Characteristics of patients
From July 2011 to September 2012, thirteen RA patients who met the inclusion criteria were recruited into the study. Baseline patient characteristics are summarized in Table 1. Most patients were female (69.2%) with a mean age of 48.2 ± 14.5 years. The mean disease duration was 11.5 ± 9.4 years. Mean TCZ-treatment duration prior to initiating spacing was 18.4 ± 7.3 months. The mean time of remission was 7.5 ± 6.2 months. Ten patients (76.9%) presented erosions at baseline and seven (53.8%) were positive for rheumatoid factor (RF) and anti-citrullinated peptide antibodies (ACPA). Prior to TCZ treatment, the mean number of previous csDMARDs and bDMARDs used were 2.1 ± 0.9 and 1.7 ± 1.2, respectively. None of the patients in the study had received rituximab prior to TCZ treatment. Six patients received concomitant treatment with methotrexate (MTX), three with leflunomide, and five received TCZ in monotherapy. At the start of the progressive spacing, the mean DAS28 score was 1.6 ± 0.9.
Table 1: Baseline RA patient’ characteristics.
|
All patients
(n=13)
|
Age, mean (SD), years; (min-max)
|
48.23± 14.5, (21-67)
|
Female, n (%)
|
9 (69.2%)
|
Disease duration, mean (SD), years; (min-max)
|
11.46 ± 9.44; (2-36)
|
TCZ exposure before tapering, months (SD) ; (min-max)
|
18.38 ±7.34; (5-29)
|
Remission achievement delay, months (SD) ; (min-max)
|
2.38 ±2.63; (0-8)
|
Remission duration, months (SD) ; (min-max)
|
7.53 ± 6.21; (3-24)
|
RF positive, n (%)
|
7 (53.8%)
|
ACPA positive, n (%)
|
7 (53.8%)
|
Erosive status, n (%)
|
10 (76.9%)
|
Previous number of sDMARDs; (min-max)
|
2.15 ± 0.9; (1-4)
|
Previous number of bDMARDs; (min-max)
|
1.7 ± 1.18; (0-3)
|
Swollen joint count; (min-max)
|
0.08 ± 0.27; (0-1)
|
Tender joint count; (min-max)
|
3.14 ± 3.4; (0-10)
|
Hyperhemia at Doppler
|
2 (15.4%)
|
Number of patients with active synovitis
|
7 (53.8%)
|
Number of active synovitis
|
|
Hands
|
14
|
Feet
|
5
|
Concomitant therapy
|
|
MTX, n (%)
|
6 (42.9%)
|
LEF, n (%)
|
3 (21.4%)
|
DAS28; mean (SD) ; (min-max)
|
1.65 ± 0.93; (0-2.6)
|
ESR, mm/h; mean (SD) ; (min-max)
|
7 ± 9.12; (1-27)
|
CRP, mg/L, mean (SD) ; (min-max)
|
2.21 ± 2.34; (0.2-7.4)
|
TCZ plasma level, mg/L; mean (SD) ; (min-max)
|
7.35± 6.41; (0-16)
|
ADAb positivity
|
0
|
RA: rheumatic arthritis, SD: standard deviation, n: number, TCZ: tocilizumab, RF: rheumatoid factor, ACPA: anti-citrullinated peptide antibodies, MTX: methotrexate, LEF: leflunomide, DAS28: Disease Activity Score in 28 joints; ESR: erythrocyte sedimentation rate, CRP: C-reactive protein, ADAb: antidrug antibodies
Patient outcomes after tapering TCZ infusions
After a 2-year follow-up, 8/13 patients remained on TCZ therapy after the spacing attempt. Successful tapering of TCZ treatment with a long-term controlled disease and a minimum 5-week interval between infusions, was achieved for six patients (46.1%) (Table 2). Among these patients, four were maintained on a RTI of eight or more weeks, and their mean DAS28 score at 24 months was 1.58 ± 0.6.
The successful long-term maintenance group (6/13) experienced on average one flare ± 0.9 during the study, with a mean delay of occurrence of 4.4 ± 4.9 months after the start of spacing. Only two patients remained on a 4-week RTI of TCZ infusions. A switch to another biologic was needed for five patients, four of which experienced a secondary failure (one was switched to anti-TNF-α and the other three to abatacept). The remaining patient developed a severe TCZ-induced neutropenia.
