Katrin Sangkuhl

All Publications

• PGxQA: A Resource for Evaluating LLM Performance for Pharmacogenomic QA Tasks. Pacific Symposium on Biocomputing. Pacific Symposium on Biocomputing Keat, K., Venkatesh, R., Huang, Y., Kumar, R., Tuteja, S., Sangkuhl, K., Li, B., Gong, L., Whirl-Carrillo, M., Klein, T. E., Ritchie, M. D., Kim, D. 2025; 30: 229-246

  Abstract

  Pharmacogenetics represents one of the most promising areas of precision medicine, with several guidelines for genetics-guided treatment ready for clinical use. Despite this, implementation has been slow, with few health systems incorporating the technology into their standard of care. One major barrier to uptake is the lack of education and awareness of pharmacogenetics among clinicians and patients. The introduction of large language models (LLMs) like GPT-4 has raised the possibility of medical chatbots that deliver timely information to clinicians, patients, and researchers with a simple interface. Although state-of-the-art LLMs have shown impressive performance at advanced tasks like medical licensing exams, in practice they still often provide false information, which is particularly hazardous in a clinical context. To quantify the extent of this issue, we developed a series of automated and expert-scored tests to evaluate the performance of chatbots in answering pharmacogenetics questions from the perspective of clinicians, patients, and researchers. We applied this benchmark to state-of-the-art LLMs and found that newer models like GPT-4o greatly outperform their predecessors, but still fall short of the standards required for clinical use. Our benchmark will be a valuable public resource for subsequent developments in this space as we work towards better clinical AI for pharmacogenetics.

  View details for PubMedID 39670373

• PharmVar Tutorial on CYP2D6 Structural Variation Testing and Recommendations on Reporting. Clinical pharmacology and therapeutics Turner, A. J., Nofziger, C., Ramey, B. E., Ly, R. C., Bousman, C. A., Agúndez, J. A., Sangkuhl, K., Whirl-Carrillo, M., Vanoni, S., Dunnenberger, H. M., Ruano, G., Kennedy, M. A., Phillips, M. S., Hachad, H., Klein, T. E., Moyer, A. M., Gaedigk, A. 2023

  Abstract

  The Pharmacogene Variation Consortium (PharmVar) provides nomenclature for the highly polymorphic human CYP2D6 gene locus and a comprehensive summary of structural variation. CYP2D6 contributes to the metabolism of numerous drugs and thus, genetic variation in its gene impacts drug efficacy and safety. To accurately predict a patient's CYP2D6 phenotype, testing must include structural variants including gene deletions, duplications, hybrid genes, and combinations thereof. This tutorial offers a comprehensive overview of CYP2D6 structural variation, terms and definitions, a review of methods suitable for their detection and characterization, and practical examples to address the lack of standards to describe CYP2D6 structural variants or any other pharmacogene. This PharmVar tutorial offers practical guidance on how to detect the many, often complex, structural variants, as well as recommends terms and definitions for clinical and research reporting. Uniform reporting is not only essential for electronic health record-keeping but also for accurate translation of a patient's genotype into phenotype which is typically utilized to guide drug therapy.

  View details for DOI 10.1002/cpt.3044

  View details for PubMedID 37669183

• Clinical Pharmacogenetics Implementation Consortium (CPIC) Guideline for CYP2D6, CYP2C19, CYP2B6, SLC6A4, and HTR2A Genotypes and Serotonin Reuptake Inhibitor Antidepressants. Clinical pharmacology and therapeutics Bousman, C. A., Stevenson, J. M., Ramsey, L. B., Sangkuhl, K., Hicks, J. K., Strawn, J. R., Singh, A. B., Ruaño, G., Mueller, D. J., Tsermpini, E. E., Brown, J. T., Bell, G. C., Leeder, J. S., Gaedigk, A., Scott, S. A., Klein, T. E., Caudle, K. E., Bishop, J. R. 2023

