Efficacy of Secukinumab and Adalimumab in Patients With Psoriatic Arthritis and Concomitant Moderate-to-severe Plaque Psoriasis

Results From EXCEED, a Randomized, Double-blind Head-to-head Monotherapy Study

A.B. Gottlieb; J.F. Merola; K. Reich; F. Behrens; P. Nash; C.E.M. Griffiths; W. Bao; P. Pellet; L. Pricop; I.B. McInnes

Disclosures

The British Journal of Dermatology. 2021;185(6):1124-1134. 

In This Article

Results

A total of 853 patients were randomized to receive secukinumab (N = 426) or adalimumab (N = 427) in the study; at baseline, there were 211 patients in the psoriasis subset [secukinumab (N = 110) and adalimumab (N = 101)]. Up to week 50, a total of six of 110 (5·5%) patients had discontinued treatment in the secukinumab group vs. 18 of 101 (17·8%) patients in the adalimumab group in the psoriasis subset. The major reasons for discontinuation of secukinumab vs. adalimumab treatment were lack of efficacy [one of 110 (0·9%) vs. seven of 101 (6·9%)] and patient/guardian decision [four of 110 (3·6%) vs. six of 101 (5·9%)] (Figure 1). The analysis of patients in the psoriasis subset for time to study treatment discontinuation (Kaplan–Meier curve) indicated that a higher proportion of patients were being retained for a longer period for secukinumab than adalimumab treatment until last dosing visit at week 50 (P = 0·0005) (Figure 2).

Figure 1.

Patient study treatment disposition up to week 52 in the psoriasis subset of patients with psoriatic arthritis. N, number of randomized patients; n, number of available patients; BSA body surface area; PASI, Psoriasis Area and Severity Index.

Figure 2.

Kaplan–Meier time to study treatment discontinuation curve in the psoriasis subset of patients with psoriatic arthritis. ADA, adalimumab; SEC, secukinumab. P-value vs. adalimumab. Numbers of patients at risk are presented for SEC and ADA.

Demographics and disease characteristics at baseline were comparable in the secukinumab and adalimumab groups, except for the proportion of patients with enthesitis, which was higher in the adalimumab group [59 of 110 (53·6%) in secukinumab vs. 69 of 101 (68·3%) in the adalimumab groups] (Table 1). ACR 20 response rates at week 52 were 76·4% in the secukinumab group vs. 68·3% in the adalimumab group [OR (vs. adalimumab) 1·53, 95% CI 0·83–2·83; P = 0·175] (Figure 3a and Table 2). At week 52, PASI 90 responses in the secukinumab and the adalimumab groups were 68·6% and 41·7%, respectively (P < 0·001) (Figure 3b and Table 2). ACR 50 responses were 54·5% (secukinumab) vs. 49·3% (adalimumab) (P = 0·386) (Figure 3c and Table 2). A total of 74·5% of patients receiving secukinumab and 66·2% of patients receiving adalimumab achieved resolution of enthesitis at week 52 (P = 0·157) (Figure 3d and Table 2) and the mean (SE) change in HAQ-DI from baseline to week 52 in the secukinumab and the adalimumab groups was −0·60 (0·049) and −0·56 (0·053), respectively (P = 0·532) (Figure 3e and Table 2).

Figure 3.

ACR 20/50, PASI 90 response rates, resolution of enthesitis, mean change from baseline in HAQ-DI through week 52 in the psoriasis subset of patients with psoriatic arthritis. N, number of active PsApatients who had body surface area (BSA) > 10% or PASI ≥ 10 affected by psoriasis at baseline. ACR, American College of Rheumatology; csDMARD, conventional synthetic disease modifyinganti-rheumatic drugs; HAQ-DI, health assessment questionnaire-disability index; PASI, psoriasis area severity index; PsA, psoriatic arthritis. P-value vs. adalimumab. Unadjusted P-values are reported at week 52. Binary outcomes (ACR 20, PASI 90, ACR 50 and resolution of enthesitis) were assessed using logistic regression; mean change from baseline in HAQ-DI was analysed using a mixed-effect model with repeated measures and HAQ-DI was analysed using logistic regression. Patients who discontinued study treatment before or at week 50 or took csDMARDs after week 36 are considered nonresponders for the visits after discontinuation or taking csDMARDs. Multiple imputation is used for all other missing data.

