Real-time qPCR is commonly used to analyze cell-free fetal DNA in maternal serum and plasma for prenatal genetic screening, but this approach tends to detect very low fetal DNA concentrations — in the range of 1 percent to 3 percent early in pregnancy and 5 percent at term.
Last year, however, researchers at the Chinese University of Hong Kong led by Dennis Lo found that microfluidics-based digital PCR can detect concentrations of cell-free fetal DNA that are more than twice that amount. The study, which used Fluidigm's BioMark system, was published in Clinical Chemistry.
Recently, Sinuhe Hahn of University Women's Hospital in Basel, Switzerland, and several of his colleagues followed up on a "discrepancy" that they identified in the Lo paper. Namely, Hahn and co-authors wrote in a brief communication published in Clinical Chemistry last month, the investigators in the prior study "used amplicons of different lengths and targets" for the real-time and digital PCR assays that they compared.
Specifically, they note, the amplicon size was 87 bp for the digital PCR assay, but it was 137 bp for the real-time qPCR assay. "This feature might not have been relevant were it not for the observation that cell-free DNA is fragmented, probably into apoptotic nucleosomal fragments, and that fetal cell-free DNA fragments are generally smaller than those of maternal origin," Hahn and his colleagues wrote. They estimate that the majority of cell-free fetal DNA molecules may be shorter than 200 base pairs in length.
In order to determine the role of amplicon length in detecting cell-free fetal DNA with qPCR, they used a method based on a universal template for probe hybridization that is linked to the 5' end of one of the PCR primers. This allowed them to develop a new real-time qPCR assay with an amplicon size of only 50 base pairs for the DYS14 locus.
They found that the short-amplicon UT assay detected around 1.6-fold more cell-free fetal DNA than the real-time assay, indicating that it was the amplicon length — not the digital PCR itself — that led to the improved detection in the prior study. They acknowledge, however, that the issue "will need to be addressed further in a more detailed analysis."
PCR Insider spoke to Hahn this week about the work and its implications. An edited transcript of the interview follows.
Can you discuss the background for this work? Were you looking to compare real-time PCR to digital PCR for this application, or was it more of a study to assess the effects of amplicon length?
I don't think it's a comparison between the two methods. Digital PCR is much more precise, but there was an intriguing feature of the previous paper on digital PCR where they showed that they were able to detect more cell-free fetal DNA using the digital PCR method than the real-time PCR method, and that feature was that the PCR amplicons — the size of the PCR products they were amplifying — was smaller for the digital PCR than for the real-time PCR procedure they were using. So for the two procedures, they were using different sized amplicons.
And because we've previously shown that cell-free fetal DNA is highly fragmented and actually consists of very small fragments, we were intrigued to know whether this was due to the digital PCR, or if it was just due to the use of smaller amplicons.
So we repeated the experiment then using real-time PCR, and we showed that if you use small amplicons you do indeed detect more fetal DNA. So that suggests to us that the fetal DNA is highly fragmented and probably of a very small size.
What are the implications of that finding in terms of real-world research?
I don't think it's that major. It's not going to change overnight what's happening in the field, but I think the interesting thing for us is that if you use small amplicons you can detect more DNA. In the clinical sense, if you can detect more, it usually improves your sensitivity and specificity. So we'll be using this approach now for some other loci and see if we can improve the clinical outcome, or the accuracy of the test.
[ pagebreak ]
I think also for people who are using real-time PCR for other fetal loci, such as the mutations in hemoglobinopathies or cystic fibrosis. I think it might be useful for them to switch to the method that we described because I think that will yield more optimal results.
And of course if you detect more, that also implies that early in pregnancy, when you want to do these tests, and where the amount of fetal DNA is very low, maybe by using this procedure you'll be able to detect sufficient quantities early in pregnancy to make a diagnosis.
Are there any particular disadvantages to the digital PCR approach that would make this real-time PCR approach with shorter amplicons a better choice?
The problem with digital PCR is that at the moment the machines are still very expensive — we're talking a couple of hundred thousand dollars — and the microfluidic devices that you use for the analysis are also quite expensive. They're a couple of hundred dollars apiece. It's nice for research if you've got a healthy budget, but if you're doing routine diagnostics, you really can't afford to spend that type of money.
So I think in the coming years, as digital PCR comes of age, you'll see the prices plummet. Because as soon as you go into the market and start producing the quantities that the market wants, the prices will come down just due to scalability.
So it's not that this method is better that than method. Digital PCR is a very powerful tool, and as soon as it becomes affordable it's going to be a very useful tool — both for research and clinically.
But in the meantime, you've shown that it's possible to get the same results with real-time PCR.
The same results using this method, yeah.
Does this have implications in terms of other digital PCR studies, and whether any improvements researchers are seeing is also due to amplicon length?
There's only one study out with digital PCR for quantification, which is the report from Dennis Lo's lab, so it will be interesting to see when other people come out now with digital PCR processes and see what their results are like, and also to see what PCR strategy they use. So I think it's too early to say, really.
Are you looking into some of the other methods under development for detecting fetal DNA, such as mass spectrometry-based or sequencing-based approaches?
We've used the mass spec. It's a very handy tool, but it's also got the downside that it's very expensive. We found that it's not as quantitative as the other methods are, so I think if you want to do quantitation, then you're going to have to go to digital PCR. The sequencing approaches — these deep sequencing approaches — are probably the most promising, but there you have the downside that it's really expensive and it's very labor intensive, so just the reading of one result takes more than a day. So that really needs to change.
How are you using the assay you described in the paper in practice?
We're expanding it now to look at some different loci, and then we'll do a bigger study and see if it's useful for the clinic. But at the moment it's still in the exploratory phase; it's not in the diagnostic phase at all. We're a research lab, we're not a diagnostic lab, so we would have to hand it over to a diagnostic partner.
Do you have plans to commercialize this method?
Not at the moment. We'll see how it develops in these ongoing studies. This was just a very short paper and it's a limited number of cases, so we would really like to validate this in a much larger study with different markers and see how that goes and then talk about commercialization.
Are there other applications for the UT assay approach? For example, are there other cases where this short amplicon size would be desirable?
It might be useful for people studying cell-free DNA in variety of other diseases or concerns, such as cancer or transplantation. Wherever people use cell-free DNA this might be a useful tool for that as well. One would have to check whether it's really useful in those conditions.
Are you looking to apply it to any different areas?
At the moment, no. Our focus is really pretty prenatal so we'll be sticking to that at the moment. You can't chase too many rabbits.
What I would like to do now is get some other partners. We had a European consortium that just finished at the end of last year and I'd like to see if I can persuade a couple of the partners to also try the approach and do a multi-center study. Because that's always the best — if lab A, B, C, and D get the same result, then it means that it's not just the Basel lab's strange PCR.