June 21, 2018 by Dale Yuzuki
Overlapping multiplex PCR gives single-well simplicity and outstanding data quality
There is an ongoing need for the clinical laboratory operations to be made more streamlined, simplified and economical. With the rise of personalized, targeted therapy to individual genetic and genomic information projected to rise at double-digit rates over the next five years. This need for efficiency and economy will only rise. A current flood of targeted therapies in development has an estimate that 73% of new oncology therapies in the drug development pipeline are targeted ones. With this plethora of additional targeted therapies comes an increasing number of companion diagnostic approvals, from the US Food and Drug Administration (FDA), and implementation of these diagnostic tests will continue to strain limited resources for clinical assay implementation.
By offering a single-well, multiplex overlapping PCR technology, Pillar Biosciences has recently made available the ONCO/Reveal BRCA1 & BRCA2 Panel. The technology has recently been described and the performance with 35 cell line DNAs and 6 patient blood DNAs reported. Here, the assay performance characteristics will be described with a comparison against other data from other commercially-available BRCA1 and BRCA2 panels.
Pillar Biosciences’ enabling SLIMamp™ technology
By using a novel stem-loop complementary for multiplex, overlapping PCR, highly specific and sensitive target enrichment has been achieved with minimal hands-on time (less than 8 hours from a purified DNA sample to sequencer-ready library) and DNA sample input requirements (as low as 2.5 ng). The simplicity of the workflow, the robustness of the assay and its affordable economics do not come at a price of data quality. For more information about SLIMamp technology, we recently discussed it here at the Pillar Post.
The ONCO/Reveal panel itself is designed around 91 amplicons and a target size of 17,769bp (the coding regions of BRCA1 and BRCA2 +/- 20bp) with an average length of 195 bp.
32 of the 36 cell line DNAs with known pathogenic BRCA1 & BRCA2 variants (available from Coriell) and two anonymized patient samples with sufficient quantity were used. Included in the 32 cell line DNAs was the Genome in a Bottle sample NA12878, a well-characterized standard reference material produced by the US National Institute for Standards and Technology now called Reference Material 8398. Both Sanger capillary electrophoresis sequencing, Illumina’s True Seq Custom Amplicon (TSCA) as well as Pillar’s ONCO/Reveal BRCA1 & BRCA2 Panel by NGS were run on these samples and examined for concordance. Two of the known positive cell lines, the known negative cell line, and one of the patient samples were also run in triplicate to examine reproducibility with the Pillar assay.
From all these 34 unique samples, a total of 409 true positives were detected, of which 234 were heterozygous SNVs, 153 were homozygous SNVs, and 22 were indels. With no false negatives, the sensitivity achieved from these 34 samples was 100%; additionally, with no false positives being called, a specificity of 100% was also achieved.
A total of 66 unique variants were detected across the 34 samples: 47 SNVs, 4 insertions and 15 deletions. The full set of data tables describing the target region, sample information, NGS statistics and unique variants detected, in addition to the reproducibility and library statistics, can be found online here.
Very high mapping and on-target rates
For those not familiar with typical enrichment technology performance, it is instructive to note that mapping rates of 70% or 80% are not unusual in the literature. (Reference: 40% to 85% average 70% on Ion Torrent, Chan and Lee Nov 2012 JMD 14(6); 53%-60% for hybridization-based capture.) While alternative multiplex PCR approaches are available, other difficulties with false-positive and false-negative results for BRCA1 and BRCA2 have been reported (https://www.ncbi.nlm.nih.gov/pubmed/28263838 and https://www.ncbi.nlm.nih.gov/pubmed/25893891).
With the Pillar Biosciences ONCO/Reveal BRCA1 & BRCA2 Panel, for the 34 samples in this dataset, both mapping and on-target rates exceeded 99%. Also, importantly, the percent of bases with coverage depth greater than 20% relative to the mean base coverage was well above 98%. Thus, a sample with a mean base coverage of 3,500x will have 98% of the target bases covered at least 700x or greater.
Screenshot of data from Supplemental Table S3 from Schenk and Wang et al PLOS One July 2017
High BRCA reproducibility
In order to assess reproducibility, two cell line DNAs and one anonymous patient DNA were run in triplicate, and the mapping and on-target rates had Coefficients of Variation (CVs) of only 1.09% and 0.30%, respectively. In addition, the DNA titration data show robust performance across a range of amounts of DNA inputs, with strong NGS run statistics with four DNA samples at seven different amounts (ranging from 5ng through 100ng) across 28 libraries without any decrease in performance metrics.
You should try this BRCA panel
With 100% concordance rates, >99% mapping and >99% on-target rates, >99% of the target bases with a coverage depth greater than 20% of the mean coverage, a workflow that goes from purified DNA to sequencer-ready library in less than eight hours, what else is limiting greater adoption of clinical next-generation sequencing?
You may be pleasantly surprised at what Pillar Biosciences can do for you. Contact us today!