Abstract
Introduction
Real-world evidence investigating treatment patterns and associated economic outcomes in patients with psoriasis (PsO) are lacking. This study evaluated treatment patterns, healthcare resource utilisation (HCRU), and costs in biologic-naïve patients with moderate-to-severe PsO initiating biologics in France.
Methods
Biologic-naïve adults with moderate-to-severe PsO initiating an anti-tumour necrosis factor (TNFα) or anti-interleukin (IL) biologic were identified from the Système National des Données de Santé. Switching and dose escalation rates, calculated using Kaplan–Meier analyses, were assessed over 42 months of follow-up. HCRU and costs were assessed during the 12 months following initiation. Comparisons between switchers/non-switchers and dose escalators/non-dose escalators were conducted using Wilcoxon rank sum, chi-square, and Fisher exact tests.
Results
Overall, 20,995 biologic-naïve adult patients initiating a biologic were included in the analysis. At 42 months, switch rates and dose escalation rates were lowest for patients initiating anti-IL-23 agents and highest for anti-TNFα agents. Specifically, the lowest rate of switching (15.0%) and dose escalation (9.0%) were observed for patients treated with the anti-IL-23 agent risankizumab. The mean number of total healthcare visits at 12 months was significantly (p < 0.001) higher for treatment switchers versus non-switchers (48.8 vs 39.0) and dose escalators versus non-dose escalators (47.4 vs 39.9). Furthermore, switchers had significantly (p < 0.001) higher mean medical (€3435 vs €2578), inpatient (€1126 vs €826), outpatient (€1751 vs €1331), and pharmacy (€13,105 vs €10,589) costs versus non-switchers. Similar trends were observed for dose escalators versus non-dose escalators (€3453 vs €2638; €965 vs €870; €1682 vs €1371; and €13,736 vs €10,683, respectively, p < 0.001).
Conclusion
Anti-IL-23 agents demonstrated the lowest rates of switching and dose escalation over 42 months, whilst anti-TNFα agents had the highest rates. More specifically, the lowest rates were observed for patients treated with risankizumab. Overall, higher rates of switching or dose escalation corresponded with higher HCRU and costs.
Keywords: Biologics, Cost, Dose escalation, France, Healthcare resource utilisation, Psoriasis, Real-world evidence, Switching, Treatment patterns
Key Summary Points
| Why carry out this study? |
| Psoriasis (PsO) is one of the most prevalent chronic, immune-mediated skin conditions in France; however, there is currently limited comprehensive, real-world evidence investigating treatment patterns and associated economic outcomes amongst patients with PsO at a national level in France. |
| This study analysed treatment patterns, healthcare resource utilisation (HCRU), and costs associated with biologic use amongst biologic-naïve patients with moderate-to-severe PsO in France using data from the Système National des Données de Santé. |
| What was learned from the study? |
| Over 42 months, switching was common amongst patients with PsO treated with biologics; rates of switching and dose escalation varied across biologics but were lowest in patients treated with anti-interleukin (IL)-23 agents, specifically risankizumab. |
| Biologic switching and dose escalation were associated with increased all-cause HCRU and costs. |
| Anti-IL-23 agents, such as risankizumab, may offer low switching rates and predictable dosing regimens in patients with PsO in France. |
Introduction
Psoriasis (PsO) is a chronic immune-mediated skin condition with varying clinical features. PsO is relatively common and one of the most prevalent immune-mediated skin conditions in France, with an estimated prevalence of approximately 4.4% [1, 2]. The clinical features of PsO vary depending on the type of PsO. Plaque PsO is the most common type of PsO, accounting for approximately 80% of cases, while less frequently observed variants include guttate PsO, erythrodermic PsO, and pustular PsO. Common clinical manifestations present across all types of PsO include erythema, thickening, and scales [3].
Treatment approaches for all types of PsO vary depending on disease severity. According to consensus amongst French dermatologists, moderate-to-severe PsO is defined by involvement of special areas of the skin (e.g. nails, genitals, armpit, face, and scalp), involvement of a greater body surface area [4], presence of uncontrolled symptoms, and negative impact on patient quality of life [5]. Moderate-to-severe PsO also often fails to respond to first-line topical treatment [4, 5].
According to European treatment guidelines, phototherapy and conventional (e.g. ciclosporin, methotrexate) treatments, as well as biologic systemic treatments, are recommended as first-line treatment for moderate-to-severe PsO [6]. Reimbursement of PsO-specific biologic agents in France is currently restricted to patients who have failed to respond to ≥ 1 conventional systemic therapies or when these therapies are contraindicated. However, in 2025, the Psoriasis Research Group of the French Society of Dermatology published updated guidelines recommending that systemic biologic agents be reimbursed for first-line treatment of PsO in France, as these therapies have higher efficacy and clinical added value compared with non-targeted systemic treatments [7]. Different classes of biologics are available in France including anti-tumour necrosis factor alpha (TNFα) agents (adalimumab, certolizumab pegol, etanercept, infliximab), anti-interleukin (IL)-17 or anti-IL-17 receptor agents (bimekizumab, secukinumab, ixekizumab, brodalumab), anti-IL-12/-23 agents (ustekinumab), and anti-IL-23 agents (guselkumab, risankizumab, tildrakizumab) [6, 8]. Network meta-analysis has shown at a class level, anti-IL-17/anti-IL-17 receptor agents and anti-IL-23 agents were significantly more effective in improving disease severity compared with anti-TNFα agents [6].
The efficacy of biologic treatments for moderate-to-severe PsO has been demonstrated in clinical trials [9, 10], but there is limited real-world data on treatment patterns, as well as the associated healthcare resource utilisation (HCRU) and costs following biologic use amongst patients with PsO in France. Previously, healthcare costs and treatment patterns amongst patients with PsO treated with biologics have been explored using French insurance claims and centre-based cohorts [8, 11–14]. However, these studies focused on a few research centres, were limited to older treatment areas, had fewer biologic comparators, and lacked data on dose escalation and economic outcomes. Therefore, further studies evaluating treatment patterns in relation to economic outcomes across broader patient populations are needed. This analysis seeks to extend on prior claim-based PsO treatment pattern studies by providing an up-to-date, comprehensive, national assessment of real-world treatment patterns, including switching and dose escalation, as well as the associated HCRU, and costs after initiating biologic therapies (anti-TNFα and anti-IL, including newer agents such as risankizumab, tildrakizumab, and bimekizumab) in biologic-naïve adults with moderate-to-severe PsO in France using the Système National des Données de Santé (SNDS).
Methods
Study Design and Patient Population
This was a retrospective, noninterventional, cohort study of biologic-naïve (defined as having no use of any biologics in the study within the 12 months prior to the index date) patients aged ≥ 18 years with moderate-to-severe PsO initiating biologics (anti-TNFα or anti-IL) in France using secondary data from the SNDS from January 1, 2017 to December 31, 2023 (Fig. 1).The SNDS is the French national health database integrating health insurance data from the Système National d’Information Inter-Régimes de l’Assurance Maladie, and hospital and health facilities data from the Programme de Médicalisation des Systèmes d’Information. The SNDS covers approximately 66 million individuals, equating to 99% of the French population [15].
