Background
Natalizumab (NTZ) (Tysabri©, Biogen Idec Inc., Cambridge, MA) is a humanized monoclonal antibody which blocks the α4 subunit of the α4 integrins (α4β1 and α4β7) which is expressed on the surface of lymphocytes and leukocytes [
1,
2]. α4β1 is required for endothelial adhesion, and it facilitates migration of peripheral blood lymphocytes to the brain via the blood-brain barrier [
1]. NTZ showed its efficacy in the multiple sclerosis (MS) treatment by reducing the annual relapse rate in two clinical trials by 68 % [
3,
4]. This monoclonal antibody is currently considered to be one of the best therapeutical options for patients with relapsing-remitting MS (RRMS) not responding to traditional therapies [
5].
The most fearsome clinical complication of NTZ treatment is progressive multifocal leukoencephalopathy (PML), a John Cunningham virus (JCV) infection of the brain [
6]. PML is a demyelinating condition characterized by the degenerative loss of cerebral white matter after infection by JCV, a normally latent polyomavirus which becomes virulent in the presence of immunosuppression [
7]. During therapy with NTZ, PML risk can be stratified depending on three risk factors: anti-JCV antibody status and level, treatment length (more or less than 2 years) and prior immunosuppressive therapy [
8,
9]. It is important to identify other factors allowing better identification of MS patients at high risk for PML. Schwab and collaborators [
10] investigated the expression of T lymphocytes surface markers in NTZ-treated patients before PML development. Their investigation suggested L-selectin (CD62L) expressed on CD4
+ T cells might help identify MS patients at high risk for PML, since all the eight patients who developed PML had a previous blood sample showing a level of CD62L expression below a hypothetical threshold [
10].
The aim of this study was to investigate whether the expression of CD62L is lower in NTZ-treated patients than in patients treated with other drugs and to evaluate levels of CD62L in a subset of NTZ-treated patients throughout treatment.
Methods
Standard protocol approval, registration and patients consent
We obtained peripheral blood samples from healthy individuals and from subjects with RRMS after they signed a written informed consent approved by the Ethics Committee of San Luigi Gonzaga University Hospital (Ethic approval number 7777, March 25, 2013). The blood samples were collected in 6 ml ethylenediaminetetraacetic acid (EDTA) Vacutainers (BD Biosciences, Milan, Italy) and processed within the following 4 h. This study was performed according to good clinical practice and the Declaration of Helsinki.
Patients and healthy controls
We enrolled 225 patients with RRMS, aged 39 ± 10 years. We excluded subjects with concomitant endocrine and metabolic disorders. Nineteen patients were untreated (UT), 113 were treated with NTZ (300 mg iv every 28 days, duration 1–86 months), 26 were treated with interferon-beta (IFNb), 26 were treated with glatiramer acetate (GA), 23 were treated with fingolimod (FTY) and 18 were treated with rituximab (RTX). At the time of enrolment, clinically definite MS was diagnosed using the revised McDonald’s criteria [
11]. Age- and sex-matched healthy donors (HD,
n = 23) with no previous history of any neurologic or immune-mediated disease served as controls (Table
1). We have also analysed samples from two PML/IRIS (immune reconstitution inflammatory syndrome) patients, who had discontinued NTZ treatment, and we collected other four samples during 9 months from a patient who developed clinical asymptomatic PML while under NTZ. All patients underwent periodic clinical and magnetic resonance imaging (MRI) evaluations.
Table 1
Patient characteristics
HD (n = 23) | 18/5 (78 %) | 33 ± 11 (20–60) | n.d. | n.d. | n.d. |
UT (n = 19) | 12/7 (63 %) | 41 ± 8 (27–57) | 1.3 (0–3) | n.d. | n.d. |
GA (n = 26) | 16/10 (62 %) | 44 ± 11 (22–60) | 1.9 (0–8.5) | 45 ± 38 (1–122) | 13/26 IFNb; 9/26 nv; 4/26 IS |
IFNb (n = 26) | 15/11 (58 %) | 36 ± 11 (24–65) | 1.6 (0–5) | 51 ± 48 (3–147) | 25/26 nv; 1/26 GA |
NTZ (n = 113) | 76/37 (67 %) | 37 ± 9 (19–60) | 2.9 (0–8) | 35 ± 24 (1–86) | 52/113 IFNb; 23/113 GA; 13/113 IS; 12/113 nv; 6/113 FTY; 7/113 other treatments |
RTX (n = 18) | 15/3 (83 %) | 41 ± 10 (30–65) | 4.4 (2–7) | 1.25 (1–3) | 1/18 IFNb; 1/18 GA; 5/18 FTY; 9/18 NTZ; 2/18 n.d. |
FTY (n = 23) | 15/8 (65 %) | 41 ± 9 (21–57) | 3.1 (0–6.5) | 15 ± 8 (1–29) | 1/23 nv; 8/23 NTZ; 5/23 GA; 5/23 IFNb; 2/23 IS; 2/23 other treatments |
Blood samples and bio-banking
CRESM (Centro di Riferimento Regionale per la Sclerosi Multipla) at University Hospital San Luigi, Orbassano, Italy, houses a biological bank collecting samples from HD and MS patients. The bio-bank stores biological samples (cerebrospinal fluid (CSF), RNA, DNA, sera, plasma and peripheral blood mononuclear cells (PBMC)) for research purposes. For this research, blood samples were drawn just before NTZ-injection or during planned routine visits in patients treated with other drugs.
