NEW YORK (GenomeWeb) – Prenatal testing appears to be at the brink of another round of major innovation, with three recent reports demonstrating the feasibility of isolating fetal cells noninvasively and analyzing their DNA for aberrations, a goal researchers had pursued for decades.
Testing DNA from fetal cells instead of cell-free DNA from the mother's blood — the approach taken by current noninvasive prenatal tests, or NIPTs — has the advantage that the genetic material is unadulterated by maternal DNA, enabling the detection of smaller changes more sensitively and possibly earlier during pregnancy, according to the developers. However, challenges exist to recover enough fetal cells per sample, increase the throughput of fetal cell isolation, and keep down costs.
Several academic and commercial teams are competing at the moment to develop clinical fetal cell-based tests that could become available as early as next year.
In a study published today in Science Translational Medicine, researchers at Wayne State University School of Medicine showed that they could isolate trophoblast cells — which are fetal in origin — from Pap smears collected as early as five weeks of gestation and profile their DNA by targeted next-generation sequencing. PerkinElmer has taken an exclusive license to the method, and a startup founded by the developers called Advanced Reproductive Testing is also working on commercial assays and applications.
Two teams led by Baylor College of Medicine have taken a different approach, focusing on fetal cells from the mother's blood. In an article published in Prenatal Diagnosis in early October, Baylor researchers, in collaboration with scientists from RareCyte in Seattle, demonstrated that they could recover fetal trophoblasts from maternal blood as early as 10 weeks of gestation and analyze their DNA by array comparative genomic hybridization (CGH) and NGS. In another study, published in Prenatal Diagnosis last month, the Baylor team and collaborators at Arcedi Biotech in Denmark showed that they could also enrich fetal trophoblasts from maternal blood using a different approach, amplify their DNA, and profile the genetic material by array CGH and NGS.
Baylor has an ongoing research study that is comparing results from cell-based NIPT with those from invasive prenatal diagnostic tests, the current gold standard, and plans to present first results early next year, according to Art Beaudet, professor and chair of molecular and human genetics at Baylor and the senior author of both studies. His team is also working on a clinical assay that could launch as early as mid-2017, and he believes that fetal cell-based NIPT will eventually supersede cell-free DNA assays. "I think this test will be done on pregnant women for the next 200 years," he said.
In addition, researchers at the University of California, Los Angeles have isolated circulating fetal nucleated cells from maternal blood, using a technique dubbed "NanoVelcro", and plan to submit their results for publication in the near future, GenomeWeb has learned.
"Ultimately, I think that cell-free DNA and fetal cells can be used in conjunction to improve the accuracy and spectrum of diagnostic applications of noninvasive prenatal testing," said Dennis Lo, chairman and professor of chemical pathology and professor of medicine at the Chinese University of Hong Kong, who was not involved in the studies. "They can be regarded as complementary technologies." Lo is one of the inventors of maternal plasma-based noninvasive prenatal testing and has been an innovator of the cell-free DNA approach. This week, he and his team published a study in the Proceedings of the National Academy of Sciences where they sequenced maternal plasma DNA to uncover fetal de novo mutations.
One goal, many approaches
To isolate fetal cells, the Wayne State team turned to the Papanicolaou smear, better known as the Pap smear. They collected samples with a nylon cytobrush from the lower part of the endocervical canal of 20 pregnant women, ranging from 5 weeks to 19 weeks of gestation, and enriched for fetal cells using a previously published protocol called trophoblast retrieval isolation from the cervix (TRIC), which relies on anti-HLA-G-coated magnetic nanoparticles. On average, they collected almost 300 trophoblast cells per sample.
They then extracted the DNA of the trophoblasts and found initially that it was contaminated with large amounts of cell-free maternal DNA. After they devised a protocol of nuclear isolation and DNAse treatment, they obtained fetal DNA ranging in purity from 85 to 99.9 percent. They then used Illumina's ForenSeq platform to analyze the DNA by NGS, targeting 59 short tandem repeats and 94 single nucleotide variants across all 24 chromosomes, and achieved 100 percent correct fetal haplotyping.
Sascha Drewlo, an associate professor of obstetric and gynecology at Wayne State and the corresponding author of the study, told GenomeWeb via email that the advantages of this method are that it can be performed very early in pregnancy and recovers more fetal cells than other approaches, providing enough material for various types of analyses. His team is currently conducting whole-genome sequencing and "various other omics approaches" on the isolated fetal cells.
Drewlo and his colleagues are also developing clinical validation trials and genetic tests to compare their targeted sequencing approach with diagnostic tests that use amniocentesis and chorionic villus sampling (CVS).
The plan is to expand the current work to methods that will allow them to screen for chromosomal aneuploidies and other genetic disorders simultaneously. "We believe that with the right funding and infrastructure, we could be able to have this technology on the market within the next three years," he said.
