Janice JN Goh, Xing Yi Woo
Bioinformatics Institute (BII), Agency for Science, Technology and Research (A*STAR), 30 Biopolis Street, #07-01 Matrix, Singapore 138671, Republic of Singapore
Objectives: Despite making up more than 60% of the world’s population, Asian genomes are under-characterized. A large-scale whole genome sequencing initiative in Singapore, SG10K, has given us better insight into the pharmacogenomic (PGx) polymorphisms that are enriched in our population.1 Currently, clear guidance for dose adjustments in Asians with PGx polymorphisms has yet to be developed. However, a typical hospital formulary in Singapore consists of over 1000 medications,2 and pharmacokinetic studies are resource-intensive, making it almost impossible to study all drugs. A framework for prioritizing drug-gene interactions to develop Asian-specific guidance for drug dosing in the context of pharmacogenomics is thus needed. In this study, we focused on prioritizing chemotherapy drugs, the majority of which have a narrow therapeutic window, that were likely to be impacted by pharmacogenomic variants in the Singaporean population. A data integration approach to put together drug disposition and genomic information was developed to evaluate the impact of these drug-gene interactions. Having such dosing guidance will be important to ensure that pharmacogenomic testing will be accompanied by implementable clinical guidance.
Methods: A database consisting of chemotherapy drugs from the Cancer Drug List,3 each drug’s major metabolism/transporter effects4 and functional variant frequencies of all annotated star alleles from SG10K1 were compiled. Biologics were excluded from this analysis as most of the pharmacogenes of interest would not impact their drug disposition.5 Drug-gene pairs were prioritized by having a major metabolism/transporter pathway where at least 10% of the Singaporean population had a functional variant. Pharmacokinetic (PK) models that have PGx covariates described were then used for Monte Carlo simulations at the recommended clinical dose. PK simulations that showed a significant change in area under the curve (AUC) of > 30% between wild type and variant highlighted that the drug-gene pair was important for further study. Alternative doses were also simulated to give dosing recommendations.
In our Tamoxifen case study, the impact of CYP2D6 polymorphisms on active metabolite endoxifen exposure was evaluated using Monte Carlo simulations via a published joint parent-metabolite population PK model6. Doses of tamoxifen 20mg and 40mg once daily via oral administration were simulated. A virtual population was designed using patient demographics of Singaporean breast cancer patients, including covariates for age and body weight.7 Each population was further tested with a different CYP2D6 polymorphism (wild type, poor, intermediate and rapid metabolizer) (total 6000 virtual patients). A total of 1,000 simulations per virtual patient was performed in NONMEM 7.5.1.
Results: A total of 205 drugs and their metabolism/transport pathways from the Cancer Drug List and 13 genes known to impact drug disposition with reported variant frequencies were compiled. Of the 13 genes, 6 of them, CYP2D6, CYP2B6, UGT1A1, SLCO1B1, NUDT15, and CYP2C9 were identified as having a proportion of variants > 10% of the population. They were responsible for major metabolism/transport pathways in the list of cancer drugs, making them pathways of interest. We identified 8 chemotherapy drugs that were impacted via these pathways, namely, tamoxifen, cyclophosphamide, doxorubicin, irinotecan, darolutamide, docetaxel, mercaptopurine, and erdafitinib.
Our tamoxifen PK simulations show CYP2D6 poor and intermediate metabolizers have a decrease in median endoxifen AUC by 4.62 and 1.65 fold respectively, indicating tamoxifen CYP2D6 was an important drug gene pair. Further simulations of 40mg tamoxifen suggest CYP2D6 intermediate metabolizers could benefit from an increased dose of tamoxifen to achieve a similar endoxifen AUC to CYP2D6 wild-type. In contrast, poor metabolizers should switch to alternative therapy.
Conclusion: This framework describes a systematic method of evaluating clinically important drug-gene interactions, and highlights the utility of using population PK models to evaluate these interactions and recommend appropriate dose adjustments, before conducting actual clinical studies.
- Wu, D. et al. Large-Scale Whole-Genome Sequencing of Three Diverse Asian Populations in Singapore. Cell 179, 736-749.e15 (2019).
- National Drug Formulary (NDF). https://www.ndf.gov.sg/.
- MOH | Cancer Drug List. https://www.moh.gov.sg/home/our-healthcare-system/medishield-life/what-is-medishield-life/what-medishield-life-benefits/cancer-drug-list.
- Lexicomp® Drug Interactions – UpToDate. https://www.uptodate.com/drug-interactions/?source=responsive_home#di-druglist.
- Zhao, L., Ren, T. & Wang, D. D. Clinical pharmacology considerations in biologics development. Acta Pharmacol. Sin. 33, 1339–1347 (2012).
- Mueller-Schoell, A. et al. Obesity Alters Endoxifen Plasma Levels in Young Breast Cancer Patients: A Pharmacometric Simulation Approach. Clin. Pharmacol. Ther. 108, 661–670 (2020).
- Lal, S. et al. Influence of ABCB1 and ABCG2 polymorphisms on doxorubicin disposition in Asian breast cancer patients. Cancer Sci. 99, 816–823 (2008).
Reference: PAGE 32 (2024) Abstr 10776 [www.page-meeting.org/?abstract=10776]
Poster: Clinical Applications