P136 SLEDAI response prediction to belimumab therapy by baseline levels of BLyS, APRIL and CD8+ effector memory T-cells

Background Systemic Lupus Erythematosus (SLE) patients show high levels of BLyS (B-lymphocyte stimulator) and other cytokines belonging to the tumor necrosis factor (TNF) superfamily¹. Belimumab, a monoclonal antibody against BLyS, mainly targets B-cells², but a BLyS-dependent T-cell activation pathway has been demonstrated³. Clinical studies showed that anti-DNA antibodies and complement levels at baseline are predictors of response to Belimumab. Our study aims at identifying other biomarkers as response predictors.

Methods Twenty-one SLE patients received Belimumab. Biomarkers belonging to the TNF superfamily (BLyS, APRIL, sBCMA, sCD40L, sTACI, TWEAK) were tested by ELISA in all patients and lymphocyte immunophenotyping was performed by flow cytometry in ten subjects at baseline and every six months. SLE-disease activity was assessed by SLEDAI-2K score.

Results BLyS and APRIL baseline serum levels and the number of CD3+CD8+ effector memory T-cells were correlated positively with SLEDAI-2K improvement after 12 months of treatment (Pearson correlation=0.535 (p=0.015), 0.504 (p=0.023) and 0.654 (p=0.040)). After backwards exclusion from linear regression analysis including SLEDAI-2K, effector T-cell relative number and BLyS or APRIL at baseline, only APRIL remained as significant independent predictor of SLEDAI-2K improvement after 12 months (adjusted R square=0.649; p=0.025). After comparing TNF-family members levels and SLEDAI-2K at baseline, only BLyS showed the best predictive value (adjusted R 0.564, p<0.001).

Conclusions In our cohort of SLE patients, baseline level of APRIL together with percentage of CD3+CD8+ effector memory T-cells, or BLyS serum level alone resulted as best predictors of response to Belimumab. Considering that immunophenotyping is rarely done in clinical practice, BLyS baseline serum levels alone could be used routinely as a good predictor of response, as suggested by post-hoc analyses of the BLISS study.

References

1. Ng LG, et al. J Immunol, 2004.

2. Ramsköld D, et al. EBioMedicine, 2019.

3. Samy E, et al. Int Rev Immunol, 2017.