Article Text
Abstract
Objective To assess the association between a single nucleotide polymorphism in the gene of FCGR3A and the response to treatment with rituximab (RTX) in rheumatoid arthritis (RA).
Methods SMART is a randomised open trial assessing two strategies of re-treatment in patients responding to 1 g infusion of RTX with methotrexate on days 1 and 15 after failure, intolerance or contraindication to tumour necrosis factor (TNF) blockers. Among the 224 patients included, 111 could be genotyped and were included in an ancillary study of SMART. Univariate and multivariate analyses adjusted on disease activity score on 28 joints were performed to assess whether FCGR3A-158V/F polymorphism was associated with European League Against Rheumatism response at week 24.
Results Among the 111 patients, 90 (81%) were responders of whom 30 (27%) were good responders. V allele carriage was significantly associated with a higher response rate (91% of responders vs 70%, OR 4.6 (95% CI 1.5 to 13.6), p=0.006). These results were also confirmed in rheumatoid factor-positive patients (93% vs 74%, p=0.025). In multivariate analysis, V allele carriage was independently associated with response to RTX (OR 3.8 (95% CI 1.2 to 11.7), p=0.023).
Conclusion The 158V/F polymorphism of FCGR3A seems to influence the response to RTX in patients with RA after failure, intolerance or contraindication to TNF blockers.
Statistics from Altmetric.com
The anti-CD20 monoclonal antibody that targets B cells, rituximab (RTX), is a chimeric immunoglobulin G1 (IgG1) monoclonal antibody and is an effective treatment of rheumatoid arthritis (RA) that is refractory to tumour necrosis factor (TNF) blockers.1
The non-synonymous 158 V/F polymorphism in FCGR3A (rs396991), which substitutes a valine (V) for a phenylalanine (F) at amino acid position 158, is known to affect both FcγRIIIa (CD16) expression, binding of RTX to FcγRIIIa and RTX-mediated cytotoxicity.
Some studies have reported that patients carrying the FCGR3A-VVgenotype had a higher biological and clinical response to RTX in the treatment of B lymphoproliferative malignancies2 3 and in cryoglobulinaemia associated with hepatitis C virus.4
The present study was carried out to evaluate the influence of the FCGR3 -158V/F polymorphism on the therapeutic response to RTX in a cohort of patients with RA who experienced an inadequate response, intolerance or with contraindication to TNF blockers.
Methods
Patients
The design of the study has already been described in detail elsewhere.5 Further details are given in the online supplement.
Briefly, a total of 224 patients with RA were included in the SMART study. This study is a 2-year national multicentre randomised open-label study evaluating the efficacy and tolerability of two doses of RTX for re-treatment after one course of RTX (1000 mg on days 1 and 15).
Each patient received a stable dose of methotrexate (MTX) (≥10 mg/week for at least 4 weeks) and had experienced an inadequate response or intolerance to TNF blockers or TNF blockers were contraindicated.
Study protocol
All patients received one course of RTX. Treatment efficacy was evaluated 24 weeks after the RTX first infusion according to the European League Against Rheumatism (EULAR) response.6
Data collected
Serum B cells markers such as rheumatoid factor (RF) and anti-cyclic citrullinated peptide (CCP) antibodies were collected before the first RTX infusion.
Genotyping of the FCGR3A-158V/F polymorphism was performed in patients who gave specific additional consent for this ancillary study of SMART.
Statistical analysis
Response rates were compared across FCGR3A genotypes using the χ2 test or the Fisher exact test, when appropriate. The relationship between the EULAR response at 24 weeks and explanatory variables was analysed by logistic regression. These variables were selected by univariate logistic regression among FCGR3A-158 V/F polymorphism carriage, patients and disease characteristics and B cell activation markers at baseline. Significant variables after univariate regression (p<0.15) were entered into a stepwise multivariate model adjusted to the disease activity score on 28 joints with C reactive protein use (DAS28-CRP). Results are expressed as the OR with 95% CI.
Results
Characteristics of the study population
Of the 224 patients with RA (age 56±11 years, disease duration 13±9 months, 84% women) who received one course of RTX, 209 had an inadequate response to anti-TNF and 16 had a contraindication to TNF blockers. Among these, 109 did not give their consent for the genetic search and four patients could not be genotyped for the FCGR3A genotype. Thus, 111 patients were analysed in the present study. The baseline characteristics of these patients according to FCGR3A genotypes are shown in table 1.
