A Clinical Laboratory's Approach to Curating Unexpected Variants in Pharmacogenes
Laboratory Genetics and Genomics
-
Primary Categories:
- Laboratory Genetics
-
Secondary Categories:
- Laboratory Genetics
Introduction:
Pharmacogenetic (PGx) results are described by haplotype blocks defined by one or more core variants inherited in cis, termed star alleles. The Association for Molecular Pathology (AMP) and American College of Medical Genetics (ACMG) issue standards for PGx variant testing and allele reporting, including a minimum list of variants needed to call key alleles and laboratory approaches to PGx testing. Although well-characterized alleles with an appreciable multiethnic frequency are considered for inclusion in these guidelines, testing may detect rare or understudied genetic variations. Examples of this variation may include an unexpected variant combination or a nucleotide change at a targeted location that is not associated with a star allele. Guidance for interpreting such unexpected PGx results is lacking at this time. Consequently, laboratories may choose to report unexpected variants in different ways. This variable reporting between labs may introduce challenges for clinicians and patients, such as inconsistent clinical guidance. Here, we describe our independently derived approach for the interpretation of variants detected using our targeted Next Generation Sequencing (tNGS) PGx panel. We present examples that illustrate our manual review and reporting process of unexpected findings. This method for PGx variant curation may be implemented by clinical laboratories without the use of more costly technologies for workup (i.e., long read sequencing).
Methods:
To interpret findings from the tNGS PGx panel, the laboratory uses an automated bioinformatics pipeline that includes a customized list of star alleles with known function and definitive evidence as described by Clinical Pharmacogenetics Implementation Consortium (CPIC) and the Pharmacogene Variation Consortium (PharmVar). Occasionally, an unexpected variant call or variant combination is flagged for manual review because it cannot be classified through the routine bioinformatics process. This review includes assessing the phase and functional impact of variants, as well as the clinical utility of candidate star alleles. For variants that are in close proximity, Integrative Genomics Viewer (IGV) is used to determine phasing when the variants are captured in the same amplicon. Additional resources such as the CPIC allele definition tables, PharmVar, PharmGKB, and peer-reviewed publications may also aid interpretation. Reports containing variants which require manual curation typically contain a comment explaining the findings, including the supporting evidence used to guide the interpretation.
Results:
From December 2023 to November 2024, over 70 unique unexpected PGx result scenarios identified by tNGS (0.2% of cases) were clarified by manual curation. Of the unique cases, 10% required curation due to an unexpected nucleotide change at a targeted site, while 90% required curation due to an unexpected combination of core variants. The most common gene curated was CYP2D6 (59.4% of curations).
Conclusion:
Overall, unexpected variant calls or combinations comprised a small proportion of the cases assessed by tNGS making manual curation of these rare events logistically feasible. The review process implemented here aims to provide additional information to maximize clinical utility while minimizing the number of unknown, uncertain, or misleading (e.g., *1 call due to a lack of reported alleles detected) results. While the clinical utility of some unexpected findings may not be established, commenting on the finding provides the opportunity to implement future clinically actionable updates. Additionally, the preference for reporting star alleles with known function allows for any downstream clinical decision support alerts, which trigger off specific alleles, to function as intended. As implementation of precision medicine expands, there is a growing need for PGx variant interpretation and reporting guidelines that provide an accessible approach for curation. In the absence of guidelines, we introduce our approach to classifying uncommon PGx variation which aims to maximize clinical utility and provide the most accurate results to patients and providers.
Pharmacogenetic (PGx) results are described by haplotype blocks defined by one or more core variants inherited in cis, termed star alleles. The Association for Molecular Pathology (AMP) and American College of Medical Genetics (ACMG) issue standards for PGx variant testing and allele reporting, including a minimum list of variants needed to call key alleles and laboratory approaches to PGx testing. Although well-characterized alleles with an appreciable multiethnic frequency are considered for inclusion in these guidelines, testing may detect rare or understudied genetic variations. Examples of this variation may include an unexpected variant combination or a nucleotide change at a targeted location that is not associated with a star allele. Guidance for interpreting such unexpected PGx results is lacking at this time. Consequently, laboratories may choose to report unexpected variants in different ways. This variable reporting between labs may introduce challenges for clinicians and patients, such as inconsistent clinical guidance. Here, we describe our independently derived approach for the interpretation of variants detected using our targeted Next Generation Sequencing (tNGS) PGx panel. We present examples that illustrate our manual review and reporting process of unexpected findings. This method for PGx variant curation may be implemented by clinical laboratories without the use of more costly technologies for workup (i.e., long read sequencing).
Methods:
To interpret findings from the tNGS PGx panel, the laboratory uses an automated bioinformatics pipeline that includes a customized list of star alleles with known function and definitive evidence as described by Clinical Pharmacogenetics Implementation Consortium (CPIC) and the Pharmacogene Variation Consortium (PharmVar). Occasionally, an unexpected variant call or variant combination is flagged for manual review because it cannot be classified through the routine bioinformatics process. This review includes assessing the phase and functional impact of variants, as well as the clinical utility of candidate star alleles. For variants that are in close proximity, Integrative Genomics Viewer (IGV) is used to determine phasing when the variants are captured in the same amplicon. Additional resources such as the CPIC allele definition tables, PharmVar, PharmGKB, and peer-reviewed publications may also aid interpretation. Reports containing variants which require manual curation typically contain a comment explaining the findings, including the supporting evidence used to guide the interpretation.
Results:
From December 2023 to November 2024, over 70 unique unexpected PGx result scenarios identified by tNGS (0.2% of cases) were clarified by manual curation. Of the unique cases, 10% required curation due to an unexpected nucleotide change at a targeted site, while 90% required curation due to an unexpected combination of core variants. The most common gene curated was CYP2D6 (59.4% of curations).
Conclusion:
Overall, unexpected variant calls or combinations comprised a small proportion of the cases assessed by tNGS making manual curation of these rare events logistically feasible. The review process implemented here aims to provide additional information to maximize clinical utility while minimizing the number of unknown, uncertain, or misleading (e.g., *1 call due to a lack of reported alleles detected) results. While the clinical utility of some unexpected findings may not be established, commenting on the finding provides the opportunity to implement future clinically actionable updates. Additionally, the preference for reporting star alleles with known function allows for any downstream clinical decision support alerts, which trigger off specific alleles, to function as intended. As implementation of precision medicine expands, there is a growing need for PGx variant interpretation and reporting guidelines that provide an accessible approach for curation. In the absence of guidelines, we introduce our approach to classifying uncommon PGx variation which aims to maximize clinical utility and provide the most accurate results to patients and providers.