Efficacy and Safety of Secukinumab in US Patients with Psoriatic Arthritis: A Subgroup Analysis of the Phase 3 FUTURE Studies

Psoriatic arthritis (PsA) is a chronic inflammatory disease that is characterized by peripheral arthritis, enthesitis, dactylitis, skin and nail psoriasis, and axial disease [1]. PsA is a progressive disease that can lead to irreversible joint damage if not treated early and appropriately [2, 3]. Additionally, PsA is associated with reduced quality of life, physical function, and work productivity [4]. In the United States, the estimated prevalence of PsA in the general population is reported to be in the range of 0.06–0.25% [5]. The prevalence of PsA is higher in people with psoriasis; of the 3.2% (95% CI 2.6–3.7%) of US adults who have psoriasis [6], 19.0% (95% CI 16.3–21.8%) also have PsA [7].

Interleukin (IL)-17A is a proinflammatory cytokine that is key to multiple biological processes characteristic of PsA, including inflammation of joints, enthesitis, cartilage and bone erosion, and pathological new bone growth [8‐10]. Secukinumab, a selective inhibitor of interleukin 17A, demonstrated rapid and significant improvement in the signs and symptoms of PsA and had a favorable safety profile in the global phase 3 FUTURE studies (FUTURE 1–5) [11‐15]. Patients treated with secukinumab achieved significantly higher response rates than those treated with placebo in various efficacy outcomes—including 20, 50, and 70% improvement per American College of Rheumatology (ACR20/50/70) criteria and 75, 90, and 100% improvement in Psoriasis Area and Severity Index (PASI75/90/100) responses—and experienced significant improvements in quality-of-life measures. Secukinumab has shown significant efficacy across each of the disease manifestations that characterize PsA, including joint and skin symptoms [16, 17]. Furthermore, secukinumab has demonstrated sustained efficacy and safety through 5 years and sustained inhibition of radiographic structural progression through 2 years [18‐20]. The efficacy of secukinumab has also been demonstrated in both biologic-naive patients and those who have been treated previously with tumor necrosis factor inhibitors (TNFis) but had inadequate response or stopped treatment for safety or tolerability reasons (TNF-IR) [11‐15].

US patients were a minority of the total population in the pivotal phase 3 FUTURE trials. These patients had baseline disease characteristics different than those of the global population, and the impact of these differences on treatment response is unknown. A preliminary analysis of the international FUTURE trials found that US patients tended to have characteristics indicating harder-to-treat disease: US patients were heavier; had higher tender and swollen joint counts; and were more likely to have enthesitis, dactylitis, and prior exposure to TNFis than patients from the rest of the world [21]. The objective of this study was to evaluate secukinumab in the US patient subpopulation of the FUTURE studies and report pooled efficacy and safety findings for secukinumab 300 mg and secukinumab 150 mg (with and without loading dose) vs. placebo in this challenging-to-treat population.

Secukinumab was efficacious in US patients with PsA in the pooled FUTURE 2–5 studies, leading to rapid improvements in clinical endpoints and quality-of-life measures. US patients constituted a minority of the FUTURE 2–5 trials, and the demographic and baseline disease-state parameters of US patients indicated that this was a challenging-to-treat subgroup relative to the total population of the FUTURE 2–5 studies [12‐15, 21] and relative to the international population in the recently published EXCEED trial of secukinumab in PsA [26]. Mean TJC78 and SJC76 were 25.2 and 13.7, respectively, which were higher than the range of means reported in the international populations of the FUTURE 2–5 and EXCEED studies (TJC78, 20.0–22.6; SJC76, 9.7–11.7) [12‐15, 26]. Prior TNFi use was also higher in the US patients: only 44.8% were TNFi naive vs. 64.8–76.2% in the full FUTURE 2–5 studies [12‐15]. The mean weight of the US patients was higher than the international populations of the FUTURE 2–5 and EXCEED studies (92.0 vs. 83.4–87.1 kg) [12‐15, 27], and pharmacokinetic studies of secukinumab have shown that its clearance is proportional to a patient’s weight [28], potentially reducing efficacy in heavier patients. The outcomes of treatment with biologics in PsA (and psoriasis and other spondyloarthropathies) have generally been worse in obese patients [29, 30], so the higher mean BMI of the US patients in the present study relative to that of the international FUTURE 2–5 study population again suggested a challenging-to-treat population. More women than men were included in the pooled US patient population (55.6 vs. 44.4%). There is some evidence that women are less likely to respond to biologic treatment for PsA than men [31, 32]. Concomitant methotrexate use was lower in the pooled US patient population compared with the international populations of the FUTURE 2–5 studies (29.7 vs. 46.6–50.1%) [12‐15].