  Abstract

  Serotonin reuptake inhibitor antidepressants, including selective serotonin reuptake inhibitors (SSRIs; i.e., citalopram, escitalopram, fluoxetine, fluvoxamine, paroxetine, and sertraline), serotonin and norepinephrine reuptake inhibitors (SNRIs; i.e., desvenlafaxine, duloxetine, levomilnacipran, milnacipran, and venlafaxine), and serotonin modulators with SSRI-like properties (i.e., vilazodone and vortioxetine) are primary pharmacologic treatments for major depressive and anxiety disorders. Genetic variation in CYP2D6, CYP2C19, and CYP2B6 influences the metabolism of many of these antidepressants, which may potentially affect dosing, efficacy, and tolerability. In addition, the pharmacodynamic genes SLC6A4 (serotonin transporter) and HTR2A (serotonin-2A receptor) have been examined in relation to efficacy and side effect profiles of these drugs. This guideline updates and expands the 2015 Clinical Pharmacogenetics Implementation Consortium (CPIC) guideline for CYP2D6 and CYP2C19 genotypes and SSRI dosing and summarizes the impact of CYP2D6, CYP2C19, CYP2B6, SLC6A4, and HTR2A genotypes on antidepressant dosing, efficacy, and tolerability. We provide recommendations for using CYP2D6, CYP2C19 and CYP2B6 genotype results to help inform prescribing these antidepressants and describe the existing data for SLC6A4 and HTR2A which do not support their clinical use in antidepressant prescribing.

  View details for DOI 10.1002/cpt.2903

  View details for PubMedID 37032427

• How to Run the Pharmacogenomics Clinical Annotation Tool (PharmCAT). Clinical pharmacology and therapeutics Li, B., Sangkuhl, K., Keat, K., Whaley, R. M., Woon, M., Verma, S., Dudek, S., Tuteja, S., Verma, A., Whirl-Carrillo, M., Ritchie, M. D., Klein, T. E. 2022

  Abstract

  Pharmacogenomics (PGx) investigates the genetic influence on drug response and is an integral part of precision medicine. While PGx testing is becoming more common in clinical practice and may be reimbursed by Medicare/Medicaid and commercial insurance, interpreting PGx testing results for clinical decision support is still a challenge. The Pharmacogenomics Clinical Annotation Tool (PharmCAT) has been designed to tackle the need for transparent, automatic interpretations of patient genetic data. PharmCAT incorporates a patient's genotypes, annotates PGx information (allele, genotype, and phenotype), and generates a report with PGx guideline recommendations from the Clinical Pharmacogenetics Implementation Consortium (CPIC) and/or the Dutch Pharmacogenetics Working Group (DPWG). PharmCAT has introduced new features in the last two years, including a VCF preprocessor, the inclusion of DPWG guidelines, and functionalities for PGx research. For example, researchers can use the VCF preprocessor to prepare biobank-scale data for PharmCAT. In addition, PharmCAT enables the assessment of novel partial and combination alleles that are composed of known PGx variants and can call CYP2D6 genotypes based on SNPs and INDELs in the input VCF file. This tutorial provides materials and detailed step-by-step instructions for how to use PharmCAT in a versatile way that can be tailored to users' individual needs.

  View details for DOI 10.1002/cpt.2790

  View details for PubMedID 36350094

• Clinical Pharmacogenetics Implementation Consortium Guideline for CYP2C19 Genotype and Clopidogrel Therapy: 2022 Update. Clinical pharmacology and therapeutics Lee, C. R., Luzum, J. A., Sangkuhl, K., Gammal, R. S., Sabatine, M. S., Stein, C. M., Kisor, D. F., Limdi, N. A., Lee, Y. M., Scott, S. A., Hulot, J. S., Roden, D. M., Gaedigk, A., Caudle, K. E., Klein, T. E., Johnson, J. A., Shuldiner, A. R. 2022

  Abstract

  CYP2C19 catalyzes the bioactivation of the antiplatelet prodrug clopidogrel, and CYP2C19 genotype impacts clopidogrel active metabolite formation. CYP2C19 intermediate and poor metabolizers who receive clopidogrel experience reduced platelet inhibition and increased risk for major adverse cardiovascular and cerebrovascular events. This guideline is an update to the 2013 Clinical Pharmacogenetics Implementation Consortium (CPIC) guideline for the use of clopidogrel based on CYP2C19 genotype and includes expanded indications for CYP2C19 genotype-guided antiplatelet therapy, increased strength of recommendation for CYP2C19 intermediate metabolizers, updated CYP2C19 genotype to phenotype translation, and evidence from an expanded literature review (updates at www.cpicpgx.org).

