Phenotype-dependent effects of IL-17A inhibitors on lipid profiles in moderate-to-severe plaque psoriasis: a retrospective cohort study

Abstract

Background

Psoriasis is associated with elevated cardiovascular risk, partly mediated through dyslipidemia and the interleukin-23/Th17 inflammatory axis. Interleukin-17A (IL-17A) inhibitors demonstrate superior dermatologic efficacy, but their effects on lipid metabolism remain controversial. Most previous studies analyzed lipid changes at the population level without stratifying by baseline lipid phenotypes. This study aimed to evaluate the phenotype-specific effects of IL-17A inhibitors on lipid profiles and identify risk factors for dyslipidemia in moderate-to-severe plaque psoriasis.

Methods

This retrospective cohort study included 353 patients with moderate-to-severe plaque psoriasis (body surface area [BSA] ≥ 3%) treated with IL-17A inhibitors (secukinumab or ixekizumab) for one year. Patients were stratified into five mutually exclusive lipid phenotypes according to the 2023 Chinese guidelines for lipid management. Multivariable logistic regression identified independent risk factors for dyslipidemia. Lipid changes were assessed using paired tests with Benjamini-Hochberg false discovery rate (FDR) correction. Sensitivity analyses were performed in the BSA ≥ 10% subgroup (n = 274).

Results

Dyslipidemia was present in 72.2% of patients, with mixed dyslipidemia (26.1%) and hypertriglyceridemia (21.2%) being predominant. BMI ≥ 28 kg/m² (OR = 3.35, 95% CI 1.79–6.28, P < 0.001) and male sex (OR = 2.37, 95% CI 1.40–4.01, P = 0.001) were independent risk factors. Phenotype-specific effects were observed: mixed dyslipidemia patients showed significant reductions in TC (0.47 mmol/L), TG (0.46 mmol/L), and LDL-C (0.18 mmol/L) (all FDR-adjusted P ≤ 0.010); hypercholesterolemia patients demonstrated reductions in TC (0.49 mmol/L) and LDL-C (0.43 mmol/L) (both FDR-adjusted P < 0.001). Among 110 patients with baseline LDL-C ≥ 3.4 mmol/L, 31.0% achieved LDL-C < 3.4 mmol/L. Hypertriglyceridemia patients showed no significant changes after FDR adjustment. Isolated low HDL-C patients exhibited modest increases in HDL-C and LDL-C. Normal lipid patients experienced mild increases in TG and LDL-C within normal ranges. Psoriasis Area and Severity Index (PASI) improvement showed no correlation with lipid changes (all FDR-adjusted P > 0.05). Sensitivity analyses confirmed the robustness of all primary findings.

Conclusions

IL-17A inhibitors exert heterogeneous, phenotype-dependent effects on lipid metabolism. Patients with elevated baseline cholesterol may derive dual dermatologic and cardiometabolic benefits, while those with normal lipids require monitoring for modest unfavorable changes. These findings support phenotype-guided treatment selection and individualized lipid monitoring strategies.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12944-026-02974-7.

Background

Psoriasis is a chronic immune-mediated inflammatory disease affecting approximately 2–3% of the global population, characterized by systemic inflammation extending beyond cutaneous manifestations [1]. Accumulating evidence demonstrates that patients with moderate-to-severe psoriasis face elevated cardiovascular risk, attributed partly to increased prevalence of traditional risk factors, including dyslipidemia, obesity, and metabolic syndrome [2]. Moreover, chronic systemic inflammation in psoriasis may independently contribute to accelerated atherosclerosis and adverse cardiovascular outcomes [3, 4].

The pathophysiological link between psoriasis and cardiometabolic comorbidities is increasingly attributed to the interleukin-23 (IL-23)/Th17 inflammatory axis. A comprehensive review by Egeberg et al. [5] elucidated how interleukin-17 (IL-17) and IL-23 drive systemic inflammation that extends beyond cutaneous disease, contributing to endothelial dysfunction, atherosclerosis, and metabolic syndrome. These cytokines promote vascular inflammation through recruitment of neutrophils, upregulation of adhesion molecules, and modulation of adipose tissue metabolism, thereby establishing a mechanistic bridge between psoriatic inflammation and cardiovascular risk.

Interleukin-17 A (IL-17A) plays a pivotal role in psoriasis pathogenesis, and IL-17A inhibitors have demonstrated superior efficacy in achieving skin clearance compared with earlier biologics [6]. Beyond their dermatologic benefits, emerging data suggest that IL-17A inhibitors may influence systemic inflammation and metabolic parameters [7–9]. However, their effects on lipid metabolism remain controversial, with studies reporting conflicting results ranging from favorable lipid reductions to unfavorable increases in atherogenic lipoproteins [8, 10–14].