Table 2: Evolution of RA patients’ disease activity during the 24-months follow-up.
|
Day 0
|
W6
|
M3
|
M6
|
M9
|
M12
|
M18
|
M24
|
P1
|
DAS 28
|
1.76
|
3.14
|
2.43
|
2.43
|
1.76
|
1.55
|
5.18
|
1.64
|
RTI
|
4
|
6
|
6
|
6
|
6
|
8
|
8
|
6
|
Dose
|
4
|
4
|
4
|
4
|
4
|
4
|
4
|
4
|
P2
|
DAS 28
|
1.69
|
2.35
|
2.79
|
2.89
|
2.69
|
3.14
|
1.34
|
2.13
|
RTI
|
4
|
6
|
6
|
8
|
6
|
8
|
4
|
4
|
Dose
|
8
|
8
|
8
|
8
|
8
|
8
|
8
|
8
|
P3
|
DAS 28
|
2.4
|
2.07
|
2.57
|
3.92
|
2.14
|
3.7
|
3.68
|
2.61
|
RTI
|
4
|
6
|
6
|
8
|
4
|
4
|
5
|
4
|
Dose
|
8
|
8
|
8
|
8
|
8
|
8
|
8
|
8
|
P4
|
DAS 28
|
2.6
|
1.42
|
1.34
|
1.94
|
1.05
|
1.12
|
3.35
|
1.49
|
RTI
|
4
|
6
|
8
|
8
|
4
|
5
|
5
|
5
|
Dose
|
8
|
8
|
8
|
8
|
8
|
8
|
8
|
8
|
P5
|
DAS 28
|
2.58
|
2.6
|
1.61
|
2.11
|
2.23
|
2.22
|
2.03
|
0.91
|
RTI
|
4
|
6
|
6
|
7
|
8
|
8
|
9
|
9
|
Dose
|
8
|
8
|
8
|
8
|
8
|
8
|
8
|
8
|
P6
|
DAS 28
|
0.48
|
NA
|
0.76
|
1.37
|
0.28
|
0.28
|
2.22
|
1.84
|
RTI
|
4
|
6
|
7
|
6
|
6
|
6
|
8
|
9
|
Dose
|
8
|
8
|
8
|
8
|
8
|
8
|
8
|
8
|
P 7
|
DAS 28
|
1.61
|
NA
|
2.57
|
1.34
|
1.94
|
3.28
|
2
|
2.5
|
RTI
|
4
|
5
|
8
|
8
|
8
|
8
|
8
|
8
|
Dose
|
6
|
8
|
8
|
6
|
6
|
6
|
8
|
8
|
P8
|
DAS 28
|
1.98
|
2.26
|
1.68
|
2.14
|
NA
|
1.81
|
2.42
|
Secondary failure Switch to infliximab
|
RTI
|
4
|
6
|
6
|
8
|
8
|
8
|
4
|
Dose
|
8
|
8
|
8
|
8
|
8
|
8
|
4
|
P9
|
DAS 28
|
0.07
|
1.45
|
1.89
|
1.89
|
1.12
|
0.76
|
0.76
|
1.12
|
RTI
|
4
|
6
|
6
|
7
|
7
|
8
|
9
|
10
|
Dose
|
8
|
8
|
8
|
8
|
8
|
8
|
8
|
8
|
P10
|
DAS 28
|
1.38
|
3.2
|
NA
|
2.85
|
2.45
|
3.71
|
Secondary failure Switch to abatacept
|
RTI
|
4
|
6
|
4
|
4
|
4
|
4
|
Dose
|
8
|
8
|
8
|
8
|
8
|
8
|
P11
|
DAS 28
|
2.31
|
2.31
|
3.48
|
0.85
|
Neutropenia
Switch to abatacept
|
RTI
|
4
|
5
|
6
|
4
|
Dose
|
4
|
4
|
6
|
6
|
P12
|
DAS 28
|
2.56
|
1.03
|
4.08
|
3.51
|
Secondary failure
Switch to abatacept
|
RTI
|
4
|
6
|
6
|
4
|
Dose
|
8
|
8
|
8
|
8
|
P13
|
DAS 28
|
0
|
1.03
|
1.55
|
2.57
|
5.35
|
Secondary failure
Switch to abatacept
|
RTI
|
4
|
6
|
6
|
8
|
6
|
Dose
|
6
|
6
|
6
|
6
|
6
|
Dose: mg/kg, W: week, M: month, P: Patient, RTI: Retreatment interval, NA: not available
Predictors of maintaining remission or flare after tapering
In order to evaluate potential predictors of maintaining remission following TCZ tapering, we compared patients experiencing secondary failure (n=4) with those experiencing successful long-term maintenance (n=6) (Table 3). Patients with secondary failure were significantly older than patients experiencing successful long-term maintenance (mean age: 60.7 ± 6.7 and 41.8 ± 15.2 respectively; p=0.038). Hand and feet US characteristics were similar in each group. TCZ plasma level was similar in both groups and none of them developed ADAb. There was a marginally non-significant tendency (p=0.07) for patients in the successful long-term maintenance group to experience fewer flares during the 2 years of the study (mean number of flares: 2 ± 0 versus 1 ± 0.9).
