Imaging single DNA molecules for high precision NIPT

Abstract

Cell-free DNA analysis is becoming adopted for first line aneuploidy screening, however for most healthcare programs, cost and workflow complexity is limiting adoption of the test. We report a novel cost effective method, the Vanadis NIPT assay, designed for high precision digitally-enabled measurement of chromosomal aneuploidies in maternal plasma. Reducing NIPT assay complexity is achieved by using novel molecular probe technology that specifically label target chromosomes combined with a new readout format using a nanofilter to enrich single molecules for imaging and counting without DNA amplification, microarrays or sequencing. The primary objective of this study was to assess the Vanadis NIPT assay with respect to analytical precision and clinical feasibility. Analysis of reference DNA samples indicate that samples which are challenging to analyze with low fetal-fraction can be readily detected with a limit of detection determined at <2% fetal-fraction. In total of 286 clinical samples were analysed and 30 out of 30 pregnancies affected by trisomy 21 were classified correctly. This method has the potential to make cost effective NIPT more widely available with more women benefiting from superior detection and false positive rates.

Conflict of interest statement

All authors affiliated with Vanadis Diagnostics were at some point during the project employed by PerkinElmer or Vanadis Diagnostics that holds the commercial rights to the technology presented herein.

Figures

Figure 1

Vanadis NIPT assay. (1) Extracted cfDNA is first subjected to specific fragmentation using a restriction enzyme. The resulting target cfDNA fragments are similar in size and GC content and are derived from the chromosomes of interest. (2) Probes, designed to hybridize to the target cfDNA fragments to form circular DNA complexes, are mixed with the target cfDNA fragments, backbone oligos and DNA ligase. (3) By allowing the target cfDNA fragments to hybridize to the probe complex and DNA ligase to seal the nicks, covalently closed circles are generated that each includes a cfDNA target fragment and a corresponding chromosomal tag. All DNA that is not circularized is removed with exonucleases. (4) The DNA circles are copied about 1000 times by rolling-circle-amplification (RCA) to generate one rolling circle replication product (RCP), a single stranded concatemer amplification product. (5) The RCPs self-assemble to submicron-sized DNA objects. Because each RCP includes copies of a chromosomal tag it can be recognized by a corresponding fluorescently labeled oligonucleotide. (6) The labelled RCPs are then deposited to a 96-well nanofilter microplate. The microplate has a nanofilter membrane in the bottom to allow the RCPs to be captured on the plate bottom, while buffer and non-hybridized fluorophores are washed through the membrane. The deposited RCPs are finally imaged through the nanofilter using the Vanadis View imaging instrument.

Figure 2

(a) Graph outlining the detection rate as a function of the assay precision for false positive rates 0.15% and 0.3%. (b) Graph outlining the fraction of samples that need to be eliminated to achieve 99% detection rate with 0.15% false positive rate at different measurement precisions.

Figure 3

To evaluate the quantification precision of the Vanadis NIPT assay counting readout, a plate was analysed with identical input. The variation of the ratio-measurement is set by the number of objects quantified. When fewer objects are imaged the precision deteriorates according to the theoretically optimal measurement precision. Standard deviations are plotted from five randomized down samplings.

Figure 4

Reference samples from SeraCare with 0%, 2%, 4% and 8% T21 DNA were analysed. All samples with 4% and 8% T21 DNA, and all but one out of 14 samples with 2% T21 DNA are separated from normal with >3 standard deviations.

Figure 5

(a) 17 clinical samples from pregnancies with T21 were analysed among a set of 182 samples. All affected samples were identified correctly. (b) 104 prospectively collected samples were collected analysed blindly. 13 positive trisomy 21 pregnancies were classified correctly with no false positives.

Figure 6

Schematic of Vanadis NIPT probe design. Correct hybridization of backbone oligonucleotide and MseI digested target cfDNA fragment generates a DNA circle, following closing of nicks by a DNA ligase. Each DNA circle contains sequence information of the incorporated target cfDNA fragment as well as a chromosomal tag enabling readout following fluorescent labeling of the chromosomal tag.