- Title
- Pharmacogenomic factors influencing individual responses to drugs used in rheumatoid arthritis in a community cohort of older Australians
- Creator
- Dias, Nanayakkara Wasam Galgodellage Thilani Hasanthi
- Relation
- University of Newcastle Research Higher Degree Thesis
- Resource Type
- thesis
- Date
- 2020
- Description
- Research Doctorate - Doctor of Philosophy (PhD)
- Description
- Responses to anti-rheumatic drugs and analgesics such as opioids and non-steroidal anti-inflammatory drugs (NSAIDs) may be affected by drug-gene interactions (DGIs), drug-drug interactions (DDIs) and the combined effect of DGIs and DDIs i.e. multifactorial drug-gene interactions (multifactorial DGIs). The overarching hypothesis of the study was that pre-emptive genotyping and information about drug and gene interactions may assist in improving medication safety and effectiveness for RA patients by identifying individuals at high risk of therapeutic failure or adverse drug reactions. Prevalence of potential drug and gene interactions for drugs used for the treatment of RA was assessed retrospectively in the Hunter Community Study (HCS) of older Australians aged 55 years or above and residing in the catchment area of the Hunter New England Area Health Service, Newcastle, Australia. Self-reported medication history was available for 2642 HCS participants. Genotype data for 2121 HCS participants were imputed using 1000 Genomes, Hap Map Phase II European and Haplotype Reference Consortium v1.1 reference panels from Affymetrix Axiom Kaiser DNA microarray data. Clinically significant DDIs relevant to the anti-rheumatic drugs of interest (methotrexate, azathioprine), and analgesics of interest (the opioids codeine, tramadol and oxycodone and the NSAIDs celecoxib and diclofenac) were identified from the Monthly Index of Medical Specialties (MIMS), FDA clinical CYP tables and the clinically relevant CYP450 Drug Interaction table compiled by Flockhart (Indiana University). Clinically significant DGIs with strong or moderate evidence levels relevant to the drugs of interest were identified from the Pharmacogenomics Knowledge Base (PharmGKB) and, where available, from the Clinical Pharmacogenetics Implementation Consortium (CPIC) guidelines and, for clinical interactions for celecoxib, the United States Food and Drug Administration (FDA). With regard to the two anti-rheumatic drugs with strong or moderate evidence for clinically significant DGIs that were used in the cohort, azathioprine and methotrexate, 58% (95% CI: 56%-60%) of HCS participants had clinically significant thiopurine S-methyltransferase genotypes for azathioprine (TPMT rs1142345, rs1800460) or methylenetetrahydrofolate reductase genotypes for methotrexate (MTHFR rs1801133) or both, which are predicted to increase the toxicity of these drugs. Only three participants were using azathioprine but of the 19 participants taking methotrexate, ~74% (95% CI: 54%-93%) were at risk of at least one potential severe or moderate DDI relating to methotrexate, with 1.6 mean (SD± 0.8) possible interactions per participant. None of the 12 methotrexate users with relevant genotype data was predicted to have potential simple DGIs but ~ 42% (95% CI: 14%-70%) had potential multifactorial DGIs, placing them at risk of more serious toxicity. Linkage data for hospital admissions, the Pharmaceutical Benefits scheme (PBS) and mortality were used to assess possible clinical effects of DGIs in the small group of participants taking methotrexate. The data showed that heterozygous or homozygous variant genotypes of MTHFR rs1801133 did not significantly corelate with whether the participants experienced potential serious methotrexate toxicity leading to hospital admission (Fisher’s exact one tailed test p value of 0.05). Also, there was no evidence of a relationship with potential methotrexate intolerance, as assessed indirectly from PBS data (p=0.6). Future studies are required with larger sample sizes and more informative clinical and PBS data. For opioids used in the cohort, there