Table 3: Comparison of baseline characteristics between successful long-term maintenance group (n=6) vs. Secondary failure group (n=4).
|
Secondary failure
(n=4)
|
Patients with successful long-term maintenance
(n=6)
|
p-value
|
Age, mean (SD), years
|
60.75± 6.7
|
41.8 ± 15.23
|
<0.05
|
Female, n (%)
|
2 (50%)
|
4 (66.7%)
|
NS
|
Disease duration, mean (SD), years
|
15±14.07
|
7.8 ± 4.35
|
NS
|
TCZ exposure before tapering, months (SD)
|
19.5±7.23
|
14.8 ± 7.5
|
NS
|
Remission achievement delay, months (SD)
|
3.5±3.4
|
2.16 ± 2.4
|
NS
|
Remission duration, months (SD)
|
8 ±5.7
|
4.8 ± 2.13
|
NS
|
RF positive, n (%)
|
3 (75%)
|
2 (33.3%)
|
NS
|
ACPA positive, n (%)
|
3 (75%)
|
2 (33.3%)
|
NS
|
Erosive status, n (%)
|
4 (100%)
|
4 (66.7%)
|
NS
|
Previous number of sDMARDs
|
2.75±1.25
|
2 ± 0.63
|
NS
|
Previous number of bDMARDs
|
1.5±1.73
|
1.66 ± 1.21
|
NS
|
Swollen joint count
|
0
|
0.166 ±0.4
|
NS
|
Tender joint count
|
2.25±2.6
|
3.16±4.57
|
NS
|
Hyperhemia at Doppler
|
1 (25%)
|
1 (16.7%)
|
NS
|
Number of patients with active synovitis
|
1 (25%)
|
4 (66.7%)
|
NS
|
Number of active synovitis
|
|
|
|
Hands
|
8
|
4
|
NS
|
Feet
|
0
|
5
|
NS
|
Concomitant therapy
|
|
|
|
MTX, n (%)
|
2(50%)
|
2(33.3%)
|
NS
|
LEF, n (%)
|
0(0%)
|
2 (33.3%)
|
NS
|
DAS28; mean (SD)
|
1.48±1.1
|
1.51 ± 1.05
|
NS
|
ESR, mm/h; mean (SD)
|
7.5± 11
|
6.66± 10.1
|
NS
|
CRP, mg/L, mean (SD)
|
2.55±3.3
|
3 ±1.87
|
NS
|
TCZ plasma level, mg/L; mean (SD)
|
7.12±7.7
|
7.35 ± 6.41
|
NS
|
ADAb positivity
|
0
|
0
|
NS
|
Flare, mean (SD)
|
2±0
|
1±0.9
|
0.07
|
SD: standard deviation, n: number, TCZ: tocilizumab, RF: rheumatoid factor, ACPA: anti-citrullinated peptide antibodies, MTX: methotrexate, LEF: leflunomide, DAS28: Disease Activity Score in 28 joints; ESR: erythrocyte sedimentation rate, CRP: C-reactive protein, ADAb: antidrug antibodies, NS: non significant
Lastly, we compared patients who experienced one flare or less during the whole study with the remaining patients who experienced two or more (Table 4). While none of the baseline clinical, biological, and imaging characteristics were associated with successful tapering of TCZ infusions, we found that RF and ACPA positivity were both associated with a greater number of flares (p=0.004). A multivariate analysis found no significant independent predictors of maintaining remission or flare occurrence after tapering.
Table 4: Comparison between patients with one flare or less vs. patients with more than a flare.
|
Patients with one flare or less
(n=5)
|
Patients with more than a flare
(n=8)
|
p-value
|
Age, mean (SD), years
|
40.8 ±15.7
|
52.9±12.5
|
NS
|
Female, n (%)
|
3 (60%)
|
6 (75%)
|
NS
|
Disease duration, mean (SD), years
|
12.4 ±8.7
|
10.9±10.4
|
NS
|
TCZ exposure before tapering, months (SD)
|
16.4±8.8
|
20.4±5.7
|
NS
|
Remission achievement delay, months (SD)
|
3 ±2.4
|
2±2.8
|
NS
|
Remission duration, months (SD)
|
5.6±2.8
|
8.8±7.5
|
NS
|
RF positive, n (%)
|
0 (0%)
|
7 (87.5%)
|
0.004
|
ACPA positive, n (%)
|
0 (0%)
|
7 (87.5%)
|
0.004
|
Erosive status, n (%)
|
3 (60%)
|
7(87.5%)
|
NS
|
Previous number of sDMARDs
|
1.8±0.8
|
2.4±1
|
NS
|
Previous number of bDMARDs
|
1.6±1.1
|
1.75±1.3
|
NS
|
Swollen joint count
|
0
|
0.125 ±0.3
|
NS
|
Tender joint count
|
2.4±4.3
|
3.6±2.97
|
NS
|
Hyperhemia at Doppler
|
0 (0%)
|
(37.5%)
|
NS
|
Number of patients with active synovitis
|
3 (60%)
|
4(50%)
|
NS
|
Number of active synovitis
|
|
|
|
Hands
|
3
|
11
|
NS
|
Feet
|
2
|
3
|
NS
|
Concomitant therapy
|
|
|
|
MTX, n (%)
|
3 (60%)
|
3 (37.5%)
|
NS
|
LEF, n (%)
|
1 (20%)
|
2 (25%)
|
NS
|
Flare, mean (SD)
|
0.6±0.5
|
2.13±0.35
|
0.001
|
SD: standard deviation, n: number, TCZ: tocilizumab, RF: rheumatoid factor, ACPA: anti-citrullinated peptide antibodies, MTX: methotrexate, LEF: leflunomide, NS: non-significant