Most previous studies have analyzed lipid changes at the population level without stratifying by baseline lipid phenotypes. Given the marked heterogeneity in lipid profiles among psoriasis patients, phenotype-specific analysis is essential to elucidate differential lipid changes and guide clinical decision-making. Furthermore, whether lipid-modulating effects of IL-17A inhibitors are mediated through reduction of systemic inflammation or direct metabolic regulation remains unclear.

This study aimed to: (1) determine the prevalence and risk factors of dyslipidemia in patients with moderate-to-severe plaque psoriasis; (2) evaluate the effects of one-year IL-17A inhibitor treatment on lipid parameters across different baseline lipid phenotypes; and (3) assess whether lipid changes correlate with dermatologic response.

Methods

Study design and setting

This retrospective cohort study was conducted at Peking Union Medical College Hospital in China. We reviewed electronic medical records of patients with plaque psoriasis treated with IL-17A inhibitors between March 2021 and March 2025. The study adhered to the principles of the Declaration of Helsinki and received institutional ethics committee approval (I-25PJ2973). Given the retrospective nature and use of de-identified data, the requirement for informed consent was waived.

Study population

Inclusion Criteria: Patients were eligible if they met all of the following criteria: (1) age ≥ 18 years at baseline; (2) clinical diagnosis of moderate-to-severe plaque psoriasis, defined as baseline body surface area (BSA) involvement ≥ 3% per the guideline for the diagnosis and treatment of psoriasis in China (2023 edition) [15]; (3) treatment with IL-17A inhibitors for at least one year; (4) complete baseline measurements of all four lipid parameters: total cholesterol (TC), triglycerides (TG), low-density lipoprotein cholesterol (LDL-C), and high-density lipoprotein cholesterol (HDL-C).

Exclusion Criteria: Patients who received lipid-lowering medications (statins, fibrates, or other hypolipidemic agents) during the one-year treatment period were excluded to avoid confounding effects.

Data collection

Data were extracted from the electronic medical record system using a standardized case report form. The following variables were collected: (1) Demographics and disease severity: age, sex, height, weight, disease duration, baseline Psoriasis Area and Severity Index (PASI), and BSA involvement. (2) Comorbidities and behavioral factors: hypertension, diabetes, current smoking status, and alcohol consumption. (3) Laboratory parameters at baseline and one-year follow-up: TC, TG, LDL-C, HDL-C, serum uric acid, and C-reactive protein (CRP). (4) Treatment response: PASI improvement rate, calculated as [(baseline PASI - one-year PASI) / baseline PASI] × 100%.

Definitions

Dyslipidemia was defined according to 2023 Chinese guidelines for lipid management [16]: TG ≥ 1.7 mmol/L, TC ≥ 5.2 mmol/L, or HDL-C < 1.0 mmol/L. Patients were stratified into mutually exclusive subtypes: (1) mixed dyslipidemia: TC ≥ 5.2 mmol/L and TG ≥ 1.7 mmol/L; (2) Hypertriglyceridemia: TG ≥ 1.7 mmol/L and TC < 5.2 mmol/L; (3) Hypercholesterolemia: TC ≥ 5.2 mmol/L and TG < 1.7 mmol/L; (4) Isolated low HDL-C: HDL-C < 1.0 mmol/L with TC < 5.2 mmol/L and TG < 1.7 mmol/L; (5) normal lipids: TC < 5.2 mmol/L, TG < 1.7 mmol/L, and HDL-C ≥ 1.0 mmol/L. Lipid parameter changes were calculated as baseline minus follow-up values. Positive values indicate reductions (favorable for TC, TG, and LDL-C), while negative values indicate increases (favorable for HDL-C). Body mass index (BMI) was classified using Chinese obesity criteria (overweight 24–27.9 kg/m², obese ≥ 28 kg/m²) [17].

Secukinumab was administered by subcutaneous injection using weight-based dosing: 150 mg for patients weighing < 60 kg and 300 mg for patients weighing ≥ 60 kg, at weeks 0, 1, 2, 3, and 4, followed by the same dose every 4 weeks. Ixekizumab was administered at 160 mg at week 0, followed by 80 mg at weeks 2, 4, 6, 8, 10, and 12, then 80 mg every 4 weeks. Concomitant topical therapies (topical corticosteroids, vitamin D analogues) were used as adjunctive treatment per routine clinical practice. No standardized dietary counseling was documented.

Outcomes

Primary outcomes: Changes in lipid parameters (TC, TG, LDL-C, and HDL-C) after one-year treatment with IL-17A inhibitors, stratified by baseline lipid phenotypes.

Secondary outcomes: (1) Independent risk factors for dyslipidemia; (2) lipid changes in patients with baseline LDL-C ≥ 3.4 mmol/L, representing a cardiovascular high-risk subgroup; (3) correlation between PASI improvement rate and lipid parameter changes.

