Session Information
Session Type: Abstract Submissions (ACR)
Background/Purpose: To compare effectiveness and costs of standard versus individually tailored reduced doses of TNF inhibitors (TNFi) in patients with Rheumatoid Arthritis (RA) after achieving low disease activity.
Methods: This was a single center prospective observational study performed within the national biologics registry. The TNFi dose tapering strategy was chosen by treating physicians, without pre-specified protocol. Patients with RA treated for at least 6 months by TNFi who reached low disease activity (LDA) were eligible for this analysis. LDA was defined as DAS28≤3.2. Firstly, we selected one “baseline” visit for each patient in the standard dosing group, which would be most comparable to baseline visits in the reduced dosing group (=start of dose reduction) using 3 parameters (duration of TNFi treatment, DAS28 and HAQ) to find the best match. Secondly, a propensity score (PS) was used to control for all other confounders associated with starting TNFi dose reduction. The co-primary outcomes were change (D) in HAQ and loss of LDA (defined as DAS28 > 3.2 & DDAS28 ≥ 0.6). Secondary outcomes were DAS28 area under the curve (AUC), HAQ AUC, and annual cost of anti-TNF therapy. The outcomes after one and two years of treatment (since baseline visit) by standard vs reduced doses of TNFi were assessed by generalized linear regression after adjustment on PS.
Results: In the reduced dosing group the mean dose of TNFi corresponded to 0.64, 0.66 and 0.69 of the standard dose initially, at 12 and 24 months resp. After PS adjustment, baseline demographic and clinical characteristics between the groups were well balanced (table 1). 48 (32) and 136 (76) patients in the reduced vs standard dosing group after 1 year (2 years) resp. were available for analysis. Both co-primary outcomes were similar between both groups after one and two years since baseline (table 2 and 3). Annual cost of TNFis per patient was lower by ~4,000 € in the reduced dosing group.
Conclusion: In RA patients after reaching LDA, a tailored approach to reduce doses of TNFi produced similar clinical outcomes at 1 and 2 years resp., but was substantially less costly.
Acknowledgements: This work was supported by IGA grant NT12437
Table 1 Baseline characteristics
|
Standard dosing group |
Reduced dosing group |
Crude p-value |
Adjusted p-value1) |
||
|
n=136 |
n=48 |
|
|
||
|
Female |
n (%) |
115 (84.6 %) |
34 (70.8 %) |
0.040 |
0.156 |
|
Age (years) |
Mean (SD) |
52.9 (11.1) |
51.4 (11.8) |
0.416 |
0.924 |
|
Weight (kg) |
Mean (SD) |
71.7 (13.5) |
73.5 (15.7) |
0.446 |
0.922 |
|
Smoking |
n (%) |
47 (41.2 %) |
17 (38.6 %) |
0.766 |
0.966 |
|
Working or employable |
n (%) |
70 (51.5 %) |
24 (50.0 %) |
0.861 |
0.898 |
|
Disease duration prior to start of anti-TNF therapy (years) |