Fig. 1.
Study design. aIncluded ≥ 1 hospitalisation with ICD-10 diagnosis codes (principal diagnostic, or related diagnostic of medical unit summary, or significant associated diagnostics of stay) for PsO, or ≥ 1 long-term affliction with ICD-10 for PsO. bPatients with multiple dates of birth were removed, as such patients cannot be differentiated in the database. cNewly used is defined on the basis of the last biologic used before January 29, 2020. For example, if the last biologic agent used before January 29, 2020 was adalimumab and the patient continues to use adalimumab for the remainder of 2020 and then switches to risankizumab in 2021, then the first risankizumab used in 2021 will be the index date [13, 16]. dBiologic-naïve, defined as having no use of any of the biologics under study within 12 months prior to the index date. eThe healthcare consumption years for a patient spanned from the first calendar year to the last calendar year with an observed healthcare activity in any dataset. ATC Anatomical Therapeutic Chemical, HCRU healthcare resource utilisation, ICD-10 International Classification of Diseases, version 10, IL interleukin, PsO psoriasis, TNF tumour necrosis factor
PsO diagnosis was determined by either ≥ 1 hospitalisation with International Classification of Diseases, version 10 (ICD-10) diagnosis codes for PsO, ≥ 1 record of long-term affliction with ICD-10 codes for PsO, or ≥ 2 distinct dispensations of topical vitamin D derivatives (Anatomical Therapeutic Chemical D05AX) within a 2-year period [13, 16].
The index date was defined as the date of first dispensation of a newly initiated anti-TNFα or anti-IL biologic (index agent) from January 29, 2020 to December 31, 2022, after first use of phototherapy, topical treatment, or any systemic therapy (disease-modifying anti-rheumatic drugs including apremilast). “Newly initiated” refers to agents started during this time frame, after the last use of any biologic prior to January 29, 2020. The cutoff for “newly initiated” was selected because some therapies (e.g. risankizumab, tildrakizumab, and bimekizumab) were not yet reimbursed for the broader population before January 2020. This approach ensures appropriate and comparable selection of eligible patients across all treatment options.
Additional inclusion criteria were ≥ 12 months of healthcare consumption prior to the index date (historical period, starting no earlier than January 1, 2017) and ≥ 12 months of healthcare consumption after the index date until the last healthcare consumption year, death, or December 31, 2023 (observation period).
Outcomes
The following outcomes were assessed over the observation period after initiation of index drug up to the 42-month follow-up time: switch rate, defined as the proportion of patients who switched to a new biologic (switchers; patients with discontinuation or nonadherence were not considered); time to switch amongst switchers (i.e. drug persistence), defined as treatment duration of the index agent spanning from index date until earliest switch date, end of continuous clinical activity, or end of data; dose escalation rate, defined as the proportion of patients (dose escalators) who had ≥ 2 consecutive claims in which the estimated average daily dose (mg/day based on the amount in mg dispensed and time to next prescription claim) was 30% higher than the expected daily dose (mg/day per label); time to dose escalation, defined as treatment duration or persistence to the index agent spanning from the initiation of the maintenance dose until the earliest date of dose escalation, end of continuous clinical activity, treatment switch, or end of data availability (December 31, 2023).
HCRU and costs were reported during a 12-month period following index date amongst switchers versus non-switchers and dose escalators versus non-dose escalators. The study period was defined as the 12-month period beginning on the index date. These analyses were restricted to the first 12 months of follow-up to ensure all patients were compared over the same follow-up duration. The proportion of HCRU visits and mean number of visits, including inpatient admissions (hospital visits where admission date is strictly before the discharge date), 1-day hospital visits (hospital visits where admission and discharge occurred on the same day), outpatient visits, emergency department (ED) visits, pharmacy claims and laboratory and imaging visits, and the total length of stay of > 1 day inpatient admissions were measured. The number of inpatient stays included both public hospitalisations using the Programme de Médicalisation des Systèmes d’Information (PMSI) and private hospitalisations using the données de consommation inter-régimes (DCIR). Mean 12-month healthcare costs, including medical, inpatient, 1-day hospitalisation, outpatient, ED, and pharmacy costs, were also measured. Pharmacy costs included costs for expensive drugs, as outlined in the liste en sus (https://sante.gouv.fr/soins-et-maladies/medicaments/professionnels-de-sante/autorisation-de-mise-sur-le-marche/la-liste-en-sus/). The costs for inexpensive drugs are included in the cost of inpatient stays and cost of the diagnosis-related group. Medication costs in the outpatient setting include costs for expensive and inexpensive drugs. The number of inpatient days and inpatient costs were estimated using the PMSI data only, because the end date of hospitalisation is not available in DCIR. HCRU and costs for PMSI-Recueil d’Informations Psychiatriques (hospital psychiatry) were not calculated.
Statistical Analysis
Demographic, clinical, and treatment characteristics were assessed using descriptive statistics. Patient demographic characteristics were reported as of the closest date to the index date and clinical characteristics were reported within 12 months prior to the index date.
Kaplan–Meier analyses were used to estimate time to switch and dose escalation. Time to switch was estimated as time interval from initiation of index treatment to initiation of next treatment or censoring, whichever occurred first. Patients without an observed event were censored at the earliest date of death, end of last healthcare consumption year, or end of data period, whichever occurred first. Time to dose escalation was estimated as time interval from initiation of maintenance dose of index treatment to dose escalation or censoring, whichever occurred first. Patients without an observed event were censored at end of follow-up or at treatment switch.
HCRU and costs were assessed using descriptive statistics amongst switchers and non-switchers and amongst dose escalators and non-dose escalators. Costs were assessed from the perspective of statutory health insurance, adjusted for inflation using the Institut National de la Statistique et des Etudes Economiques general inflation index and are presented in 2023 euro (€) (https://www.insee.fr/en/statistiques/serie/001769682).
Comparisons between switchers versus non-switchers and dose escalators versus non-dose escalators were conducted using Wilcoxon rank sum tests for continuous variables, chi-square tests for categorical variables when the expected frequency was ≥ 5, and Fisher exact tests for categorical variables when the expected frequency was < 5.
Ethics
Ethical approval issued by the Comité Éthique et Scientifique pour les Recherches, les Études et les Évaluations dans le domaine de la Santé (CESREES) on June 15, 2023 (TPS No. 12242458).
