Secukinumab compared to ixekizumab in biologic-naïve Chinese individuals afflicted with moderate-to-severe plaque psoriasis: A retrospective single-center analysis

Abstract

The clinical benefits of interleukin (IL)-17A therapies for psoriasis are well-substantiated. However, efficacy outcomes for moderate-to-severe plaque psoriasis (pPS) may vary across IL-17A treatments. Practitioners require evidence to guide treatment selection within the IL-17A class for biologic-naïve patients with moderate-to-severe pPS. This study compares the effectiveness and safety of ixekizumab and secukinumab in such patients. A retrospective analysis of 5 years of electronic medical records (demographics, clinical characteristics, treatment outcomes, and adverse effects) was conducted on biologic-naïve patients receiving IL-17A treatments for moderate-to-severe stable pPS. The study comprised 70 ixekizumab users (ET cohort) and 55 secukinumab users (SB cohort) from a single center. Concurrent therapies were evenly distributed across groups. Both cohorts exhibited significant reductions in percentage body surface area involvement and Psoriasis Area and Severity Index scores posttreatment (P < .001 for all). However, the SB cohort demonstrated greater reductions in percentage body surface area involvement (P < .001) and Psoriasis Area and Severity Index scores compared with the ET cohort. Physician Global Assessment of treatment response favored the SB cohort (1.1 [1.4–1] vs 1.1 [1.5–1]). Systemic organ infections were more prevalent in the SB cohort (P = .009). Following treatment, 26 (37%) and 24 (44%) patients in the ET and SB cohorts, respectively, achieved mild disease activity (≤1 on Physician Global Assessment). The SB cohort experienced a higher incidence of adverse effects (P < .001, risk ratio = 0.52; 95% confidence interval = 0.37–0.73, I² = 0%). In Chinese biologic-naïve patients with moderate-to-severe pPS, secukinumab may be associated with a superior response but a higher incidence of adverse effects compared with ixekizumab.

1. Introduction

Plaque psoriasis (pPS) of moderate-to-severe manifestation constitutes a systemic inflammatory dermatological condition,^[1] potentially affecting appendages and joints, further complicated by systemic comorbidities such as cardiometabolic syndrome and inflammatory bowel diseases, thus representing a significant disease burden.^[2] A considerable proportion of pPS patients remains inadequately treated.^[3] Moderate-to-severe pPS is prevalent in China, affecting 4.5% to 4.8% of the Chinese populace,^[4] with pPS being the predominant form.^[5]

Current systemic therapeutic options for moderate-to-severe pPS include retinoids, phosphodiesterase-4 inhibitors (e.g., apremilast), methotrexate, fumarates, and calcineurin inhibitors (e.g., cyclosporine).^[6,7] Retinoids, methotrexate, and cyclosporine are broadly acting, orally administered molecules effective for moderate-to-severe pPS.^[3] However, their long-term use is associated with severe adverse effects, particularly in specific conditions (e.g., pregnancy) and drug–drug interactions, necessitating meticulous monitoring, thereby limiting their application in treating moderate-to-severe pPS.^[6,8,9]

Further systemic treatments for moderate-to-severe pPS encompass biologics, such as interleukin (IL)-17 inhibitors, IL-23 inhibitors, IL-12 inhibitors, and antitumor necrosis factor agents.^[6,7] While these agents enhance the quality of life for patients with moderate-to-severe pPS,^[10] they carry the risk of secondary serious infections.^[3] Moreover, their efficacy may diminish over time, and repeated administration can lead to administration site reactions.^[6,8] Data concerning long-term treatment persistence with these agents remain limited.^[11] Consequently, there is an unmet need for convenient, safer, and patient-managed systemic treatments for moderate-to-severe pPS.^[3]

Anti-IL-17A antibodies have demonstrated efficacy in pPS,^[12] with ixekizumab and secukinumab being prominent examples.^[13] Asian patients with moderate-to-severe pPS treated with anti-IL-17A agents (ixekizumab or secukinumab) exhibit significantly higher odds of achieving favorable outcomes compared with those treated with other biologics and other populations.^[14]

Ixekizumab (LY2439821) is a humanized IgG4 monoclonal antibody that neutralizes IL-17A.^[15] A phase 2 trial has indicated short-term efficacy and safety of ixekizumab in moderate-to-severe pPS.^[16] Both the United States Food and Drug Administration and the European Medicines Agency approved ixekizumab in 2016 for the treatment of moderate-to-severe pPS.^[15] Secukinumab, another anti-IL-17A antibody, is also the United States Food and Drug Administration-approved.^[2] Retrospective studies suggest comparable efficacies between ixekizumab and secukinumab.^[17] However, ixekizumab may offer improved treatment adherence and reduced discontinuation and switching rates compared with secukinumab in patients with moderate-to-severe pPS.^[18] Continuous treatment is generally required for individuals with moderate-to-severe pPS.^[17] Notably, head-to-head comparisons of ixekizumab and secukinumab in biologic-experienced patients with moderate-to-severe pPS are scarce in the Chinese context. Hence, there is a need for concrete evidence to guide practitioners in making informed treatment decisions within the IL-17A biologic class for biologic-naïve patients suffering from moderate-to-severe pPS. While real-world studies have been conducted on these agents,^[19–28] direct comparative data of these IL-17A inhibitors remain limited. Details of comparative pPS treatment studies in varying settings are summarized in Table 1.