This analysis found that there were numerical differences in responses to the three dosing regimens among the US population, including in the achievement of MDA components. In general, US patients treated with secukinumab 300 mg and secukinumab 150 mg with loading dose achieved the highest response rates, including ACR50 and ACR70 responses and the proportions of patients showing at least an MCID improvement in the health-related quality-of-life measure HAQ-DI. The responses to secukinumab observed in the US patients in this analysis, particularly those who received secukinumab 300 mg or 150 mg with loading dose, were largely similar to the responses observed for the total FUTURE 2–5 population [12‐15]. Considering that treatment with secukinumab 150 mg with and without a loading dose produced similar ACR20/50/70 responses, resolution of enthesitis and dactylitis, and improvements in HAQ-DI in the total FUTURE 4 population [14], the lower efficacy observed with secukinumab 150 mg without loading dose in US patients may be explained partially by higher mean BMI at baseline and slightly higher disease activity at baseline vs. non-US patients. The similar ACR20/50 responses between patients who received 150 mg secukinumab without a loading dose and those receiving placebo are also likely due to the low ACR response rates in patients who were TNFi inadequate responders (Fig. 3). Therefore, a secukinumab loading dose may be particularly important for patients with disease characteristics indicating challenging-to-treat PsA, such as higher body weight, higher tender and swollen joint counts, the presence of dactylitis or enthesitis, and previous TNFi exposure. Radiographic progression among US patients at week 24 as determined by mean change in vdH-mTSS was similar to that observed in the overall population of FUTURE 5, with patients receiving secukinumab experiencing less change from baseline compared with patients receiving placebo [15]. Overall, efficacy and safety results from this analysis are consistent with primary results from the recent CHOICE study, a phase 3 trial evaluating secukinumab in a biologic-naive population of US patients with PsA [33]. These results demonstrate that secukinumab was effective for the treatment of PsA in US patients with clinical characteristics indicating harder-to-treat disease and suggest that other patient populations with similar characteristics (such as high levels of obesity or prior TNFi exposure) would also benefit from treatment with secukinumab.

The TNFi subgroup analysis, which was compatible with previous findings from the global FUTURE 5 population [15], showed that ACR response rates were generally higher at week 16 in TNFi-naive patients than in TNF-IR patients, suggesting that secukinumab is effective as first-line biologic therapy in the US population. In the present study, despite being a more challenging-to-treat population, the proportion of TNFi-naive patients in the secukinumab 300-mg dose group who achieved ACR20 at week 16 was comparable to that in the EXCEED study, which used the same dosing in TNFi-naive patients (63.3 vs. 66%) [27]. The BMI subgroup analysis was equivocal, which may have been due to the relatively small subgroup sizes combined with the fact that patients were not stratified on the basis of weight or BMI at randomization. In the CHOICE study by comparison, secukinumab resulted in similar achievement of ACR responses in patients with BMI > 30 kg/m² or BMI ≤ 30 kg/m², with numerically greater ACR50/70 responses among patients in the lower BMI subgroup [33].

Our logistic regression analysis of the US cohort at week 16 suggested that using secukinumab with a loading dose may be important to ensure optimal efficacy, regardless of BMI, baseline disease state, or TNFi status. The regression analysis also suggested that again, regardless of BMI, baseline disease state, or TNFi status, patients with PsA plus skin symptoms might benefit more from secukinumab 300 mg than from secukinumab 150 mg. US healthcare providers should be aware that patients may require a loading-dose regimen to achieve optimal treatment outcomes and lower associated long-term healthcare costs.

This study had the limitations inherent to post hoc analyses. Logistic regression analyses were performed without adjustment for multiple comparisons, and nominal P values were calculated for hypothesis generation. In addition, some of the subgroups analyzed were quite small (n < 20). Interpretation of the effect of BMI was limited by the fact that patients were not stratified on the basis of weight or BMI at randomization. Therefore, caution should be exercised in drawing conclusions from these subgroup analyses. The description of US patients in this analysis as challenging-to-treat was intended as a comparison to the total FUTURE 2–5 population and is not intended to refer to any specific definitions or criteria for PsA. Radiographic progression data were only collected in FUTURE 5, and the earliest time of analysis was week 24. Additionally, while patients in all three secukinumab groups experienced lower mean changes from baseline in vdH-mTSS at week 24 vs. placebo, there was variability in these scores within each group. As such, these radiographic findings should be interpreted with caution.

These results provide valuable insight into the efficacy and safety of secukinumab in a subgroup of US patients with PsA in the FUTURE 2–5 trials who were heavier and had clinical characteristics that indicated more challenging-to-treat disease compared with patients from the rest of the world. Our findings are consistent with those from previous studies and show that secukinumab is an effective treatment for this challenging subpopulation of patients with PsA, with a safety profile similar in US patients to that observed in the full study population. These results suggest that other patient populations who have disease characteristics indicating harder-to-treat disease (such as obesity, higher tender and swollen joint counts, the presence of dactylitis or enthesitis, or previous TNFi exposure) may also benefit from secukinumab and that patients with PsA, particularly those with active psoriasis, may benefit more from secukinumab 300 vs. 150 mg. This analysis also suggests that a loading-dose regimen—particularly for patients receiving secukinumab 150 mg—increases the odds of optimal outcomes in US patients with PsA treated with secukinumab.