The FCGR3A genotyping was in Hardy–Weinberg equilibrium (p=0.314). The demographic characteristics did not differ among the different genotypes except for the subgroup of patients carrying VV genotype who were more often RF-positive in comparison with the FF genotype (93% vs 66%, p=0.05).
The patients included in the study had longer disease duration, more active disease and more often erosions (p=0.02, p=0.03 and p=0.04, respectively) in comparison with the patients without available DNA (see table S1 in online supplement).
EULAR response rates according to FCGR3A genotypes
Twenty-four weeks after the first course of RTX, 90 patients (81%) were responders of whom 30 (27%) were good responders; 10 patients (9%) were in remission according to the EULAR criteria (DAS28 <2.6) and 21 patients (19%) were in low disease activity (DAS28 2.6–3.2). The EULAR response rate 24 weeks after a first course of RTX according to the FCGR3A genotype is shown in figure 1.
Ninety-three percent of patients with FCGR3A-158 VV genotype and 91% of patients with FCGR3A-158 VF experienced a response (good or moderate) to the first course of RTX in comparison with a response rate of 70% in those with FF genotype (p=0.096 and p=0.011, respectively).
The proportion of good responders did not differ between the three genotypes (21% for VV, 32% for VF and 25% for FF genotypes, p>0.43)
EULAR response rates according to FCGR3A alleles
When performing allele analysis, FCGR3A-158 V allele carriage was significantly associated with a higher response rate (91% vs 70%, OR 4.6 (95% CI 1.5 to 13.6), p=0.006). These results were not significantly different in RF-positive and RF-negative patients (93% vs 74%, OR 4.5 (95% CI 1.1 to 18.2), p=0.025 in RF-positive patients and 87% vs 61%, OR 4.1 (95% CI 0.7 to 24.2), p=0.13 in RF-negative patients).
Multivariate analysis was performed to investigate whether V allele carriage was an independent factor of response to RTX by logistic regression where the response to RTX was the assessed outcome. Univariate analyses showed that baseline DAS28-CRP >5.1, oral corticosteroid therapy and high serum IgG level (>12.67 g/l) were associated with RTX response (p<0.15) and was then included in the model as potential confounding factors beside FCGR3A-158 V allele carriage. The results of univariate analysis and the logistic regression are presented in table 2. The final model confirmed that FCGR3A-158 V allele carriage increased the chances of response to RTX (OR 3.8 (95% CI 1.2 to 11.7), p=0.023).
Discussion
Biological disease-modifying antirheumatic drugs (DMARDs) have led to remarkable benefits for the treatment of RA in patients with an inadequate response to MTX or to a TNF inhibitor. After failure with a TNF inhibitor, there is evidence for the efficacy of switching to another TNF inhibitor or to another class of biological DMARD such as abatacept, RTX or tocilizumab.7 Since there are several therapeutic options with no clear differences in efficacy or safety concerns, clinicians need predictive factors for the response to a biological DMARD in an individual patient to make a tailored decision.8 In the present study we show that FCGR3A-158 V allele carriage is an independent factor associated with the response to a first course of RTX in patients who have had an inadequate response or a contraindication to a TNF inhibitor.
This non-synonymous 158 V/F polymorphism in FCGR3A gene was previously reported as being associated with response to RTX in patients with non-Hodgkin's lymphomas and in cryoglobulinaemia associated with hepatitis C virus in whom the homozygous FCGR3A-158V genotype was independently associated with clinical and molecular responses to a first course of RTX.2 4 The relationship between this functional single nucleotide polymorphism of the FCGR3A gene and the response to anti-TNF treatment was assessed in patients with RA, with discrepancies in the results.9,–,11 Preliminary data from a cohort of 57 patients with RA managed with RTX suggested that patients carrying the FCGR3A-158VV genotype had a better response based on American College of Rheumatology criteria than those carrying the FCGR3A-158VF or the FCGR3A-158FF genotype.12
Little is known about the basic mechanisms for the contribution of the FCGR3A-158V/F polymorphism to the response to RTX in patients with non-Hodgkin's lymphomas or RA. In these two pathological conditions the response to RTX appears to be better in patients carrying one or two copies of the FCGR3A-158 V allele. This amino acid residue directly interacts with the lower hinge region of IgG1. The variable response to RTX among FCGR3A polymorphic groups is likely to be the result of qualitative differences such as antibody affinity.2 Furthermore, the variability of response to RTX among FCGR3A polymorphic groups may result from quantitative differences by increasing FcγRIIIa expression.13 Recent work has suggested that low levels of FcγRIIIa/CD16 expression may predict non-response to RTX.14 Such qualitative and quantitative differences may explain the relationship of the FCGR3A genotype to the degree of B cell depletion by RTX observed in the treatment of systemic lupus eythematosus.15 Thus, quantitative and qualitative changes in FcγRIIIa/CD16 induced by FCGR3A genotype could influence B cell depletion and explain these differences in the response to RTX.