Table 1.

Details of the comparative studies on the treatments of the psoriatic disease in different settings.

Study	Published year	Patients’ ethnicity	Sample size (N; patients)	Age (yr)	Main conclusions
Retrospective study, Benites et al[1]	2022	Ecuador	94	Mean: 52.2 (range 1–88)	The biologics used for the treatment of moderate and severe psoriasis.
Phase 3 trial, Korman et al[3]	2024	Caucasian	1264	≥18	Ducravacitinib shows significant efficacy in the first week for moderate-to-severe plaque psoriasis, maintaining clinical response for up to 52 wk.
Prospective, Observational Study, Ying et al[4]	2023	Chinese	666	Adults (≥18)	Ixekizumab is well-tolerated and effective in real-world clinical practice in Chinese adults with moderate-to-severe plaque psoriasis.
Survey, Chen et al[5]	2017	Chinese	12,031	39.42 ± 18.4 (1–86)	Characterizing psoriasis in the Han Chinese population can be used as basic information for future studies.
Population-based survey, Lebwohl et al[9]	2014	Multinational	139,948	18–97	Disease severity, psoriasis screening, patient awareness, and therapies influence treatment outcomes in adults with moderate-to-severe plaque psoriasis.
Multicenter retrospective real-world study, Gargiulo et al[11]	2022	Italian	5300	53.74 ± 14.85	IL-23 inhibitors, and risankizumab in particular, have a higher probability of drug survival overall during a 4-yr follow-up.
Prospective study, Lynde et al[14]	2023	Asin + White	1981	Adults (≥18)	Comorbidity of patients with psoriasis may affect a clinical response.
Meta-analysis, Augustin et al[15]	2020	Real-world	43 studies	Adult patients (≥18)	Secukinumab has good outcomes, safety, and high survival in patients with moderate-to-severe psoriasis.
Phase 2 trial, Leonardi et al[16]	2012	European	142	≥18	Ixekizumab is improved the clinical symptoms of psoriasis.
Retrospective study, Blauvelt et al[17]	2020	The North American	645 ixekizumab users and 1152 secukinumab users	≥18	Ixekizumab shows better durability, lower discontinuation rates, higher adherence, and similar switching risk compared with secukinumab.
Retrospective study, Blauvelt et al[18]	2021	The North American	411 ixekizumab users and 780 secukinumab users	≥18	Ixekizumab has higher adherence rates and significantly lower non-persistence, discontinuation, and switching compared with secukinumab.
Real-life study, Valenti et al[19]	2024	Italy	145 ixekizumab users and 1096 secukinumab users	53.03 ± 14.74	Ixekizumab has the long-term effectiveness and safety.
Observational study, Li et al[20]	2024	Chinese	663	≥18	Ixekizumab is well-tolerated in Chinese patients with moderate-to-severe plaque psoriasis.
Phase 3b trial, Bagel et al[23]	2017	The North American	102	≥18	Secukinumab is efficacious and well-tolerated for patients with extensive moderate-to-severe scalp psoriasis.
Trial, Simpson et al[24]	2015	The North American	445	Adults (≥18)	The Lattice System Physician’s Global Assessment weights body surface area involved with psoriasis and aspects of plaque morphology.
Retrospective study, Bernardini et al[25]	2024	Italy	11	53.36 ± 19.38	Secukinumab may be an effective treatment option for patients with an inadequate response or loss of efficacy to ixekizumab.
Observational study, Li et al[26]	2024	Chinese	612	≥18	Ixekizumab is similarly effective in patients with plaque psoriasis across different special body areas.
Observational study, Pinter et al[27]	2024	Caucasian	1981	Adults (≥18)	6- and 12-months treatments of biologics are effective in psoriasis.
A systematic literature review, Armstrong et al[28]	2021	Real-world	195 studies	All age group	Ixekizumab, risankizumab, and brodalumab have the highest short-term efficacy, and risankizumab has the highest long-term efficacy.

Variables are reported as mean ± standard deviation.

Given the substantial number of pPS patients and the limitations of current treatments, including an increased risk of secondary infections, severe adverse effects, resistance, and loss of efficacy, and considering that anti-IL-17A antibodies, specifically Ixekizumab and Secukinumab, have emerged as highly effective therapeutic strategies against pPS, a study comparatively evaluating the efficacy of these biologics in pPS patients is warranted.

The objective of this single-center retrospective study is to compare the effectiveness (Psoriasis Area and Severity Index [PASI] score, physician global assessment [PGA] score, and body surface area [BSA] involvement) and safety (adverse effects during treatment and throughout a 1-year follow-up) of ixekizumab and secukinumab in Chinese patients with moderate-to-severe pPS.

