Session Information
Title: Rheumatoid Arthritis - Clinical Aspects: Novel Biomarkers and Other Measurements of Disease Activity
Session Type: Abstract Submissions (ACR)
Background/Purpose The prediction of treatment outcomes based on early response could guide treatment decisions in patients (pts) with rheumatoid arthritis (RA). The objective was to determine if disease state at wk 12 in pts treated with a lower dose of methotrexate (MTX) + adalimumab (ADA), assessed by CDAI, DAS28(CRP) and RAPID3, was predictive of an SDAI response at a later time point, comparable to that achieved by pts receiving a high dose of MTX+ADA.
Methods
Data for this post hoc analysis originated from 2 randomized controlled trials. In CONCERTO, pts with early RA received ADA + 2.5, 5, 10 or 20 mg/wk MTX. In MUSICA, pts with moderate to severe RA and an inadequate response to MTX, received ADA + 7.5 or 20 mg/wk MTX. “Achievers” were defined as pts achieving optimal SDAI responses comparable to those in the top 40thpercentile (pct) receiving 20 mg/wk MTX + ADA. For CONCERTO, achievers had SDAI scores at wk 16 ≤6.8, wk 20 ≤5.0 and wk 26 ≤4.2; for MUSICA, achievers had SDAIs at wk 16 ≤15.4, wk 20 ≤11.8 and wk 24 ≤11. The following outcomes at wk 12 were compared in achievers vs non-achievers: CDAI, DAS28(CRP) and RAPID3. The likelihood of predicting optimal SDAI responses was assessed by ROC analysis using logistic regression with wk 12 CDAI, DAS28(CRP) and RAPID3 as predictors. Optimal ROC thresholds based on sensitivity and specificity, as well as those with satisfactory negative and positive predictive values (N/PPV), were calculated.
Results
In CONCERTO, 11/98 (11.2%), 10/100 (10%), 21/99 (21.2%) and 21/98 (21.4%) of pts receiving 2.5, 5, 10 and 20mg/wk MTX, respectively, were achievers; In MUSICA, 29/154 (18.8%) and 28/155 (18.1%) in 7.5 and 20 mg /wk MTX groups, respectively, were achievers. ROC-AUC, PPV and NPV indicated that wk 12 CDAI, DAS28(CRP), and RAPID3 were good predictors of optimal SDAI responses (table). Across selected thresholds, all three criteria provided good NPV for all MTX groups. In CONCERTO, non-achievers in CDAI remission/LDA at wk 12 achieved SDAI remission (≤3.3) or LDA (≤11) at wk 26; most pts in CDAI MDA or HDA at wk 12 did not. In MUSICA, in the 7.5 mg MTX treatment group, non-achievers in CDAI remission/LDA at wk 12 reached SDAI LDA (≤11) or MDA (11-26) at wk 24; 72.5% of pts in wk 12 CDAI MDA did not achieve optimal SDAI responses.
|
Table: AUC, PPV and NPV of wk 12 CDAI, RAPID3 and DAS28 (CRP) for prediction of optimal SDAI response |
|
|||||||
|
MTX dose (mg/wk) |
Wk 12 criterion |
Area under curve (95%CI) |
ROC thres-hold |
Sensitivity |
1-specificity |
PPV |
NPV |
|
|
CONCERTO |
|
|||||||
|
2.5 |
CDAI |
0.907 (0.837, 0.977) |
3.6 |
0.455 |
0.038 |
0.625 |
0.926 |
|
|
5* |
|
0.758 (0.562, 0.953) |
7.3 |
0.700 |
0.190 |
0.304 |
0.958 |
|
|
10 |
|
0.902 (0.841, 0.963) |
