Overall, 2148 international patients were originally randomized in the 4 phase 3 studies. A total of 279 US patients (13.0%) were included in this pooled analysis. The patients from FUTURE 2 who received treatment with secukinumab 75 mg were not included, as it is not an FDA- or EC-approved dose for adults.

Baseline characteristics for the US cohort were generally similar across treatment groups (Table 1). Of US patients, 55.6% were women and 55.2% had been previously treated with TNFis. The mean time since diagnosis was 7.0 years. The mean body weight and BMI of US patients included in this study (92.0 kg and 32.3 kg/m2, respectively) were higher than that of non-US patients from FUTURE 2–5 (83.5 kg and 29.1 kg/m2) and indicated an obese population on average (Supplementary Material: Table S1). Enthesitis was present in 69.9% of US patients and dactylitis was present in 38.7%, both of which were slightly higher than the rates in non-US patients (61.5 and 35.4%, respectively). Mean TJC78 was 25.2 and mean SJC76 was 13.7, both higher than in non-US patients (20.6 and 10.5, respectively). US patients more frequently had previous TNFi exposure and less frequently used concomitant methotrexate at baseline than non-US patients.

At week 16, ACR20 response rates in US patients were significantly higher with secukinumab 300 mg (59.7% [P < 0.0001]) and secukinumab 150 mg with loading dose (43.4% [P < 0.0001]) than placebo (15.6%) (Fig. 1). Response rates with secukinumab 150 mg without loading dose were numerically higher than with placebo but did not achieve significance (23.5% [P = 0.30]). Responses with secukinumab were seen as early as week 2. ACR50 and ACR70 responses in US patients were also higher with any dose of secukinumab than with placebo. Likewise, a larger proportion of patients who received secukinumab than those treated with placebo had a 100% reduction in TJCs and SJCs and resolution of enthesitis and dactylitis (Table 2).

Achievement of ACR responses among US patients through week 16. ACR American College of Rheumatology, LD loading dose, SEC secukinumab. * P < 0.05 vs. placebo

The week 16 PASI response rates in US patients were higher with secukinumab than with placebo, more so with secukinumab 300 mg than either secukinumab 150 mg regimen (Fig. 2a). At week 16, PASI90/100 response rates were 47.1%/23.5% with secukinumab 300 mg, 22.2%/11.1% with secukinumab 150 mg with loading dose, and 18.2%/9.1% with secukinumab 150 mg without loading dose vs. 5.3%/2.6% with placebo. Secukinumab also led to benefits in other disease domains of PsA (Table 2). A larger proportion of patients treated with secukinumab than placebo experienced improved nail disease: rates of mNAPSI75 were 36.4, 24.6, and 15.0% in the groups receiving secukinumab 300, 150, and 150 mg without loading dose, respectively, vs. 9.1% in the placebo group. Greater rates of improvements in health-related quality of life at week 16 were observed in US patients treated with secukinumab vs. placebo (Fig. 2b). Higher proportions of patients treated with secukinumab achieved MCIDs in HAQ-DI, SF-36 PCS, and SF-36 MCS scores than patients receiving placebo. Similar results were observed when evaluating treatment response across individual MDA components (Supplementary Material: Fig. S2). Overall, secukinumab 300 mg tended to lead to higher response rates than secukinumab 150 mg. For most outcomes, higher response rates were associated with secukinumab 150 mg with loading dose than with secukinumab 150 mg without loading dose.

Achievement of A PASI75, PASI90, and PASI100 and B improvements ≥ MCID in health-related quality-of-life measures in US patients through week 16. HAQ-DI Health Assessment Questionnaire Disability Index, LD loading dose, MCID minimal clinically important difference, PASI Psoriasis Area and Severity Index, SEC secukinumab, SF-36 MCS 36-Item Short Form Health Survey Mental Component. Score; SF-36 PCS, 36-Item Short Form Health Survey Physical Component Score. *P < 0.05 vs. placebo

Among the US patients who were TNFi naive, all three secukinumab dose groups had significantly higher response rates vs. placebo for ACR20, ACR50, and ACR70 (Fig. 3). TNF-IR patients generally had lower response rates than those who were TNFi naive, although the groups receiving secukinumab 300 mg and 150 mg with loading dose still had significantly higher response rates than the placebo group for ACR20 and ACR50.