Results
Patient Demographics and Baseline Characteristics
Overall, 20,995 biologic-naïve patients were included in the analysis and were treated with the anti-TNFα agents adalimumab (n = 5311), etanercept (n = 993), infliximab (n = 885), and certolizumab (n = 654); the anti-IL-17/IL-17 receptor agents secukinumab (n = 2498), ixekizumab (n = 1768), brodalumab (n = 466), and bimekizumab (n = 70); the anti-IL-23 agents guselkumab (n = 2707), risankizumab (n = 2456), and tildrakizumab (n = 626); and the anti-IL-12/23 agent ustekinumab (n = 2561; Fig. 2, Table 1).
Fig. 2.
Sample selection. aPatients with multiple dates of birth were removed, as such patients cannot be differentiated in the database [13, 16]. bNewly used is defined based on the last biologic used before January 29, 2020. For example, if the last biologic agent used before January 29, 2020, was adalimumab and the patient continues to use adalimumab for the remainder of 2020 and then switches to risankizumab in 2021, then the first risankizumab used in 2021 will be the index date. cThe healthcare consumption years for a patient spanned from the first calendar year to the last calendar year with an observed healthcare activity in any dataset. dBiologic-naïve, defined as having no use of any of the biologics under study within 12 months prior to the index date. ATC Anatomical Therapeutic Chemical, DMARD disease-modifying antirheumatic drug, ICD-10 International Classification of Diseases, version 10, IL interleukin, TNF tumour necrosis factor
Table 1.
Patient demographics of biologic-naïve population at index
| Patient demographics at indexa | Biologic-naïveb | ||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| All biologic-naïve patientsb, N = 20,995 | Anti-TNFα | Anti-IL-17/IL-17 receptor | Anti-IL-23 | Anti-IL-12/23 | |||||||||
| Adalimumab, n = 5311 | Etanercept, n = 993 | Infliximab, n = 885 | Certolizumab, n = 654 | Secukinumab, n = 2498 | Ixekizumab, n = 1768 | Brodalumab, n = 466 | Bimekizumab, n = 70 | Guselkumab, n = 2707 | Risankizumab, n = 2456 | Tildrakizumab, n = 626 | Ustekinumab, n = 2561 | ||
| Age, years, mean ± SD | 50.0 ± 15.0 | 50.0 ± 14.2 | 56.1 ± 14.6 | 49.7 ± 15.2 | 40.5 ± 14.1 | 50.7 ± 14.2 | 50.5 ± 14.4 | 50.8 ± 15.6 | 50.0 ± 15.2 | 49.4 ± 15.0 | 49.2 ± 15.2 | 49.7 ± 16.2 | 50.8 ± 15.8 |
| Male, n (%) | 10,750 (51.2) | 2496 (47.0) | 398 (40.1) | 414 (46.8) | 84 (12.8) | 1252 (50.1) | 923 (52.2) | 282 (60.5) | 46 (65.7) | 1549 (57.2) | 1496 (60.9) | 371 (59.3) | 1439 (56.2) |
| Insurance type, n (%) | |||||||||||||
| Universal health insurance | 19,269 (91.8) | 4903 (92.3) | 921 (92.7) | 757 (85.5) | 593 (90.7) | 2311 (92.5) | 1619 (91.6) | 419 (89.9) | 63 (90.0) | 2489 (91.9) | 2280 (92.8) | 569 (90.9) | 2345 (91.6) |
| CMU-c, ACS or C2S | 1673 (8.0) | 381 (7.2) | 69 (6.9) | 120 (13.6) | 60 (9.2) | 185 (7.4) | 146 (8.3) | 47 (10.1) | 7 (10.0) | 218 (8.1) | 175 (7.1) | 57 (9.1) | 208 (8.1) |
| AME | 6 (0.0) | 2 (0.0) | 0 | 3 (0.3) | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 1 (0.0) |
| Unknown | 47 (0.2) | 25 (0.5) | 3 (0.3) | 5 (0.6) | 1 (0.2) | 2 (0.1) | 3 (0.2) | 0 | 0 | 0 | 1 (0.0) | 0 | 7 (0.3) |
| Region, n (%) | |||||||||||||
| Auvergne-Rhône-Alpes | 1994 (9.5) | 467 (8.8) | 100 (10.1) | 54 (6.1) | 64 (9.8) | 211 (8.4) | 209 (11.8) | 64 (13.7) | 7 (10.0) | 298 (11.0) | 337 (13.7) | 49 (7.8) | 134 (5.2) |
| Bourgogne-Franche-Comté | 901 (4.3) | 236 (4.4) | 33 (3.3) | 32 (3.6) | 16 (2.4) | 148 (5.9) | 79 (4.5) | 12 (2.6) | 4 (5.7) | 128 (4.7) | 126 (5.1) | 29 (4.6) | 58 (2.3) |
| Bretagne | 1179 (5.6) | 257 (4.8) | 39 (3.9) | 22 (2.5) | 27 (4.1) | 95 (3.8) | 114 (6.4) | 25 (5.4) | 3 (4.3) | 255 (9.4) | 167 (6.8) | 33 (5.3) | 142 (5.5) |
| Centre-Val de Loire | 613 (2.9) | 152 (2.9) | 36 (3.6) | 26 (2.9) | 16 (2.4) | 91 (3.6) | 64 (3.6) | 6 (1.3) | 0 | 75 (2.8) | 89 (3.6) | 9 (1.4) | 49 (1.9) |
| Corse | 128 (0.6) | 32 (0.6) | 19 (1.9) | 4 (0.5) | 3 (0.5) | 37 (1.5) | 12 (0.7) | 1 (0.2) | 0 | 6 (0.2) | 3 (0.1) | 3 (0.5) | 8 (0.3) |
| Grand Est | 1985 (9.5) | 395 (7.4) | 118 (11.9) | 75 (8.5) | 79 (12.1) | 362 (14.5) | 136 (7.7) | 49 (10.5) | 2 (2.9) | 247 (9.1) | 245 (10.0) | 77 (12.3) | 200 (7.8) |
| Hauts-de-France | 2376 (11.3) | 854 (16.1) | 98 (9.9) | 79 (8.9) | 85 (13.0) | 239 (9.6) | 162 (9.2) | 40 (8.6) | 9 (12.9) | 232 (8.6) | 177 (7.2) | 60 (9.6) | 341 (13.3) |
| Île-de-France | 3324 (15.8) | 749 (14.1) | 143 (14.4) | 238 (26.9) | 110 (16.8) | 453 (18.1) | 253 (14.3) | 66 (14.2) | 7 (10.0) | 323 (11.9) | 363 (14.8) | 60 (9.6) | 559 (21.8) |
| Normandie | 1043 (5.0) | 227 (4.3) | 45 (4.5) | 46 (5.2) | 30 (4.6) | 110 (4.4) | 83 (4.7) | 26 (5.6) | 7 (10.0) | 242 (8.9) | 125 (5.1) | 43 (6.9) | 59 (2.3) |
| Nouvelle-Aquitaine | 1923 (9.2) | 578 (10.9) | 98 (9.9) | 82 (9.3) | 49 (7.5) | 128 (5.1) | 184 (10.4) | 24 (5.2) | 5 (7.1) | 209 (7.7) | 217 (8.8) | 54 (8.6) | 295 (11.5) |
| Occitanie | 2057 (9.8) | 401 (7.6) | 92 (9.3) | 74 (8.4) | 48 (7.3) | 217 (8.7) | 203 (11.5) | 35 (7.5) | 7 (10.0) | 295 (10.9) | 197 (8.0) | 42 (6.7) | 446 (17.4) |
| Overseas | 325 (1.5) | 119 (2.2) | 15 (1.5) | 22 (2.5) | 5 (0.8) | 40 (1.6) | 8 (0.5) | 1 (0.2) | 1 (1.4) | 46 (1.7) | 25 (1.0) | 0 | 43 (1.7) |