Statistical analysis

Continuous variables were assessed for normality using Shapiro-Wilk test and presented as mean ± standard deviation (SD) or median (25th, 75th percentile). Categorical variables were reported as frequencies and percentages. Baseline characteristics between dyslipidemia and normal lipid groups were compared using independent t-test or Mann-Whitney U test for continuous variables, and chi-square or Fisher’s exact test for categorical variables, as appropriate.

Multivariable logistic regression was performed to identify independent predictors of dyslipidemia. Candidate predictors included age (≥ 45 vs. <45 years), sex, BMI (≥ 28 vs. <28 kg/m²), disease duration, baseline PASI score, hypertension, hyperuricemia, diabetes, current smoking, and alcohol consumption. Variables with P ≤ 0.10 in univariable analysis or clinical relevance were included in multivariable analysis using backward elimination method. Model performance was assessed using area under the receiver operating characteristic curve (AUC) and Hosmer-Lemeshow test. Multicollinearity was evaluated using variance inflation factors (VIF < 5). Sensitivity analyses compared four variable selection methods (enter, forward, backward, and stepwise) to assess model robustness.

Within each lipid subgroup, paired t-test or Wilcoxon signed-rank test were used for baseline-to-follow-up comparisons, as appropriate. Spearman correlation assessed relationships between PASI improvement and lipid changes, given the non-normal distribution of PASI improvement. To account for multiple testing, the Benjamini-Hochberg false discovery rate (FDR) procedure was applied, controlling FDR at 5%. As a sensitivity analysis, all primary analyses were repeated in the subgroup of patients with baseline BSA ≥ 10%.

Patients with missing follow-up values were excluded from respective analyses (complete case analysis). Extreme outliers were retained given small proportions and the use of robust statistical methods. All analyses were performed using R software (version 4.5.1). Two-tailed P < 0.05 was considered statistically significant.

Results

Study population and dyslipidemia prevalence

Of 395 patients initiating IL-17A inhibitor therapy, 353 completed one-year follow-up and met inclusion criteria (Fig. 1). Dyslipidemia was highly prevalent, affecting 72.2% (n = 255) of the cohort. Mixed dyslipidemia (26.1%, n = 92) and hypertriglyceridemia (21.2%, n = 75) were the predominant subtypes, followed by hypercholesterolemia (13.6%) and isolated low HDL-C (11.3%). Only 27.8% (n = 98) had normal lipid profiles at baseline.

Fig. 1.

Patient flow diagram for the study. Abbreviations: HDL-C, high-density lipoprotein cholesterol; IL-17A, interleukin-17A

Baseline characteristics

The baseline characteristics of the cohort are presented in Table 1. Patients were predominantly male (75.6%) with a mean age of 42 ± 12 years and long disease duration (14.7 ± 10.0 years). Baseline psoriasis severity was substantial, with mean PASI score of 18.3 ± 11.4 and BSA involvement of 22.2 ± 19.5%. Nail involvement was present in 186 patients (58.5%). Most patients (89.0%) received secukinumab, with 11.0% receiving ixekizumab.

Table 1.

Baseline characteristics stratified by dyslipidemia status

Characteristic	Overall (N = 353)	Dyslipidemia (N = 255)	Normal lipid (N = 98)	P value
Demographics
Age, years	42 ± 12	42 ± 12	40 ± 13	0.162
Male sex	267 (75.6%)	207 (81.2%)	60 (61.2%)	< 0.001
Weight, kg	78.4 ± 15.0	80.5 ± 14.2	72.9 ± 15.6	< 0.001
BMI, kg/m²	26.5 ± 4.0	27.0 ± 3.9	25.0 ± 3.8	< 0.001
Disease Characteristics
Disease duration, years	14.7 ± 10.0	14.2 ± 9.5	16.0 ± 11.0	0.144
Baseline PASI score	18.3 ± 11.4	18.6 ± 11.7	17.5 ± 10.3	0.413
Baseline BSA, %	22.2 ± 19.5	22.5 ± 19.9	21.2 ± 18.4	0.572
Nail involvement	186 (58.5%)	133 (58.3%)	53 (58.9%)	1.000
Family history	118 (33.4%)	81 (31.8%)	37 (37.8%)	0.346
Past systemic therapy	95 (26.9%)	66 (25.9%)	29 (29.6%)	0.569
Past biologic therapy	63 (17.8%)	42 (16.5%)	21 (21.4%)	0.350
Comorbidities
Hypertension	74 (21.0%)	54 (21.2%)	20 (20.4%)	0.990
Diabetes mellitus	69 (19.5%)	57 (22.4%)	12 (12.2%)	0.046
Hyperuricemia	196 (55.5%)	154 (60.4%)	42 (42.9%)	0.004
Current smoking	123 (34.8%)	100 (39.2%)	23 (23.5%)	0.008
Alcohol consumption	44 (12.5%)	35 (13.7%)	9 (9.2%)	0.329
Baseline Laboratory Parameters
TC, mmol/L	4.98 ± 1.06	5.23 ± 1.11	4.35 ± 0.50	< 0.001
TG, mmol/L	1.61 (0.99, 2.54)	2.06 (1.32, 2.95)	0.89 (0.69, 1.10)	< 0.001
HDL-C, mmol/L	1.13 ± 0.26	1.07 ± 0.24	1.29 ± 0.23	< 0.001
LDL-C, mmol/L	3.06 ± 0.88	3.23 ± 0.92	2.61 ± 0.53	< 0.001
Uric acid, µmol/L	409 ± 105	422 ± 103	374 ± 100	< 0.001
CRP, mg/L	1.56 (0.70, 3.33)	1.76 (0.79, 3.37)	1.22 (0.51, 3.19)	0.064
Treatment and response
Secukinumab	314 (89.0%)	223 (87.5%)	91 (92.9%)	0.207
PASI improvement, %	92 (83, 100)	92 (83, 100)	93 (83, 100)	0.741