2. Materials and methods

2.1. Ethical oversight and participation consent

The authors formulated the study’s protocols, which received approval from the Beijing Shijitan Hospital’s human ethics committee (IIT2025-015-003, dated June 17, 2025). Conducted in accordance with Chinese law and the Declaration of Helsinki (v2008), this retrospective study was granted a waiver for informed participation consent and Chinese clinical trial registry registration by the ethics committee. Participants provided informed consent for publication of study findings.

2.2. Design, setting, and temporal scope

A retrospective, single-center analysis was executed at Beijing Shijitan Hospital, China, utilizing medical records spanning January 1, 2019, to August 15, 2023.

2.3. Inclusionary prerequisites

Eligible subjects were ≥18 years, exhibiting moderate-to-severe stable pPS, fulfilling all criteria:

1. PASI score ≥ 12

2. Static PGA score 1–3

3. BSA involvement ≥ 10%

4. Disease duration ≥ 6 months.

2.4. Exclusionary requisites

Subjects with prior biological therapy exposure, incomplete records, chronic comorbidities, pregnant individuals, or previous ixekizumab exposure before secukinumab treatment were excluded.

2.5. Sample size determination

Assuming ≥30% of patients achieve ≤1 (mild disease activity per PGA) post-psoriasis treatment at 12 weeks, with α = 0.05, β = 0.2, and 80% power, a minimum cohort size of 55 patients was required.^[29] Sample size was computed via OpenEpi 3.01 software.

2.6. Cohorts

The ET cohort comprised 70 patients who commenced with a subcutaneous administration of 160-mg ixekizumab (two 80-mg injections), succeeded by 80 mg at weeks 2, 4, 6, 8, 10, and 12, and subsequently 80 mg every 4 weeks, culminating in 12 weeks of ixekizumab therapy^[15] The SB cohort consisted of 55 patients who received 300 mg of subcutaneous secukinumab initially, then weekly for the first 3 weeks, and thereafter every 4 weeks from week 4 through week 12.^[23] Concurrent with subcutaneous ixekizumab or secukinumab, patients received adjunctive psoriasis therapies and adverse event management (as deemed necessary by the physician). Our institution offers both treatment modalities to patients, with adjunctive therapies encompassing topical medications (e.g., corticosteroids, vitamin D analogues), phototherapy, or alternative systemic agents.

2.7. Treatment outcome metrics

2.7.1. PASI score (secondary endpoint)

The PASI score assesses psoriasis severity and extent by evaluating lesion type and affected skin area on the head, arms/upper extremities, trunk, and legs/lower extremities. Subgrades for erythema (redness), induration (thickness), and desquamation (scaling) are scored as follows: 0 = none, 1 = slight, 2 = moderate, 3 = severe, and 4 = very severe. The affected skin area in each region is graded as: 0 = 0%, 1 = <10%, 2 = 10%–29%, 3 = 30%–49%, 4 = 50%–69%, 5 = 70%–89%, and 6 = ≥90%. The total score ranges from 0 to 72, where 72 indicates maximum disease severity and 0 indicates disease absence.^[24]

2.7.2. PGA score (primary endpoint)

The PGA score measures the overall treatment response as assessed by the physician. Erythema, induration, and scaling are key factors in determining the PGA score, which ranges from 0 to 3: 0 to 0.4 = no disease activity; ≥0.5 to ≤1 = mild disease activity; >1 to ≤2 = moderate disease activity; >2 to <3 = severe disease activity; and 3 = most severe disease activity.^[29]

2.7.3. BSA involvement

BSA involvement is expressed as the percentage of body surface affected by psoriasis, calculated from 4 regions (head, arms/upper extremities, trunk, and legs/lower extremities). The range is from 0% to 100%.^[3]

PASI score and BSA involvement were evaluated before (BT) and after (AT) treatment. The PGA score was evaluated posttreatment (AT).

2.7.4. Adverse events

Serious drug-induced adverse events (e.g., liver toxicity) and nonserious adverse events were collected and analyzed during treatment and a 1-year follow-up. Serious adverse events were defined as those requiring treatment discontinuation or reported after treatment completion. Nonserious adverse events were those that did not necessitate treatment cessation. The Medical Dictionary for Regulatory Activities (MedDRA version 25.0) was used to code all adverse events by consultants.^[4]

2.8. Clinical benefits for pPS treatments

The clinical benefits of different treatments for patients with pPS were evaluated based on beneficial scores, calculated from the risk of undertreatment, as expressed in Eq. (1). Treatment risk was defined by a calculation involving the PGA of treatment response for pPS (Eq. (2)). The PGA of treatment response was considered a numerical value from 0 to 3. The beneficial score of the treatment response is the area above the curve of the treatments adopted, and the working area is the area under the curve of the adopted treatments. For all adopted treatments, ≤1 (mild disease activity), the PGA after treatments for pPS was used as the reference standard.^[30]

$$\text{Beneficial score }=\frac{\text{ Number of patients with }\le 1\text{ the PGA of response after treatments }}{\text{Total number of patients that cohort }}-$$

$$(\frac{>1\text{ the PGA of response after treatments }}{\text{ Total number of patients in that cohort }}\times \text{ Risk of under treatments })$$ (1)

$$\text{Risk of under treatments}=\frac{\text{ PGA of response after treatments }}{\text{1 + PGA of response after treatments }}.$$ (2)

