Comparative effectiveness of biologic agents for the treatment of psoriasis in a real-world setting: Results from a large, prospective, observational study (Psoriasis Longitudinal Assessment and Registry [PSOLAR])

Open AccessPublished:February 04, 2016DOI:https://doi.org/10.1016/j.jaad.2015.12.017

      Background

      Comparing effectiveness of biologics in real-world settings will help inform treatment decisions.

      Objectives

      We sought to compare therapeutic responses among patients initiating infliximab, adalimumab, or etanercept versus ustekinumab during the Psoriasis Longitudinal Assessment and Registry (PSOLAR).

      Methods

      Proportions of patients achieving a Physician Global Assessment score of clear (0)/minimal (1) and mean decrease in percentage of body surface area with psoriasis were evaluated at 6 and 12 months. Adjusted logistic regression (Physician Global Assessment score 0/1) and analysis of covariance (percentage of body surface area with psoriasis) were performed to determine treatment factors associated with effectiveness.

      Results

      Of 2541 new users on registry, 2076 had efficacy data: ustekinumab (n = 1041), infliximab (n = 116), adalimumab (n = 662), and etanercept (n = 257). Patients receiving tumor necrosis factor-alpha(-α) inhibitors were significantly less likely to achieve Physician Global Assessment score 0/1 versus ustekinumab (infliximab [odds ratio {OR} 0.396, P < .0001], adalimumab [OR 0.686, P = .0012], etanercept [OR 0.554, P = .0003] at 6 months and infliximab [OR 0.449, P = .0040] at 12 months). Mean decrease in percentage of body surface area with psoriasis was significantly greater for ustekinumab versus adalimumab (point estimate 1.833, P = .0020) and etanercept (point estimate 3.419, P < .0001) at 6 months and versus infliximab (point estimate 3.945, P = .0005) and etanercept (point estimate 2.778, P = .0007) at 12 months.

      Limitations

      Treatment selection bias and limited data for doing adjustments are limitations.

      Conclusions

      In PSOLAR, effectiveness of ustekinumab was significantly better versus all 3 tumor necrosis factor-α inhibitors studied for the majority of comparisons at 6 and 12 months.

      Key words

      Abbreviations used:

      %BSA (percentage of body surface area with psoriasis), CI (confidence interval), DLQI (Dermatology Life Quality Index), HRQoL (health-related quality of life), OR (odds ratio), PGA (Physician Global Assessment), PSOLAR (Psoriasis Longitudinal Assessment and Registry), SD (standard deviation), TNF (tumor necrosis factor)
      • Several biologics are effective for the treatment of psoriasis.
      • In a real-world setting, effectiveness was significantly better for ustekinumab than for tumor necrosis factor-α inhibitors for the majority of comparisons at 6 and 12 months.
      • These comparative effectiveness analyses of currently available biologic therapies will help inform treatment decisions in patients with psoriasis.
      Comparative effectiveness data are important to both the physician and patient for differentiating between treatment options when selecting a therapy, especially as the options for biologic therapy continue to expand.
      • Puig L.
      On clinical thresholds, clinical equivalents and indirect comparisons of biological treatments for moderate-to-severe psoriasis.
      Consequently, there is increasing demand for comparative effectiveness data in the current health care environment to better inform patients and physicians when choosing appropriate treatments for psoriasis.
      • Conway P.H.
      • Clancy C.
      Comparative-effectiveness research–implications of the Federal Coordinating Council's report.
      • Iglehart J.K.
      Prioritizing comparative-effectiveness research–IOM recommendations.
      • Abuabara K.
      • Wan J.
      • Troxel A.B.
      • et al.
      Variation in dermatologist beliefs about the safety and effectiveness of treatments for moderate to severe psoriasis.
      Selection of therapy depends on numerous factors, including efficacy, safety, response over time, convenience, and affordability. Furthermore, some studies have demonstrated a correlation between measures of health-related quality of life (HRQoL) and effectiveness, indicating the importance of patient-reported outcomes.
      • Krueger G.G.
      • Langley R.G.
      • Finlay A.Y.
      • et al.
      Patient-reported outcomes of psoriasis improvement with etanercept therapy: results of a randomized phase III trial.
      • Revicki D.A.
      • Willian M.K.
      • Menter A.
      • et al.
      Impact of adalimumab treatment on patient-reported outcomes: results from a phase III clinical trial in patients with moderate to severe plaque psoriasis.
      • Revicki D.
      • Willian M.K.
      • Saurat J.H.
      • et al.
      Impact of adalimumab treatment on health-related quality of life and other patient-reported outcomes: results from a 16-week randomized controlled trial in patients with moderate to severe plaque psoriasis.
      • Reich K.
      • Nestle F.O.
      • Papp K.
      • et al.
      Improvement in quality of life with infliximab induction and maintenance therapy in patients with moderate-to-severe psoriasis: a randomized controlled trial.
      • Lebwohl M.
      • Papp K.
      • Han C.
      • et al.
      Ustekinumab improves health-related quality of life in patients with moderate-to-severe psoriasis: results from the PHOENIX 1 trial.
      • Langley R.G.
      • Feldman S.R.
      • Han C.
      • et al.
      Ustekinumab significantly improves symptoms of anxiety, depression, and skin-related quality of life in patients with moderate-to-severe psoriasis: results from a randomized, double-blind, placebo-controlled phase III trial.
      To date, a number of head-to-head clinical trials and meta-analyses have been performed to compare the efficacy of various biologic agents for psoriasis; results pertain largely to short-term outcomes and do not always reflect findings in typical practice.
      • Griffiths C.E.
      • Strober B.E.
      • van de Kerkhof P.
      • et al.
      Comparison of ustekinumab and etanercept for moderate-to-severe psoriasis.
      • Langley R.G.
      • Elewski B.E.
      • Lebwohl M.
      • et al.
      Secukinumab in plaque psoriasis–results of two phase 3 trials.
      • Reich K.
      • Burden A.D.
      • Eaton J.N.
      • et al.
      Efficacy of biologics in the treatment of moderate to severe psoriasis: a network meta-analysis of randomized controlled trials.
      • Schmitt J.
      • Rosumeck S.
      • Thomaschewski G.
      • et al.
      Efficacy and safety of systemic treatments for moderate-to-severe psoriasis: meta-analysis of randomized controlled trials.
      • Galvan-Banqueri M.
      • Marin Gil R.
      • Santos Ramos B.
      • et al.
      Biological treatments for moderate-to-severe psoriasis: indirect comparison.
      • Lin V.W.
      • Ringold S.
      • Devine E.B.
      Comparison of ustekinumab with other biological agents for the treatment of moderate to severe plaque psoriasis: a Bayesian network meta-analysis.
      • Lucka T.C.
      • Pathirana D.
      • Sammain A.
      • et al.
      Efficacy of systemic therapies for moderate-to-severe psoriasis: a systematic review and meta-analysis of long-term treatment.
      • Griffiths C.E.
      • Reich K.
      • Lebwohl M.
      • et al.
      Comparison of ixekizumab with etanercept or placebo in moderate-to-severe psoriasis (UNCOVER-2 and UNCOVER-3): results from two phase 3 randomised trials.
      • Sullivan P.
      • Goldmann D.
      The promise of comparative effectiveness research.
      In addition, real-world evidence of comparative effectiveness of biologics is generally lacking, with few such studies reported in the literature.
      • Gelfand J.M.
      • Wan J.
      • Callis Duffin K.
      • et al.
      Comparative effectiveness of commonly used systemic treatments or phototherapy for moderate to severe plaque psoriasis in the clinical practice setting.
      • Naldi L.
      • Svensson A.
      • Zenoni D.
      • et al.
      Comparators, study duration, outcome measures and sponsorship in therapeutic trials of psoriasis: update of the EDEN Psoriasis Survey 2001-2006.
      • Takeshita J.
      • Wang S.
      • Shin D.B.
      • et al.
      Comparative effectiveness of less commonly used systemic monotherapies and common combination therapies for moderate to severe psoriasis in the clinical setting.
      Psoriasis Longitudinal Assessment and Registry (PSOLAR) is an ongoing, longitudinal, prospective, international, observational study that follows patients with psoriasis who are receiving, or are eligible to receive, systemic or biologic therapies.
      • Papp K.A.
      • Strober B.
      • Augustin M.
      • et al.
      PSOLAR: design, utility, and preliminary results of a prospective, international, disease-based registry of patients with psoriasis who are receiving, or are candidates for, conventional systemic treatments or biologic agents.
      • Kimball A.B.
      • Leonardi C.
      • Stahle M.
      • et al.
      Demography, baseline disease characteristics and treatment history of patients with psoriasis enrolled in a multicentre, prospective, disease-based registry (PSOLAR).
      In addition to generating safety data on biologic therapies, this registry also captures data relevant to the effectiveness of biologics for patients with psoriasis. The objective of this analysis is to compare the effectiveness of tumor necrosis factor (TNF)-α inhibitors (infliximab, adalimumab, and etanercept) with effectiveness of ustekinumab based on standard clinical and HRQoL measures (ie, Physician Global Assessment [PGA], percentage of body surface area with psoriasis [%BSA], and Dermatology Life Quality Index [DLQI]) after 6 and 12 months of treatment.

