4 Discussion
The effectiveness of rituximab depends on several factors, including its activity and concentration in the environment of the tumor. To maximize therapeutic efficacy and to optimize dosing, a reliable assay for accurate measurement of rituximab levels in the bloodstream of patients is needed.18 Here, we show that serum rituximab levels vary highly among patients, who received standard doses of rituximab. In different studies, higher serum rituximab levels correlated with more favorable clinical response.19 In addition, initially measured tumor burden was reported to inversely correlate with serum rituximab levels.20 We show here that patients with limited disease (risk groups R1 and R2) have higher serum rituximab levels compared to patients of advanced disease (risk groups R3 and R4) at day five after rituximab administration. This correlates with previous findings of significantly better rituximab response in patients of risk group R3/R4.21 The advanced disease stage in R3/R4 patients translates into more available CD20 epitopes, where rituximab can attach. Consequently, less unbound rituximab molecules can be measured in the R3/R4 patient’s serum. This rituximab ELISA is a reliable method to measure serum rituximab levels on a routinely basis in the future and can easily be implemented in daily routine.
ADCC, mediated by FcγRs of NK cells, is thought to be the main rituximab mechanism of action.22 FcγRI expression on classical monocytes was shown to be increased at day five after rituximab treatment, while FcγRII expression significantly decreased. FcyRII can consist of two isoforms, FcyRIIa and FcyRIIb.23 The antibody binding to FcγRII used for these experiments binds against both isoforms, FcγRIIa and FcγRIIb. However, both isoforms have different functions in B cell depletion, FcγRIIa being activatory on target effector cells and FcγRIIb being inhibitory.23 It is tempting to speculate that monocytes expressing the activatory FcγRIIa might be exhausted at day five after rituximab treatment. In DLBCL patients, low FcγRIIa expression levels at diagnosis were described to lead to impaired rituximab-mediated cytotoxicity of NK cells.24 Also, expression of inhibitory FcγRIIb might be reduced on the circulating effector cells after treatment since this Fc receptor type was described to negatively influence rituximab-mediated B cell depletion.23
FcγRI is the main receptor that mediates trogocytosis, describing the process where effector cells remove rituximab-CD20 complexes, which can lead to rituximab resistance but also to target cell death.25 The mechanism was described to be less effective in rituximab mediated B cell depletion but is initiated when other mechanisms are exhausted.26 This supports the upregulated expression of FcγRI at day five after treatment with downregulated FcγRII expression, due to exhaustion.
Although being the main receptor mediating ADCC, expression of FcγRIII remained unchanged in monocytes and NK cells. This might indicate that FcγRIII mediated B cell depletion via ADCC or ADP is not exhausted at day five after rituximab treatment and plays a role in an earlier time of the treatment. FcγR expression of neutrophils did not change, probably because neutrophils are not the predominant effector cells at day 5 after treatment. In neutrophils, only FcγRIIIb is expressed on the surface, rather mediating phagocytosis than ADCC.27
S100A9 and S100A12 belong to the S100 family and act via Toll-like receptors (TLRs) and receptors for advanced glycation end products (RAGEs).28-30 Expression of S100A9 is highest in monocytes and neutrophils, while S100A12 is mainly expressed by neutrophils.31,32 High serum levels of S100A8, S100A9 and S100A12 are associated with enhanced proinflammatory activity, predominantly found in patients with inflammatory or autoimmune diseases.33,34 Therefore, S100A9 and S100A12 can be used as valuable serum inflammatory markers.33,34Furthermore, S100A9 could be used as a prognostic factor of survival in advanced ENKL patients.35 Additionally, a combination of serum S100A8/A9 and S100A12 was shown to predict rituximab treatment response in lupus nephritis.36 The expression of the proinflammatory proteins S100A9 and S100A12 was found to be significantly increased in monocytes and neutrophils at day five after rituximab treatment. Conversely, S100A8/A9 and S100A12 serum levels significantly decreased and therefore secretion by effector cells is lower at day five after treatment. Serum S100A8/A9 and S100A12 levels are similar to healthy individuals after rituximab treatment, concluding they are slightly increased before treatment indicating signs of inflammation.37,38 Increase in intracellular expression is accompanied with decrease in secretion of S100A9 and S100A12 by effector cells and vice versa, probably depicting the location of the proteins at the measured timepoint.
Among others, S100 proteins may, directly or indirectly, lead to recruitment and activation of NK cells that are the main effector cells that deplete B cells. It will be interesting to examine these mechanisms in future studies to point to new possibilities of activation and inhibition.
CD57+ cells express more perforin (PRF1) and granzyme B (GrzB), which are crucial for lytic activity of NK cells and are primed effectors for ADCC.39 Interestingly, expression of CD57, PRF1 and GrzB was significantly decreased at day five after rituximab administration. In other studies, high expression of CD57, PRF1 and GrzB correlated with highly active NK cells.39,40 Since GrzB and PRF1 both predominantly contribute to ADCC and we identified a decrease in expression, it is tempting to speculate that ADCC by NK cells might be the main mechanism of action in patients suffering from B-NHL.40 This could indicate that the NK cell population with high B cell depletion activity might be exhausted at day five after rapid NK cell recruitment in the first days after rituximab administration. By measuring the lytic activity of NK cells from patient samples, this effect could be studied further. However, the predominant NK cell population after rituximab treatment might be less mature since expression of GrzB and PRF1 directly correlates with cellular maturity.41 Additionally, different mechanisms of action may play a role at various timepoints. In future studies, it would be feasible to monitor the expression after administration of rituximab over time to improve rituximab therapy and to identify new diagnostic markers.