2.9. Statistical analyses

Statistical analyses were conducted utilizing InStat 3.01 (GraphPad, San Diego). Categorical variables are presented as frequencies (percentages), normally distributed continuous variables as mean ± standard deviation, and non-normally distributed continuous variables as median (Q3–Q1). Categorical variables underwent analysis via Fisher exact test or Chi-squared test (χ²; with or without Yates’ correction). Quartile values were computed employing the Soup calculator (Soup calculator, LLB, USA). Continuous variable normality was assessed via the Kolmogorov–Smirnov method. Normally distributed continuous variables were statistically analyzed using unpaired (between-cohort) or paired t-tests (within-cohort, with or without Welch correction), repeated measures analysis of variance (ANOVA; within-cohort), or 1-way ANOVA (between-cohort). Bartlett test verified the normality of continuous variables. Non-normally distributed continuous variables were analyzed via the Mann–Whitney test (between-cohorts) or Friedman test (nonparametric repeated ANOVA; within-cohort). Post hoc analyses of continuous variables employed Dunnett (normal distribution) or Dunn (non-normal distribution) multiple comparisons test. Statistical significance was set at P < .05. Forest and funnel plots were generated via metaanalysisonline.com (2024-2025, ELIXIR Hungary). I² > 50% denoted moderate heterogeneity. VisualGPT (https://visualgpt.io/mind-map-maker/create) AI flowchart generator was used to generate flowcharts. https://viscript.app/sign_up AI generator was used to create Graphical Abstract. https://anythingtranslate.com/translators/premium-english-translator/ used for Premium English. Sentence Shortener (https://www.junia.ai/tools/sentence-shortener) AI tool used to condense text.

3. Results

3.1. Cohort profile

From January 1, 2019, to August 15, 2023, a cohort of 135 patients undergoing therapies for moderate-to-severe stable pPS at Beijing Shijitan Hospital, China, was assembled. Exclusions comprised 5 patients with prior biologic exposure, 1 with incomplete records, and 4 withdrawals due to adverse events (two per cohort). Ergo, data from these 10 subjects were excised from analysis. The electronic health records of 125 patients on IL-17A inhibitors for moderate-to-severe stable pPS, entailing PASI, PGA, BSA, and adverse effects, were amassed and scrutinized. The retrospective study flowchart is delineated in Figure 1.

Figure 1.

Retrospective study flowchart.

3.2. Patient demographics and presentation

The analysis encompassed patients aged 39 to 64 years, with a body mass index of 23–26 kg/m², disease duration spanning 6 to 36 months, predominantly Han Chinese, and a male-to-female ratio of 3:2. Initial BSA involvement ranged from 16% to 49%, with PASI scores of 15 to 42 (Fig. 2). Mean baseline PASI was 28.41 ± 7.3 (Fig. 3A and B). Concurrent therapies were evenly distributed across groups. No significant disparities existed in gender, age, ethnicity, body mass index, disease duration, % BSA, or PASI scores between the ET and SB cohorts at baseline. Detailed patient characteristics are summarized in Table 2.

Figure 2.

Biologic-naïve patients exhibiting moderate-to-severe plaque psoriasis prior to a 1-week treatment-free period. Body surface area (BSA) involvement (plaques affecting the upper body).

Figure 3.

Patient presenting with plaque psoriasis prior to a 1-week treatment-free interval. (A) Lower extremity. (B) Posterior aspect and scalp.

Table 2.

Demographical and clinical characteristics of patients before 1 week of treatment.

Characteristics	Cohorts		Comparisons
	ET	SB
Treatments	Ixekizumab	Secukinumab
Number of patients	70	55	P-value	Test value	df	95% CI
Gender
Male	42 (60)	35 (64)	.6782 (χ2 test)	0.172	1	0.6829–1.280
Female	28 (40)	20 (36)
Age (yr)	51.81 ± 6.99	51.22 ± 6.22	.6202 (unpaired t-test)	0.497	123	2.971 to 1.779
Body mass index (kg/m2)	25 (25.5–24)	25 (24.5–24)	.9424 (Mann–Whitney test)	1.910	NA	NA
Duration of diseases (mo)	19.67 ± 8.11	18.69 ± 9.28	.5301 (unpaired t-test)	0.63	123	4.063 to 2.102
Body surface area involvement (%)	33.07 ± 8.09	33.55 ± 9.06	.7584 (unpaired t-test)	0.308	123	2.569 to 3.517
Psoriasis Area and Severity Index score*	29.01 ± 7.33	27.64 ± 7.25	.2966 (unpaired t-test)	1.048	123	3.980 to 1.224

Categorical variables, continuous normally distributed variables, and continuous non-normally distributed variables are presented as frequencies (%), mean ± SD, and median (Q3–Q1), respectively.

The Soup calculator (Soup calculator LLB) facilitated quartile calculations.