      Methods

      The design and utility of PSOLAR have been reported previously.
      • Papp K.A.
      • Strober B.
      • Augustin M.
      • et al.
      PSOLAR: design, utility, and preliminary results of a prospective, international, disease-based registry of patients with psoriasis who are receiving, or are candidates for, conventional systemic treatments or biologic agents.
      • Kimball A.B.
      • Leonardi C.
      • Stahle M.
      • et al.
      Demography, baseline disease characteristics and treatment history of patients with psoriasis enrolled in a multicentre, prospective, disease-based registry (PSOLAR).
      A core set of data (including disease activity, HRQoL, medications, and adverse events) are collected every 6 months. Physicians prescribe treatments based on routine clinical practice. PSOLAR enrollment is complete; data for these analyses cover the period from June 20, 2007, to August 23, 2013. In all, 93 institutional review boards or ethics committees approved the registry protocol, and all patients provided written informed consent before study procedures were initiated.

       Study population

      The study population included 4 treatment groups of patients initiating a new biologic (ustekinumab, infliximab, adalimumab, or etanercept); only the first biologic started during registry participation was analyzed. Patients may have been bionaive or may have been exposed before enrollment to a biologic other than their newly initiated treatment in the registry. Patients restarting a biologic received before enrollment were excluded from the analyses. Evaluations were limited to patients who had baseline data and continued to receive their initiated therapy at their 6- and/or 12-month visits. Baseline psoriasis severity was assessed at the closest visit before the first dose of the newly initiated biologic. Patients receiving concomitant treatment with another systemic therapy for psoriasis (eg, methotrexate, cyclosporine, mycophenolate mofetil, fumarates, oral retinoids, apremilast, or systemic corticosteroids) or phototherapy were excluded, although patients receiving topical therapy (eg, retinoids and corticosteroids) were included in the analyses.

       Outcome measures

      Clinical effectiveness was measured based on PGA (ie, induration, scaling, and erythema of lesions graded on a scale ranging from clear [0] to severe [5]) and %BSA. HRQoL was assessed using the DLQI, which is a skin disease-specific, patient-reported questionnaire assessing 10 items, with an overall potential score ranging from 0 (not affected at all) to 30 (very much affected).
      • Finlay A.Y.
      • Khan G.K.
      Dermatology Life Quality Index (DLQI)–a simple practical measure for routine clinical use.
      A DLQI score of 0 or 1 indicates no impact on a patient's life, and a decrease of 5 or more points is considered clinically meaningful.
      • Samsa G.
      • Edelman D.
      • Rothman M.L.
      • et al.
      Determining clinically important differences in health status measures: a general approach with illustration to the Health Utilities Index Mark II.

       Statistical analyses

      Baseline information captured at the time of starting a new biologic is presented for each of the biologics. Clinical effectiveness end points included the proportion of patients achieving a PGA score of 0/1, the proportion of patients achieving a reduction of 2 or more points in PGA score, and the mean decrease in %BSA across treatment groups at 6 and 12 months. HRQoL end points included mean improvement in DLQI score and the proportion of patients achieving a clinically meaningful change (ie, reduction ≥5 points) in DLQI score stratified by baseline score (ie, <10 and ≥10) among treatment groups at 6 and 12 months.
      Data were captured for each study visit within a maximum window of ±3 months; if the visit occurred outside of this window, data were not included in the analyses. If 2 visits fell within a given follow-up time window, data from the visit closest to the designated time point were used. Patients who discontinued because of lack of effectiveness before a scheduled time-point visit were considered treatment failures (ie, PGA score was entered as not achieving clear [0] or minimal [1], and the change from baseline was entered as 0 for %BSA and DLQI score) for that and all subsequent scheduled visits. Data from patients who left the registry or were lost to follow-up were not included in the analyses.
      The proportions of patients achieving PGA score of 0/1 at 6 and 12 months were evaluated using a logistic regression model. Adjusted odds ratios (OR), 95% confidence intervals (CI), and corresponding P values (Wald χ2 test) were calculated for the treatment effect of each TNF-α inhibitor (infliximab, adalimumab, and etanercept) compared with ustekinumab at 6 and 12 months. Covariates other than treatment effect were also included in the analyses (Table I). Using the same covariates, an analysis of covariance was performed to compare the decrease in %BSA and improvement in DLQI score from baseline at 6 and 12 months for each TNF-α inhibitor versus ustekinumab. The changes in %BSA and DLQI score for each treatment were also summarized using the adjusted least squares means from the analysis of covariance analysis for treatment effects.
      Table IFactors other than treatment effects included in multivariate analyses of effectiveness in Psoriasis Longitudinal Assessment and Registry
      • Age at diagnosis/10 y
        To evaluate odds ratio for every 10-y increase in age or every 5 y of disease, baseline continuous variables of age and duration of disease were transformed to age divided by 10 and duration of disease divided by 5.
      • Duration of disease/5 y
        To evaluate odds ratio for every 10-y increase in age or every 5 y of disease, baseline continuous variables of age and duration of disease were transformed to age divided by 10 and duration of disease divided by 5.
      • Region: European Union/Latin America vs North America
      • Race: non-white vs white
      • Gender: male vs female
      • Overweight/obesity class I (25 < BMI < 35) vs normal weight (BMI <25)
      • Obesity class II-III (BMI ≥35) vs normal weight (BMI <25)
      • Psoriatic arthritis vs no psoriatic arthritis at baseline
      • Smoking (current/past vs never) at baseline
      • Alcohol (current/past vs never) at baseline
      • PGA score 2/3 vs 0/1 at baseline
      • PGA score 4/5 vs 0/1 at baseline
      • Percentage of BSA with psoriasis at baseline
      • DLQI score ≥10 vs <10 at baseline
      • Discontinued because of adverse event vs continued in registry
      • Discontinued insurance vs continued in registry
      • Discontinued (other reason)
        Discontinued (other reason) included all reasons except adverse event and insurance.
        vs continued in registry
      • Prior treatment with tumor necrosis factor-α inhibitors vs bionaive
      • Prior treatment with ustekinumab vs bionaive
      • Prior treatment with other biologics
        Prior other biologics included any biologic other than ustekinumab or a tumor necrosis factor-α inhibitor.
        vs bionaive
      BMI, Body mass index; BSA, body surface area; DLQI, Dermatology Life Quality Index; PGA, Physician Global Assessment.
      To evaluate odds ratio for every 10-y increase in age or every 5 y of disease, baseline continuous variables of age and duration of disease were transformed to age divided by 10 and duration of disease divided by 5.
      Discontinued (other reason) included all reasons except adverse event and insurance.
      Prior other biologics included any biologic other than ustekinumab or a tumor necrosis factor-α inhibitor.