PBMC were isolated from EDTA-treated blood by Lymphoprep density gradient centrifugation. Then, cells were cryopreserved in liquid nitrogen using freezing medium: 60 % RPMI 1640 medium (Invitrogen Life Technologies, Grand Island, NY, USA), 30 % heat-inactivated foetal bovine serum (FBS, Invitrogen Life Technologies) and 10 % dimethyl sulfoxide (DMSO, Sigma-Aldrich, St Louis, MO). All assays were performed on fresh frozen cells. After gentle thawing at 37 °C, the cells were immediately added to 5 mL RPMI 1640 supplemented with 10 % heat-inactivated FBS and centrifuged to remove DMSO. The samples were re-suspended in RPMI 1640 medium supplemented with 10 % heat-inactivated FBS and counted.
Flow cytometry
Non-specific sites of 2.5 × 10
5 cells were blocked with rabbit immunoglobulins G (IgG, Sigma-Aldrich). Then, the cells were incubated with fluorochrome-conjugated monoclonal Ab (mAb) and isotype-matched negative controls for 20 min at 4 °C. The following mAbs were used: anti-human CD3 APC-Vio770 (Miltenyi Biotec, Bergisch Gladbach, Germany), anti-human CD4 PE-Cy7 and anti-human CD62L FITC allophycocyanin (both from BD Pharmingen, San Diego, CA) and isotype-matched mAb (Miltenyi Biotec). After staining, the cells were washed and re-suspended in PBS (Sigma-Aldrich) supplemented with 0.2 % bovine serum albumin and 0.01 % sodium azide (both from Sigma-Aldrich). The samples were collected and analysed using a CyAn ADP, running Summit 4.3 analysis software (Beckman Coulter, Brea, CA, USA). The living cells identified by propidium iodide (Sigma-Aldrich) exclusion were gated according to their light scatter properties to exclude cell debris. Quality experiments have been performed (Additional file
1: Figure S1) in order to evaluate whether the freeze-thawing cycle could affect the quantification of CD62L protein expression and to check the reproducibility of the flow cytometric procedure and the stability of the expression of CD62L on CD4
+ T cells in frozen samples.
Statistical analysis
Statistical analysis was performed using GraphPad Prism software (GraphPad Software, version 4; San Diego, CA, USA). The differences between the two groups were calculated with an unpaired, two-tailed, nonparametric Mann-Whitney U test. The Student’s paired t test was used to evaluate differences in the longitudinal studies.
Discussion
The main goal of this research was to test whether NTZ treatment has a specific impact on CD62L expression in CD4
+ T cells as suggested by Schwab and collaborators [
10]. This issue is relevant as low levels of CD62L were suggested as a potential marker for PML risk stratification in NTZ-treated patients [
10].
The CD62L is a key adhesion molecule which regulates both the migration of leukocytes at sites of inflammation and the recirculation of lymphocytes between blood and lymphoid tissues. In our work, we analysed separately the first-line drugs (IFNb and GA). The data are consistent with those already reported since the patients treated with NTZ showed a lower percentage of CD62LCD4
+ T cells than the patients treated with IFNb (
p = 0.0454) and GA (
p = 0.0001). Although data in literature show that the GA had no impact on the expression of the T cell surface activation markers such as CD62L in animal model [
12], in our study, a GA-treated patient showed a low percentage of CD62LCD4
+ T cells for no apparent reason. Monitoring the medical record, 4 months later CD62L analysis, the patient had a serious infection resulting in prolonged hospitalization. This data shows how this might be a correlation between the low percentage of CD62L and the subsequent development of infections even though this is not closely related to MS. We have also studied in our research the impact of two second-line DMTs on CD62L. RTX does not influence CD62L expression on CD4
+ T cells, since we have not detected any differences with HD controls, UT, GA- and IFNb-treated patients (Fig.
2). FTY induced a decreased CD62L expression, similarly to NTZ (Fig.
2). The peculiarity of CD62L is that it acts as a “homing receptor” for leukocytes to enter secondary lymphoid tissues via high endothelial venules. The function of FTY is the segregation of lymphocytes in the lymph nodes. Previous reports showed FTY mainly influences naïve and central memory cells [
13], and probably for these reasons, a high percentage of CD62LCD4
+ T cells are absent in the blood of our patients. Furthermore, different patients were previously treated with NTZ and this could influence the expression of CD62L on CD4
+ T cells. The phenomenon we observed is very interesting and deserves a dedicated study to evaluate the expression of CD4CD62L
+ T cells on a broader population of FTY-treated patients and its stability during time.
In Fig.
2, we took a picture of the first sampling by evaluating all the available patients we had at that time. Then, we stratified the patients according to the number of given infusions, and we noticed only the first year was statistically different, while the next was no longer (Fig.