  View details for DOI 10.1002/cpt.2526

  View details for PubMedID 35034351

• An evidence-based framework for evaluating pharmacogenomics knowledge for personalized medicine. Clinical pharmacology and therapeutics Whirl-Carrillo, M., Huddart, R., Gong, L., Sangkuhl, K., Thorn, C. F., Whaley, R., Klein, T. E. 2021

  Abstract

  Clinical annotations are one of the most popular resources available on the Pharmacogenomics Knowledgebase (PharmGKB). Each clinical annotation summarizes the association between variant-drug pairs, shows relevant findings from the curated literature and is assigned a level of evidence (LOE) to indicate the strength of support for that association. Evidence from the pharmacogenomic literature is curated into PharmGKB as variant annotations, which can be used to create new clinical annotations or added to existing clinical annotations. This means that the same clinical annotation can be worked on by multiple curators over time. As more evidence is curated into PharmGKB, the task of maintaining consistency when assessing all the available evidence and assigning a level of evidence becomes increasingly difficult. To remedy this, a scoring system has been developed to automate LOE assignment to clinical annotations. Variant annotations are scored according to certain attributes including study size, reported p-value and whether the variant annotation supports or fails to find an association. Clinical guidelines or FDA-approved drug labels which give variant-specific prescribing guidance are also scored. The scores of all annotations attached to a clinical annotation are summed together to give a total score for the clinical annotation, which is used to calculate a LOE. Overall, the system increases transparency, consistency and reproducibility in LOE assignment to clinical annotations. In combination with increased standardization of how clinical annotations are written, use of this scoring system helps to ensure that PharmGKB clinical annotations continue to be a robust source of pharmacogenomic information.

  View details for DOI 10.1002/cpt.2350

  View details for PubMedID 34216021

• PharmVar GeneFocus: CYP2C9. Clinical pharmacology and therapeutics Sangkuhl, K., Claudio-Campos, K., Cavallari, L. H., Agundez, J., Whirl-Carrillo, M., Duconge, J., Del Tredici, A. L., Wadelius, M., Botton, M. R., Woodahl, E. L., Scott, S. A., Klein, T. E., Pratt, V. M., Daly, A. K., Gaedigk, A. 2021

  Abstract

  The Pharmacogene Variation Consortium (PharmVar) catalogues star (*) allele nomenclature for the polymorphic human CYP2C9 gene.Genetic variation within the CYP2C9 gene locusimpacts the metabolism or bioactivation of many clinically important drugs including NSAIDs, phenytoin, anti-diabetic agents and angiotensin receptor blocker. Variable CYP2C9 activity is of particular importance regarding efficacy and safety of warfarin and siponimod as indicated in their package inserts. This GeneFocus provides a comprehensive overview and summary of CYP2C9 and describes how haplotype information catalogued by PharmVar is utilized by the Pharmacogenomics Knowledgebase (PharmGKB) and the Clinical Pharmacogenetics Implementation Consortium (CPIC).

  View details for DOI 10.1002/cpt.2333

  View details for PubMedID 34109627

• Transfer learning enables prediction of CYP2D6 haplotype function. PLoS computational biology McInnes, G., Dalton, R., Sangkuhl, K., Whirl-Carrillo, M., Lee, S., Tsao, P. S., Gaedigk, A., Altman, R. B., Woodahl, E. L. 2020; 16 (11): e1008399

  Abstract

  Cytochrome P450 2D6 (CYP2D6) is a highly polymorphic gene whose protein product metabolizes more than 20% of clinically used drugs. Genetic variations in CYP2D6 are responsible for interindividual heterogeneity in drug response that can lead to drug toxicity and ineffective treatment, making CYP2D6 one of the most important pharmacogenes. Prediction of CYP2D6 phenotype relies on curation of literature-derived functional studies to assign a functional status to CYP2D6 haplotypes. As the number of large-scale sequencing efforts grows, new haplotypes continue to be discovered, and assignment of function is challenging to maintain. To address this challenge, we have trained a convolutional neural network to predict functional status of CYP2D6 haplotypes, called Hubble.2D6. Hubble.2D6 predicts haplotype function from sequence data and was trained using two pre-training steps with a combination of real and simulated data. We find that Hubble.2D6 predicts CYP2D6 haplotype functional status with 88% accuracy in a held-out test set and explains 47.5% of the variance in in vitro functional data among star alleles with unknown function. Hubble.2D6 may be a useful tool for assigning function to haplotypes with uncurated function, and used for screening individuals who are at risk of being poor metabolizers.