      Results

       Patient characteristics

      Of the 2541 patients who initiated a new biologic on registry, 2076 (ustekinumab [n = 1041], infliximab [n = 116], adalimumab [n = 662], and etanercept [n = 257]) had adequate data at baseline and at 6 and/or 12 months and were included in the analyses (Table II). The mean age was 46.5 years; slightly over half of the patients were male (57.4%) and most were white (83.5%). The mean duration of psoriasis was 17.5 years, ranging from 14.7 (etanercept group) to 19.1 (ustekinumab group) years. Approximately one third of patients (34.9%) had psoriatic arthritis at baseline; the infliximab group (44.0%) had the highest proportion of patients with psoriatic arthritis.
      Table IIDemographic and patient characteristics at enrollment or start of a new biologic (baseline) during the registry
      Ustekinumab

      N = 1041
      Infliximab

      N = 116
      Adalimumab

      N = 662
      Etanercept

      N = 257
      Total

      N = 2076
      Total includes patients who continued to receive their initiated therapy at the 6- and/or 12-mo visit and for whom data were available for at least 1 visit. Patients who discontinued because of lack of effectiveness were included as treatment failures (ie, not achieving PGA score of 0/1 and no changes from baseline for %BSA and DLQI) at subsequent visits.
      Age, y, N10411166622572076
       Mean ± SD46.3 ± 13.5447.9 ± 13.1546.7 ± 13.5946.8 ± 13.9446.5 ± 13.58
      Age at onset of psoriasis,
      Data were collected at enrollment in the Psoriasis Longitudinal Assessment and Registry; all other data were collected at the start of therapy.
      y, N
      10351166582552064
       Mean ± SD27.7 ± 15.3131.2 ± 14.2531.1 ± 15.5732.6 ± 16.4129.6 ± 15.59
      Male gender,
      Data were collected at enrollment in the Psoriasis Longitudinal Assessment and Registry; all other data were collected at the start of therapy.
      N
      10411166622572076
       n (%)591 (56.8)73 (62.9)384 (58.0)144 (56.0)1192 (57.4)
      White race,
      Data were collected at enrollment in the Psoriasis Longitudinal Assessment and Registry; all other data were collected at the start of therapy.
      N
      10411166622562075
       n (%)900 (86.5)106 (91.4)511 (77.2)216 (84.4)1733 (83.5)
      BMI, kg/m2, N10321166582522058
       Mean ± SD31.2 ± 7.1532.6 ± 7.5730.5 ± 6.7030.0 ± 7.0430.9 ± 7.04
      Years since psoriasis diagnosis, y, N10351166582552064
       Mean ± SD19.1 ± 13.0717.2 ± 12.3516.1 ± 12.1714.7 ± 13.1917.5 ± 12.86
      Psoriatic arthritis,
      Data were collected at enrollment in the Psoriasis Longitudinal Assessment and Registry; all other data were collected at the start of therapy.
      N
      10411166622572076
       n (%)349 (33.5)51 (44.0)232 (35.0)92 (35.8)724 (34.9)
      Prior therapy
       Biologic agents, N10411166622572076
      Ustekinumab016 (13.8)43 (6.5)10 (3.9)69 (3.3)
      Infliximab179 (17.2)047 (7.1)18 (7.0)244 (11.8)
      Adalimumab389 (37.4)42 (36.2)039 (15.2)470 (22.6)
      Etanercept497 (47.7)48 (41.4)339 (51.2)0884 (42.6)
       Other therapies, N10401156622572074
      Phototherapy648 (62.3)60 (52.2)344 (52.0)106 (41.2)1158 (55.8)
      Immunomodulators538 (51.7)64 (55.7)258 (39.0)100 (38.9)960 (46.3)
      Methotrexate430 (41.3)49 (42.6)214 (32.3)81 (31.5)774 (37.3)
      Cyclosporine247 (23.8)25 (21.7)93 (14.0)28 (10.9)393 (18.9)
      Data are presented as no. of patients (%), unless otherwise indicated.
      BMI, Body mass index; SD, standard deviation.
      Total includes patients who continued to receive their initiated therapy at the 6- and/or 12-mo visit and for whom data were available for at least 1 visit. Patients who discontinued because of lack of effectiveness were included as treatment failures (ie, not achieving PGA score of 0/1 and no changes from baseline for %BSA and DLQI) at subsequent visits.
      Data were collected at enrollment in the Psoriasis Longitudinal Assessment and Registry; all other data were collected at the start of therapy.
      Of patients included in the analyses, 42.6% had been exposed to etanercept, 22.6% to adalimumab, 11.8% to infliximab, and 3.3% to ustekinumab before enrollment. Before entering the registry, 55.8% of patients had used phototherapy (numerically higher in the ustekinumab group) and 46.3% of patients had used nonbiologic immunomodulators (mostly methotrexate [37.3%]). Baseline PGA, %BSA, and DLQI values for patients considered in the 6-month analyses were comparable with those for patients in the 12-month analyses (Table III). Generally, baseline clinical values numerically reflected more severe disease in the infliximab group. During participation in the registry, patients generally received treatment according to the prescribing information for the biologic agent they initiated, and the proportion of patients receiving standard doses at regular intervals was similar at 6 and 12 months (Supplemental Table I; available at http://www.jaad.org). Effectiveness data were missing for 465 patients who could not be included in any analyses. The infliximab group had the highest proportion of patients with missing data (30%), whereas other groups were missing data for 14% to 21% of patients. In general, characteristics of the patients with missing data were comparable with those of patients evaluated in the analyses.
      Table IIIDisease characteristics and patient–reported assessments at start of new biologic treatment (baseline) and at 6 and 12 months
      6 mo
      Analyses were limited to patients who continued to receive their initiated therapy at the 6-mo evaluation visit and for whom data were available at baseline.
      Patients discontinuing because of lack of effectiveness were included in the analysis but were considered treatment failures (ie, not achieving PGA score of 0/1 and no changes from baseline for %BSA and DLQI) at subsequent visits.
      12 mo
      Patients discontinuing because of lack of effectiveness were included in the analysis but were considered treatment failures (ie, not achieving PGA score of 0/1 and no changes from baseline for %BSA and DLQI) at subsequent visits.
      Analyses were limited to patients who continued to receive their initiated therapy at the 12-mo evaluation visit and for whom data were available at baseline.
      UstekinumabInfliximabAdalimumabEtanerceptUstekinumabInfliximabAdalimumabEtanercept
      PGA, N94411059523176369421165
       PGA 0/1 at baseline, n (%)102 (10.8)6 (5.5)98 (16.5)43 (18.6)74 (9.7)4 (5.8)77 (18.3)32 (19.4)
       PGA 0/1 on registry, n (%)539 (57.1)40 (36.4)298 (50.1)117 (50.6)452 (59.2)29 (42.0)238 (56.5)95 (57.6)
      PGA, N
      § Patients with a baseline PGA score less than 2 could not be included in the analysis of PGA reduction by 2 or more points.
      83510449218568365344132
       ≥2-Point reduction, n (%)436 (52.2)37 (35.6)199 (40.4)74 (40.0)378 (55.3)26 (40.0)146 (42.4)62 (47.0)
      %BSA, N91310457020775067406155
       Mean ± SD at baseline19.9 ± 19.9225.8 ± 22.9917.0 ± 20.4118.6 ± 20.2620.4 ± 19.9726.4 ± 24.9216.8 ± 19.7019.7 ± 20.81
       Mean decrease ± SD−14.7 ± 19.65−17.4 ± 21.50−10.6 ± 18.74−11.4 ± 18.29−16.3 ± 18.53−17.6 ± 21.23−12.3 ± 19.14−13.8 ± 18.77
      DLQI, N8239051018865953352146
       Mean ± SD at baseline10.3 ± 7.0012.1 ± 8.088.6 ± 6.7710.4 ± 7.3310.2 ± 6.9711.7 ± 7.688.2 ± 6.459.8 ± 7.44
      Improvement in DLQI, N8239051218865953352146
       Mean improvement ± SD6.9 ± 7.336.5 ± 8.104.5 ± 6.656.2 ± 7.217.5 ± 7.216.9 ± 9.114.9 ± 6.735.4 ± 7.72
      DLQI, N8239051018865953352146
       DLQI 0/1 at baseline, n (%)50 (6.1)7 (7.8)54 (10.6)13 (6.9)38 (5.8)1 (1.9)41 (11.6)11 (7.5)
       DLQI 0/1 on registry, n (%)392 (47.6)32 (35.6)222 (43.5)64 (34.0)361 (54.8)23 (43.4)176 (50.0)64 (43.8)
      Data are presented as n (%) unless otherwise noted.
      %BSA, Percentage of body surface area with psoriasis, DLQI, Dermatology Life Quality Index, PGA, Physician Global Assessment; SD, standard deviation.
      Analyses were limited to patients who continued to receive their initiated therapy at the 6-mo evaluation visit and for whom data were available at baseline.
      Patients discontinuing because of lack of effectiveness were included in the analysis but were considered treatment failures (ie, not achieving PGA score of 0/1 and no changes from baseline for %BSA and DLQI) at subsequent visits.
      Analyses were limited to patients who continued to receive their initiated therapy at the 12-mo evaluation visit and for whom data were available at baseline.
      § Patients with a baseline PGA score less than 2 could not be included in the analysis of PGA reduction by 2 or more points.