Reported are test statistic values (χ² for χ² test; t-value for unpaired t-test; Mann–Whitney U statistic for Mann–Whitney test; relative risk for Fisher exact test).

CI = confidence interval (Katz approximation for categorical variables), df = degrees of freedom, NA = not applicable.

^*Scores ranged from 0 to 72, where 72 indicates maximum disease severity and 0 indicates absence of disease.

Statistical significance was defined as P < .05.

3.3. Treatment efficacy

At the study endpoint, 37% (n = 26) and 44% (n = 24) of the ET and SB cohorts, respectively, exhibited mild disease (PGA ≤ 1). The SB cohort demonstrated superior, albeit statistically equivalent, PGA scores compared with the ET cohort (1.1 [1.4–1] vs 1.1 [1.5–1], P = .204; Mann–Whitney U = 1670). PGA outcomes are illustrated in Figure 4.

Figure 4.

Physician’s global assessment (PGA) of patient treatment response, graded on a scale of 0 to 3: 0 to 0.4: disease remission; ≥0.5 and ≤1: mild disease activity; >1 and ≤2: moderate disease activity; >2 and <3: severe disease activity; and 3: most severe disease activity.

Both cohorts experienced reductions in % BSA and PASI scores posttreatment relative to baseline. The SB cohort exhibited more pronounced decreases in % BSA and PASI (Fig. 5A and B) relative to the ET cohort (Fig. 6A and B). However, PASI score improvements were statistically indistinguishable between cohorts. Comprehensive outcome data are presented in Table 3.

Figure 5.

Percentage of body surface area (BSA) affected in the SB cohort. (A) Pretreatment (one-week treatment-free). (B) Posttreatment.

Figure 6.

Percentage of body surface area (BSA) affected in the ET cohort posttreatment. (A) Posterior aspect. (B) Scalp.

Table 3.

Treatments outcome measures.

Characteristics	Cohorts										Comparisons between cohorts in AT conditions
	ET					SB
Treatments	Ixekizumab					Secukinumab
Level	BT	AT	Comparison between BT and AT			BT	AT	Comparison between BT and AT
Numbers of patients	70	70	P-value	Test value	df	55	55	P-value	Test value		p-value	Test value	df	CI
% BSA involvement	33.07 ± 8.09	19.79 ± 3.86	<.001 (ANOVA/Tukey test)	9.6	209	32 (41–27)	13 (17–12)	<.001 (Friedman test/Dunn test)	104	N/A	<0.001 (Kruskal-Wallis’ test/Dunn test)	120	N/A	N/A
*PASI score	29.01 ± 7.33	19.5 ± 5.68	<.001 (ANOVA/Tukey test)	8.4	209	27.64 ± 7.25	18.55 ± 6.85	<.001 (ANOVA/Tukey test	42.3	164	0.931 (Unpaired t-test)	0.0869	123	-2.316 to 2.121

Variables are presented as median (Q3–Q1).

Test statistic (Friedman statistic; Kruskal–Wallis’ statistic; F-statistic). All results were deemed significant at P < .05.

ANOVA = analysis of variance, AT = posttreatment, BSA = body surface area involvement, BT = before therapy, CI = confidence interval, df = degrees of freedom, N/A = not applicable, PASI = psoriasis area and severity index.

^*Scores range from 0 to 72, with 72 indicating maximum disease severity and 0 indicating absence of disease.

3.4. Adverse events

No mortality was observed during the trial or the ensuing year. Serious adverse events were infrequent, occurring in 2% to 3% of each cohort. Manifestations such as diabetes mellitus, injection site edema, systemic infections, integumentary and subcutaneous infections, as well as upper respiratory infections, were documented across both cohorts. Notably, incidences of diabetes mellitus and systemic (P = .009) and upper respiratory infections were disproportionately higher in the SB cohort. Diabetes, while an adverse event, may represent a comorbidity or medication-induced condition. Given the absence of substantiating evidence linking the medications to the onset of diabetes, it is not deemed a primary adverse effect. A granular breakdown of adverse events during the 12-week treatment phase and the subsequent year is available in Table 4.

Table 4.

Reported adverse effects during treatments and 1 year of follow-up.

Event	Cohorts		Comparisons
	ET	SB
Treatments	Ixekizumab	Secukinumab
Numbers of patients	70	55	P-value	Relative risk	% CI
Serious adverse effects but treatment completed	2 (3)	1 (2)	>.999	1.196	0.5290–2.704
Hypersensitivity	1 (1)	2 (4)	.582	0.5894	0.1180–2.943
Diabetes mellitus (in controlled condition)	13 (19)	17 (31)	.140	0.7222	0.4647–1.123
Pustular psoriasis	1 (1)	2 (4)	.582	0.5894	0.1180–2.943
Injection administration site swelling	6 (9)	7 (13)	.559	0.8077	0.4394–1.485
Pruritus	3 (4)	3 (5)	>.999	0.8881	0.3928–2.008
Urticaria	2 (3)	4 (7)	.404	0.5833	0.1861–1.829
Systemic organ infection(s)	9 (13)	18 (33)*	.009	0.5355	0.3073–0.9332
Upper respiratory tract infections	3 (4)	7 (13)	.196	0.5597	0.2141–1.463
Nausea	4 (6)	6 (11)	.333	0.697	0.3210–1.514
Vomiting	1 (1)	1 (2)	>.999	0.8913	0.2209–3.597
Skin rashes	1 (1)	2 (4)	.582	0.5894	0.1180–2.943

Variables are presented as frequencies with percentages enclosed in parentheses.

