P628: A multiplexing droplet digital PCR approach to obtain accurate CYP2D6 copy number in the presence of rare variants*

Introduction:The pharmacogene CYP2D6 is responsible for the metabolism of over 25% of prescribed medications.Due to the complexity of the genetic region and the presence of highly similar pseudogenes CYP2D7 and CYP2D8, CYP2D6 gene deletions and multiplications may occur, which influence enzyme activity.Moreover, cross-overs between CYP2D6 and CYP2D7 may lead to inactive fusions.It is therefore important to accurately assess the number of intact CYP2D6 copies.Primer/probe-based methods are most commonly used to assess CYP2D6 copy number.However, rare variants in the assayed region may interfere with primer or probe binding and detection of the affected allele, leading to under-assessment of the number of intact CYP2D6 copies.Using our previously validated duplex droplet digital Polymerase Chain Reaction (ddPCR), assay patterns indicative of variants interfering with primer or probe binding were identified.We have now characterized 8 different variants that cause these patterns using Sanger sequencing.The ddPCR patterns are now used to confidently identify these cases and accurately assess CYP2D6 copy number, despite the presence of these variants.In an effort to optimize workflows and implement cost savings, we evaluated an all-in-one multiplex ddPCR reaction on a 2-channel (VIC/FAM) ddPCR system to assess copy number at two sites along with a reference assay.While validating this assay, we ensured that patterns indicative of variants that interfere with assay binding were still discernable.Methods: In this work, multiplexing ddPCR assay for both CYP2D6 intron 2 and exon 9 against the reference TERT was performed in one reaction using a 2channel (VIC/FAM) instrument.Separation of the three targets, CYP2D6 intron 2, CYP2D6 exon 9 and the reference TERT was achieved using amplitudebased multiplexing.Sanger sequencing of samples containing distinct ddPCR patterns identified several different variants that interfered with assay binding.In addition, data from our targeted next generation sequencing pharmacogenetics (tNGS-PGx) panel was interrogated for the presence of variants in the CYP2D6 assay binding sites to determine allele frequency.Results: The implementation of our newly validated multiplexing ddPCR allowed us to accurately assess up to 9 copies of CYP2D6.Suboptimal amplification, due to the presence of variants in the assay binding regions, was discernable permitting accurate CYP2D6 copy number assessment.Multiplexing CYP2D6 intron 2 and exon 9 with TERT on a 2-channel instrument led to a 2-fold reduction in cost and time relative to the duplex assay.Additionally, as all data is evaluated simultaneously, this multiplexing leads to improved data review and reliability.Retrospective evaluation of over 45,000 tNGS-PGx cases allowed us to determine allele frequency for the 8 previously detected variants.The distribution of CYP2D6 deletions, duplications, and CYP2D6/CYP2D7 fusions was also determined. Conclusion:We have developed an efficient multiplexing ddPCR approach for CYP2D6 copy number assessment.Eight unique variants that may interfere with copy number assessment have been characterized.Laboratories testing for CYP2D6 copy number using probe-based assays should be aware of the potential of variants that may interfere with copy number assessment.