Data presented as mean ± SD, median (25th, 75th percentile), or n (%)

Abbreviations: BMI body mass index, BSA body surface area, CRP C-reactive protein, HDL-C high-density lipoprotein cholesterol, LDL-C low-density lipoprotein cholesterol, PASI Psoriasis Area and Severity Index, TC total cholesterol, TG triglycerides

Compared to patients with normal lipids, those with dyslipidemia exhibited a distinct metabolic risk profile (Table 1). Notable differences included higher male predominance (81.2% vs. 61.2%, P < 0.001), elevated BMI (27.0 vs. 25.0 kg/m², P < 0.001), and increased serum uric acid (422 vs. 374 µmol/L, P < 0.001). The dyslipidemia group also showed higher prevalence of hyperuricemia (60.4% vs. 42.9%, P = 0.004), diabetes (22.4% vs. 12.2%, P = 0.046), and current smoking (39.2% vs. 23.5%, P = 0.008). Disease severity and treatment response were comparable between groups, with both achieving approximately 92% PASI improvement (P = 0.741). Twenty-two patients (6.2%) who achieved complete skin clearance at one year had their dosing intervals extended to every 8 weeks, while maintaining the same dose per injection.

Independent risk factors for dyslipidemia

Univariable analysis identified five factors significantly associated with dyslipidemia: male sex, BMI ≥ 28 kg/m², diabetes, hyperuricemia, and current smoking (Table 2). After adjustment for potential confounders in multivariable analysis, only two factors remained independently associated with dyslipidemia: BMI ≥ 28 kg/m² (adjusted OR = 3.35, 95% CI 1.79–6.28, P < 0.001) and male sex (adjusted OR = 2.37, 95% CI 1.40–4.01, P = 0.001) (Table 2). The model demonstrated good discrimination (AUC = 0.672) and calibration (Hosmer-Lemeshow P = 1.000), with findings validated across alternative variable selection methods (Supplementary Table 1).

Table 2.

Logistic regression analysis of risk factors for dyslipidemia

Variable	Univariable Analysis		Multivariable Analysis‡
	OR (95% CI)	P Value	OR (95% CI)	P Value
Age ≥ 45 years	1.16 (0.70–1.91)	0.562
Male sex	2.73 (1.63–4.56)	< 0.001	2.37 (1.40–4.01)	0.001
BMI ≥ 28 kg/m²	3.75 (2.02–6.96)	< 0.001	3.35 (1.79–6.28)	< 0.001
Disease duration, per year	0.98 (0.96–1.01)	0.145
Baseline PASI score, per point	1.01 (0.99–1.03)	0.412
Hypertension	1.05 (0.59–1.86)	0.874
Hyperuricemia	2.03 (1.27–3.26)	0.003
Diabetes	2.06 (1.05–4.04)	0.035
Current smoking	2.10 (1.24–3.58)	0.006
Alcohol consumption	1.57 (0.73–3.41)	0.250

Abbreviations: AIC Akaike Information Criterion, AUC area under the receiver operating characteristic curve, BMI body mass index, CI confidence interval, OR odds ratio, PASI Psoriasis Area and Severity Index, VIF variance inflation factor

‡Backward elimination selection based on AIC (N = 353). Model performance: AIC = 391.89, AUC = 0.672, Hosmer-Lemeshow P = 1.00. All VIF < 5 (no multicollinearity)

Phenotype-specific effects on lipid parameters

Mixed dyslipidemia (n=92)

Patients with mixed dyslipidemia experienced significant reductions across multiple atherogenic parameters (Table 3). TC decreased by 0.47 mmol/L (95% CI 0.31–0.63, FDR-adjusted P < 0.001), TG by 0.46 mmol/L (95% CI 0.23–0.78, FDR-adjusted P = 0.001), and LDL-C by 0.18 mmol/L (95% CI 0.05–0.32, FDR-adjusted P = 0.010), while HDL-C remained stable.

Table 3.