| Pays de la Loire | 997 (4.7) | 341 (6.4) | 63 (6.3) | 39 (4.4) | 21 (3.2) | 57 (2.3) | 55 (3.1) | 24 (5.2) | 2 (2.9) | 139 (5.1) | 149 (6.1) | 41 (6.5) | 66 (2.6) |
| Provence-Alpes-Côte d'Azur | 2090 (10.0) | 478 (9.0) | 91 (9.2) | 86 (9.7) | 100 (15.3) | 305 (12.2) | 203 (11.5) | 91 (19.5) | 16 (22.9) | 211 (7.8) | 234 (9.5) | 126 (20.1) | 149 (5.8) |
| Other | 1 (0.0) | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 1 (0.0) |
| Unknown | 47 (0.2) | 25 (0.5) | 3 (0.3) | 5 (0.6) | 1 (0.2) | 2 (0.1) | 3 (0.2) | 0 | 0 | 0 | 1 (0.0) | 0 | 7 (0.3) |
| Multiple regions | 12 (0.1) | 0 | 0 | 1 (0.1) | 0 | 3 (0.1) | 0 | 2 (0.4) | 0 | 1 (0.0) | 1 (0.0) | 0 | 4 (0.2) |
| Index year, n (%) | |||||||||||||
| 2020 | 5209 (24.8) | 1505 (28.3) | 342 (34.4) | 225 (25.4) | 137 (20.9) | 688 (27.5) | 407 (23.0) | 118 (25.3) | 0 | 600 (22.2) | 411 (16.7) | 8 (1.3) | 768 (30.0) |
| 2021 | 7755 (36.9) | 2038 (38.4) | 351 (35.3) | 365 (41.2) | 277 (42.4) | 924 (37.0) | 616 (34.8) | 245 (52.6) | 0 | 1045 (38.6) | 795 (32.4) | 187 (29.9) | 912 (35.6) |
| 2022 | 8031 (38.3) | 1768 (33.3) | 300 (30.2) | 295 (33.3) | 240 (36.7) | 886 (35.5) | 745 (42.1) | 103 (22.1) | 70 (100.0) | 1062 (39.2) | 1250 (50.9) | 431 (68.8) | 881 (34.4) |
| Study period duration, months, mean ± SD | 28.1 ± 9.7 | 29.2 ± 9.7 | 30.3 ± 9.9 | 29.0 ± 9.7 | 28.3 ± 9.1 | 29.1 ± 9.7 | 27.2 ± 9.9 | 30.4 ± 8.9 | 14.0 ± 1.1 | 27.6 ± 9.5 | 25.4 ± 9.2 | 20.9 ± 6.2 | 29.2 ± 9.9 |
ACS Aide à l’acquisition d’une complémentaire santé, AME Aide médicale de l’État, C2S Complémentaire Santé Solidarité, CMU-c couverture maladie universelle complémentaire, IL interleukin, PsO psoriasis, SD standard deviation, TNFα tumour necrosis factor alpha
aPatient characteristics were reported as of the closest date to the index date
bBiologic-naïve patients did not use any of the biologics under study in the 12 months before the index date (baseline period)
The mean (standard deviation) age for biologic-naïve patients was 50.0 (15.0) years, and 51.2% were male (Table 1). Almost all biologic-naïve patients (99.7%) were prescribed topical corticosteroids at baseline, and 59.3% were prescribed topical vitamin D derivatives for the treatment of PsO (Table 2). Baseline characteristics were largely consistent amongst patients treated with individual biologics.
Table 2.
Patient clinical characteristics present in ≥ 5% of biologic-naïve population at index date
| Patient characteristics at indexa | Biologic-naïveb | ||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| All biologic-naïve patientsb, N = 20,995 | Anti-TNFα | Anti-IL-17/IL-17 receptor | Anti-IL-23 | Anti-IL-12/23 | |||||||||
| Adalimumab, n = 5311 | Etanercept, n = 993 | Infliximab, n = 885 | Certolizumab, n = 654 | Secukinumab, n = 2498 | Ixekizumab, n = 1768 | Brodalumab, n = 466 | Bimekizumab, n = 70 | Guselkumab, n = 2707 | Risankizumab, n = 2456 | Tildrakizumab, n = 626 | Ustekinumab, n = 2561 | ||
| PsO treatments and other medications, n (%) | |||||||||||||
| Corticosteroids, plain | 20,927 (99.7) | 5304 (99.9) | 990 (99.7) | 882 (99.7) | 652 (99.7) | 2489 (99.6) | 1760 (99.5) | 464 (99.6) | 69 (98.6) | 2698 (99.7) | 2445 (99.6) | 621 (99.2) | 2553 (99.7) |
| Topical vitamin D derivatives (indicated only for the treatment of psoriasis) | 12,458 (59.3) | 2530 (47.6) | 333 (33.5) | 311 (35.1) | 304 (46.5) | 1495 (59.8) | 1132 (64.0) | 341 (73.2) | 47 (67.1) | 1952 (72.1) | 1809 (73.7) | 448 (71.6) | 1756 (68.6) |
| NSAIDs | 9581 (45.6) | 3041 (57.3) | 603 (60.7) | 417 (47.1) | 405 (61.9) | 1303 (52.2) | 851 (48.1) | 159 (34.1) | 32 (45.7) | 950 (35.1) | 788 (32.1) | 204 (32.6) | 828 (32.3) |
| Methotrexate | 9746 (46.4) | 2575 (48.5) | 556 (56.0) | 295 (33.3) | 244 (37.3) | 1161 (46.5) | 877 (49.6) | 205 (44.0) | 31 (44.3) | 1256 (46.4) | 1127 (45.9) | 270 (43.1) | 1149 (44.9) |
| Anxiolytic | 5529 (26.3) | 1489 (28.0) | 276 (27.8) | 318 (35.9) | 158 (24.2) | 641 (25.7) | 476 (26.9) | 131 (28.1) | 23 (32.9) | 640 (23.6) | 569 (23.2) | 139 (22.2) | 669 (26.1) |
| Antidepressant | 4039 (19.2) | 1110 (20.9) | 200 (20.1) | 237 (26.8) | 120 (18.3) | 531 (21.3) | 352 (19.9) | 81 (17.4) | 16 (22.9) | 454 (16.8) | 375 (15.3) | 100 (16.0) | 463 (18.1) |
| Apremilast | 3574 (17.0) | 629 (11.8) | 84 (8.5) | 47 (5.3) | 58 (8.9) | 475 (19.0) | 308 (17.4) | 123 (26.4) | 11 (15.7) | 609 (22.5) | 578 (23.5) | 145 (23.2) | 507 (19.8) |
| Retinoids | 1247 (5.9) | 172 (3.2) | 19 (1.9) | 33 (3.7) | 4 (0.6) | 178 (7.1) | 99 (5.6) | 52 (11.2) | 6 (8.6) | 225 (8.3) | 212 (8.6) | 48 (7.7) | 199 (7.8) |
| Comorbid conditions, n (%) | |||||||||||||
| Psoriatic arthritis | 2001 (9.5) | 799 (15.0) | 217 (21.9) | 131 (14.8) | 95 (14.5) | 331 (13.3) | 214 (12.1) | 13 (2.8) | 1 (1.4) | 74 (2.7) | 28 (1.1) | 11 (1.8) | 87 (3.4) |
| Inflammatory bowel disease | 1158 (5.5) | 581 (10.9) | 7 (0.7) | 245 (27.7) | 19 (2.9) | 6 (0.2) | 9 (0.5) | 1 (0.2) | 1 (1.4) | 16 (0.6) | 19 (0.8) | 7 (1.1) | 247 (9.6) |
| All cancers | 1493 (7.1) | 349 (6.6) | 91 (9.2) | 88 (9.9) | 26 (4.0) | 182 (7.3) | 98 (5.5) | 34 (7.3) | 4 (5.7) | 166 (6.1) | 155 (6.3) | 42 (6.7) | 258 (10.1) |
