Predictive genetic biomarkers for the efficacy of methotrexate in rheumatoid arthritis: a systematic review.

Multiple pharmacogenetic studies investigated the effectiveness of methotrexate. However, due to the use of nonvalidated outcomes, lack of validation or conflicting results it remains unclear if genetic markers can help to predict response to MTX treatment. Therefore, a systematic review was performed. PubMed was searched for articles reporting potential pharmacogenetic biomarkers associated (p < 0.05) with MTX efficacy using the validated endpoints DAS(28), EULAR, or ACR response criteria. The PICO method was used for study selection, and PRISMA guidelines to prepare the report. Thirty-five studies met the inclusion criteria, providing 39 potential genetic biomarkers in 19 genes. After Bonferroni correction, six genetic biomarkers were associated with the efficacy of MTX: ATIC rs7563206; SLC19A1 rs1051266; DHFR rs836788; TYMS rs2244500, rs2847153, and rs3786362 in at least one study. Only SLC19A1 rs1051266 was replicated in an independent cohort and promising for predicting methotrexate efficacy.

Validation of a clinical pharmacogenetic model to predict methotrexate nonresponse in rheumatoid arthritis patients.

AIM: To study the performance of a clinical pharmacogenetic model for the prediction of nonresponse in rheumatoid arthritis (RA) patients treated with methotrexate (MTX) in combination with other synthetic or biologic disease-modifying anti-rheumatic drugs . This prediction model includes gender, smoking status, rheumatoid factor positivity and four genetic variants in AMPD1 (rs17602729), ATIC (rs2372536), ITPA (rs1127354) and MTHFD1 (rs17850560). METHODS: A total of 314 RA patients from three Dutch studies were retrospectively included. Eligible patients were adults diagnosed with RA and had a treatment duration with MTX and follow-up for at least two study evaluation visits. Prediction model risk scores at the first and second evaluation were calculated and compared with the actual nonresponse (disease activity score >2.4). Regression and receiver operating characteristic curve analyses of the prediction model were performed. Also, the sensitivity, specificity and the positive and negative predictive values (PPV and NPV) were determined. RESULTS: The receiver operating characteristic area under the curve was 75% at first and 70% after second evaluation. At the second evaluation, prediction nonresponse had a sensitivity of 67% (CI: 54-78%), specificity of 69% (CI: 60-77%), PPV of 52% (CI: 45-60%) and NPV of 80% (CI: 73-85%). CONCLUSIONS: This study demonstrates that the clinical pharmacogenetic model has an inadequate performance for the prediction of nonresponse to MTX in RA patients treated with combination therapies.