Adverse events, as determined by consultants, were coded according to the Medical Dictionary for Regulatory Activities (MedDRA version 25.0).

Fisher exact test was employed for statistical analysis.

A P-value of <.05 was the threshold for statistical significance.

Patients may exhibit 1 or more adverse effects.

^*Significantly elevated in comparison to the ET cohort.

CI = Confidence interval (Katz approximation).

In total, 12 adverse effects were analyzed, involving 70 subjects in the ET cohort and 55 in the SB cohort (Fig. 7). A Mantel–Haenszel random effects model revealed a statistically significant divergence between the cohorts, with a collective risk ratio of 0.52 (95% CI: 0.37–0.73). The P < .05 significance level underscores the disparity in overall adverse effect profiles. The absence of significant heterogeneity intimates that adverse effects manifested with similar intensity and direction across the cohorts. The funnel plot furnishes no indicia of reporting bias for adverse events (Fig. 8). Egger test similarly fails to support any funnel plot asymmetry (intercept: 0.11, 95% CI: −0.55 to 0.76, t: 0.315, P = .76). The Z-score plot (Fig. 9) suggests that more research is requisite to attain an optimal adverse event tally for conclusive findings. Synthesis tables, heterogeneity quantifications, and heterogeneity tests for adverse effects are documented in Tables 5.

Figure 7.

Forest plot depicting adverse events. Heterogeneity: Tau² = 0; Chi-squared value = 2.91, degree of freedom = 11 (P = .9918); I² = 0.0%. Overall effect: Z = −3.70 (P = .0002).

Figure 8.

Funnel plot analyzing adverse events.

Figure 9.

Z-score plot evaluating adverse events.

Table 5.

Summary table using random effects model with Mantel–Haenszel method for adverse effects.

S. No.	Adverse effect	ET cohort	SB cohort	Relative risk	95% CI	Weight	t-value	P-value
1	Serious adverse effects but treatment completed	2/70	1/55	1.571	0.146–16.883	2.14
2	Hypersensitivity	1/70	2/55	0.393	0.037–4.221	2.14
3	Diabetes mellitus (in controlled condition)	13/70	17/55	0.601	0.32–1.128	30.44
4	Pustular psoriasis	1/70	2/55	0.393	0.037–4.221	2.14
5	Injection administration site swelling	6/70	7/55	0.673	0.24–1.89	11.35
6	Pruritus	3/70	3/55	0.786	0.165–3.742	4.96
7	Urticaria	2/70	4/55	0.393	0.075–2.067	4.38
8	Systemic organ infection(s)	9/70	18/55	0.393	0.192–0.805	23.43
9	Upper respiratory tract infections	3/70	7/55	0.337	0.091–1.242	7.09
10	Nausea	4/70	6/55	0.524	0.155–1.765	8.18
11	Vomiting	1/70	1/55	0.786	0.05–12.281	1.6
12	Skin rashes	1/70	2/55	0.393	0.037–4.221	2.14
	Random effects model	46/840	70/660	0.519	0.367–0.735	99.99	−3.7	<.001

Heterogeneity was quantified as follows: τ² = 0 (95% CI = N/A–N/A), τ = 0 (95% CI = N/A–N/A), I² = 0 (95% CI = 0–0.583), and H = 1 (95% CI = 1–1.549). The test for heterogeneity yielded: Q = 2.91, df = 11, P = .99.

CI = Confidence interval, df = Degree of freedom, N/A = Not applicable.

Table 6 recapitulates the results of the assumption tests for the current study.

Table 6.

Results of assumptions test of the current study.