       Physician Global Assessment

      At 6 and 12 months, respectively, the proportions of patients achieving a PGA score of 0/1 were 57.1% and 59.2% for ustekinumab, 50.1% and 56.5% for adalimumab, and 50.6% and 57.6% for etanercept; rates for infliximab were 36.4% and 42.0% (Fig 1 and Table III). The PGA score was reduced from baseline by 2 or more points for 52.2% of ustekinumab, 35.6% of infliximab, 40.4% of adalimumab, and 40.0% of etanercept group patients at 6 months, and 55.3%, 40.0%, 42.4%, and 47.0% of patients in these groups, respectively, at 12 months.
      Figure thumbnail gr1
      Fig 1Psoriasis. Proportion of patients with Physician Global Assessment (PGA) score of 0 (clear) or 1 (minimal) at 6 and 12 months; patients initiating ustekinumab, infliximab, adalimumab, or etanercept on registry.
      Adjusted logistic regression analyses showed that patients using TNF-α inhibitors were less likely to achieve a PGA score of 0/1 at 6 months compared with ustekinumab (infliximab vs ustekinumab [OR 0.396, 95% CI 0.255-0.617, P < .0001], adalimumab vs ustekinumab [OR 0.686, 95% CI 0.547-0.861, P = .0012], and etanercept vs ustekinumab [OR 0.554, 95% CI 0.400-0.765, P = .0003]) (Table IV). Similar OR estimates were observed at 12 months, although only the infliximab vs ustekinumab comparison was statistically significant (Table IV). Factors other than treatment effect (ie, duration of disease, region, weight, and baseline PGA score) were significantly associated with achieving a PGA score of 0/1 (Supplemental Table II; available at http://www.jaad.org).
      Table IVAdjusted multivariate analyses of treatment effects: logistic regression for proportion of patients achieving a Physician Global Assessment score of clear (0) or minimal (1) at 6 and 12 months
      6-mo Analysis12-mo Analysis
      Odds ratio (95% CI)P valueOdds ratio (95% CI)P value
      Infliximab vs ustekinumab0.396 (0.255-0.617)<.00010.449 (0.260-0.774).0040
      Adalimumab vs ustekinumab0.686 (0.547-0.861).00120.841 (0.645-1.097).2025
      Etanercept vs ustekinumab0.554 (0.400-0.765).00030.686 (0.466-1.009).0557
      CI, Confidence interval.

       Body surface area

      At 6 and 12 months, respectively, the mean decrease in %BSA from baseline was −14.7 and −16.3 for ustekinumab, −17.4 and −17.6 for infliximab, −11.4 and −13.8 for etanercept, and −10.6 and −12.3 for adalimumab (Fig 2 and Table III). Analysis of covariance showed that from baseline to 6 months patients in the ustekinumab group demonstrated significantly better improvement (ie, decreases) in %BSA compared with adalimumab (point estimate 1.832%, 95% CI 0.670-2.995, P = .0020) and etanercept (point estimate 3.419%, 95% CI 1.728-5.110, P < .0001), but not infliximab (point estimate 1.807%, 95% CI −0.368 to 3.982, P = .1033) (Table V). At 12 months, the improvement in %BSA for infliximab and etanercept was significantly less (point estimate 3.945%, 95% CI 1.717-6.172, P = .0005 and point estimate 2.778%, 95% CI 1.180-4.376, P = .0007, respectively) compared with ustekinumab (but not for adalimumab [point estimate 1.038%, 95% CI −0.068 to 2.144, P = .0659] vs ustekinumab) (Table V). Other factors (ie, ethnicity, weight, alcohol use, baseline %BSA, and previous TNF-α inhibitor use) also significantly affected %BSA response (Supplemental Table III; available at http://www.jaad.org).
      Figure thumbnail gr2
      Fig 2Psoriasis. Mean decrease in percentage of body surface area (%BSA) with psoriasis at 6 and 12 months; patients initiating ustekinumab, infliximab, adalimumab, or etanercept on registry.
      Table VAdjusted multivariate analyses of treatment effects: analysis of covariance and least squares mean decrease from baseline for percentage of body surface area with psoriasis and mean improvement from baseline for Dermatology Life Quality Index scores at 6 and 12 mo
      6-mo Analysis12-mo Analysis
      LSM decrease from baselinePoint estimate
      Difference in LSM change from baseline compared with ustekinumab.


      (95% CI)
      P valueLSM decrease from baselinePoint estimate
      Difference in LSM change from baseline compared with ustekinumab.


      (95% CI)
      P value
      Mean decrease in %BSA
       Infliximab−12.446%1.807% (−0.368 to 3.982).1033−11.754%3.945% (1.717 to 6.172).0005
       Adalimumab−12.421%1.833% (0.670 to 2.995).0020−14.661%1.038% (−0.06 to 2.144).0659
       Etanercept−10.835%3.419% (1.728 to 5.110)<.0001−12.920%2.778% (1.180 to 4.376).0007
       Ustekinumab−14.253%−15.699%
      Mean improvement in DLQI
       Infliximab−5.3051.270 (0.075 to 2.465).0372−5.5631.365 (−0.168 to 2.899).0810
       Adalimumab−5.5221.053 (0.430 to 1.676).0009−6.1850.743 (0.025 to 1.462).0427
       Etanercept−5.5141.061 (0.156 to 1.966).0216−5.0111.917 (0.909 to 2.925).0002
       Ustekinumab−6.575−6.928
      %BSA, Percentage of body surface area with psoriasis; CI, confidence interval; DLQI, dermatology quality life index; LSM, least squares mean.
      Difference in LSM change from baseline compared with ustekinumab.

       Dermatology Life Quality Index

      At 6 and 12 months, respectively, the mean improvement in DLQI score from baseline was 6.9 and 7.5 for ustekinumab, 6.5 and 6.9 for infliximab, 6.2 and 5.4 for etanercept, and 4.5 and 4.9 for adalimumab (Table III). Among patients with a DLQI score of less than 10 at baseline, the proportion achieving a clinically meaningful reduction (≥5) at 6 months was 54.7% (111 of 203) for ustekinumab, 53.8% (14 of 26) for infliximab, 43.9% (65 of 148) for adalimumab, and 41.5% (22 of 53) for etanercept (Fig 3, A). At 12 months, responses in each group were generally similar to those observed at 6 months (ustekinumab, 58.7% [88 of 150]; adalimumab, 49.1% [55 of 112], and etanercept, 45.0% [18 of 40]), except for infliximab (37.5% [6 of 16]) (Fig 3, B). Among patients with higher DLQI scores (≥10) at baseline, the proportions achieving a reduction of at least 5 points at 6 months were 86.7% [351 of 405] for ustekinumab, 76.1% [35 of 46] for infliximab, 76.9% [143 of 186] for adalimumab, and 82.8% [72 of 87] for etanercept (Fig 3, A). Responses at 12 months were 91.8% (302 of 329) for ustekinumab, 74.1% (20 of 27) for infliximab, 88.9% (104 of 117) for adalimumab, and 72.1% (44 of 61) for etanercept (Fig 3, B). At 6 months, the proportion of patients achieving a DLQI score of 0/1 (no impact on HRQoL) was 47.6% in the ustekinumab group, 35.6% in the infliximab group, 43.5% in the adalimumab group, and 34.0% in the etanercept group; at 12 months, the proportions were slightly higher across all treatment groups (Fig 4 and Table III).
      Figure thumbnail gr3
      Fig 3Psoriasis. Proportion of patients achieving clinically meaningful improvement in Dermatology Life Quality Index (DLQI) score (≥5) by DLQI level (<10 and ≥10) at the time of initiating ustekinumab, infliximab, adalimumab, or etanercept on registry: 6 (A) and 12 (B) months.
      Figure thumbnail gr4
      Fig 4Psoriasis. Proportions of patients achieving Dermatology Life Quality Index (DLQI) score of 0 or 1 (no impact on health-related quality of life) at baseline and at 6 and 12 months; patients initiating ustekinumab, infliximab, adalimumab, or etanercept on registry.
      The multivariate analysis of DLQI improvement from baseline to 6 months demonstrated that improvement in the ustekinumab group was significantly better than that in the other treatment groups. The point estimates for improvements in DLQI scores for ustekinumab compared with adalimumab, etanercept, and infliximab were 1.053 (95% CI 0.430-1.676, P = .0009), 1.061 (95% CI 0.156-1.966, P = .0216), and 1.270 (95% CI 0.075-2.465, P = .0372), respectively (Table V). At 12 months, DLQI improvement for ustekinumab was better compared with both etanercept (1.917, 95% CI 0.909-2.925, P = .0002) and adalimumab (0.743, 95% CI 0.025-1.462, P = .0427), but was not significantly different compared with infliximab (1.365, 95% CI −0.168 to 2.899, P = .0810) (Table V). In addition, other factors (ie, age at diagnosis, disease duration, ethnicity, weight, region, and baseline PGA score, %BSA, DLQI) significantly affected DLQI improvement (Supplemental Table IV; available at http://www.jaad.org).