The researchers have filed patents on the isolation and use of endocervical trophoblast cells for fetal diagnosis and other applications, and Drewlo said PerkinElmer has exclusively licensed IP directly related to the method and is working on developing commercial tests.
In addition, Drewlo and his Wayne State colleague Randall Armant founded Advanced Reproductive Testing (ART), which is developing technology to analyze isolated trophoblasts, including genetic tests. ART is working closely with PerkinElmer but is also developing its own assays and applications, he said.
Experts in the field have raised concerns that the method used by the Wayne State team to sample cells from the cervix is not truly noninvasive and might increase the risk of pregnancy loss, similar to amniocentesis and CVS. However, so far, that does not seem to be the case, Drewlo said. "We did over 450 cases in our lab and we did not see an increase in any abnormal outcome compared to the national average," he said, adding that those data will be published shortly.
Baylor's team has been working with both RareCyte and Arcedi Biotech for some time and presented early results of these collaborations at a prenatal molecular diagnostics meeting in Boston a year ago. Its Prenatal Diagnosis study with RareCyte is the first published report that uses array CGH and whole-genome sequencing to detect chromosomal abnormalities in fetal trophoblast cells from maternal blood.
For this study, the researchers collected blood from several dozen women at 10 to 16 weeks of gestation, many of whom were scheduled for invasive diagnostic testing. They isolated the fetal cells using RareCyte's trophoblast enrichment and staining method, which starts by separating the red blood cells by density fractionation, followed by immunostaining to identify trophoblasts, which are cytokeratin-positive and CD45-negative. Stained cells were then transferred to slides and individual trophoblast cells were picked using the company's CytePicker. Overall, the scientists were able to recover fetal cells from most samples, including from women carrying fetuses with chromosomal aneuploidies, chromosomal deletions, and a woman with confined placental mosaicism. The researchers then performed whole-genome amplification on the DNA and analyzed samples with known abnormalities and several cases with male fetuses by array CGH and NGS to look for chromosomal aneuplodies, chromosomal deletions, and to confirm fetal sex.
For the other study, conducted in collaboration with Arcedi, the Baylor team collected blood samples from 111 pregnant women at 10 to 17 weeks of gestation. After lysing red blood cells, fetal cells were enriched and stained using a cocktail of antibodies that Arcedi holds a US patent for. Fetal cells spread on a slide were then picked using either a CellCelector from ALS Automated Lab Solutions or RareCyte's CytePicker. After whole-genome amplification of their genomes, the researchers performed array CGH and NGS.
Beaudet told GenomeWeb that Baylor continues to pursue several different approaches for isolating fetal trophoblasts and declined to say whether RareCyte's and Arcedi's methods have advantages over each other. He said Baylor is currently conducting a clinical study, processing about five samples per week right now from women about to undergo amniocentesis or CVS, and comparing the results of its test with the diagnostic test. The goal is to present initial data at the American College of Medical Genetics and Genomics annual meeting in March 2017.
The lab is also focusing on increasing fetal cell recovery and improving throughput. Right now, Beaudet said, the researchers recover three or four fetal cells per sample on average, but sometimes none, and the aim is to get up to 10 to 15 fetal cells per sample. In terms of throughput, the goal is to be able to process 100 samples per week at the time the test launches, he said.
Baylor Genetics Laboratories may launch the test as early as the summer of 2017, he said, though there are still "significant uncertainties." The price of the test is also currently unclear, though Beaudet said it would likely be somewhat higher than that of current cell-free NIPTs, with an upper limit of $3,000 for a successful launch.
Baylor has not filed any patents on its test methods, he said, and unless it chose Arcedi's patented enrichment for its test, "we have not felt like we would be infringing on anybody's patents if we launched with our current methodologies," he said.
Beaudet noted that unlike the highly litigious cell-free NIPT field, there does not appear to be a lot of IP around fetal cell-based NIPT at the moment, though this might change as the field gets more active.
"I think that it is still early days with regard to cell-based noninvasive prenatal testing," said Lo, noting that the number of cases reported in the two Prenatal Diagnosis papers is relatively small, and that the robustness of the test needs to be validated in large-scale studies across many centers. He also pointed out that the imaging-based steps of the Baylor methods may be difficult to scale.
But the positive predictive value of test results from pure fetal DNA may ultimately be better than that of cell-free fetal DNA tests with their large background of maternal DNA and highly fragmented genetic material, according to Hsian-Rong Tseng, a professor of molecular and medical pharmacology at UCLA.
His team has been developing a method to enrich another type of fetal cell, circulating fetal nucleated cells, from maternal blood. It uses a patented technology called NanoVelcro that employs thermo-responsive chips with a "hairy" surface of antibody-coated temperature-sensitive polymer brushes and was described at last year's prenatal molecular diagnostics meeting.
Tseng said he plans to submit a manuscript on his fetal cell-based NIPT in the near future. Overall, he said, his method is able to generate better whole-genome profiling data and has a faster turnaround time than the approaches described by the Baylor team.