| Hypertension | 1368 (6.5) | 347 (6.5) | 105 (10.6) | 116 (13.1) | 13 (2.0) | 166 (6.6) | 108 (6.1) | 41 (8.8) | 7 (10.0) | 146 (5.4) | 112 (4.6) | 27 (4.3) | 180 (7.0) |
IL interleukin, NSAID nonsteroidal anti-inflammatory drugs, PsO psoriasis, TNFα tumour necrosis factor alpha
aPatient clinical characteristics were reported during the 12 months prior to the index date (historical period). This table reports characteristics present in ≥ 5% of patients
bBiologic-naïve patients did not use any of the biologics under study in the 12 months before the index date (baseline period)
Drug Persistence
During the maximum follow-up period of 42 months, biologic-naïve patients treated with anti-IL-23 agents had the highest rate of drug persistence compared with other classes of biologics, followed by anti-IL-12/23 and anti-IL-17/IL-17 receptor agents. The highest persistence rate was observed specifically for patients treated with risankizumab (85%, 95% confidence interval [CI] 82–87%). Patients treated with anti-TNFα agents had the lowest rate of persistence with the lowest persistence observed for patients treated with certolizumab (43%, 95% CI 37–49%). At 12 months (the longest observation period for patient treatment with bimekizumab), patients treated with risankizumab had the highest persistence rate (94%, 95% CI 93–95%) compared with all therapies and bimekizumab (90%, 95% CI 80–95%), whilst patients treated with etanercept had the lowest persistence rate (71%, 95% CI 68–74%) (Fig. 3).
Fig. 3.
Time to next treatment amongst biologic-naïve patients with PsO over 42 months.a aBiologic-naïve patients did not use any of the biologics under study in the 12 months before the index date (baseline period). IL interleukin, PsO psoriasis, TNF tumour necrosis factor
Switch Rate
Overall, switching was common amongst patients across the 42 months, with biologic switch rates varying from 15% to 57% (Fig. 4). At 42 months, biologic-naïve patients who were treated with anti-IL-23 agents had the lowest switch rate, whilst patients treated with anti-TNFα agents had the highest switch rates. Specifically, the lowest switch rates were observed for patients treated with risankizumab and guselkumab, 15% and 19%, respectively, whilst certolizumab and adalimumab had the highest switch rates, 57% and 52%, respectively (Fig. 4). Trends in switch rates between therapies were similar at 12, 24, and 36 months. At 12 months, the switch rate for patients treated with bimekizumab was 10% compared with 6% for risankizumab.
Fig. 4.
Switch rates across 42 months amongst biologic-naïve patients with PsO.a aBiologic-naïve patients did not use any of the biologics under study in the 12 months before the index date (baseline period). IL interleukin, PsO psoriasis, TNF tumour necrosis factor
Dose Escalation
Biologic-naïve patients treated with the anti-IL-23 inhibitors risankizumab and guselkumab had the longest time to dose escalation compared with other biologics in the study during the 42-month follow-up period. The shortest time to dose escalation was reported for patients treated with the anti-TNFα agents infliximab and adalimumab (Fig. 5).
Fig. 5.
Time to dose escalation amongst biologic-naïve patients with PsO over 42 months.a,b aBiologic-naïve patients did not use any of the biologics under study in the 12 months before the index date (baseline period). b172 patients were excluded because they switched treatment during the induction period. IL interleukin, PsO psoriasis, TNF tumour necrosis factor
Dose escalation rates at 42 months ranged from 9% to 56% (Fig. 6), and biologic-naïve patients treated with the anti-TNFα agents infliximab (56%) and adalimumab (32%) had the highest rates of dose escalation. The lowest rates of dose escalation at 42 months were observed for biologic-naïve patients treated with the anti-IL-23 agents, risankizumab and guselkumab (9% and 12%, respectively). The next lowest rate of dose escalation was seen for the anti-TNFα agent etanercept (15.0%). Trends in dose escalation rates between therapies were similar at 12, 24, and 36 months. At 12 months, 11% of biologic-naïve patients treated with bimekizumab had a dose escalation compared with 3% of patients treated with risankizumab.
Fig. 6.
Dose escalation rates across 42 months amongst biologic-naïve patients with PsO.a,b aBiologic-naïve patients did not use any of the biologics under study in the 12 months before the index date (baseline period). b172 patients were excluded because they switched treatment during the induction period. IL interleukin, PsO psoriasis, TNF tumour necrosis factor
HCRU and Medical Costs
Overall, patients who switched advanced therapies during the observation period compared with non-switchers had a significantly higher proportion of inpatient admissions (20.2% vs 13.4%, p < 0.001), 1-day hospital visits (57.0% vs 45.9%, p < 0.001), and ED visits (12.5% vs 10.2%, p < 0.001) within 12 months (Fig. 7a). Similar trends were observed for patients with or without dose escalations (Fig. 8a). Furthermore, switchers and dose escalators had a significantly higher number of HCRU visits (total length of stay was not significantly different between dose escalators compared with non-dose escalators; Figs. 7b and 8b).
