GENType: all-in-one preimplantation genetic testing by pedigree haplotyping and copy number profiling suitable for third-party reproduction

Abstract

Abstract STUDY QUESTION Is it possible to develop a comprehensive pipeline for all-in-one preimplantation genetic testing (PGT), also suitable for parents-only haplotyping and, for the first time, third-party reproduction? SUMMARY ANSWER Optimized reduced representation sequencing (RRS) by GENType, along with a novel analysis platform (Hopla), enables cheap, accurate and comprehensive PGT of blastocysts, even without the inclusion of additional family members or both biological parents for genome-wide embryo haplotyping. WHAT IS KNOWN ALREADY Several haplotyping strategies have proven to be effective for comprehensive PGT. However, these methods often rely on microarray technology, whole-genome sequencing (WGS) or a combination of strategies, hindering sample throughput and cost-efficiency. Moreover, existing tools (including other RRS-based strategies) require both prospective biological parents for embryo haplotyping, impeding application in a third-party reproduction setting. STUDY DESIGN, SIZE, DURATION This study included a total of 257 samples. Preliminary technical validation was performed on 81 samples handpicked from commercially available cell lines. Subsequently, a clinical validation was performed on a total of 72 trophectoderm biopsies from 24 blastocysts, tested for a monogenic disorder (PGT-M) (n = 15) and/or (sub)chromosomal aneuploidy (PGT-SR/PGT-A) (n = 9). Once validated, our pipeline was implemented in a diagnostic setting on 104 blastocysts for comprehensive PGT. PARTICIPANTS/MATERIALS, SETTING, METHODS Samples were whole-genome amplified (WGA) and processed by GENType. Quality metrics, genome-wide haplotypes, b-allele frequencies (BAFs) and copy number profiles were generated by Hopla. PGT-M results were deduced from relative haplotypes, while PGT-SR/PGT-A results were inferred from read-count analysis and BAF profiles. Parents-only haplotyping was assessed by excluding additional family members from analysis and using an independently diagnosed embryo as phasing reference. Suitability for third-party reproduction through single-parent haplotyping was evaluated by excluding one biological parent from analysis. Results were validated against reference PGT methods. MAIN RESULTS AND THE ROLE OF CHANCE Genome-wide haplotypes of single cells were highly accurate (mean > 99%) compared to bulk DNA. Unbalanced chromosomal abnormalities (>5 Mb) were detected by GENType. For both PGT-M as well as PGT-SR/PGT-A, our technology demonstrated 100% concordance with reference PGT methods for diverse WGA methods. Equally, for parents-only haplotyping and single-parent haplotyping (of autosomal dominant disorders and X-linked disorders), PGT-M results were fully concordant. Furthermore, the origin of trisomies in PGT-M embryos was correctly deciphered by Hopla. LIMITATIONS, REASONS FOR CAUTION Intrinsic to linkage-analysis strategies, de novo single-nucleotide variants remain elusive. Moreover, parents-only haplotyping is not a stand-alone approach and requires prior diagnosis of at least one reference embryo by an independent technology (i.e. direct mutation analysis) for haplotype phasing. Using a haplotyping approach, the presence of a homologous recombination site across the chromosome is biologically required to distinguish meiotic II errors from mitotic errors during trisomy origin investigation. WIDER IMPLICATIONS OF THE FINDINGS We offer a generic, fully automatable and accurate pipeline for PGT-M, PGT-A and PGT-SR as well as trisomy origin investigation without the need for personalized assays, microarray technology or WGS. The unique ability to perform single-parent assisted haplotyping of embryos paves the way for cost-effective PGT in a third-party reproduction setting. STUDY FUNDING/COMPETING INTEREST(S) L.D.W. is supported by the Research Foundation Flanders (FWO; 1S74619N). L.R. and B.M. are funded by Ghent University and M.B., S.S., K.T., F.V.M. and A.D. are supported by Ghent University Hospital. Research in the N.C. lab was funded by Ghent University, VIB and Kom op Tegen Kanker. A.D.K and N.C. are co-inventors of patent WO2017162754A1. The other authors have no conflicts of interest. TRIAL REGISTRATION NUMBER N/A.