  View details for DOI 10.1371/journal.pcbi.1008399

  View details for PubMedID 33137098

• Review and Consensus on Pharmacogenomic Testing in Psychiatry. Pharmacopsychiatry Bousman, C. A., Bengesser, S. A., Aitchison, K. J., Amare, A. T., Aschauer, H. n., Baune, B. T., Asl, B. B., Bishop, J. R., Burmeister, M. n., Chaumette, B. n., Chen, L. S., Cordner, Z. A., Deckert, J. n., Degenhardt, F. n., DeLisi, L. E., Folkersen, L. n., Kennedy, J. L., Klein, T. E., McClay, J. L., McMahon, F. J., Musil, R. n., Saccone, N. L., Sangkuhl, K. n., Stowe, R. M., Tan, E. C., Tiwari, A. K., Zai, C. C., Zai, G. n., Zhang, J. n., Gaedigk, A. n., Müller, D. J. 2020

  Abstract

  The implementation of pharmacogenomic (PGx) testing in psychiatry remains modest, in part due to divergent perceptions of the quality and completeness of the evidence base and diverse perspectives on the clinical utility of PGx testing among psychiatrists and other healthcare providers. Recognizing the current lack of consensus within the field, the International Society of Psychiatric Genetics assembled a group of experts to conduct a narrative synthesis of the PGx literature, prescribing guidelines, and product labels related to psychotropic medications as well as the key considerations and limitations related to the use of PGx testing in psychiatry. The group concluded that to inform medication selection and dosing of several commonly-used antidepressant and antipsychotic medications, current published evidence, prescribing guidelines, and product labels support the use of PGx testing for 2 cytochrome P450 genes (CYP2D6, CYP2C19). In addition, the evidence supports testing for human leukocyte antigen genes when using the mood stabilizers carbamazepine (HLA-A and HLA-B), oxcarbazepine (HLA-B), and phenytoin (CYP2C9, HLA-B). For valproate, screening for variants in certain genes (POLG, OTC, CSP1) is recommended when a mitochondrial disorder or a urea cycle disorder is suspected. Although barriers to implementing PGx testing remain to be fully resolved, the current trajectory of discovery and innovation in the field suggests these barriers will be overcome and testing will become an important tool in psychiatry.

  View details for DOI 10.1055/a-1288-1061

  View details for PubMedID 33147643

• Standardizing CYP2D6 Genotype to Phenotype Translation: Consensus Recommendations from the Clinical Pharmacogenetics Implementation Consortium and Dutch Pharmacogenetics Working Group. Clinical and translational science Caudle, K. E., Sangkuhl, K. n., Whirl-Carrillo, M. n., Swen, J. J., Haidar, C. E., Klein, T. E., Gammal, R. S., Relling, M. V., Scott, S. A., Hertz, D. L., Guchelaar, H. J., Gaedigk, A. n. 2019

  Abstract

  Translating CYP2D6 genotype to metabolizer phenotype is not standardized across clinical laboratories offering pharmacogenetic (PGx) testing and PGx clinical practice guidelines, such as the Clinical Pharmacogenetics Implementation Consortium (CPIC) and the Dutch Pharmacogenetics Working Group (DPWG). The genotype to phenotype translation discordance between laboratories and guidelines can cause discordant cytochrome P450 2D6 (CYP2D6) phenotype assignments and, thus lead to inconsistent therapeutic recommendations and confusion among patients and clinicians. A modified-Delphi method was used to obtain consensus for a uniform system for translating CYP2D6 genotype to phenotype among a panel of international CYP2D6 experts. Experts with diverse involvement in CYP2D6 interpretation (clinicians, researchers, genetic testing laboratorians, and PGx implementers; n = 37) participated in conference calls and surveys. After completion of 7 surveys, a consensus (> 70%) was reached with 82% of the CYP2D6 experts agreeing to the final CYP2D6 genotype to phenotype translation method. Broad adoption of the proposed CYP2D6 genotype to phenotype translation method by guideline developers, such as CPIC and DPWG, and clinical laboratories as well as researchers will result in more consistent interpretation of CYP2D6 genotype.