Variables		Assumption test adopted with reasons
Categorical variables
2 × 2 Table, 3 × 2 Table		Fisher exact test or Chi-square test. Chi-square test was performed if for sample size > 5 and total sample population ≥ 40. Chi-square test with Yate corrections was performed if any individual sample was <10. Yates correction is designed for continuity. The Fisher test is performed only if N ≤ 90. Otherwise, the Chi-square test with independence. Fisher was performed if there were fewer than 6 elements in the contingency table per group. The total sample size might be large, but if in 1 category there were not many elements, the correct test was Fisher test. (for example, in Table 1, the analysis for Ethnicity). For tables larger than 2 × 2, Fisher-Freeman-Halton tests was preferred.
Larger tables		Chi-square test for independence when all expected values are >1.0 and at least 20% of the expected values are >5 otherwise Chi-square test for trends.
Continuous variables
Demographical and clinical characteristics
Age (yr)		Both columns passed normality test (P-values > .1 for both). Therefore, the parametric test was selected. F = 1.264. The P-value is .186. This test suggests that the difference between the 2 standard deviations (SDs) is not significant. Therefore, an unpaired 2-tailed t-test was preferred.
Body mass index (kg/m2)		One column failed in the normality test (P-values were .238 and >.1). Therefore, the Mann–Whitney test was performed (2-tailed).
Duration of diseases (mo)		Both columns passed normality test (P-values > .1 for both). Therefore, the parametric test was selected. F = 1.308. The P-value is .1459. This test suggests that the difference between the 2 SDs is not significant. Therefore, an unpaired 2-tailed t-test was preferred.
% Body surface area involvement		Both columns passed normality test (P-values > .1 for both). Therefore, the parametric test was selected. F = 1.253. The P-value is .1872. This test suggests that the difference between the 2 SDs is not significant. Therefore, an unpaired 2-tailed t-test was preferred.
Psoriasis area and severity index score		Both columns passed normality test (P-values > .1 for both). Therefore, the parametric test was selected. F = 1.023. The P-value is .4687. This test suggests that the difference between the 2 SDs is not significant. Therefore, an unpaired 2-tailed t-test was preferred.
% Body surface area involvement	Within ET cohort	All columns passed normality test (P-values > .1 for all). Therefore, the Repeated Measures Analysis of Variance (ANOVA) was preferred.
	Within SB cohort	One column failed normality test (P-values were >.1, .0025, and >.1). Therefore, Friedman test was preferred.
	Between ET and SB cohorts	One column failed normality test (P-values were >.1, .0031, and >.1). Therefore, Kruskal–Wallis’ test was preferred.
Psoriasis area and severity index score	Within ET cohort	All columns passed normality test (P-values were >.1, .0732, and .0998. Therefore, the Repeated Measures of ANOVA was preferred.
	Within SB cohort	All columns passed normality test (P-values > .1 for all). Furthermore, Bartlett statistic (corrected) = 0.1877. The P-value is .9104. Bartlett test suggests that the differences among the SDs is not significant. Therefore, the Repeated Measures of ANOVA was preferred.
	Between ET and SB cohorts	Both columns passed normality test (P-values were .0732 and >.1). F = 1.457. The P-value is .0700. This test suggests that the difference between the 2 SDs is not quite significant. Therefore, an unpaired 2-tailed t-test was preferred.
Physician global assessment score		Both columns failed normality test (P-values were .0063 and .0277). Therefore, the Mann–Whitney test was performed (2-tailed).
Forrest plot and funnel plot		Setting: Random effect model, Method: Inverse, Summary measures: Standard mean difference, Between study heterogeneity estimator: DerSimonian-Laird

3.5. Clinical therapeutic efficacy in pPS

For the ET cohort, treatments proved optimally effective within a PASI range of 1 to 1.43. Beyond this threshold, especially amidst before therapy (BT) scenarios, the ET cohort risked undertreatment. Analogously, the SB cohort demonstrated efficacy within a PASI range of 1 to 2.1, with an undertreatment risk at scores exceeding this, notably during BT. A graphical portrayal of these therapeutic benefits is available in Figure 10. Detailed clinical benefit analyses for pPS treatments are presented in Table 7.

Figure 10.

Clinical treatment benefits for psoriasis.

Table 7.

The details of clinical benefits for treatments of psoriasis.

Parameters		Cohorts
		ET	SB
Treatments		Ixekizumab	Secukinumab
Total number of patients		70	55
Numbers of patients with ≤1 psoriasis area and severity index score (mild disease activity)		26	24
Numbers of patients with >1 psoriasis area and severity index score		44	31
Psoriasis area and severity index score	Risk of undertreatment	-	-
0	0	0.37	0.38
0.1	0.09	0.31	0.33
0.2	0.17	0.27	0.29
0.3	0.23	0.23	0.25
0.4	0.29	0.19	0.22
0.5	0.33	0.16	0.19
0.6	0.38	0.14	0.17
0.7	0.41	0.11	0.15
0.8	0.44	0.09	0.13
0.9	0.47	0.07	0.11
1	0.5	0.06	0.1
1.1	0.52	0.04	0.09
1.2	0.55	0.03	0.07
1.3	0.57	0.02	0.06
1.4	0.58	0.01	0.05
1.5	0.6	−0.01	0.04
1.6	0.62	−0.02	0.03
1.7	0.63	−0.02	0.03
1.8	0.64	−0.03	0.02
1.9	0.66	−0.04	0.01
2	0.67	−0.05	0.01
2.1	0.68	−0.05	0
2.2	0.69	−0.07	−0.06
2.3	0.7	−0.07	−0.01
2.4	0.71	−0.07	−0.02
2.5	0.71	−0.08	−0.02
2.6	0.72	−0.08	−0.03
2.7	0.73	−0.09	−0.03
2.8	0.74	−0.09	−0.03
2.9	0.74	−0.1	−0.04
3	0.75	−0.1	−0.04
Working area (psoriasis area and severity index score before treatment)		1–1.43	1–2.1
The risk of undertreatment (psoriasis area and severity index score before treatment)		>1.43	>2.1

Effect size: Achievement of ≤ 1 (mild disease activity; physician’s global assessment) at 12 weeks post-psoriasis treatment in ≥ 30% of patients.