Fig. 7.
a, b All-cause healthcare resource utilisation and c costs during the 12-month follow-up period amongst biologic-naïve PsO switchers versus non-switchers.a,b aBiologic-naïve patients did not use any of the biologics under study in the 12 months before the index date (baseline period). bThe switchers cohort includes patients who changed their index biologic treatment during the first 12 months following the index date. ED emergency department, PsO psoriasis, SD standard deviation
Fig. 8.
a, b All-cause healthcare resource utilisation and c costs during the 12-month follow-up period amongst biologic-naïve PsO dose escalators versus non-dose escalators.a,b aBiologic-naïve patients did not use any of the biologics under study in the 12 months before the index date (baseline period). bThe dose escalators cohort includes patients who increased the dosage of their index treatment during the first year following the first maintenance dose of the index treatment. ED emergency department, PsO psoriasis, SD standard deviation
Mean medical costs at 12 months were significantly higher for patients who switched advanced therapy (€3435) or had a dose escalation (€3453) compared with non-switchers (€2578) or non-dose escalators (€2638; all comparisons p < 0.001) (Fig. 7c and 8c). Mean pharmacy costs were the largest medical costs for patients who switched therapy (€13,105) or had a dose escalation (€13,736). ED costs were significantly higher for non-switchers compared with switchers (mean, €51 vs €45; p < 0.001), although the absolute difference was modest (€6).
Discussion
To our knowledge this is the first national real-world study of biologic-naïve patients with moderate-to-severe PsO in France to comprehensively investigate treatment patterns, including switching and dose escalation, as well as the associated economic outcomes, such as HCRU and cost after initiating biologic therapies.
Our findings showed that rates of switching or dose escalations were common for several of the biologics studied in patients with PsO in France. This observation is in line with data from the USA, which observed that 23.0% of patients with PsO switched to another biologic agent over 12 months [17], demonstrating consistency in outcomes across different countries. In this study, the lowest switch rates were seen for anti-IL-23 therapies, followed by anti-IL-12/23 agents and anti-IL-17 agents, with anti-TNFα therapies having the highest switch rate for patients in France. Similar to our findings, a real-world study in the USA found that patients with PsO treated with the anti-IL agents ixekizumab (15%), secukinumab (16%), ustekinumab (15%), or guselkumab (11%) were associated with a lower switch rate compared with those treated with the anti-TNFα agents adalimumab (25%), etanercept (28%), or infliximab (26%) over 12 months [17]. Specifically, we observed that over 42 months, patients treated with the anti-IL-23 antibody risankizumab had the lowest rates of switching compared with patients treated with other biologics. Similar to the results in this study, a real-world analysis in Japan found that 10.7% of patients treated with risankizumab experienced switching at 24 months, followed by ustekinumab (19.29%), ixekizumab (21.5%), guselkumab (23.2%), secukinumab (25.7%), and brodalumab (26.8%) [18]. As these analyses were descriptive and not adjusted for possible confounding variables, further investigation is warranted to understand why anti-IL-23 therapies, and more specifically risankizumab, have lower switch rates in patients with PsO compared with other biologics.
Previous studies have suggested that higher switch rates are likely due to the lack of initial effectiveness, reduced effectiveness over time, or patient experience of adverse events [19, 20]. The switching of biologic agents in these prior studies was used as a strategy to improve effectiveness or minimise safety outcomes in patients with PsO [19, 20].
In this study, the anti-TNFα agent certolizumab was associated with high rates of switching and low rates of persistence in patients in France. This finding may correspond to the fact that during pregnancy, women switch to certolizumab as an alternative to a previous biologic agent for safety reasons, as certolizumab is only able to cross the placenta by passive means, and evidence suggests it does not transfer to breast milk. After delivery and/or breastfeeding, these women are likely to switch back to their previous biologic agent [21]. This study was unable to identify the reasons for why patients switched therapies, and therefore the high switch rate for certolizumab may be confounded by patients who experienced a pregnancy during the observational period. Future analysis comparing treatment patterns between male and female patients treated with certolizumab may help to determine if the higher rates of switching observed for certolizumab are linked to pregnancy.
At 42 months, anti-IL-23 agents (risankizumab and guselkumab) provided the lowest rates of dose escalation in the study population, followed by etanercept (anti-TNFα), brodalumab and secukinumab (anti-IL-17/IL-17 receptor agents), certolizumab (anti-TNFα), ixekizumab (anti-IL-17/IL-17 receptor), and ustekinumab (anti-IL-12/23). The anti-TNFα agents adalimumab and infliximab had the highest rates of dose escalation. Patients treated with tildrakizumab had a higher dose escalation rate compared with the other anti-IL-23 agents at 36 months. A systematic review and meta-analysis found that tildrakizumab had the lowest efficacy among anti-IL-23 therapies, with 37.2% of patients treated with tildrakizumab having a ≥ 90% improvement from baseline in PsO Area and Severity Index (PASI) compared with 66.4% and 70.8% of patients treated with guselkumab and risankizumab, respectively [22]. A possible explanation is that patients treated with tildrakizumab may require a dose escalation to increase the therapeutic effect. However, it should be noted that as clinical data were not available within the claims database, we were unable to evaluate the potential confounding effect of differences in disease severity between treatment groups.
Overall, patients treated with risankizumab had the lowest rates of dose escalation in this study, similar to a real-world study based in the USA, which showed that the percentage of patients with PsO with dose escalation was significantly lower with risankizumab (2.0%) compared with other anti-IL-23, anti-IL-12/23, anti-TNF, or anti-IL-17 agents during the full maintenance periods of each therapy [23]. This is further supported by Japanese claims data showing 0% of patients treated with risankizumab had dose escalation compared with 37.2% of patients treated with ixekizumab at 12 months [24]. This previous study suggested lower dose escalation rates may indicate greater therapeutic benefit at per-label dosing allowing for longer-term sustained disease control [24], although future adjusted analysis accounting for disease severity and reasons for dose escalation would be important to understand these results. In contrast, patients treated with the anti-TNFα agents adalimumab and infliximab had the highest rates of dose escalation in this study. A systematic review comparing PASI response for PsO therapies observed that 44.2% and 55.8% of patients treated with adalimumab and infliximab, respectively, achieved ≥ 90% improvement from baseline in PASI compared with ixekizumab (71.6%), risankizumab (70.8%), brodalumab (70.1%), guselkumab (66.4%), and secukinumab (62.0%) [22]. These results indicate that anti-TNFα agents may have lower efficacy compared with other therapies, and this difference warrants further study to aid in identifying the best treatment option for patients with PsO in France and other regions.