  View details for DOI 10.1111/cts.12692

  View details for PubMedID 31647186

• The Evolution of PharmVar. Clinical pharmacology and therapeutics Gaedigk, A., Sangkuhl, K., Whirl-Carrillo, M., Twist, G. P., Klein, T. E., Miller, N. A., PharmVar Steering Committee 2018

  View details for PubMedID 30536702

• Clinical Pharmacogenetics Implementation Consortium (CPIC) guideline for the use of potent volatile anesthetic agents and succinylcholine in the context of RYR1 or CACNA1S genotypes. Clinical pharmacology and therapeutics Gonsalves, S. G., Dirksen, R. T., Sangkuhl, K., Pulk, R., Alvarellos, M., Vo, T., Hikino, K., Roden, D., Klein, T., Mark Poler, S., Patel, S., Caudle, K. E., Gordon, R., Brandom, B., Biesecker, L. G. 2018

  Abstract

  The identification in a patient of one of the 50 variants in the RYR1 or CACNA1S genes reviewed here should lead to a presumption of malignant hyperthermia susceptibility. Malignant hyperthermia susceptibility can lead to life-threatening reactions to potent volatile anesthetic agents or succinylcholine. We summarize evidence from the literature supporting this association and provide therapeutic recommendations for the use of these agents in patients with these RYR1 or CACNA1S variants (updates at https://cpicpgx.org/guidelines/ and www.pharmgkb.org). This article is protected by copyright. All rights reserved.

  View details for PubMedID 30499100

• The International SSRI Pharmacogenomics Consortium (ISPC): a genome-wide association study of antidepressant treatment response. Translational psychiatry Biernacka, J. M., Sangkuhl, K., Jenkins, G., WHALEY, R. M., Barman, P., Batzler, A., Altman, R. B., Arolt, V., Brockmöller, J., Chen, C. H., Domschke, K., Hall-Flavin, D. K., Hong, C. J., Illi, A., Ji, Y., Kampman, O., Kinoshita, T., Leinonen, E., Liou, Y. J., Mushiroda, T., Nonen, S., Skime, M. K., Wang, L., Baune, B. T., Kato, M., Liu, Y. L., Praphanphoj, V., Stingl, J. C., Tsai, S. J., Kubo, M., Klein, T. E., Weinshilboum, R. 2016; 6 (11)

  View details for DOI 10.1038/tp.2016.187

  View details for PubMedID 27801898

  View details for PubMedCentralID PMC5314112

• Pharmacogenetic allele nomenclature: International workgroup recommendations for test result reporting. Clinical pharmacology & therapeutics Kalman, L. V., Agúndez, J., Appell, M. L., Black, J. L., Bell, G. C., Boukouvala, S., Bruckner, C., Bruford, E., Caudle, K., Coulthard, S. A., Daly, A. K., Tredici, A. D., den Dunnen, J. T., Drozda, K., Everts, R. E., Flockhart, D., Freimuth, R. R., Gaedigk, A., Hachad, H., Hartshorne, T., Ingelman-Sundberg, M., Klein, T. E., Lauschke, V. M., Maglott, D. R., McLeod, H. L., McMillin, G. A., Meyer, U. A., Müller, D. J., Nickerson, D. A., Oetting, W. S., Pacanowski, M., Pratt, V. M., RELLING, M. V., Roberts, A., Rubinstein, W. S., Sangkuhl, K., Schwab, M., Scott, S. A., Sim, S. C., Thirumaran, R. K., Toji, L. H., Tyndale, R. F., van Schaik, R., Whirl-Carrillo, M., Yeo, K., Zanger, U. M. 2016; 99 (2): 172-185

  Abstract

  This article provides nomenclature recommendations developed by an international workgroup to increase transparency and standardization of pharmacogenetic (PGx) result reporting. Presently, sequence variants identified by PGx tests are described using different nomenclature systems. In addition, PGx analysis may detect different sets of variants for each gene, which can affect interpretation of results. This practice has caused confusion and may thereby impede the adoption of clinical PGx testing. Standardization is critical to move PGx forward.