Lipid parameter changes in mixed dyslipidemia subgroup (n = 92)

Parameter	Baseline	One-Year Follow-up	Change	95% CI	P Value	FDR-Adjusted P
TC, mmol/L (n = 85)	6.03 ± 0.78	5.56 ± 0.82	0.47 ± 0.73	0.31 to 0.63	< 0.001	< 0.001
TG, mmol/L† (n = 85)	2.92 (2.18, 3.99)	2.51 (1.77, 3.60)	0.46 (-0.17, 1.14)	0.23 to 0.78	< 0.001	0.001
LDL-C, mmol/L (n = 84)	3.63 ± 0.79	3.45 ± 0.76	0.18 ± 0.62	0.05 to 0.32	0.007	0.010
HDL-C, mmol/L (n = 84)	1.09 ± 0.19	1.06 ± 0.22	0.03 ± 0.15	-0.01 to 0.06	0.100	0.100

Data presented as mean ± SD or median (25th, 75th percentile). Change calculated as baseline minus follow-up. Mixed dyslipidemia defined as TC ≥ 5.2 mmol/L and TG ≥ 1.7 mmol/L 

Abbreviations: CI confidence interval, FDR false discovery rate, HDL-C high-density lipoprotein cholesterol, LDL-C low-density lipoprotein cholesterol, TC total cholesterol, TG triglycerides

†Three extreme outliers retained. Maximum 8 patients excluded due to missing values

Hypertriglyceridemia (n = 75)

TG showed a modest reduction of 0.27 mmol/L (95% CI 0.00-0.53), which did not maintain statistical significance after FDR adjustment (P = 0.123). Other lipid parameters remained unchanged (Table 4).

Table 4.

Lipid parameter changes in hypertriglyceridemia subgroup (n = 75)

Parameter	Baseline	One-Year Follow-up	Change	95% CI	P value	FDR-Adjusted P
TC, mmol/L (n = 73)	4.46 ± 0.61	4.53 ± 0.78	−0.06 ± 0.65	−0.21 to 0.09	0.427	0.427
TG, mmol/L (n = 73)	2.67 ± 1.03	2.40 ± 1.26	0.27 ± 1.13	0.00 to 0.53	0.047	0.123
LDL-C, mmol/L (n = 73)	2.57 ± 0.55	2.67 ± 0.66	−0.10 ± 0.52	−0.22 to 0.02	0.092	0.123
HDL-C, mmol/L (n = 73)	0.99 ± 0.19	1.02 ± 0.19	−0.03 ± 0.13	−0.06 to 0.00	0.062	0.123

Data presented as mean ± SD. Change calculated as baseline minus follow-up. Hypertriglyceridemia defined as TG ≥ 1.7 mmol/L and TC < 5.2 mmol/L. Maximum 2 patients excluded due to missing values

Abbreviations as in Table 3

Hypercholesterolemia (n = 48)

Patients with hypercholesterolemia demonstrated robust improvements in cholesterol parameters (Table 5). TC declined by 0.49 mmol/L (95% CI 0.31–0.68, FDR-adjusted P < 0.001) and LDL-C by 0.43 mmol/L (95% CI 0.26–0.61, FDR-adjusted P < 0.001). However, HDL-C decreased by 0.06 mmol/L (FDR-adjusted P = 0.013) and TG showed a small increase (median change − 0.09 mmol/L, FDR-adjusted P = 0.029).

Table 5.

Lipid parameter changes in hypercholesterolemia subgroup (n = 48)

Parameter	Baseline	One-Year Follow-up	Change	95% CI	P value	FDR-Adjusted P
TC, mmol/L (n = 44)	5.89 ± 0.93	5.40 ± 0.98	0.49 ± 0.60	0.31 to 0.68	< 0.001	< 0.001
TG, mmol/L† (n = 44)	1.23 (1.01, 1.34)	1.23 (0.93, 1.64)	−0.09 (− 0.44, 0.07)	−0.38 to − 0.02	0.029	0.029
LDL-C, mmol/L (n = 44)	3.97 ± 0.91	3.53 ± 0.95	0.43 ± 0.57	0.26 to 0.61	< 0.001	< 0.001
HDL-C, mmol/L (n = 44)	1.29 ± 0.30	1.23 ± 0.34	0.06 ± 0.15	0.02 to 0.11	0.010	0.013

Data presented as mean ± SD or median (25th, 75th percentile). Change calculated as baseline minus follow-up. Hypercholesterolemia defined as TC ≥ 5.2 mmol/L and TG < 1.7 mmol/L. Maximum 4 patients excluded due to missing values

Abbreviations as in Table 3

†Two extreme outliers retained

Isolated low HDL-C (n = 40)

This subgroup demonstrated increases across all lipid parameters (Table 6), with only HDL-C (-0.05 mmol/L, FDR-adjusted P = 0.026) and LDL-C (-0.21 mmol/L, FDR-adjusted P = 0.041) reaching statistical significance.

Table 6.