      Discussion

      Our analyses were based on data from more than 2000 patients with psoriasis who initiated a new biologic during their participation in PSOLAR. Effectiveness was compared between ustekinumab, which inhibits interleukin-12/23, and 3 TNF-α inhibitors (ie, infliximab, adalimumab, and etanercept). At 6 months, 5 of the 6 comparisons for PGA and %BSA showed significantly better effectiveness for ustekinumab versus the TNF-α inhibitors, taking into account several variables such as baseline psoriasis severity and prior treatment. At 12 months, 3 of the 6 PGA and %BSA comparisons demonstrated statistically significant better effectiveness for ustekinumab, whereas all comparisons were numerically better for ustekinumab. That fewer comparisons reached statistical significance at 12 months than at 6 months may reflect that treatment groups at 12 months were smaller (by as much as one-third) than those at 6 months, thereby reducing the statistical power to distinguish significant differences.
      Several factors were found to be associated with effectiveness outcomes for biologic therapy. As expected, patients of lower weight generally experienced more favorable results compared with heavier patients.
      • Bremmer S.
      • Van Voorhees A.S.
      • Hsu S.
      • et al.
      Obesity and psoriasis: from the Medical Board of the National Psoriasis Foundation.
      • Puig L.
      Obesity and psoriasis: body weight and body mass index influence the response to biological treatment.
      More severe disease at the start of therapy was associated with lower PGA response, but higher %BSA response. This may reflect differences in the biology of response to therapy for psoriasis. Small areas of moderate-to-severe plaque can develop, even in patients who have only minimal areas of overall involvement. The clinical relevance of the magnitude of adjusted differences in %BSA noted among the biologic groups, although significant for several comparisons, is not clear. Of note, patients with prior TNF-α inhibitor use had worse PGA and %BSA responses compared with bionaive patients, whereas prior ustekinumab treatment did not have an effect on response outcomes.
      Ustekinumab showed significantly better DLQI responses compared with each TNF-α inhibitor at 6 months, and compared with adalimumab and etanercept at 12 months. Improvements from baseline at both 6 and 12 months were more apparent among patients with higher DLQI scores (≥10) compared with those with lower DLQI scores (<10) across all treatment groups. This supports the idea that patients reporting a very large impact of psoriasis (DLQI score ≥10) on daily functioning may experience more meaningful HRQoL improvement after appropriate treatment.
      The effectiveness of biologics in PSOLAR was less than what has been reported in randomized controlled trials, which is consistent with other comparative studies of psoriasis therapies.
      • Gelfand J.M.
      • Wan J.
      • Callis Duffin K.
      • et al.
      Comparative effectiveness of commonly used systemic treatments or phototherapy for moderate to severe plaque psoriasis in the clinical practice setting.
      • Takeshita J.
      • Wang S.
      • Shin D.B.
      • et al.
      Comparative effectiveness of less commonly used systemic monotherapies and common combination therapies for moderate to severe psoriasis in the clinical setting.
      • Leonardi C.L.
      • Kimball A.B.
      • Papp K.A.
      • et al.
      Efficacy and safety of ustekinumab, a human interleukin-12/23 monoclonal antibody, in patients with psoriasis: 76-week results from a randomised, double-blind, placebo-controlled trial (PHOENIX 1).
      • Papp K.A.
      • Langley R.G.
      • Lebwohl M.
      • et al.
      Efficacy and safety of ustekinumab, a human interleukin-12/23 monoclonal antibody, in patients with psoriasis: 52-week results from a randomised, double-blind, placebo-controlled trial (PHOENIX 2).
      • Reich K.
      • Nestle F.O.
      • Papp K.
      • et al.
      Infliximab induction and maintenance therapy for moderate-to-severe psoriasis: a phase III, multicentre, double-blind trial.
      • Menter A.
      • Feldman S.R.
      • Weinstein G.D.
      • et al.
      A randomized comparison of continuous vs. intermittent infliximab maintenance regimens over 1 year in the treatment of moderate-to-severe plaque psoriasis.
      • Gordon K.B.
      • Langley R.G.
      • Leonardi C.
      • et al.
      Clinical response to adalimumab treatment in patients with moderate to severe psoriasis: double-blind, randomized controlled trial and open-label extension study.
      • Menter A.
      • Tyring S.K.
      • Gordon K.
      • et al.
      Adalimumab therapy for moderate to severe psoriasis: a randomized, controlled phase III trial.
      • Papp K.A.
      • Tyring S.
      • Lahfa M.
      • et al.
      A global phase III randomized controlled trial of etanercept in psoriasis: safety, efficacy, and effect of dose reduction.
      • Bagel J.
      • Lynde C.
      • Tyring S.
      • et al.
      Moderate to severe plaque psoriasis with scalp involvement: a randomized, double-blind, placebo-controlled study of etanercept.
      Interpretation of our results in the context of other postapproval studies is difficult because of differences in baseline study population characteristics, study design, adherence, and clinical end points (eg, global assessment scales) and reference treatment groups.
      Some limitations of this study design should be considered when evaluating these results, as PSOLAR data are derived from a day-to-day clinical setting with wide-ranging inclusion criteria. In particular, there may be treatment selection bias. Consistent with this limitation, baseline demography and disease characteristics varied among treatment groups (eg, patients receiving infliximab had more severe psoriasis). We attempted to adjust results for relevant and identifiable confounding factors, but residual confounding could exist for unmeasured variables. Data regarding dosing adjustments were not captured uniformly across treatment groups thereby eliminating the possibility of stratifying results by dose and dose interval. In addition, relatively few patients started infliximab in PSOLAR, thus limiting statistical power for assessing effectiveness of infliximab in these analyses. Also notable is the fact that more patients had missing data in the smallest treatment group than in the other groups.
      Overall, effectiveness of ustekinumab was significantly better than all 3 of the TNF-α inhibitors studied for the majority of comparisons at 6 and 12 months. Correlation with findings related to patient-reported outcomes showed benefit with ustekinumab treatment in nearly all comparisons at both time points. Of note, a recent PSOLAR analysis comparing drug survival across biologic therapies found that ustekinumab had better drug survival compared with all 3 TNF-α inhibitors.

      Menter A, Papp KA, Gooderham M, et al. Drug survival of biologic therapy in a large, disease-based registry of patients with psoriasis: results from the Psoriasis Longitudinal Assessment and Registry (PSOLAR). J Eur Acad Dermatol Venereol. doi: 10.1111/jdv.13611. Published online March 30, 2016.

      This may reflect the findings of generally better effectiveness and patient-reported outcomes with ustekinumab in the current study, although several other factors may also contribute to drug survival. In conclusion, these analyses of PSOLAR data expand our understanding of the comparative effectiveness of biologic agents beyond clinical trials and help better characterize their impact on clinical outcomes, including HRQoL, in patients with psoriasis treated in a real-world setting.
      The authors thank Cynthia Arnold, Samantha Simpson, and Christine Wolkin (Janssen Scientific Affairs, LLC, Spring House, PA) for their editorial assistance and writing support of this manuscript; Kezhen L. Tang (Janssen Research and Development, LLC, Horsham, PA) for her statistical analysis support; and Joel Gelfand, MD, MSCE (Hospital of the University of Pennsylvania, Philadelphia) and members of the PSOLAR Scientific Advisory Committee who provided critical review of the analytical plan for the manuscript.