2b). These data suggest an implication of NTZ in the variation of CD62L expression on CD4
+ T cells. To evaluate the effect of NTZ on the expression of CD62L in the first year, we have separated the first 6 months of treatment and the second ones. Our results show a persistent CD62L low expression if you continue the treatment, and it disappears after treatment discontinuation. In the group of patients who had discontinued NTZ, the patients who showed a significant increase of CD62L percentage belonged to the 25th percentile group (pt #1, pt #2, pt #5). Moreover, the patient (pt #6) who had a decrease of CD62L expression remained in the medium group. Only one patient (pt #3), who had an unchanged value of CD62L, had a relapse 4 months after NTZ suspension (MRI pos, see Table
2): this is an indication that CD62L is a marker of infections and not of the MS relapse activity. However, all patients were monitored by MRI 2–3 months after discontinuation, and this exam was unchanged compared to the previous result in five patients out of six.
In the whole subsets of patients, freeze-thawed cells showed a lower percentage of CD62L expression than fresh cells, but the decrease was more relevant in NTZ-treated patients (Additional file
1: Figure S1). Our data confirmed prolonged interaction between NTZ and PBMC expressing α4 integrins induces increased cell activation causing easier shedding of CD62L [
14,
15] or reduced cell viability.
As freeze-thawing strongly decreases the percentage of PBMC CD62L
+ cells, a phenomenon already observed on polymorphonuclear leukocytes [
16], the procedures of PBMC collection, banking and freeze-thawing are crucial and small variations could explain the lower percentage of CD62LCD4
+ T cells we found in comparison with the Schwab study [
10]. Staining and gating of positive cells do not seem to influence the results as the reproducibility of flow cytometry quantification of CD62L-positive cells was very good in the same frozen sample (Additional file
1: Figure S1).
The seminal work on the role of CD62L as a biomarker for PML risk stratification showed all the eight patients, who later developed PML, had a value of CD62L below a tentative threshold [
10]. In this work, the longitudinal data of a patient who later developed PML support the role of CD62L as a risk biological marker. In order to get a precise comparison with the work of Schwab and colleagues [
10], we should calculate the threshold value as the mean minus twice the standard deviation of the percentage of CD62L in long-term NTZ-treated patients. A value of CD62L nearby to the tentative threshold was detected in the clinically asymptomatic phase of the disease, and a value in the lower 25th percentile was detected when MRI was judged negative for PML (Fig.
4). This observation opens a question about the establishment of threshold risk. In this research, we found a correlation between the low percentage of CD62L and the development of a severe infection. A greater number of observations are required to establish a threshold, and it is also necessary to evaluate critically whether the values obtained could be arranged in the 25th percentile or below the long-term NTZ-treated patients threshold or if it is better to calculate the threshold based on the first-line drug group since the majority of NTZ patients (75 patients out of 113, Table
1) starts NTZ after the first-line drug treatment.
Previous studies have shown that NTZ-treated MS patients have an increased pool of T cells and other lymphocytes in the blood compartment [
17,
18] which may be enriched in T cells with an increased activation state. The percentage of CD62L in each patient fluctuates in time; in fact, the longitudinal measurements in the same patients showed 3 % of the patients tested three times 4 months apart had one single value below the empirically established threshold (Fig.
3), and the value of CD62L decreased during time. Considering PML risk increases with time on treatment [
19], in our analysis, we found patients with a percentage of CD62L below threshold in the fourth and seventh year (Fig.
2b). The patients with CD62L low expression will be closely monitored by MRI and CD62L analysis. In addition to the existing parameters, the analysis of CD62L could be another parameter to evaluate patients’ health and their risk to develop PML.
Competing interests
MS received a travel grant and speaker honoraria from Biogen Idec. MC received a travel grant from TEVA and Merck Serono and speaker honoraria from Biogen Idec. FM received a travel grant from Euroimmun and speaker honoraria from Biogen Idec. AL is a Bayer Schering, Biogen Idec, Genzyme/Sanofi, Merck Serono Advisory Board Member. She received travel grants and honoraria from Bayer Schering, Biogen Idec, Genzyme, Merck Serono, Novartis, Sanofi and Teva and research grants from Bayer Schering, Biogen Idec, Merck Serono, Novartis, Sanofi and Teva. Prof. Lugaresi has also received travel and research grants from the Associazione Italiana Sclerosi Multipla and was a Consultant of “Fondazione Cesare Serono”. AB received honoraria for serving in the scientific advisory boards of Almirall, Bayer Healthcare, Biogen Idec, Sanofi Genzyme and received speaker honoraria from Biogen Idec, Sanofi Genzyme, Novartis and TEVA. His institution has received grant support from Bayer Healthcare, Biogen Idec, Merck Serono, Novartis, TEVA, Italian MS Society, Associazione Ricerca Biomedica ONLUS, and San Luigi ONLUS.
Authors’ contributions
MS performed the research and collected and analysed the data. MC and FM collected and analysed the data and helped with patient recruitment. AL collected the data and helped with patient recruitment. AB designed the research, analysed the data, and generated funding. All the authors wrote the manuscript. All authors read and approved the final manuscript.