Switching and dose-escalated use of biologics was associated with increased HCRU, including inpatient admissions and ED visits, which tend to be costly because of the need for complex and specialised care and highly trained staff [25]. As observed at 12 months post-index in this study, increased HCRU coincided with higher all-cause costs, including mean medical, inpatient, 1-day hospital, outpatient, ED, and pharmacy costs. Biologic-naïve patients with PsO who switched or received dose escalation incurred 33.2% or 30.9% higher mean annual medical costs compared with non-switchers or non-dose escalators at 12 months. It is possible that the higher mean costs associated with switching and dose escalation could be attributed to the need for additional diagnostic testing, dose titration, and hospital visits seeking to address adverse events that may have occurred during initiation of a new treatment. However, this association may be confounded by disease severity and other patient characteristics, such as patients with more severe or uncontrolled disease being more likely to switch or escalate, as well as incur higher costs. Therefore, further adjusted analysis is needed to confirm this relationship. Despite this, these findings are supported by a previous study conducted in the USA among patients with PsO that demonstrated that therapy switching compared with non-switching was associated with an increase in ED and outpatient visits and significantly higher mean all-cause total healthcare costs, which may be because switchers experience more uncontrolled disease or safety issues [26]. Together these findings highlight that the selection of biologic agents in patients with PsO may significantly impact HCRU and economic outcomes for patients and the healthcare system.
A major strength of this study is it provides a comprehensive analysis of patients with PsO from a large real-world database including demographic, medical claim, and pharmacy claim data for 99% of the French population [27]. Further, the combination of both treatment pattern and economic data, and the inclusion of bimekizumab, which was not reported in past studies, provide unique insights.
This study did have limitations. As a result of the descriptive nature of this study, causal conclusions cannot be drawn. In addition, because the SNDS database included limited information on potential covariates, additional adjusted analyses will be necessary to confirm these findings and evaluate associations such as those between biologic agent and time to switch or dose escalation. We were unable to determine reasons for switching or dose escalation from the claims database which would provide additional context to the interpretation of the data; it is also possible that reasons for switching may differ between different biologics, and as a result no conclusions on effectiveness or safety could be made. We did not have access to clinical notes detailing disease severity; therefore, moderate-to-severe PsO was based on a proxy of patients initiating a biologic (anti-TNFα, anti-IL) with prior use of phototherapy, topical treatment, or systemic therapy. Furthermore, as a result of a lack of severity data, channelling bias may have contributed to the observed difference in HCRU and costs and limited our ability to make direct comparisons between biologic agents. For example, some biologics, such as infliximab, may be chosen over other biologic agents for the most severe or difficult-to-treat patients because these therapies can easily be adjusted to fit patients’ needs, but this study was unable to capture that data. Whilst the inclusion of newly approved biologics such as bimekizumab provides unique insights that have not previously been captured in claims-based studies, the relatively small sample size available for analysis may limit the robustness of the findings; therefore, these results should be interpreted with caution. Similarly, differences in available follow-up times for outcomes (42 months for treatment patterns and 12 months for HCRU and costs) and smaller sample sizes at later timepoints limited our ability to interpret our findings. This study did not include oral medications (e.g. deucravacitinib), making the range of therapies available for the treatment of PsO not fully represented. Additionally, claims data may be subject to data errors (e.g. miscoding), and the presence of a claim for a filled prescription may not indicate actual use of the drug by a patient. Future studies identifying factors influencing switch rates and dose escalations in the biologic-naïve PsO population in France may help to optimise clinical decision-making related to disease management and improve treatment outcomes for patients.
Conclusion
In this real-world study, treatment switching and dose escalation amongst patients with PsO using biologics was common over 42 months. Patients treated with anti-IL-23 agents had the lowest rates of treatment switching and dose escalation compared with other classes of biologics. At an individual drug level, risankizumab had the lowest rates of treatment switching and dose escalation. Furthermore, patients who experienced a switch or dose escalation subsequently experienced higher HCRU and costs. Therefore, reducing the rate of switching or dose escalation may indirectly help reduce economic burden for patients and the healthcare system. These findings may inform clinical decision-making related to the selection of the first biologic treatment for patients with PsO.
Acknowledgments
Medical Writing/Editorial Assistance
Medical writing services provided by Ellie Sweatman, PhD, of Avalere Health Ltd, and funded by AbbVie.
Author Contributions
Substantial contributions to conception and design: Thierry Passeron, Richard Ta, Keith A. Betts, Maryaline Catillon. Acquisition of data: Keith A. Betts. Data analysis: Keith A. Betts, Maryaline Catillon, Calvin Plays. Interpretation of data: Thierry Passeron, Richard Ta, Keith A. Betts, Maryaline Catillon, Calvin Plays. Involved in drafting/revising critically for intellectual content: Thierry Passeron, Richard Ta, Véronique Hospital, Maryaline Catillon, Denis Jullien.
Funding
AbbVie funded this study and participated in the study design, research, analysis, data collection, interpretation of data, reviewing, and approval of the publication. All authors had access to relevant data and participated in the drafting, review, and approval of this publication. No honoraria or payments were made for authorship. AbbVie funded the Rapid Service Fee for publication.
Data Availability
The datasets generated and/or analysed during the current study are not publicly available, in accordance with the Order of 22 March 2017, on the security repository applicable to the National Health Data System (SNDS). Access to SNDS data is restricted to qualified and authorized personnel only. No individual data can be extracted from the secured platform.
Declarations
Conflict of Interest
Thierry Passeron has received personal fees from AbbVie, ACM Pharma, Almirall, Boehringer Ingelheim, Bristol Myers Squibb, Calypso, Celgene, Galderma, Genzyme/Sanofi, GlaxoSmithKline, Incyte, Janssen, LEO Pharma, Eli Lilly, Novartis, Roivant Sciences Ltd, Sanofi, Sun Pharmaceuticals, and VYNE Therapeutics and is a co-founder of NIKAIA Pharmaceuticals and founder of SUNLUTION. Richard Ta and Véronique Hospital are employees of AbbVie and may hold stocks/stock options. Keith Betts and Calvin Plays are employees of Analysis Group, Inc., which received payment from AbbVie Inc. for participation in this research. Maryaline Catillon is a former employee of Analysis Group, Inc., which received payment from AbbVie Inc. for participation in this research. Denis Jullien is a consultant for and has received personal fees from AbbVie, Almirall, Amgen, Boehringer Ingelheim, Bristol Myers Squibb, Celgene, Janssen-Cilag, Leo, Lilly, MEDAC, Novartis, and UCB.
Ethical Approval
Ethical approval issued by the Comité Éthique et Scientifique pour les Recherches, les Études et les Évaluations dans le domaine de la Santé (CESREES) on June 15, 2023 (TPS No. 12242458).