was strong or moderate evidence of clinically significant DGIs for codeine-cytochrome P450 family 2 subfamily D member 6 (CYP2D6) gene variants (rs16947, rs769258, rs1065852, rs1135840, rs3892097, rs28371717 and rs28371725) and tramadol-CYP2D6 and oxycodone-CYP2D6 gene variants (e.g. rs3892097). Approximately 37% (95% CI: 35%-39%) of the HCS participants had clinically significant CYP2D6 poor or intermediate metaboliser genotypes that may increase risks of therapeutic failure for codeine, tramadol or oxycodone. Although no participants using tramadol or oxycodone had these genotypes, there were 96 codeine users, of whom 54 had relevant genotype data and of these, ~30% (95% CI: 17%-42%) had potential simple DGIs and 9% (95% CI: 2%-17%) had potential multifactorial DGIs. The results may underestimate the prevalence of clinically significant genotypes affecting opioids as no data were available for CYP2D6 deletion variants, which can also increase risk of therapeutic failure, or copy number variants (ultra-rapid metaboliser genotypes), which can increase risk of toxicity. Of the NSAIDs used by study participants, there are potentially clinically significant interactions relating to gastrointestinal (GI) bleeding risk for celecoxib and diclofenac with CYP2C9 variant rs1057910. The total prevalence of genotypes potentially increasing GI bleeding risks (CYP2C9*1/*3 or *3/*3) for these NSAIDs was ~13% (95% CI: 11%-14%), with ~1% (95% CI: 0%-1%) having clinically significant poor metaboliser genotypes (CYP2C9*3/*3). Of the 137 celecoxib users in the study, 79 had genotype data, with predicted simple and multifactorial DGIs of 6% (95% CI: 1%-12%) and ~9% (95% CI: 3%-15%), respectively. There was a surprisingly high prevalence of GI bleeding or ulcers (8%, 95% CI: 3%-13%) among celecoxib users. Relationships between GI bleeding and simple or multifactorial DGIs could not be effectively assessed due to the small sample size of participants with risk genotypes but further studies in larger groups appear warranted. Overall across all drugs investigated, ~73% (95% CI: 71%-75%) of all participants with genotype data had clinically significant risk genotypes relevant to at least one gene variant of interest relevant to toxicity or therapeutic failure. Of the 136 participants with genotype data taking methotrexate, codeine or celecoxib, 27% (95% CI: 20%-35%) had potential simple or multifactorial DGIs with prevalences of 15% (95% CI: 9%-22%) and ~12% (95% CI: 6%-17%) respectively. In conclusion, this PhD study has provided evidence that approximately three quarters of RA patients are at risk of experiencing adverse drug reactions or therapeutic failure due to clinically significant gene variants relevant to anti-rheumatic drugs such as azathioprine or methotrexate or analgesics such as codeine, tramadol, oxycodone, celecoxib or diclofenac if they are prescribed the corresponding drug. The findings also suggest at least 10% of RA patients are at increased risk due to multifactorial DGIs. Identifying this restricted subset of patients for close monitoring and high levels of caution in prescription of high-risk interacting drugs may improve clinical outcomes without excessive clinical alert fatigue. Prospective, appropriately designed and powered clinical studies of RA patients are required to assess the therapeutic benefits and economic feasibility of taking drug and gene interactions into account in routine practice.
- Subject
- anti-rheumatic drugs; opioids; drug-gene interactions; older Australians
- Identifier
- http://hdl.handle.net/1959.13/1503868
- Identifier
- uon:55422
- Rights
- Copyright 2020 Nanayakkara Wasam Galgodellage Thilani Hasanthi Dias
- Language
- eng
- Full Text
- Hits: 140
- Visitors: 131
- Downloads: 11
Thumbnail | File | Description | Size | Format | |||
---|---|---|---|---|---|---|---|
View Details Download | ATTACHMENT01 | Thesis | 159 MB | Adobe Acrobat PDF | View Details Download | ||
View Details Download | ATTACHMENT02 | Abstract | 316 KB | Adobe Acrobat PDF | View Details Download |