  View details for DOI 10.1002/cpt.280

  View details for PubMedID 26479518

  View details for PubMedCentralID PMC4724253

• PharmGKB summary: venlafaxine pathway PHARMACOGENETICS AND GENOMICS Sangkuhl, K., Stingl, J. C., Turpeinen, M., Altman, R. B., Klein, T. E. 2014; 24 (1): 62-72

  View details for DOI 10.1097/FPC.0000000000000003

  View details for Web of Science ID 000328629800008

  View details for PubMedID 24128936

• PharmGKB: UNDERSTANDING THE EFFECTS OF INDIVIDUAL GENETIC VARIANTS DRUG METABOLISM REVIEWS Sangkuhl, K., Berlin, D. S., Altman, R. B., Klein, T. E. 2008; 40 (4): 539-551

  Abstract

  The Pharmacogenetics and Pharmacogenomics Knowledge Base (PharmGKB: http://www.pharmgkb.org) is devoted to disseminating primary data and knowledge in pharmacogenetics and pharmacogenomics. We are annotating the genes that are most important for drug response and present this information in the form of Very Important Pharmacogene (VIP) summaries, pathway diagrams, and curated literature. The PharmGKB currently contains information on over 500 drugs, 500 diseases, and 700 genes with genotyped variants. New features focus on capturing the phenotypic consequences of individual genetic variants. These features link variant genotypes to phenotypes, increase the breadth of pharmacogenomics literature curated, and visualize single-nucleotide polymorphisms on a gene's three-dimensional protein structure.

  View details for DOI 10.1080/03602530802413338

  View details for Web of Science ID 000260325500002

  View details for PubMedID 18949600

  View details for PubMedCentralID PMC2677552

• Identification of receptors and signaling pathways for orphan bone morphogenetic protein/growth differentiation factor ligands based on genomic analyses JOURNAL OF BIOLOGICAL CHEMISTRY Mazerbourg, S., Sangkuhl, K., Luo, C. W., Sudo, S., Klein, C., Hsueh, A. J. 2005; 280 (37): 32122-32132

  Abstract

  There are more than 30 human transforming growth factor beta/bone morphogenetic protein/growth differentiation factor (TGFbeta/BMP/GDF)-related ligands known to be important during embryonic development, organogenesis, bone formation, reproduction, and other physiological processes. Although select TGFbeta/BMP/GDF proteins were found to interact with type II and type I serine/threonine receptors to activate downstream Smad and other proteins, the receptors and signaling pathways for one-third of these TGFbeta/BMP/GDF paralogs are still unclear. Based on a genomic analysis of the entire repertoire of TGFbeta/BMP/GDF ligands and serine/threonine kinase receptors, we tested the ability of three orphan BMP/GDF ligands to activate a limited number of phylogenetically related receptors. We characterized the dimeric nature of recombinant GDF6 (also known as BMP13), GDF7 (also known as BMP12), and BMP10. We demonstrated their bioactivities based on the activation of Smad1/5/8-, but not Smad2/3-, responsive promoter constructs in the MC3T3 cell line. Furthermore, we showed their ability to induce the phosphorylation of Smad1, but not Smad2, in these cells. In COS7 cells transfected with the seven known type I receptors, overexpression of ALK3 or ALK6 conferred ligand signaling by GDF6, GDF7, and BMP10. In contrast, transfection of MC3T3 cells with ALK3 small hairpin RNA suppressed Smad signaling induced by all three ligands. Based on the coevolution of ligands and receptors, we also tested the role of BMPRII and ActRIIA as the type II receptor candidates for the three orphan ligands. We found that transfection of small hairpin RNA for BMPRII and ActRIIA in MC3T3 cells suppressed the signaling of GDF6, GDF7, and BMP10. Thus, the present approach provides a genomic paradigm for matching paralogous polypeptide ligands with a limited number of evolutionarily related receptors capable of activating specific downstream Smad proteins.

  View details for DOI 10.1074/jbc.M504629200

  View details for Web of Science ID 000231794800012

  View details for PubMedID 16049014