Lipid parameter changes in isolated low HDL-C subgroup (n = 40)

Parameter	Baseline	One-Year Follow-up	Change	95% CI	P value	FDR-Adjusted P
TC, mmol/L (n = 35)	4.03 ± 0.55	4.22 ± 0.69	−0.19 ± 0.56	−0.38 to 0.00	0.048	0.053
TG, mmol/L (n = 35)	1.11 ± 0.30	1.26 ± 0.57	−0.15 ± 0.46	−0.31 to 0.00	0.053	0.053
LDL-C, mmol/L (n = 34)	2.65 ± 0.57	2.86 ± 0.69	−0.21 ± 0.50	−0.39 to − 0.03	0.020	0.041
HDL-C, mmol/L (n = 34)	0.91 ± 0.08	0.96 ± 0.10	−0.05 ± 0.10	−0.09 to − 0.02	0.006	0.026

Data presented as mean ± SD. Change calculated as baseline minus follow-up. Isolated low HDL-C defined as HDL-C < 1.0 mmol/L with TC < 5.2 mmol/L and TG < 1.7 mmol/L. Maximum 6 patients excluded due to missing values

Abbreviations as in Table 3

Normal lipid profile (n = 98)

Patients with baseline normal lipids experienced mild elevations in TG (-0.19 mmol/L, FDR-adjusted P = 0.005) and LDL-C (-0.12 mmol/L, FDR-adjusted P = 0.038), though values remained within normal ranges (Table 7).

Table 7.

Lipid parameter changes in normal lipid profile subgroup (n = 98)

Parameter	Baseline	One-Year Follow-up	Change	95% CI	P value	FDR-Adjusted P
TC, mmol/L (n = 90)	4.37 ± 0.48	4.48 ± 0.66	−0.11 ± 0.59	−0.23 to 0.02	0.088	0.117
TG, mmol/L (n = 90)	0.95 ± 0.32	1.14 ± 0.62	−0.19 ± 0.54	−0.30 to − 0.08	0.001	0.005
LDL-C, mmol/L (n = 89)	2.64 ± 0.51	2.76 ± 0.59	−0.12 ± 0.48	−0.22 to − 0.02	0.019	0.038
HDL-C, mmol/L† (n = 89)	1.26 (1.13, 1.38)	1.22 (1.10, 1.37)	0.00 (− 0.05, 0.12)	−0.01 to 0.07	0.162	0.162

Data presented as mean ± SD or median (25th, 75th percentile). Change calculated as baseline minus follow-up. Normal lipid profile defined as TC < 5.2 mmol/L and TG < 1.7 mmol/L and HDL-C ≥ 1.0 mmol/L. . Maximum 9 patients excluded due to missing values

Abbreviations as in Table 3

†Two extreme outliers retained

Cardiovascular high-risk patients

Among 110 patients (31.2%) with baseline LDL-C ≥ 3.4 mmol/L—a threshold indicating elevated cardiovascular risk—IL-17A inhibitors demonstrated lipid-lowering effects (Table 8). TC decreased by 0.48 mmol/L (95% CI 0.33–0.62, FDR-adjusted P < 0.001), LDL-C by 0.34 mmol/L (95% CI 0.22–0.46, FDR-adjusted P < 0.001); however, HDL-C also decreased by 0.05 mmol/L (95% CI 0.02–0.08, FDR-adjusted P = 0.004). Approximately 90% of these high-risk patients had mixed dyslipidemia or hypercholesterolemia phenotypes. Notably, 31.0% achieved the LDL-C target of < 3.4 mmol/L with IL-17A inhibitor monotherapy.

Table 8.

Lipid parameter changes in patients with baseline LDL-C ≥ 3.4 mmol/L (n = 110)

Parameter	Baseline	One-Year Follow-up	Change	95% CI	P Value	FDR-Adjusted P
TC, mmol/L (n = 100)	6.03 ± 0.89	5.55 ± 0.89	0.48 ± 0.73	0.33 to 0.62	< 0.001	< 0.001
TG, mmol/L† (n = 100)	1.94 (1.26, 2.66)	1.81 (1.30, 2.55)	0.00 (-0.37, 0.58)	-0.37 to 0.58	0.474	0.474
LDL-C, mmol/L (n = 100)	4.03 ± 0.69	3.70 ± 0.77	0.34 ± 0.60	0.22 to 0.46	< 0.001	< 0.001
HDL-C, mmol/L (n = 100)	1.15 ± 0.20	1.10 ± 0.23	0.05 ± 0.15	0.02 to 0.08	0.003	0.004

Data presented as mean ± SD or median (25th, 75th percentile). Change calculated as baseline minus follow-up. †Two extreme outliers retained. Maximum 10 patients excluded due to missing values

Abbreviations as in Table 3

Correlation between PASI improvement and lipid changes

Spearman correlation analyses revealed no significant associations between PASI improvement rate and changes in any lipid parameter across all dyslipidemia subgroups after FDR adjustment (all P > 0.05) (Supplementary Table 2). This finding suggests that lipid-modulating effects occur independently of dermatologic response.