2.5 |
0.571 |
0.056 |
0.750 |
0.881 |
|
|
20 |
|
0.923 (0.870,0.976) |
2.6 |
0.631 |
0.082 |
0.666 |
0.905 |
|
|
2.5 |
RAPID3 |
0.879 (0.796, 0.962) |
0.6 |
0.450 |
0.038 |
0.625 |
0.925 |
|
|
5* |
|
0.807 (0.711, 0.902) |
6.3 |
0.900 |
0.265 |
0.290 |
0.984 |
|
|
10 |
|
0.863 (0.784, 0.943) |
2.6 |
0.570 |
0.100 |
0.600 |
0.872 |
|
|
20 |
|
0.839 (0.738, 0.941) |
1.9 |
0.631 |
0.111 |
0.600 |
0.901 |
|
|
2.5 |
DAS28 |
0.870 (0.790, 0.950) |
1.6 |
0.180 |
0.012 |
0.667 |
0.895 |
|
|
5* |
|
0.755 (0.566, 0.944) |
2.8 |
0.700 |
0.167 |
0.333 |
0.959 |
|
|
10 |
|
0.890 (0.818, 0.960) |
2.0 |
0.619 |
0.098 |
0.650 |
0.888 |
|
|
20 |
|
0.923 (0.870, 0.976) |
2.5 |
0.894 |
0.150 |
0.608 |
0.968 |
|
|
MUSICA |
|
|||||||
|
7.5 |
CDAI |
0.822 (0.749, 0897) |
4.1 |
0.222 |
0.19 |
0.750 |
0.833 |
|
|
20 |
|
0.946 (0.908, 0.983) |
8.7 |
0.851 |
0.056 |
0.793 |
0.962 |
|
|
7.5* |
RAPID3 |
0.765 (0.673, 0.858) |
7.9 |
0.740 |
0.243 |
0.444 |
0.919 |
|
|
20 |
|
0.839 (0.764, 0.914) |
3.4 |
0.444 |
0.077 |
0.600 |
0.865 |
|
|
7.5 |
DAS28 |
0.814 (0.740, 0.888) |
2.7 |
0.330 |
0.050 |
0.600 |
0.849 |
|
|
20 |
|
0.940 (0.901, 0.979) |
3.3 |
0.925 |
0.121 |
0.658 |
0.979 |
|
|
For CONCERTO, optimal SDAI response was SDAI at wk 16 ≤6.8, wk 20 ≤5.0 and wk 26 ≤4.2; for MUSICA, optimal SDAI response was SDAI at wk16 ≤15.4, wk20 ≤11.8 and wk 24 ≤11. ROC thresholds with NPV≥0.8 and PPV≥0.6 are presented, except for the groups indicated by the asterisk, where the optimal ROC cut-off is presented.
|
||||||||
Conclusion
CDAI and RAPID3 are quick, convenient tools to assess treatment response, and correlated well with DAS28(CRP) as predictors of later outcome in the CONCERTO and MUSICA trials. In pts with early RA receiving lower doses of MTX, the achievement of remission or LDA targets at wk 12 was predictive of a subsequent optimal SDAI treatment response comparable to pts receiving higher MTX doses. Based on these data, pts in MDA/HDA at wk 12 might benefit from an adjustment of therapy.
Disclosure:
G. Burmester,
AbbVie, Pfizer, UCB, Roche,
2,
AbbVie, BMS, Pfizer, Merck, MedImmune, UCB, Roche,
5,
AbbVie, BMS, Pfizer, Merck, UCB, Roche,
8;
G. S. Kaeley,
AbbVie,
5;
J. R. Curtis,
AbbVie, Amgen, Genentech, BMS, Janssen and CORRONA,
2,
Genentech, UCB, Janssen, Amgen and CORRONA,
5;
Y. Yazici,
AbbVie, BMS, Celgene, Genentech, Horizon, UCB and Pfizer,
5;
B. Guerette,
AbbVie,
1,
AbbVie,
3;
X. Wang,
AbbVie,
1,
AbbVie,
3;
A. Friedman,
AbbVie, Inc.,
1,
AbbVie, Inc.,
3;
V. Strand,
AbbVie, Afferent, Amgen, Biogen Idec, Bioventus, BMS, Carbylan, Celgene, Celltrion, CORRONA, Crescendo, Genentech/Roche, GSK, Hospira, Iroko, Janssen, Lilly, Merck, Novartis, Pfizer, Regeneron, Sanofi, SKK, Takeda, UCB, Vertex,
5.
« Back to 2014 ACR/ARHP Annual Meeting
ACR Meeting Abstracts - https://acrabstracts.org/abstract/the-use-of-week-12-cdai-rapid3-and-das28crp-responses-to-predict-optimal-response-to-methotrexate/