Digital PCR (dPCR) is an increasingly popular manifestation of PCR that offers a number of unique advantages when applied to preclinical research, particularly when used to detect rare mutations and in the precise quantification of nucleic acids. As is common with many new research methods, the application of dPCR to potential clinical scenarios is also being increasingly described.1
A recently published study in Analytical Chemistry set out to explore the clinical utility of dPCR. Twenty-one independent laboratories quantified a rare single nucleotide variant using digital PCR with high reproducibility between labs, demonstrating the method’s potential for routine clinical testing. The study was the first to ever look at the reproducibility and highlights several potential advantages of the technique in measuring clinically important mutations.
To find out more about dPCR and what this study means we spoke to Jim Huggett, Principal Scientist, LGC, Senior Lecturer, The University of Surrey.
What prompted LGC to conduct this study and what did you find?
This study was performed as part of the European Metrology Research Progamme funded project, BioSITrace, which was focused on developing (SI) traceable quantification of biological entities through the concept of enumeration. dPCR has the potential to be a highly reproducible method for the quantification of DNA and this study investigated this reproducibility by comparing the performance of 21 laboratories when quantifying DNA materials containing mutations associated with cancer. The study demonstrated that dPCR was indeed highly reproducible in the absence of a calibrator when measuring these sequences; the vast majority of laboratories got the same result. For the three that did not, this was clearly due to analysis of the results rather than the instrument.
What implications does this study have for the adoption of digital PCR in clinical testing?
These findings demonstrate that quantification using dPCR can be reproducible without calibration in this example. This makes it much simpler for laboratories to obtain similar results which are crucial for a clinical test to be applied across different laboratories on large numbers of patients; a major challenge when performing quantitative analysis using qPCR is calibration to ensure harmonisation between laboratories. This study demonstrates that dPCR could have an important role in advancing molecular diagnostics either by supporting qPCR calibration or as a diagnostic method in its own right. I suspect more advanced diagnostic methods demanding improved sensitivity or quantitative performance are going to be needed if the promise of precision medicine is to be maximised.
Does it provide all the evidence decision makers need?
While I am not a clinician it is my understanding that clinical decisions are rarely made on a single result. However, for such a methodology to be clinically useful in contributing to such a decision it must ideally be accurate, robust and reproducible. These findings contribute to a growing body of evidence from our group, and others, that suggest dPCR is a robust and reproducible approach that could lend itself to supporting accurate diagnostics in the future.
Looking into the future can you see any applications that digital PCR may be able to support in a way that qPCR has not been able to?
Yes, areas, where qPCR cannot currently compete, is in the detection of rare genetic variants in cell-free DNA, such as those analysed in this study. Another important question is whether next generation sequencing will eclipse other methods including dPCR. NGS allows you to measure many more mutations than PCR-based approaches and is already the method of choice for genotyping in many clinical laboratories. However, NGS is still under development when it comes to the need for quantification or for more sensitive analyses. Our study demonstrates that dPCR may be a robust alternative. dPCR is also likely to be a valuable tool for ensuring quantitative measurements associated with NGS are accurate even if the later becomes more established in this context.
Are there any domains that you believe will remain firmly entrenched with qPCR?
Given that changes in clinical testing can be compared to an oil tanker altering direction then in the short to medium term qPCR is certainly here to stay. It is relatively cheap, fast, has been developed in high throughput formats as well as point of care and, crucially, is automated, which was a major advantage when it came along. While qPCR and dPCR were invented at a similar time, it was qPCR that was extensively developed whereas dPCR had to wait over 10 years for the advances in microfluidics and emulsions chemistries that make it a practical approach. Had this not been the case then I suspect dPCR would be used far more routinely due to the characteristics demonstrated by our study; so I think dPCR has a lot to offer in advancing molecular diagnostics in the coming years.
Jim Huggett was speaking to Jack Rudd, Senior Editor for Technology Networks.
1. Huggett, J. F., Cowen, S., & Foy, C. A. (2015). Considerations for digital PCR as an accurate molecular diagnostic tool. Clinical chemistry, 61(1), 79-88.