Sensitivity analysis: baseline BSA ≥ 10% subgroup

To validate the robustness of our findings, we performed sensitivity analyses restricted to the baseline BSA ≥ 10% subgroup (n = 274, 77.6% of the full cohort). Dyslipidemia prevalence was 73.4% in this subgroup, comparable to the full cohort (72.2%). Phenotype-specific lipid changes were consistent in direction and significance with the primary analyses: mixed dyslipidemia patients showed significant reductions in TC (0.48 mmol/L, FDR-adjusted P < 0.001) and LDL-C (0.19 mmol/L, FDR-adjusted P = 0.027); hypercholesterolemia patients demonstrated robust improvements in TC (0.45 mmol/L, FDR-adjusted P < 0.001) and LDL-C (0.40 mmol/L, FDR-adjusted P < 0.001). Among 84 patients with baseline LDL-C ≥ 3.4 mmol/L in this subgroup, 30.3% achieved LDL-C < 3.4 mmol/L, consistent with the full cohort finding (31.0%). In the multivariable logistic regression, both BMI ≥ 28 kg/m² (OR = 4.81, 95% CI 2.17–10.64, P < 0.001) and male sex (OR = 2.31, 95% CI 1.25–4.25, P = 0.007) remained independently associated with dyslipidemia (Supplementary Tables 3–5).

Discussion

Principal findings

This retrospective cohort study demonstrates that IL-17A inhibitors exert phenotype-specific effects on lipid metabolism in patients with moderate-to-severe plaque psoriasis. Several key findings merit emphasis. First, dyslipidemia was highly prevalent (72.2%) in this population, with elevated BMI (≥ 28 kg/m²) and male sex identified as independent risk factors. Second, lipid changes varied substantially by baseline phenotype: patients with mixed dyslipidemia or hypercholesterolemia experienced significant reductions in atherogenic lipids, while those with normal lipids exhibited mild unfavorable changes. Third, among cardiovascular high-risk patients with LDL-C ≥ 3.4 mmol/L, IL-17A inhibitors achieved clinically meaningful LDL-C reduction (0.34 mmol/L), with 31% reaching target levels without additional lipid-lowering therapy. Finally, the lipid-modulating effects appeared independent of PASI improvement, suggesting direct metabolic mechanisms beyond systemic inflammation reduction.

Interpretation in context of existing literature

The prevalence of dyslipidemia in our cohort aligns with previous studies reporting high rates of metabolic abnormalities in moderate-to-severe psoriasis [18]. Our identification of BMI ≥ 28 kg/m² as the strongest predictor (OR = 3.35) is consistent with established associations between obesity and dyslipidemia, likely mediated through adipose tissue inflammation and insulin resistance [19], and further reflects the well-established bidirectional relationship between obesity and psoriasis [20]: obesity promotes systemic inflammation through adipokine dysregulation—including increased leptin and decreased adiponectin—which exacerbates psoriasis severity, while psoriasis-related inflammation and behavioral factors contribute to weight gain. This interplay has therapeutic implications, as metabolic syndrome can compromise biologic responses and influence treatment outcomes. The male predominance in dyslipidemia (OR = 2.37) may reflect sex-specific differences in lipid metabolism, body fat distribution, and hormone profiles [21].

Prior studies of IL-17A inhibitors have reported conflicting lipid findings, ranging from favorable reductions to unfavorable increases (Supplementary Table 6) [8, 10–14, 22–24]. Large cohort studies have demonstrated comparable cardiometabolic safety across biologic classes, with no significant adverse lipid changes during therapy [10, 25, 26]. Our phenotype-stratified approach reveals that both favorable and unfavorable patterns coexist within cohorts but manifest in different lipid subgroups: patients with baseline mixed dyslipidemia or hypercholesterolemia experienced reductions in atherogenic lipids, while those with normal baseline lipids showed modest unfavorable changes within normal ranges.

The absence of correlation between PASI improvement and lipid changes suggests that IL-17A may influence lipid metabolism through direct pathways independent of cutaneous disease activity. IL-17A exerts opposing effects on lipid metabolism through distinct tissue-specific mechanisms. In hepatic tissue, IL-17A exacerbates steatosis by inhibiting fatty acid β-oxidation via the JNK-PPARα pathway, and anti-IL-17A treatment attenuates hepatic lipid accumulation by restoring β-oxidation enzyme expression (PPARα, ECHS1) [27, 28]; this hepatic mechanism may explain the TC and LDL-C reductions observed in our hypercholesterolemia and mixed dyslipidemia subgroups. Conversely, IL-17 negatively regulates adipogenesis by suppressing adipogenic transcription factors (C/EBPα, PPARγ) and downstream genes including lipoprotein lipase, with IL-17-deficient mice developing accelerated diet-induced obesity [29]; IL-17A inhibition may thus promote adipogenesis and lipid uptake into adipose tissue, explaining the modest unfavorable changes in our normal lipid and isolated low HDL-C subgroups. These opposing hepatic and adipose effects suggest that the net impact depends on baseline metabolic state: patients with pre-existing hyperlipidemia may predominantly benefit from restored hepatic lipid handling, while those with normal lipids may experience a predominant pro-adipogenic effect [27–32]. Further mechanistic studies incorporating serial adipokine measurements are needed to validate this framework.