Achievement of ACR responses through week 16 among US patients grouped by TNFi and BMI status. ACR American College of Rheumatology, BMI body mass index, IR inadequate response, LD loading dose, SEC secukinumab, TNFi tumor necrosis factor inhibitor. *P < 0.05 vs. placebo

In patients with BMI > 30 kg/m2, the ACR20, ACR50, and ACR70 response rates were numerically higher in the secukinumab 300-mg than both secukinumab 150-mg dose groups (Fig. 3). Among patients with BMI ≤ 30 kg/m2, the response rates in the group receiving secukinumab 150 mg without loading dose were notably lower than those in the groups receiving secukinumab 300 mg and 150 mg with loading dose. Comparison between the two BMI subgroups showed no clear trends.

For all of the ACR binary outcomes, the logistic regression analysis of responses in the US cohort found that the odds ratios significantly (P < 0.05) favored the groups receiving secukinumab 300 mg and secukinumab 150 mg with loading dose over placebo (Fig. 4). The odds ratios for the groups receiving secukinumab 150 mg without loading dose vs. placebo were > 1 but were not significant.

Odds ratios (95% CI) of secukinumab vs. placebo for ACR and PASI response rates from baseline to week 16 in the US population from a logistic regression model with treatment as a factor and BMI, Disease Activity Score 28-joint count using C-reactive protein, Disease Activity Score 28-joint count using erythrocyte sedimentation rate, SJC76, TJC78, and TNFi status (naive vs. inadequate response) as covariates (nonresponder imputation). ACR American College of Rheumatology, BMI body mass index, CI confidence interval, LD loading dose, PASI Psoriasis Area and Severity Index, PBO placebo, SEC secukinumab, SJC76 swollen joint count of 76 joints, TJC78 tender joint count of 78 joints, TNFi tumor necrosis factor inhibitor. Error bars indicate 95% CI

For the PASI75 outcomes, the logistic regression analysis found that the odds ratios significantly (P < 0.05) favored all three secukinumab groups over placebo (Fig. 4). For the PASI90 and PASI100 outcomes, only the secukinumab 300-mg group was significantly (P < 0.05) favored over placebo (Fig. 4).

Radiographic progression among US patients was assessed using available data collected at week 24 in FUTURE 5 (Supplementary Material: Fig. S3). No trend was observed among treatment groups for the proportion of patients with no structural progression at week 24, defined as vdH-mTSS ≤ 0. However, patients in all three secukinumab groups experienced lower mean (SD) change from baseline in vdH-mTSS at week 24 (300 mg, – 0.01 [0.84]; 150 mg with loading dose, – 0.05 [1.19]; 150 mg without loading dose, – 0.28 [0.81]) compared with placebo (0.66 [2.20]).

The frequency of all treatment-emergent AEs through week 16 was similar for patients receiving secukinumab 300 mg (51.4%), secukinumab 150 mg with loading dose (54.2%), secukinumab 150 mg without loading dose (55.9%), and placebo (64.4%). The most frequent treatment-emergent AEs in the groups receiving secukinumab 300 mg, secukinumab 150 mg with loading dose, secukinumab 150 mg without loading dose, and placebo, respectively, were upper respiratory tract infection (5.6, 9.6, 8.8, and 10.0%), nasopharyngitis (1.4, 2.4, 8.8, and 7.8%), nausea (1.4, 7.2, 0, and 6.7%), and sinusitis (2.8, 6.0, 2.9, and 2.2%) (Supplementary Material: Table S2). No cases of inflammatory bowel disease, uveitis, major adverse cardiovascular events, venous thromboembolism, or tuberculosis were observed. Treatment-emergent AEs only led to discontinuation in one instance: an AE of chronic lymphocytic leukemia in the groups receiving secukinumab 150 mg with loading dose.

Serious AEs were reported in one patient in the secukinumab 300-mg group (n = 1 each of dehydration, traumatic amputation of the limb, and osteomyelitis); four patients in the secukinumab 150-mg group (n = 1 each of spontaneous abortion, biliary dyskinesia, chronic lymphocytic leukemia, coronary artery disease, ectopic pregnancy, gastritis, and suicidal ideation); three patients in the placebo group (n = 1 each of cellulitis, Escherichia urinary tract infection, and infectious mononucleosis). Safety in this cohort appeared similar to that observed in the full study population.