High BMI is associated with lower TNF-α inhibitor serum trough levels and higher disease activity in patients with axial spondyloarthritis

BACKGROUND

TNF-α inhibitor (TNFi) serum trough levels have previously been found to be related to disease activity in axial spondyloarthritis (axSpA). However, most research regarding serum trough levels has been conducted in patients who only recently started TNFi therapy. Therefore, our objective was to explore TNFi serum trough level measurements in relation to disease activity and BMI in the total axSpA population in daily clinical practice, also including patients on long-term TNFi therapy.

METHODS

Consecutive patients from the Groningen Leeuwarden Axial Spondyloarthritis (GLAS) cohort were approached for a TNFi serum trough level measurement during their regular outpatient visit at the UMCG. Spearman's correlation coefficient was used to analyse the relation of serum trough levels with disease activity and BMI. Logistic regression was performed to analyse the relation between therapeutic drug levels and disease activity, corrected for potential confounders, including BMI.

RESULTS

Thirty-four patients on adalimumab and 21 patients on etanercept were included. Mean age was 45 ± 12 years, 47% were male, median BMI was 26.4 (IQR 23.9-32.5) and median treatment duration was 41 months (range 2-126). According to definitions of Sanquin, 47% of patients had therapeutic serum trough levels. No significant correlations were found between TNFi levels and disease activity (ASDAS-CRP: adalimumab: ρ = -0.16, p = 0.39; etanercept: ρ = -0.29, p = 0.20). TNFi levels were moderately correlated with BMI (adalimumab: ρ = -0.48, p = 0.004; etanercept: ρ = -0.50, p = 0.021). Patients with active disease (ASDAS ≥ 2.1) showed higher BMI than patients with inactive disease (median 29.7 vs. 24.6, p = 0.015). In multivariable regression analyses, BMI was identified as the only confounder for the relationship between therapeutic drug levels and ASDAS.

CONCLUSION

In this cross-sectional, observational study of axSpA patients mainly on long-term treatment with TNFi, higher BMI was significantly associated with lower adalimumab and etanercept serum trough levels and higher disease activity.

An optimized method for plasma extracellular vesicles isolation to exclude the copresence of biological drugs and plasma proteins which impairs their biological characterization

Extracellular vesicles (EVs) are cell membrane-derived phospholipid bilayer nanostructures that contain bioactive proteins, enzymes, lipids and polymers of nucleotides. They play a role in intercellular communication and are present in body fluids. EVs can be isolated by methods like ultracentrifugation (UC), polyethylene-glycol-precipitation (PEG) or size exclusion chromatography (SEC). The co-presence of immunoglobulins (Ig) in EV samples isolated from plasma (pEVs) is often reported and this may influence the assessment of the biological function and phenotype of EVs in bio- and immunoassay. Here, we studied the presence of an Ig-based therapeutic (etanercept) in pEV samples isolated from rheumatoid arthritis (RA) patients and improved the isolation method to obtain purer pEVs. From plasma of etanercept (Tumor-necrosis-factor (TNF)-α antibodies)-treated RA patients pEVs were isolated by either UC, PEG or SEC. SEC isolated pEVs showed the highest particle-to-protein ratio. Strong TNF-α inhibition determined in a TNF-α sensitive reporter assay was observed by pEVs isolated by UC and PEG, and to a lesser extent by SEC, suggesting the presence of functional etanercept. SEC isolation of etanercept or labelled immunoglobulin G (IgG) showed co-isolation of these antibodies in the pEV fraction in the presence of plasma or a high protein (albumin) concentration. To minimize the presence of etanercept or immunoglobulins, we extended SEC (eSEC) column length from 56mm to 222mm (total stacking volume unchanged). No effect on the amount of isolated pEVs was observed while protein and IgG content were markedly reduced (90%). Next, from six etanercept- treated RA patients, pEVs were isolated on a eSEC or standard SEC column, in parallel. TNF-α inhibition was again observed in pEVs isolated by conventional SEC but not by eSEC. To confirm the purer pEVs isolated by eSEC the basal IL-8 promoter activation in human monocytes was determined and in 4 out of 5 SEC isolated pEVs activation was observed while eSEC isolated pEVs did not. This study shows that extended SEC columns yielded pEVs without detectable biologicals and with low protein and IgG levels. This isolation method will improve the characterization of pEVs as potential biomarkers and mediators of disease.

Application of nSMOL coupled with LC-MS bioanalysis for monitoring the Fc-fusion biopharmaceuticals Etanercept and Abatacept in human serum

The principle of nano-surface and molecular-orientation limited (nSMOL) proteolysis has a unique characteristic Fab-selective proteolysis for antibody bioanalysis that is independent of a variety of monoclonal antibodies by the binding antibody Fc via Protein A/G in a pore with 100 nm diameter and modified trypsin immobilization on the surface of nanoparticles with 200 nm diameter. Since minimizing peptide complexity and protease contamination while maintaining antibody sequence specificity enables a rapid and broad development of optimized methods for liquid chromatography-mass spectrometry (LC-MS) bioanalysis, the application of regulatory LC-MS for monitoring antibody biopharmaceuticals is expected. nSMOL is theoretically anticipated to be applicable for representative Fc-fusion biopharmaceuticals, because Protein A/G-binding site Fc exists on the C-terminus, and its functional domain is available to orient and interact with the reaction solution. In this report, we describe the validated LC-MS bioanalysis for monitoring Ethanercept and Abatacept using nSMOL technology. The quantitation range of Ethanercept in human serum was from 0.195 to 100 μg/mL using the signature peptide VFCTK (aa.43-47), and that of Abatacept was from 0.391 to 100 μg/mL using the signature peptide MHVAQPAVVLASSR (aa.1-14). Both proteins fulfilled the guideline criteria for low-molecular-weight drug compounds. The results indicate that the clinical and therapeutic monitoring for antibody and Fc-fusion biopharmaceuticals are adequately applicable using nSMOL proteolysis coupled with LC-MS bioanalysis.