4. Discussions

Within the SB cohort, a greater proportion of patients exhibited mild disease activity (≤1 on PGA of treatment response) compared with the ET cohort. Furthermore, the SB cohort demonstrated a reduced risk of undertreatment relative to the ET cohort. Future study designs should prioritize patients devoid of comorbidities and complicating factors such as pregnancy.^[14] These factors contribute to the enhanced efficacy of secukinumab in patients afflicted with moderate-to-severe pPS.^[31] Efficacy outcomes observed in the SB cohort are consistent with findings from a meta-analysis,^[15] a Phase 3b trial,^[23] and a retrospective study.^[25] The elevated proportion of males, the inclusion of older patients, and the substantial number experiencing high BSA involvement within the present study correlated with diminished ixekizumab efficacy in the Chinese population.^[14] Causality cannot be definitively established, only correlation. Biologic-naïve Chinese patients afflicted with moderate-to-severe pPS, particularly older males with extensive BSA involvement, are anticipated to exhibit a superior response to secukinumab compared with ixekizumab. Determining whether subgroup analyses were executed based on clinical characteristics such as gender, age, and BSA involvement is crucial. In the absence of these subgroup analyses and their subsequent reporting within the study, the conclusion excessively relies on external literature,^[14] which may lack direct applicability to our specific patient cohort.

Within the SB cohort, a greater incidence of adverse events, encompassing systemic organ infection, was observed with secukinumab compared with the ET cohort. The advanced age of a significant patient subset within the SB cohort contributes substantially to this elevated adverse event rate.^[31] Furthermore, the study’s inclusion of the treatment duration plus a 1-year follow-up period also accounts for the higher incidence of adverse effects. Secukinumab is also associated with a comparatively elevated risk of hepatic impairment relative to other available biologic therapies.^[31] Biologic-naïve Chinese patients exhibiting moderate-to-severe peripheral psoriatic arthritis are anticipated to experience a higher frequency of adverse events with secukinumab than with ixekizumab.

Ixekizumab demonstrates notable efficacy coupled with a tolerable adverse effect profile. Its effectiveness mirrors both observational studies^[20,26,27] and expert consensus.^[15] In biologic-naïve Chinese patients afflicted with moderate-to-severe pPS, ixekizumab proves efficacious and exhibits a favorable tolerability profile.

Patients with prior ixekizumab exposure were excluded from secukinumab treatment cohorts. However, biologic-naïve and switching patients demonstrated comparable aggregate outcomes.^[17] Ixekizumab’s superior adherence generally yields greater therapeutic efficacy in psoriasis patients compared with secukinumab.^[18,28] Consequently, to mitigate bias, biologic-naïve patients were preferentially selected.

Acknowledged limitations encompass the retrospective nature of single-center electronic medical record data, absent multicenter trials, although rigorous trial inclusion criteria may inherently limit generalizability, whereas clinical practice data offer valuable insights into real-world effectiveness and adverse events. Furthermore, independent endpoint evaluation, including both response and adverse effects, was not performed. Sample size calculation, based on physician’s global assessment of treatment response rather than severe adverse event incidence as employed in prior studies,^[4] and the exclusion of participants discontinuing treatment due to significant adverse effects, constitute additional constraints. Consequently, the PGA served as the basis for sample size determination. Notwithstanding these modest constraints, this study furnishes salient insights, thereby establishing a foundation for subsequent multicenter investigations. The present retrospective study’s sample size precludes definitive validation of the reported findings.

5. Conclusions

Ixekizumab demonstrates efficacy and tolerability in biologic-naïve Chinese patients afflicted with moderate-to-severe pPS. Secukinumab may elicit a heightened response alongside an elevated incidence of adverse events compared with ixekizumab within this cohort. This study furnishes pertinent real-world perspectives on IL-17 inhibitor utilization within a Chinese demographic. Nevertheless, assertions regarding secukinumab’s superior efficacy warrant further substantiation, and the adverse event analysis necessitates enhanced contextualization. With refined methodology and statistical rigor in subsequent investigations, these research findings could substantially augment the existing body of knowledge.

Acknowledgments

The authors are thankful to the medical and nonmedical staff of the Beijing Shijitan Hospital, Capital Medical University, Beijing, China. All the data and related metadata underlying the reported findings already provided as part of the submitted article. There are no supplementary files (supplementary tables, supplementary figures, and others) referred to in the manuscript. Therefore, there are nothing to deposit in appropriate public data repositories.

Author contributions

Conceptualization: Dayu Zhao.

Data curation: Dayu Zhao.

Formal analysis: Dayu Zhao.

Investigation: Dayu Zhao.

Methodology: Dayu Zhao.

Project administration: Dayu Zhao.

Resources: Dayu Zhao.

Software: Dayu Zhao.

Supervision: Dayu Zhao.

Validation: Dayu Zhao.

Visualization: Dayu Zhao.

Writing – original draft: Dayu Zhao.

Writing – review & editing: Dayu Zhao.