      Appendix

      Supplemental Table IMost common doses and dose frequencies by treatment for patients initiating new therapy during the registry
      UstekinumabInfliximabAdalimumabEtanercept
      6 mo
       No. of administrations20122431187470
      Dose45 mg90 mg5 mg/kg40 mg50 mg
      1242 (61.7)768 (38.2)175 (72.0)1113 (93.8)468 (99.6)
      Dose frequencyEvery 12 wkEvery 8 wk
      Other common dose frequencies for infliximab were other (39 [16.0%]), every 6 wk (22 [9.1%]), and every 4 wk (24 [9.9%]).
      Every other wkWeekly
      In addition, a large proportion (44.5%, n = 209) of etanercept doses were administered twice weekly.
      996 (49.5)603 (30.0)97 (39.9)969 (81.6)241 (51.3)
      12 mo
       No. of administrations1680169909372
      Dose45 mg90 mg5 mg/kg40 mg50 mg
      1064 (63.3)614 (36.5)117 (69.2)859 (94.5)371 (99.7)
      Dose frequencyEvery 12 wkEvery 8 wk
      Other common dose frequencies for infliximab were other (28 [16.6%]), every 6 wk (16 [9.5%]), and every 4 wk (16 [9.5%]).
      Every other wkWeekly
      § In addition, a large proportion (43.8%, n = 163) of etanercept doses were administered twice weekly.
      849 (50.5)478 (28.5)65 (38.5)752 (82.7)202 (54.3)
      Data are reported as no. of doses administered (%).
      Other common dose frequencies for infliximab were other (39 [16.0%]), every 6 wk (22 [9.1%]), and every 4 wk (24 [9.9%]).
      In addition, a large proportion (44.5%, n = 209) of etanercept doses were administered twice weekly.
      Other common dose frequencies for infliximab were other (28 [16.6%]), every 6 wk (16 [9.5%]), and every 4 wk (16 [9.5%]).
      § In addition, a large proportion (43.8%, n = 163) of etanercept doses were administered twice weekly.
      Supplemental Table IIAdjusted multivariate analyses of demographic, disease, and prior treatment factors: logistic regression for proportion of patients achieving a Physician Global Assessment score of clear (0) or minimal (1) at 6 and 12 months
      6-mo Analysis12-mo Analysis
      Odds ratio (95% CI)P valueOdds ratio (95% CI)P value
      Age at psoriasis diagnosis/10 y1.019 (0.943-1.101).62821.055 (0.966-1.154).2344
      Duration of disease/5 y
      At 12 mo, patients with longer duration of psoriasis (with every 5 y of disease) were more likely to achieve PGA 0/1.
      1.012 (0.965-1.061).61831.067 (1.011-1.127).0194
      Region: European Union/Latin America vs North America
      At 6 and 12 mo, patients living in Europe and Latin America (vs North America) were more likely to achieve PGA 0/1.
      2.100 (1.548-2.850)<.00012.479 (1.726-3.561)<.0001
      Non-white vs white1.061 (0.809-1.391).66931.060 (0.775-1.451).7136
      Male vs female0.999 (0.815-1.226).99600.927 (0.731-1.176).5317
      Overweight/obesity class I (25 < BMI < 35) vs normal (BMI <25)
      At 6 and 12 mo, overweight and obese patients (vs normal weight) at baseline were less likely to achieve PGA 0/1.
      0.638 (0.487-0.837).00120.607 (0.445-0.830).0017
      Obesity class II-III (BMI ≥35) vs normal (BMI <25)
      At 6 and 12 mo, overweight and obese patients (vs normal weight) at baseline were less likely to achieve PGA 0/1.
      0.396 (0.290-0.541)<.00010.463 (0.323-0.664)<.0001
      PsA vs no PsA0.887 (0.721-1.091).25650.859 (0.675-1.094).2180
      Smoke (current/past vs never)0.893 (0.730-1.091).26821.069 (0.848-1.347).5730
      Alcohol (current/past vs never)1.159 (0.907-1.481).23800.878 (0.665-1.160).3602
      Baseline PGA 2/3 vs 0/1
      § At 6 and 12 mo, patients with more severe baseline disease (PGA 2/3 and 4/5 vs 0/1) were less likely to achieve PGA 0/1.
      0.439 (0.324-0.594)<.00010.480 (0.333-0.692)<.0001
      Baseline PGA 4/5 vs 0/1
      § At 6 and 12 mo, patients with more severe baseline disease (PGA 2/3 and 4/5 vs 0/1) were less likely to achieve PGA 0/1.
      0.397 (0.269-0.585)<.00010.401 (0.252-0.636).0001
      Baseline %BSA1.003 (0.998-1.009).27270.998 (0.992-1.004).4656
      DLQI ≥10 vs <100.860 (0.697-1.061).15800.821 (0.642-1.048).1138
      Discontinued because of AE vs continued0.589 (0.287-1.208).14840.873 (0.386-1.974).7440
      Discontinued insurance vs continued1.277 (0.616-2.645).51121.224 (0.483-3.104).6701
      Discontinued (other reason) vs continued1.085 (0.668-1.762).74110.769 (0.439-1.345).3565
      Prior TNF-α inhibitor vs bionaive
      // At 6 mo, patients with prior TNF-α inhibitor usage (vs bionaive) at baseline were less likely to achieve PGA 0/1.
      0.550 (0.357-0.846).00660.744 (0.453-1.221).2415
      Prior ustekinumab vs bionaive0.879 (0.508-1.518).64311.433 (0.652-3.150).3711
      Prior other biologics vs bionaive0.800 (0.597-1.073).13650.891 (0.629-1.260).5128
      AE, Adverse event; BMI, body mass index; %BSA, percentage of body surface area with psoriasis; CI, confidence interval; DLQI, Dermatology Life Quality Index; PGA, Physician Global Assessment; PsA, psoriatic arthritis; TNF, tumor necrosis factor.
      At 12 mo, patients with longer duration of psoriasis (with every 5 y of disease) were more likely to achieve PGA 0/1.
      At 6 and 12 mo, patients living in Europe and Latin America (vs North America) were more likely to achieve PGA 0/1.
      At 6 and 12 mo, overweight and obese patients (vs normal weight) at baseline were less likely to achieve PGA 0/1.
      § At 6 and 12 mo, patients with more severe baseline disease (PGA 2/3 and 4/5 vs 0/1) were less likely to achieve PGA 0/1.
      // At 6 mo, patients with prior TNF-α inhibitor usage (vs bionaive) at baseline were less likely to achieve PGA 0/1.
      Supplemental Table IIIAdjusted multivariate analyses of demographic, disease, and prior treatment factors: analysis of covariance for decrease from baseline in percentage of body surface area with psoriasis at 6 and 12 months
      6-mo Analysis12-mo Analysis
      Point estimate (95% CI)P valuePoint estimate (95% CI)P value
      Age at psoriasis diagnosis/10 y0.129 (−0.262 to 0.520).51720.098 (−0.265 to 0.460).5973
      Duration of disease/5 y0.008 (−0.235 to 0.251).9490−0.002 (−0.227 to 0.223).9876
      Region: European Union/Latin America vs North America−0.790 (−2.271 to 0.690).2952−0.944 (−2.313 to 0.426).1768
      Non-white vs white
      At 6 mo, non-white ethnicity (vs white) was significantly associated with worse %BSA response.
      1.760 (0.352 to 3.167).01431.160 (−0.163 to 2.483).0856
      Male vs female0.441 (−0.594 to 1.476).40340.103 (−0.873 to 1.078).8367
      Overweight/obesity class I (25 < BMI < 35) vs normal (BMI <25)0.963 (−0.378 to 2.305).15920.663 (−0.567 to 1.892).2907
      Obesity class II-III (BMI ≥35) vs normal (BMI <25)
      At 6 and 12 mo, heavier weight (obesity class II/III vs normal weight) was significantly associated with worse %BSA response.
      3.343 (1.778 to 4.908)<.00012.518 (1.053 to 3.982).0008
      PsA vs no PsA0.541 (−0.527 to 1.609).3205−0.125 (−1.133 to 0.883).8074
      Smoke (current/past vs never)−0.537 (−1.555 to 0.481).3010−0.420 (−1.371 to 0.531).3865
      Alcohol (current/past vs never)
      At 6 mo, current/past alcohol use (vs never used) was associated with better %BSA response.
