Background ABCB1 (P-gp) and ABCG2 (BCRP1) are part of the adenine triphosphate (ATP)-binding cassette (ABC) transporter proteins. These proteins mediate efflux of drugs from inside the cells and can confer a multidrug resistance phenotype. Drugs included among its substrates are indicated in RA patients. Accordingly, an association between disease activity and ABCB1 and ABCG2 function might be hypothesized.
Objectives The study was undertaken to compare the drug efflux function of both transporters in patients with active and inactive RA and to define the impact of disease activity on transporter function.
Methods We included 17 active RA patients (DAS28≥3.2), paired (age, gender and disease duration) to 17 patients in remission (DAS28<2.6), from an early RA cohort. All patients had a complete medical evaluation and serum sample obtained at study inclusion and 27 of them had an additional sample and evaluation at six-month follow-up.
ABCB1 and ABCG2 functional activity was measured in peripheral mononuclear cells by flow cytometry. The percentage of cells able to extrude substrates for ABCB1 (daunorubicin) and ABCG2 (mitoxantrone) were recorded in the presence or absence of the selective ABCB1 (verapamil) or ABCG2 (KO143) inhibitors. Efflux activity was defined as a percentage greater than the mean in 30 healthy controls plus two standard deviations, i.e. 1.67% (range 0.26% to 1.98%).
The study was approved by our internal review board.
For statistical analysis appropriate tests were used according to variable distribution and linear regression model to establish the association between disease activity and transporter's function.
Results Data from 34 patients (94.1% women) were analyzed. Their mean age was 41.6±11 years and disease duration was 6.3±3.5 years; 85.3% had RF and 97.5% ACCP. Patients with active disease ([mean ± SD] DAS28: 4.8±1.3) had significantly higher cumulative corticosteroids doses than inactive patients ([mean ± SD] DAS28: 1.2±0.6). The formers had also higher efflux function of both transporters: median (25-75 IQR) ABCB1: 7.1% (1.4-29.3) vs. 1.6% (0.7-3.5), p=0.02 and ABCG2 6.2% (1.3-22.4) vs. 1.3% (0.7-2), p=0.007. Percentage of cells able to extrude daunorubicin or mitoxantrone correlated with DAS28: rho=0.45, p=0.008 for ABCB1 and rho=0.52, p=0.002, for ABCG2; no correlation was found with cumulative corticosteroids or other treatment.
Linear regression model applied to the 34 patients at baseline showed DAS28 as the only predictor of both ABCB1 (beta coefficient: 0.50; 95%CI: 1.6-6.7; p=0.002; R2=0.23) and ABCG2 (beta coefficient: 0.48; 95CI%: 1.4-6.8; p=0.004; R2=0.21) function.
Finally, disease activity increased in 14.8% of the patients at 6 month follow-up, while 48.1% remained stable and 37.1% improved. DAS28 changes at this time correlated with shift in ABCB1 (r=0.35, p=0.07) and ABCG2 (r=0.33, p=0.09) function.
Conclusions Patients with active RA have a higher efflux function of ABCB1 and ABCG2 compared with those in remission at baseline. What seems to be conclusive is that ABCB1 and ABCG2 behavior is conditioned by disease activity.
Disclosure of Interest None declared