Footnotes
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
References
- 1.Richard M-A, Corgibet F, Beylot-Barry M, et al. Sex- and age-adjusted prevalence estimates of five chronic inflammatory skin diseases in France: results of the « OBJECTIFS PEAU » study. J Eur Acad Dermatol Venereol. 2018;32(11):1967–71. [DOI] [PubMed] [Google Scholar]
- 2.Richard MA, Sei JF, Philippe C, et al. Prevalence of comorbidities in atopic dermatitis and psoriasis in the French population. Ann Dermatol Venereol. 2021;148(1):28–33. [DOI] [PubMed] [Google Scholar]
- 3.Armstrong AW, Read C. Pathophysiology, clinical presentation, and treatment of psoriasis: a review. JAMA. 2020;323(19):1945–60. [DOI] [PubMed] [Google Scholar]
- 4.Richard M-A, Aubin F, Beneton N, et al. Moderate psoriasis in clinical practice: French expert consensus using a modified Delphi method. Adv Ther. 2022;39(11):5203–15. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Meyer N, Paul C, Feneron D, et al. Psoriasis: an epidemiological evaluation of disease burden in 590 patients. J Eur Acad Dermatol Venereol. 2010;24(9):1075–82. [DOI] [PubMed] [Google Scholar]
- 6.Nast A, Smith C, Spuls PI, et al. EuroGuiDerm guideline on the systemic treatment of psoriasis vulgaris - part 1: treatment and monitoring recommendations. J Eur Acad Dermatol Venereol. 2020;34(11):2461–98. [DOI] [PubMed] [Google Scholar]
- 7.Masson Regnault M, Brenaut E, Marniquet M-E, et al. French guidelines on systemic treatments for moderate-to-severe psoriasis in adults: update 2025. J Eur Acad Dermatol Venereol. 2025;00:1–17. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Curmin R, Guillo S, De Rycke Y, et al. Switches between biologics in patients with moderate-to-severe psoriasis: results from the French cohort PSOBIOTEQ. J Eur Acad Dermatol Venereol. 2022;36(11):2101–12. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Armstrong AW, Puig L, Joshi A, et al. Comparison of biologics and oral treatments for plaque psoriasis: a meta-analysis. JAMA Dermatol. 2020;156(3):258–69. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Sbidian E, Chaimani A, Guelimi R, et al. Systemic pharmacological treatments for chronic plaque psoriasis: a network meta-analysis. Cochrane Database Syst Rev. 2025;8:Cd011535. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Pina Vegas L, Penso L, Claudepierre P, Sbidian E. Long-term persistence of first-line biologics for patients with psoriasis and psoriatic arthritis in the French health insurance database. JAMA Dermatol. 2022;158(5):513–22. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Roche H, Bouiller K, Puzenat E, et al. Efficacy and survival of biologic agents in psoriasis: a practical real-life 12-year experience in a French dermatology department. J Dermatolog Treat. 2019;30(6):540–4. [DOI] [PubMed] [Google Scholar]
- 13.Sbidian E, Mezzarobba M, Weill A, Coste J, Rudant J. Persistence of treatment with biologics for patients with psoriasis: a real-world analysis of 16 545 biologic-naïve patients from the French National Health Insurance database (SNIIRAM). Br J Dermatol. 2019;180(1):86–93. [DOI] [PubMed] [Google Scholar]
- 14.Le Moigne M, Sommet A, Lapeyre-Mestre M, et al. Healthcare cost impact of biological drugs compared with traditional systemic treatments in psoriasis: a cohort analysis in the French insurance database. J Eur Acad Dermatol Venereol. 2014;28(9):1235–44. [DOI] [PubMed] [Google Scholar]
- 15.Maillard O, Bun R, Laanani M, et al. Use of the French National Health Data System (SNDS) in pharmacoepidemiology: a systematic review in its maturation phase. Therapies. 2024;79(6):659–69. [DOI] [PubMed] [Google Scholar]
- 16.Sbidian E, Mezzarobba M, Shourick J, et al. Choice of systemic drugs for the management of moderate-to-severe psoriasis: a cross-country comparison based on national health insurance data. Acta Derm Venereol. 2021;101(6):adv00473. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Thai S, Zhuo J, Zhong Y, et al. Real-world treatment patterns and healthcare costs in patients with psoriasis taking systemic oral or biologic therapies. J Dermatolog Treat. 2023;34(1):2176708. [DOI] [PubMed] [Google Scholar]
- 18.Tada Y, Soliman AM, Ishii K, et al. Real-world discontinuation and switching patterns for interleukin-inhibitor treatments in patients with moderate-to-severe psoriasis in Japan. Dermatol Ther. 2024;14(1):99–114. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Bagit A, Sood S, Alli S, et al. The efficacy and safety of switching between interleukin inhibitors in psoriasis therapy: a systematic review. JAAD Rev. 2024;1:97–9. [Google Scholar]
- 20.Zhang L, Zhao X, Li W, Sun Y, Liu H. Switching biologic agent in patients with psoriasis: a systematic review and meta-analysis. J Dermatolog Treat. 2025;36(1):2521082. [DOI] [PubMed] [Google Scholar]
- 21.Esposito M, Calianno G, Lappi A, Fargnoli MC. Treatment of severe psoriasis during pregnancy and breastfeeding: a therapeutic challenging case. Int J Womens Dermatol. 2021;7(5, Part B):832–4. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Sawyer LM, Malottki K, Sabry-Grant C, et al. Assessing the relative efficacy of interleukin-17 and interleukin-23 targeted treatments for moderate-to-severe plaque psoriasis: a systematic review and network meta-analysis of PASI response. PLoS ONE. 2019;14(8):e0220868. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Wu JJ, Patel M, Zeng F, et al. Real-world dose escalation of biologics for moderate-to-severe psoriasis in the United States. J Dermatolog Treat. 2023;34(1):2200869. [DOI] [PubMed] [Google Scholar]
- 24.Tada Y, Soliman AM, Ishii K, et al. Dose escalation of biologic treatment in patients with moderate-to-severe psoriasis in Japan. Exp Dermatol. 2024;33(5):e15067. [DOI] [PubMed] [Google Scholar]
- 25.Scott KW, Liu A, Chen C, et al. Healthcare spending in U.S. emergency departments by health condition, 2006-2016. PLoS ONE. 2006;16(10):e0258182. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Feldman SR, Tian H, Wang X, Germino R. Health care utilization and cost associated with biologic treatment patterns among patients with moderate to severe psoriasis: analyses from a large U.S. claims database. J Manag Care Spec Pharm. 2019;25(4):479–88. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Blin P, Lasalle R, Thurin N, et al. PRM93 - SNDS, the French nationwide claims database: a powerful tool for pharmacoeconomy and pharmacoepidemiology. Value Health. 2018;21:S371. [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Data Availability Statement
The datasets generated and/or analysed during the current study are not publicly available, in accordance with the Order of 22 March 2017, on the security repository applicable to the National Health Data System (SNDS). Access to SNDS data is restricted to qualified and authorized personnel only. No individual data can be extracted from the secured platform.