amplifiable target is determined by applying the Poisson distribution law.

What sets dPCR apart from other PCR techniques is, firstly, absolute quantification. There is no need to rely on calibration curves. The technology also allows high precision in the measurement. It’s renowned for its ability to discriminate low-prevalence variants – ‘to find the needle in a haystack’ – because when you segregate the different DNA molecules into individual partitions, then the target sequence is amplified independently (and without competition) from the vast array of similar sequences.

For example, if you are trying to detect a rare sequence containing a point mutation within a mixture containing many copies of the wild-type sequence, PCR amplification of the mutant will not be very efficient and you would have a bias in amplification. dPCR is the perfect tool for such an application.

Another great feature of dPCR is its tolerance to inhibitors. Since inhibitors and targets are often captured in separate partitions, dPCR enables scientists to extract useful data from challenging samples.

Last but not least, dPCR measurements are quantitative, absolute and therefore, highly reproducible between labs. Lack of reproducibility between different laboratories has been reported with other PCR techniques. If you can’t rely on a reproducible quantification, then what do your measurements actually mean, especially in cases of longitudinal monitoring?

Why is it such a vital life science tool?

Valérie: It’s a great tool in life science in general because of its high accuracy, high precision and high sensitivity. It’s also quantitative, so you can obtain a precise measurement of the molecule (i.e. nucleic acids) that you want to detect with very good intra- and inter-laboratory reproducibility.

As Magali pointed out, it’s a really powerful technology, because it is not sensitive to inhibitors, and this has really been instrumental in its adoption. It’s also very simple to perform; you basically just count molecules.

As a result, it has a lot of powerful applications. In oncology, the field I’ve worked in for more than a decade, and in medicine in general, its high sensitivity and reproducibility from a tool that anyone is able to use, has led to wide adoption both in research and routine hospital laboratories.

What are some advances in the technology in the last 5 or so years?

Magali: In the last 5 years, a lot of new instruments have come onto the market. The offer, both for droplet-based or well-based dPCR, is now quite extensive.

Many of the newer machines offer more fluorescence channels for increased multiplexing. There have also been some interesting new developments in fluorophores with longer Stokes shift for probes. Expansion of the number of fluorescence channels and reducing the interference of probe signals both increase the amount of multiplexing that is feasible, which means more markers can be detected from a single reaction.

The throughput capacities have also been increased and different types of microfluidic or partitioning chips mean that you can choose the optimal throughput, volume and number of partitions depending on your application. For example, if you are looking for low-abundance mutants, you may need to analyze a large volume, to make sure that you are able to detect the rare events, but in other instances, you may want to discriminate between two similar populations. In that case, you may need more partitions for accurate quantification.

Another notable development, I think, in the last 5 years, is that advances have not only been carried out on the machines or the consumables, etc. There have also been advances in reagent performance and assay development. For example, new PCR primer/probe systems enable faster development of assays and better discrimination between positive and negative populations.

Valérie: You can also now have access to all-in-one machines, where, with one machine, you input your sample and then receive the results, whereas before you had separate machines for different steps in the process. Such advances permit a significant increase in throughput but also reduce the hands-on time for your experiments. Also, this level of automation enables more and more applications for the clinical laboratory.

What are the main challenges dPCR faces today?

Magali: There are challenges in implementation for more cost-sensitive applications. Even if the price of machines has decreased and they’re more affordable, running the experiments may still be too costly for some applications.

Again, it depends on the application; an acceptable price in oncology may not be acceptable for the food industry or for environmental monitoring. We certainly hope to see the price of the consumables come down.

Another challenge in some dPCR experiments is robust and automated data analysis. There’s this phenomenon called ‘rain’, which are droplets of intermediate fluorescence between negative and positive populations that can be hard to classify. Are they positive? Are they negative? Sometimes misclassification has no impact on the result, or it’s negligible, and sometimes it is impactful. When working with multiplex dPCR assays, challenges also include: can you use robust automated clustering? Can you ensure that sparse measurements are correctly identified? These elements are crucial to standardization, which is key for clinical implementation, for example.

What does the future hold for dPCR?

Magali: We believe that the future is bright for dPCR. It is a powerful tool that definitely still has a lot of potential for growth and implementation in different applications and markets. It is useful on many levels and complements other techniques that are currently used.

The technology is very versatile. dPCR is well-established in research and translational settings and we will see applications continue to expand. Also, from our perspective at METHYS Dx, there are a lot of avenues for clinical implementation, which will truly reveal the power of dPCR.

I encourage anyone who hasn’t tried dPCR yet to give it some thought, it’s a wonderful technology!

Valérie: I agree, and I do look forward to new developments in dPCR capabilities. I see that some companies are proposing to enable real-time (qPCR) analysis within compartments, which is interesting. Some other systems will enable an increase in the sample volume for testing or a greater number of analyzed compartments, etc. Today, there are a lot of strong and high-level developments in the field – both in academic and industrial labs. Moreover, several well-organized initiatives are establishing guidelines for data analysis and reporting, which will further facilitate adoption of this technology.

Valérie is CEO, Scientific Director and co-founder of METHYS Dx, and Magali is the Chief Operating Officer of METHYS Dx, as well as the co-founder of STILLA Technologies.

The opinions expressed in this interview are those of the interviewees and do not necessarily reflect the views of BioTechniques or Taylor & Francis Group.