Although a previous study of the SMART trial found an association between RF positivity and the response to RTX,5 RF positivity was not associated with the EULAR response in the present ancillary study. In this sample of 111 patients the difference between patients with and without RF antibodies in terms of the EULAR response was only 11% (84% vs 73% of EULAR response) whereas, in the sample of 208 patients, this difference was about 21% (79% vs 58%). The small difference in this sample can explain the absence of a statistically significant difference. Another explanation could be a selection bias of patients who gave their consent for the genetic study, with a better response rate in the subgroup of RF-negative patients. However, we assume that this selection bias could not influence the FCGR3A genotype distribution and its association with the EULAR response.
If the association between FCGR3A-158 V allele carriage and the response to RTX could be replicated in independent studies, this pharmacogenetic marker, as well as B cell activation biomarkers,5 could help clinicians to make a tailored decision in the choice of a biological DMARD in individual patients with RA in respect of personalised medicine.8
Acknowledgments
The authors thank all the SMART investigators: Dr I Azais, Poitiers; Dr J C Balblanc, Belfort; Dr F Berenbaum, Paris; Dr P Bertin, Limoges; Dr M C Boissier, Bobigny; Dr P Bourgeois, Paris; Dr A Cantagrel, Toulouse; Dr P Carli, Toulon; Dr P Y Chouc, Marseille; Dr M Couret, Valence; Dr L Euller-Ziegler, Nice; Dr P Fardellone, Amiens; Dr P Fauquert, Berck/Mer; Dr R M Flipo, Lille; Dr P Gaudin, Echirolles; Dr J L Grauer, Aix en Provences; Dr A Heraud, Libourne; Dr P Hilliquin, Corbeil; Dr S Hoang, Vannes; Dr E Houvenagel, Lomme; Dr D Keita, Paris; Dr K Lassoued, Cahors; Dr L Le Dantec, Lievin; Dr J M Le Parc, Boulogne; Dr L Lequen, Pau; Dr F Lioté, Paris; Dr C Marcelli, Caen; Dr O Meyer, Paris; Dr J L Pellegrin, Pessac; Dr A Perdriger, Rennes; Dr G Rajzbaum, Paris; Dr S Redeker, Abbeville; Dr J M Ristori, Clermont-Ferrand; Dr A Saraux, Brest; Dr G Tanguy, La Roche sur Yon; Dr T Thomas, Saint-Priest-en-Jarez; Dr L Zabraniecki, Toulouse, Dr C Zarnitski, Montivilliers, France. The authors also thank Dr Rosemary Jourdan, Dr Nadine Mackenzie (Roche, France) and Dr Jamila Filipecki (previously at Roche).
References
Supplementary materials
Supplementary Data
This web only file has been produced by the BMJ Publishing Group from an electronic file supplied by the author(s) and has not been edited for content.
Files in this Data Supplement:
- Web Only Data - This web only file has been produced by the BMJ Publishing Group from an electronic file supplied by the author(s) and has not been edited for content.
- Web Only Data - This web only file has been produced by the BMJ Publishing Group from an electronic file supplied by the author(s) and has not been edited for content.
Footnotes
-
Funding Roche France sponsored the study but was not involved in the interpretation of the data or in the preparation of the manuscript.
-
Competing interests The authors belonging to the scientific committee of the SMART study (ARW, BC, XLL, JT, JSi, MD, XM, AC) received honoraria from Roche (less than $10 000 each) for this study.
-
Provenance and peer review Not commissioned; externally peer reviewed.