      −1.590 (−2.820 to −0.360).0113−0.862 (−1.991 to 0.268).1348
      Baseline PGA 2/3 vs 0/10.968 (−0.639 to 2.574).2376−0.030 (−1.556 to 1.497).9697
      Baseline PGA 4/5 vs 0/10.875 (−1.118 to 2.867).3894−0.320 (−2.218 to 1.578).7406
      Baseline %BSA
      § At 6 and 12 mo, higher (worse) baseline %BSA at baseline was associated with better %BSA response.
      −0.792 (−0.819 to −0.765)<.0001−0.812 (−0.838 to −0.787)<.0001
      DLQI ≥10 vs DLQI<10−0.710 (−1.766 to 0.346).1874−0.946 (−1.944 to 0.052).0632
      Discontinued because of AE vs continued−2.718 (−6.365 to 0.930).1441−1.549 (−4.948 to 1.849).3713
      Discontinued insurance vs continued
      // At 12 mo, discontinuation from the registry because of insurance (vs continuing) was associated with better %BSA response.
      −3.212 (−6.948 to 0.524).0919−4.231 (−8.070 to −0.392).0308
      Discontinued (other reason) vs continued−0.663 (−3.177 to 1.851).6051−1.636 (−4.026 to 0.755).1797
      Prior TNF-α inhibitor vs bionaive
      At 6 and 12 mo, previous treatment with TNF-α inhibitors (vs bionaive) was significantly associated with worse %BSA response.
      2.954 (0.699 to 5.210).01033.506 (1.363 to 5.650).0014
      Prior ustekinumab vs bionaive0.208 (−2.608 to 3.023).8850−0.917 (−4.165 to 2.331).5796
      Prior other biologics vs bionaive0.797 (−0.711 to 2.305).30001.279 (−0.183 to 2.742).0864
      AE, Adverse event; BMI, body mass index; %BSA, percentage of body surface area with psoriasis; CI, confidence interval; DLQI, Dermatology Life Quality Index; PGA, Physician Global Assessment; PsA, psoriatic arthritis; TNF, tumor necrosis factor.
      At 6 mo, non-white ethnicity (vs white) was significantly associated with worse %BSA response.
      At 6 and 12 mo, heavier weight (obesity class II/III vs normal weight) was significantly associated with worse %BSA response.
      At 6 mo, current/past alcohol use (vs never used) was associated with better %BSA response.
      § At 6 and 12 mo, higher (worse) baseline %BSA at baseline was associated with better %BSA response.
      // At 12 mo, discontinuation from the registry because of insurance (vs continuing) was associated with better %BSA response.
      At 6 and 12 mo, previous treatment with TNF-α inhibitors (vs bionaive) was significantly associated with worse %BSA response.
      Supplemental Table IVAdjusted multivariate analyses of demographic, disease, and prior treatment factors: analysis of covariance for improvement from baseline in Dermatology Life Quality Index at 6 and 12 months
      6-mo Analysis12-mo analysis
      Point estimate (95% CI)P valuePoint estimate (95% CI)P value
      Age at psoriasis diagnosis/10 y
      At 6 and 12 mo, older age at psoriasis diagnosis (by every 10 y) and non-white ethnicity (vs white) were significantly associated with a worsened DLQI response.
      0.391 (0.181 to 0.602).00030.255 (0.024 to 0.486).0302
      Duration of disease/5 y
      At 6 mo, longer duration of disease (by every 5 y) was associated with a worsened DLQI response.
      0.131 (0.000 to 0.262).04960.086 (−0.060 to 0.232).2482
      Region: European Union/Latin America vs North America
      At 6 mo, European and Latin American location (vs North America) was significantly associated with a better DLQI response.
      −1.064 (−1.856 to −0.272).0085−0.885 (−1.780 to 0.009).0524
      Non-white vs white
      At 6 and 12 mo, older age at psoriasis diagnosis (by every 10 y) and non-white ethnicity (vs white) were significantly associated with a worsened DLQI response.
      0.785 (0.027 to 1.542).04230.819 (−0.024 to 1.662).0567
      Male vs female0.383 (−0.174 to 0.941).17770.364 (−0.265 to 0.993).2563
      Overweight/obesity class I (25 < BMI < 35) vs normal (BMI <25)0.613 (−0.105 to 1.331).09420.695 (−0.103 to 1.492).0878
      Obesity class II-III (BMI ≥35) vs normal (BMI <25)
      § At 12 mo, obesity class II-III (vs normal weight) was associated with a worsened DLQI response.
      0.568 (−0.269 to 1.405).18301.004 (0.065 to 1.943).0361
      PsA vs no PsA−0.297 (−0.869 to 0.275).30840.228 (−0.415 to 0.872).4860
      Smoke (current/past vs never)−0.475 (−1.024 to 0.074).0900−0.272 (−0.884 to 0.340).3838
      Alcohol (current/past vs never)−0.535 (−1.202 to 0.131).1155−0.467 (−1.188 to 0.254).2043
      Baseline PGA 2/3 vs 0/1
      // At 6 and 12 mo, more severe disease at baseline (as indicated by PGA 4/5 and 2/3 vs 0/1; larger [worse] %BSA; and larger [worse] DLQI score [≥10 vs <10]) were significantly associated with a better DLQI response.
      −1.278 (−2.163 to −0.394).0047−0.965 (−1.937 to 0.007).0518
      Baseline PGA 4/5 vs 0/1
      // At 6 and 12 mo, more severe disease at baseline (as indicated by PGA 4/5 and 2/3 vs 0/1; larger [worse] %BSA; and larger [worse] DLQI score [≥10 vs <10]) were significantly associated with a better DLQI response.
      −2.021 (−3.108 to −0.935).0003−2.701 (−3.911 to −1.491)<.0001
      Baseline %BSA
      // At 6 and 12 mo, more severe disease at baseline (as indicated by PGA 4/5 and 2/3 vs 0/1; larger [worse] %BSA; and larger [worse] DLQI score [≥10 vs <10]) were significantly associated with a better DLQI response.
      −0.041 (−0.056 to −0.027)<.0001−0.048 (−0.064 to −0.032)<.0001
      DLQI ≥10 vs DLQI <10
      // At 6 and 12 mo, more severe disease at baseline (as indicated by PGA 4/5 and 2/3 vs 0/1; larger [worse] %BSA; and larger [worse] DLQI score [≥10 vs <10]) were significantly associated with a better DLQI response.
      −8.049 (−8.613 to −7.486)<.0001−8.493 (−9.132 to −7.853)<.0001
      Discontinued because of AE vs continued0.410 (−1.546 to 2.365).68120.137 (−2.142 to 2.416).9061
      Discontinued insurance vs continued0.140 (−1.831 to 2.111).8890−0.197 (−2.708 to 2.314).8778
      Discontinued (other reason) vs continued0.820 (−0.535 to 2.175).23561.201 (−0.338 to 2.741).1260
      Prior TNF-α inhibitor vs bionaive0.842 (−0.374 to 2.058).17460.146 (−1.231 to 1.522).8357
      Prior ustekinumab vs bionaive−0.163 (−1.752 to 1.427).84110.104 (−2.069 to 2.277).9251
      Prior other biologics vs bionaive0.284 (−0.535 to 1.102).49660.029 (−0.930 to 0.988).9530
      AE, Adverse event; BMI, body mass index; %BSA, percentage of body surface area with psoriasis; CI, confidence interval; DLQI, Dermatology Life Quality Index; PGA, Physician Global Assessment; PsA, psoriatic arthritis; TNF, tumor necrosis factor.
      At 6 and 12 mo, older age at psoriasis diagnosis (by every 10 y) and non-white ethnicity (vs white) were significantly associated with a worsened DLQI response.
      At 6 mo, longer duration of disease (by every 5 y) was associated with a worsened DLQI response.
      At 6 mo, European and Latin American location (vs North America) was significantly associated with a better DLQI response.
      § At 12 mo, obesity class II-III (vs normal weight) was associated with a worsened DLQI response.
      // At 6 and 12 mo, more severe disease at baseline (as indicated by PGA 4/5 and 2/3 vs 0/1; larger [worse] %BSA; and larger [worse] DLQI score [≥10 vs <10]) were significantly associated with a better DLQI response.