Mean (SD) |
13.4 (8.5) |
11.7 (6.2) |
0.537 |
0.899 |
|
Duration of anti-TNF therapy (months) |
Mean (SD) |
33.1 (22.0) |
46.4 (30.6) |
0.006 |
0.594 |
|
Number of previous sDMARD |
Mean (SD) |
3.1 (1.3) |
2.9 (1.4) |
0.337 |
0.800 |
|
RF + |
n (%) |
112 (84.8 %) |
40 (85.1 %) |
0.966 |
0.895 |
|
Anti-CCP+ |
n (%) |
82 (76.6 %) |
28 (84.8 %) |
0.319 |
0.548 |
|
Steinbrocker stage I |
n (%) |
18 (13.3 %) |
8 (17.0 %) |
0.535 |
0.775 |
|
DAS28 |
Mean (SD) |
2.19 (0.59) |
1.93 (0.58) |
0.009 |
0.533 |
|
DAS28 < 2.6 |
n (%) |
97 (71.3 %) |
41 (85.4 %) |
0.057 |
0.592 |
|
Number of swollen joints/28 |
Mean (SD) |
0.9 (1.28) |
0.3 (0.57) |
0.001 |
0.436 |
|
Number of tender joints/28 |
Mean (SD) |
0.9 (1.2) |
0.9 (1.4) |
0.568 |
0.189 |
|
CRP (mg/l) |
Mean (SD) |
4.1 (7.0) |
2.4 (2.8) |
0.050 |
0,068 |
|
ESR (mm/h) |
Mean (SD) |
18.0 (13.2) |
13.2 (9.1) |
0.096 |
0.187 |
|
Patients global assessment |
Mean (SD) |
17.2 (13.4) |
15.8 (13.0) |
0.259 |
0.298 |
|
HAQ |
Mean (SD) |
0.77 (0.62) |
0.64 (0.65) |
0.214 |
0.799 |
|
HAQ ≤ 0.5 |
n (%) |
61 (44.9 %) |
26 (54.2 %) |
0.268 |
0.775 |
|
EuroQol |
Mean (SD) |
0.73 (0.15) |
0.74 (0.21) |
0.876 |
0.793 |
|
Concomitant glucocorticoids |
n (%) |
81 (59.6 %) |
13 (27.1 %) |
<0.001 |
0,056 |
|
Anti-TNF monotherapy |
n (%) |
33 (24.3 %) |
13 (27.1 %) |
0.698 |
0.974 |
|
Concomitant methotrexate |
n (%) |
84 (61.8 %) |
32 (66.7 %) |
0.546 |
0.642 |
|
Concomitant leflunomide |
n (%) |
15 (11.0 %) |
1 (2.1 %) |
0.092 |
0,147 |
|
Concomitant sulfasalazine |
n (%) |
9 (6.6 %) |
2 (4.2 %) |
0.542 |
0.267 |
|
Etanercept |
n (%) |
60 (44.1 %) |
22 (45.8 %) |
0.837 |
0.142 |
|
Adalimumab |
n (%) |
43 (31.6 %) |
20 (41.7 %) |
0.209 |
0.440 |
|
Infliximab |
n (%) |
33 (24.3 %) |
6 (12.5 %) |
0.092 |
0.097 |
|
First anti-TNF |
n (%) |
114 (83.8 %) |
45 (93.8 %) |
0.097 |
0.592 |
|
1) p-values are adjusted to propensity score (PS) using multivariate generalized linear regression.
|
|||||
Table 2
|
Primary and secondary outcomes after 1 year of standard vs reduced dosing of TNFis |
||||
|
|
|
Standard dosing group |
Reduced dosing group |
Adjusted p-value1) |
|
|
|
n=136 |
n=48 |
|
|
HAQ at baseline |
Mean (SD) |
0.77 (0.62) |
0.64 (0.65) |
0.799 |
|
HAQ at 12 months |
Mean (SD) |
0.88 (0.65) |
0.72 (0.70) |
0.388 |
|
Change in HAQ after 12 months |
Mean (SD) |
0.11 (0.38) |
0.08 (0.30) |
|
|
Adjusted difference of changein HAQ after 12 months |
Difference (95% CI) |
reference |
-0.082 (-0.220; 0.057) |
0.248 |
|
Loss of LDA during 12 months (DAS28 > 3.2 & DDAS28 ≥ 0.6) |
n (%) |
61 (44.9 %) |
22 (45.8 %) |
|
|
Adjusted OR of LDA loss during 12 months |
OR (95% CI) |
reference |
1.164 (0.374; 3.621) |
0.793 |
|
DAS28 AUC during 12 months |
Mean (SD) |
2.72 (0.83) |
2.27 (0.71) |
|
|
Adjusted difference of DAS28 AUC |