A total of 28·2% of patients achieved improvement in combined ACR 50 and PASI 100 response with secukinumab vs. 17·7% with adalimumab through week 52 (P = 0·06) (Figure 4 and Table 2). Other musculoskeletal outcomes such as ACR 70 response rate and the proportion of patients achieving resolution of dactylitis in the psoriasis subset are shown in Figure S1 (see Supporting Information). At week 52, PASI 100 responses in the secukinumab and the adalimumab groups were 39·1% and 23·8%, respectively (P = 0·013) (Figure 5a and Table 2). The additional skin-specific endpoints including PASI 75 (Figure 5b), quality of life (DLQI) (Figures 5c, d), and composite indices outcomes including achievement of LDA and/or REM responses/targets at week 52 in the secukinumab and the adalimumab groups (as assessed by minimal disease activity, very low disease activity, DAPSA and PASDAS) are shown in Table 2. The mean change from baseline in DLQI score in the secukinumab and the adalimumab groups was −10·27 [from the mean (SD) baseline score of 13.8] and −8·32 (from the mean baseline score of 13·3) at week 52, respectively; DLQI 0/1 response was 49·1% with secukinumab vs. 37·1% with adalimumab (P = 0·071) at week 52 (Figure 5c, d and Table 2). The mean change from baseline in SF-36 PCS/MCS and FACIT-F score in the psoriasis subset is presented Figure S2 (see Supporting Information).

Figure 4.

Combined ACR 50 + PASI 100 response rate through week 52 in the psoriasis subset of patients with psoriatic arthritis. N, number of active PsA patients who had body surface area (BSA) > 10% or PASI ≥ 10 affected by psoriasis at baseline. ACR, American College of Rheumatology; csDMARD, conventional synthetic disease modifyinganti-rheumatic drugs; PASI, psoriasis area severity index; PsA, psoriatic arthritis. P-value vs. adalimumab. Unadjusted P-values are reported at week 52. Patients who discontinued study treatment before or at week 50 or took csDMARDs after week 36 are considered nonresponders for the visits after discontinuation or taking csDMARDs. Multiple imputation is used for all other missing data.

Figure 5.

PASI 100 and PASI 75 response rates and mean change from baseline in DLQI and DLQI 0/1 response through week 52 in the psoriasis subset of patients with psoriatic arthritis. N, number of active PsA patients who had body surface area (BSA) > 10% or PASI ≥ 10 affected by psoriasis at baseline. csDMARD, conventional synthetic disease modifying anti-rheumatic drugs; DLQI, dermatology life quality index; PASI, Psoriasis Area Severity Index; PsA, psoriatic arthritis. P-value vs. adalimumab. Unadjusted P-values are reported at week 52. Patients who discontinued study treatment before or at week 50 or took (csDMARDs) after week 36 are considered nonresponders for the visits after discontinuation or taking csDMARDs. Multiple imputation is used for all other missing data. A mixed-effect model with repeated measures was used to assess mean change from baseline in DLQI.

Although the EXCEED study was designed to evaluate the efficacy of secukinumab vs. adalimumab at week 52, higher responses were observed with secukinumab from earlier timepoints. ACR 20 response rates at week 24 were 73·5% in the secukinumab group vs. 66·0% in the adalimumab group (Figure 3a). PASI 90 responses were 70·5% and 42·6% in the secukinumab and the adalimumab groups, respectively (Figure 3b). A total of 20·9% of patients achieved improvement in combined ACR 50 and PASI 100 response with secukinumab vs. 6·9% with adalimumab at week 24 (Figure 4).

The interaction testing for the efficacy outcomes between psoriasis status (stratified into 'active PsA with concomitant psoriasis' and 'nonpsoriasis subset of PsA') and the secukinumab treatment effect at week 52 was conducted and the results are shown in Table S1 (see Supporting Information).

Both secukinumab and adalimumab exhibited similar safety profiles in the overall population, which was consistent with previously published reports.[17] Separate safety analyses for this subset were not conducted.

processing....