      References

        • Puig L.
        On clinical thresholds, clinical equivalents and indirect comparisons of biological treatments for moderate-to-severe psoriasis.
        J Clin Pharm Ther. 2015; 40: 131-134
        • Conway P.H.
        • Clancy C.
        Comparative-effectiveness research–implications of the Federal Coordinating Council's report.
        N Engl J Med. 2009; 361: 328-330
        • Iglehart J.K.
        Prioritizing comparative-effectiveness research–IOM recommendations.
        N Engl J Med. 2009; 361: 325-328
        • Abuabara K.
        • Wan J.
        • Troxel A.B.
        • et al.
        Variation in dermatologist beliefs about the safety and effectiveness of treatments for moderate to severe psoriasis.
        J Am Acad Dermatol. 2013; 68: 262-269
        • Krueger G.G.
        • Langley R.G.
        • Finlay A.Y.
        • et al.
        Patient-reported outcomes of psoriasis improvement with etanercept therapy: results of a randomized phase III trial.
        Br J Dermatol. 2005; 153: 1192-1199
        • Revicki D.A.
        • Willian M.K.
        • Menter A.
        • et al.
        Impact of adalimumab treatment on patient-reported outcomes: results from a phase III clinical trial in patients with moderate to severe plaque psoriasis.
        J Dermatolog Treat. 2007; 18: 341-350
        • Revicki D.
        • Willian M.K.
        • Saurat J.H.
        • et al.
        Impact of adalimumab treatment on health-related quality of life and other patient-reported outcomes: results from a 16-week randomized controlled trial in patients with moderate to severe plaque psoriasis.
        Br J Dermatol. 2008; 158: 549-557
        • Reich K.
        • Nestle F.O.
        • Papp K.
        • et al.
        Improvement in quality of life with infliximab induction and maintenance therapy in patients with moderate-to-severe psoriasis: a randomized controlled trial.
        Br J Dermatol. 2006; 154: 1161-1168
        • Lebwohl M.
        • Papp K.
        • Han C.
        • et al.
        Ustekinumab improves health-related quality of life in patients with moderate-to-severe psoriasis: results from the PHOENIX 1 trial.
        Br J Dermatol. 2010; 162: 137-146
        • Langley R.G.
        • Feldman S.R.
        • Han C.
        • et al.
        Ustekinumab significantly improves symptoms of anxiety, depression, and skin-related quality of life in patients with moderate-to-severe psoriasis: results from a randomized, double-blind, placebo-controlled phase III trial.
        J Am Acad Dermatol. 2010; 63: 457-465
        • Griffiths C.E.
        • Strober B.E.
        • van de Kerkhof P.
        • et al.
        Comparison of ustekinumab and etanercept for moderate-to-severe psoriasis.
        N Engl J Med. 2010; 362: 118-128
        • Langley R.G.
        • Elewski B.E.
        • Lebwohl M.
        • et al.
        Secukinumab in plaque psoriasis–results of two phase 3 trials.
        N Engl J Med. 2014; 371: 326-338
        • Reich K.
        • Burden A.D.
        • Eaton J.N.
        • et al.
        Efficacy of biologics in the treatment of moderate to severe psoriasis: a network meta-analysis of randomized controlled trials.
        Br J Dermatol. 2012; 166: 179-188
        • Schmitt J.
        • Rosumeck S.
        • Thomaschewski G.
        • et al.
        Efficacy and safety of systemic treatments for moderate-to-severe psoriasis: meta-analysis of randomized controlled trials.
        Br J Dermatol. 2014; 170: 274-303
        • Galvan-Banqueri M.
        • Marin Gil R.
        • Santos Ramos B.
        • et al.
        Biological treatments for moderate-to-severe psoriasis: indirect comparison.
        J Clin Pharm Ther. 2013; 38: 121-130
        • Lin V.W.
        • Ringold S.
        • Devine E.B.
        Comparison of ustekinumab with other biological agents for the treatment of moderate to severe plaque psoriasis: a Bayesian network meta-analysis.
        Arch Dermatol. 2012; 148: 1403-1410
        • Lucka T.C.
        • Pathirana D.
        • Sammain A.
        • et al.
        Efficacy of systemic therapies for moderate-to-severe psoriasis: a systematic review and meta-analysis of long-term treatment.
        J Eur Acad Dermatol Venereol. 2012; 26: 1331-1344
        • Griffiths C.E.
        • Reich K.
        • Lebwohl M.
        • et al.
        Comparison of ixekizumab with etanercept or placebo in moderate-to-severe psoriasis (UNCOVER-2 and UNCOVER-3): results from two phase 3 randomised trials.
        Lancet. 2015; 386: 541-551
        • Sullivan P.
        • Goldmann D.
        The promise of comparative effectiveness research.
        JAMA. 2011; 305: 400-401
        • Gelfand J.M.
        • Wan J.
        • Callis Duffin K.
        • et al.
        Comparative effectiveness of commonly used systemic treatments or phototherapy for moderate to severe plaque psoriasis in the clinical practice setting.
        Arch Dermatol. 2012; 148: 487-494
        • Naldi L.
        • Svensson A.
        • Zenoni D.
        • et al.
        Comparators, study duration, outcome measures and sponsorship in therapeutic trials of psoriasis: update of the EDEN Psoriasis Survey 2001-2006.
        Br J Dermatol. 2010; 162: 384-389
        • Takeshita J.
        • Wang S.
        • Shin D.B.
        • et al.
        Comparative effectiveness of less commonly used systemic monotherapies and common combination therapies for moderate to severe psoriasis in the clinical setting.
        J Am Acad Dermatol. 2014; 71: 1167-1175
        • Papp K.A.
        • Strober B.
        • Augustin M.
        • et al.
        PSOLAR: design, utility, and preliminary results of a prospective, international, disease-based registry of patients with psoriasis who are receiving, or are candidates for, conventional systemic treatments or biologic agents.
        J Drugs Dermatol. 2012; 11: 1210-1217
        • Kimball A.B.
        • Leonardi C.
        • Stahle M.
        • et al.
        Demography, baseline disease characteristics and treatment history of patients with psoriasis enrolled in a multicentre, prospective, disease-based registry (PSOLAR).
        Br J Dermatol. 2014; 171: 137-147
        • Finlay A.Y.
        • Khan G.K.
        Dermatology Life Quality Index (DLQI)–a simple practical measure for routine clinical use.
        Clin Exp Dermatol. 1994; 19: 210-216
        • Samsa G.
        • Edelman D.
        • Rothman M.L.
        • et al.
        Determining clinically important differences in health status measures: a general approach with illustration to the Health Utilities Index Mark II.
        Pharmacoeconomics. 1999; 15: 141-155
        • Bremmer S.
        • Van Voorhees A.S.
        • Hsu S.
        • et al.
        Obesity and psoriasis: from the Medical Board of the National Psoriasis Foundation.
        J Am Acad Dermatol. 2010; 63: 1058-1069
        • Puig L.
        Obesity and psoriasis: body weight and body mass index influence the response to biological treatment.
        J Eur Acad Dermatol Venereol. 2011; 25: 1007-1011
        • Leonardi C.L.
        • Kimball A.B.
        • Papp K.A.
        • et al.
        Efficacy and safety of ustekinumab, a human interleukin-12/23 monoclonal antibody, in patients with psoriasis: 76-week results from a randomised, double-blind, placebo-controlled trial (PHOENIX 1).
        Lancet. 2008; 371: 1665-1674
        • Papp K.A.
        • Langley R.G.
        • Lebwohl M.
        • et al.
        Efficacy and safety of ustekinumab, a human interleukin-12/23 monoclonal antibody, in patients with psoriasis: 52-week results from a randomised, double-blind, placebo-controlled trial (PHOENIX 2).
        Lancet. 2008; 371: 1675-1684
        • Reich K.
        • Nestle F.O.
        • Papp K.
        • et al.
        Infliximab induction and maintenance therapy for moderate-to-severe psoriasis: a phase III, multicentre, double-blind trial.
        Lancet. 2005; 366: 1367-1374
        • Menter A.
        • Feldman S.R.
        • Weinstein G.D.
        • et al.
        A randomized comparison of continuous vs. intermittent infliximab maintenance regimens over 1 year in the treatment of moderate-to-severe plaque psoriasis.
        J Am Acad Dermatol. 2007; 56: e1-e15
        • Gordon K.B.
        • Langley R.G.
        • Leonardi C.
        • et al.
        Clinical response to adalimumab treatment in patients with moderate to severe psoriasis: double-blind, randomized controlled trial and open-label extension study.
        J Am Acad Dermatol. 2006; 55: 598-606
        • Menter A.
        • Tyring S.K.
        • Gordon K.
        • et al.
        Adalimumab therapy for moderate to severe psoriasis: a randomized, controlled phase III trial.
        J Am Acad Dermatol. 2008; 58: 106-115
        • Papp K.A.
        • Tyring S.
        • Lahfa M.
        • et al.
        A global phase III randomized controlled trial of etanercept in psoriasis: safety, efficacy, and effect of dose reduction.
        Br J Dermatol. 2005; 152: 1304-1312
        • Bagel J.
        • Lynde C.
        • Tyring S.
        • et al.
        Moderate to severe plaque psoriasis with scalp involvement: a randomized, double-blind, placebo-controlled study of etanercept.
        J Am Acad Dermatol. 2012; 67: 86-92
      1. Menter A, Papp KA, Gooderham M, et al. Drug survival of biologic therapy in a large, disease-based registry of patients with psoriasis: results from the Psoriasis Longitudinal Assessment and Registry (PSOLAR). J Eur Acad Dermatol Venereol. doi: 10.1111/jdv.13611. Published online March 30, 2016.