Difference (95% CI) |
reference |
-0.341 (-0.648; -0.034) |
0.030 |
|
HAQ AUC during 12 M |
Mean (SD) |
0.84 (0.61) |
0.66 (0.65) |
|
|
Adjusted difference of HAQ AUC |
Difference (95% CI) |
reference |
-0.106 (-0.345; 0.133) |
0.383 |
|
EuroQol AUC during 12 M |
Mean (SD) |
0.70 (0.17) |
0.72 (0.19) |
|
|
Adjusted difference of EuroQol AUC |
Difference (95% CI) |
reference |
0.031 (-0.037; 0.098) |
0.369 |
|
Annual cost of anti-TNF therapy (€) |
Mean (SD) |
12 000 (-) |
8 080 (1 785) |
|
|
Adjusted difference of annual cost (€) |
Difference (95% CI) |
reference |
-3 963 (-4 311; -3 615) |
<0.001 |
|
1) p-values are adjusted to propensity score (PS) using multivariate generalized linear regression. |
||||
Table 3
|
Primary and secondary outcomes after 2 years of standard vs reduced dosing of TNFis |
||||
|
|
|
Standard dosing group |
Reduced dosing group |
Adjusted p-value1) |
|
|
|
n=76 |
n=32 |
|
|
HAQ at baseline |
Mean (SD) |
0.83 (0.62) |
0.63 (0.68) |
0.888 |
|
HAQ at 24 months |
Mean (SD) |
1.01 (0.72) |
0.72 (0.75) |
0.583 |
|
Change in HAQ after 24 months |
Mean (SD) |
0.18 (0.45) |
0.09 (0.29) |
|
|
Adjusted difference of change in HAQ after 24 months |
Difference (95% CI) |
reference |
-0.130 (-0.346; 0.086) |
0.237 |
|
Loss of LDA during 24 months (DAS28 > 3.2 & DDAS28 ≥ 0.6) |
n (%) |
52 (68.4 %) |
16 (50.0 %) |
|
|
Adjusted OR of LDA loss during 24 months |
OR (95% CI) |
reference |
0.615 (0.169; 2.236) |
0.461 |
|
DAS28 AUC during 24 months |
Mean (SD) |
3.06 (0.81) |
2.19 (0.70) |
|
|
Adjusted difference of DAS28 AUC |
Difference (95% CI) |
reference |
-0.566 (-0.969; -0.163) |
0.006 |
|
HAQ AUC during 24 months |
Mean (SD) |
0.95 (0.62) |
0.68 (0.71) |
|
|
Adjusted difference of HAQ AUC |
Difference (95% CI) |
reference |
-0.070 (-0.409; 0.269) |
0.684 |
|
EuroQol AUC during 24 months |
Mean (SD) |
0.65 (0.20) |
0.71 (0.21) |
|
|
Adjusted difference of EuroQol AUC |
Difference (95% CI) |
reference |
0.039 (-0.069; 0.147) |
0.481 |
|
Annual cost of anti-TNF therapy (€) |
Mean (SD) |
12 000 (-) |
8 221 (1 958) |
|
|
Adjusted difference of annual cost (€) |
Difference (95% CI) |
reference |
-3 550 (-4 106; -2 995) |
<0.001 |
|
1) p-values are adjusted to propensity score (PS) using multivariate generalized linear regression. |
||||
Disclosure:
J. Zavada,
None;
M. Uher,
None;
K. Sisol,
None;
S. Forejtova,
None;
K. Jarosova,
None;
L. Sedova,
None;
Z. Urbanova,
None;
O. Sleglova,
None;
J. Vencovsky,
None;
K. Pavelka,
MSD, AbbVie, Pfizer, UCB, Roche, Amgen, Menarini, BMS,
5.
« Back to 2014 ACR/ARHP Annual Meeting
ACR Meeting Abstracts - https://acrabstracts.org/abstract/a-tailored-approach-to-reduce-dose-of-tnf-inhibitors-is-equally-effective-but-substantially-less-costly-than-standard-dosing-in-patients-with-rheumatoid-arthritis-over-one-and-two-years-a-prospectiv/