Clinical implications

These findings support a phenotype-guided approach to IL-17A inhibitor therapy. We recommend: (1) baseline lipid screening before initiating IL-17A inhibitors, particularly in obese male patients at higher dyslipidemia risk; (2) preferential selection of IL-17A inhibitors for patients with mixed dyslipidemia or hypercholesterolemia (especially those with LDL-C ≥ 3.4 mmol/L), who may derive dual dermatologic and cardiometabolic benefits through significant atherogenic lipid reductions; (3) concurrent lipid-lowering therapy consideration for patients with hypertriglyceridemia [33], given limited TG reduction with IL-17A inhibitors alone; (4) periodic lipid monitoring for patients with isolated low HDL-C and normal baseline lipids to detect unfavorable trends; and (5) since lipid effects occur independently of PASI improvement, cardiometabolic monitoring should not be neglected in patients achieving skin clearance.

Strengths and limitations

This study’s strengths include phenotype-stratified analysis providing granular insights into differential lipid changes, comprehensive lipid parameter assessment with robust FDR correction, and inclusion of a cardiovascular high-risk subgroup analysis. The one-year follow-up duration allows assessment of sustained treatment effects, and the real-world setting provides evidence applicable to clinical practice.

Several limitations merit acknowledgment. First, the retrospective single-center design limits generalizability, and residual confounding from unmeasured variables—including body weight changes, dietary habits, and circulating adipokines—cannot be excluded despite multivariable adjustment. Second, lipid parameters were assessed only at baseline and one year, without intermediate measurements, and cardiovascular events were not systematically ascertained. Third, the predominantly male cohort (75.6%) may limit generalizability of sex-specific findings. Fourth, although the observed lipid changes were statistically significant, the absolute magnitudes were modest compared to dedicated lipid-lowering therapy. Finally, modest sample sizes in some subgroups may limit statistical power for detecting subtle differences.

Conclusions

IL-17A inhibitors exert heterogeneous, phenotype-dependent effects on lipid metabolism in patients with moderate-to-severe plaque psoriasis. High BMI and male sex are independent risk factors for dyslipidemia in this population. Patients with mixed dyslipidemia or hypercholesterolemia may derive favorable lipid-lowering effects, while those with normal baseline lipids or isolated low HDL-C require monitoring for potential unfavorable changes. Among cardiovascular high-risk patients with elevated baseline LDL-C, IL-17A inhibitors provide clinically meaningful atherogenic lipid reduction. The lipid-modulating effects appear independent of dermatologic response, suggesting direct metabolic mechanisms. These findings support phenotype-guided treatment selection and individualized lipid monitoring strategies to optimize both dermatologic and cardiometabolic outcomes. Future multicenter prospective studies incorporating serial weight and adipokine measurements, longer follow-up, and systematic cardiovascular event ascertainment are needed to validate these findings and establish clinical significance.

Abbreviations

AUC

Area Under the Curve

BMI

Body Mass Index

BSA

Body Surface Area

CI

Confidence Interval

CRP

C-Reactive Protein

FDR

False Discovery Rate

HDL-C

High-Density Lipoprotein Cholesterol

IL-17A

Interleukin-17A

IL-23

Interleukin-23

LDL-C

Low-Density Lipoprotein Cholesterol

OR

Odds Ratio

PASI

Psoriasis Area and Severity Index

SD

Standard Deviation

TC

Total Cholesterol

TG

Triglycerides

VIF

Variance Inflation Factors

Authors' contributions

All authors contributed to the study conception and design. Material preparation, data collection and analysis were performed by Chao Wu, Chun-Xia He, Si-Fan Wang, and Dian-Mo Li. The first draft of the manuscript was written and commented by all authors. All authors have read and approved the final manuscript.

Funding

This study was supported by

1. Peking Union Medical College Hospital Talent Cultivation Program Category D (UHB11983).

2. National Natural Science Foundation of China (82573978).

3. Beijing Key Clinical Specialty Construction Project.

4. National Key Clinical Specialty Project of China.

Data availability

The datasets generated and/or analyzed during the current study are available from the corresponding author on reasonable request.

Declarations

Ethics approval and consent to participate

This study was reviewed and approved by the institutional review board of Peking Union Medical College Hospital (Approval number: I-25PJ2973). The requirement for informed patient consent was waived because of the anonymous nature of the data.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.