Four Novel NUDT15 Haplotypes Relevant for Treatment of Acute Lymphoblastic Leukemia
Turner A, Aggarwal P, Scharer G, Broeckel U.
Presented at The American Society of Pediatric Hematology/Oncology (ASPHO) 2019.
Genetic variants in TPMT and NUDT15 have been associated with thiopurine-related myelotoxicity requiring dose adjustment during treatment of acute lymphoblastic leukemia (ALL). The frequency of variants in these genes varies across different populations and comprehensive, accurate genotyping, with a short turnaround time (TAT) is critical to adjust thiopurine dosing.
We aim to establish a novel targeted next generation sequencing (NGS) methodology to comprehensively assay TPMT and NUDT15 haplotypes with a TAT that meets thiopurine dose optimization window.
Eight genomic DNA samples from the 1000 Genomes Project were selected as controls for assay validation. Sequencing libraries were prepared using a custom AmpliSeq assay and NGS was performed on the Illumina iSeq™ system. Sequence reads were aligned to the human reference genome (GRCh38). Haplotypes were determined based on overlapping read alignment. Known polymorphisms were further confirmed with real-time qPCR (RT-PCR).
All samples were prepared and sequenced within three days and showed 100% concordance for sequence variants with both the 1000 Genomes data and the confirming RT-PCR in TPMT and NUDT15. Deep NGS allowed the assignment of variants into haplotypes. Interestingly, in addition to identifying two samples with complex heterozygous NUDT15 haplotypes (NUDT15*2/*3) we found six samples with four previously undescribed NUDT15 haplotypes. These included four heterozygous amino acid changes: p.Val93Ile, p.Pro12Leu, p.Gly13Ala, and p.Lys33Asn, two of which protein modeling has predicted to be deleterious to enzyme activity (p.Pro12Leu and p.Lys33Asn, minor allele frequency= 0.01 to 0.20, 0.003 to 0.08 respectively).
Genotyping of TPMT and NUDT15 is recognized as a critical test to assess the risk of thiopurine associated adverse reactions, particularly in ALL patients. Currently, CPIC® and PharmVar have annotated 41 TPMT and 19 NUDT15 haplotypes, including many with reduced function. Our NGS assay accurately captured all regions containing variants known to affect enzyme activity and identified all known TPMT and NUDT15 haplotypes. Notably, we were able to identify four novel NUDT15 haplotypes which may affect enzyme activity. These results support the use of an NGS testing methodology for detection of known and novel TPMT and NUDT15 polymorphisms. This analysis approach is particularly important for genes in which many functional haplotypes have not yet been identified, to avoid enzyme activity misclassification. Importantly, the assay can be performed in a time frame that enables dose adjustment prior to or during treatment of ALL.
Click here to download poster.
Identification of Novel CYP2D6 Haplotypes that Interfere with TaqMan Copy Number Analysis
Turner A, Aggarwal P, Boone EC, Haidar CE, Relling M., Broeckel U, Gaedig, A.
Presented at American Society of Human Genetics (ASHG) 2018.
Cytochrome P450 2D6 (CYP2D6) is a highly polymorphic gene encoding an enzyme critical in the metabolism of up to 25% of commonly prescribed drugs. There are over 100 described star allele (*) haplotypes, including frequent gene conversions, deletions, duplications and gene-fusions (D6/D7 hybrid), many of which have altered enzyme activity. Accurate copy number (CN) calling is critical in determining patient drug response. TaqMan CNV assays are a commonly used methodology for CN analysis, utilizing specific primers and labeled probes to evaluate the CN state of a genomic region of interest. Polymorphisms within the primer or probe target sequence can generate a false positive for a CN loss. Currently, three CYP2D6 intra-gene regions can be tested using commercially available TaqMan assays: intron 2, intron 6 and exon 9. Gene-fusions or conversions present as discrepant CN calling between the three regions. During clinical testing, we identified samples with CNV inconsistent with known haplotypes. To characterize the haplotypes that lead to these CNV results, we performed long-range PCR, Next-Gen sequencing and allele-specific Sanger sequencing. Sequence analysis of 20 patients identified six novel sub-alleles containing polymorphisms within the TaqMan assay probe binding sites, causing a false-positive copy number loss. In total, four novel haplotypes (sub-alleles) with SNPs within the intron 2 and 6 probes were identified (*1 sub-allele 1: rs770138443, rs78854695; sub-allele 2: rs781257354, rs37522240. *2 sub-allele 1: rs180847475, rs186133763; sub-allele 2: rs370010370). Analysis of samples which showed CN loss in only exon 9, identified new *10 and *17 sub-allele haplotypes. Two additional samples with exon 9 CNV contained hybrid/deletion arrangements within the probe region that affected CN analysis. In summary, performance of widely used TaqMan CNV assays is affected by the presence of sequence variation within the probe-target regions. The identification of these haplotypes and further characterization of these structural variations will be important for clinical interpretation and could directly impact patient care.
Click here to download poster.
Implementation of Affymetrix PharmacoScanTM for Comprehensive Preemptive Clinical Pharmacogenetic Testing
Turner A, Lorier R, Aggarwal, P Matter A, Broeckel U.
Presented at American Society of Human Genetics (ASHG)-Pharmacogenomics Research Network (PGRN) symposium, 2016.
Genotyping of relevant pharmacogenetic (PGx) genes (e.g. single nucleotide polymorphisms, copy number variants) and HLA typing for known associations with drug metabolism and hypersensitivity allows for personalized drug selection and dosing prior to administration. This enables a clinician to provide a patient with an optimized treatment regimen by maximizing drug efficacy and limiting adverse reactions.
Currently, there are over 300 actionable alleles and genes with known dosing recommendations. Affected drugs span a wide range of categories from pain management to heart disease and cancer treatment (e.g. codeine, warfarin, allopurinol). With improved PGx knowledge, such as the identification of new alleles and the dependence of drug metabolism on multi-gene interactions (e.g. warfarin: VKORC1 and CYP2C9), there is a need for an updated and comprehensive PGx genotyping platform. Affymetrix’s PharmacoScanTM enables genotyping of over 4000 markers in 900 pharmacogenetic relevant genes to be done as a single test.
In our CLIA-certified clinical laboratory, we have tested approximately 5,000 samples on a variety of PGx platforms. To address the need for improved PGx testing, we clinically validated the PharmacoScanTM platform as part of our repertoire. We validated the accuracy and detection of critical PGx variants utilizing a diverse reference set of 96 Coriell samples. This set was previously genotyped on multiple PGx platforms by our group, as well as others in a collaborative effort by the CDC. The PharmacoScanTM data generated for this reference set will be available as a public resource.
Data generated on comprehensive platforms can present challenges for interpretation and practical application for both clinicians and patients. As part of our complete diagnostic approach, we developed a workflow to generate quality-assured data that is translated in a meaningful and actionable way. In addition to the raw data, we provide translational reports, which include the patient’s haplotype-specific metabolizer status based on the CPIC dosing guidelines. Downstream inclusion of this data in the EMR will enable clinicians to preemptively make the most informed drug choices and dosing decisions, providing cost-effective and better-individualized patient care.
Click here to download poster.