Association between several immune response‐related genes and the effectiveness of biological treatments in patients with moderate‐to‐severe psoriasis

Biological therapies are safer and more effective against psoriasis than conventional treatments. Even so, 30–50% of psoriatic patients show an inadequate response, which is associated with individual genetic heterogeneity. Pharmacogenetic studies have identified several single nucleotide polymorphisms (SNPs) as possible predictive and prognostic biomarkers for psoriasis treatment response. The objective of this study was to determine the link between several SNPs and the clinical response to biological therapies in patients with moderate–severe psoriasis. A set of 21 SNPs related to psoriasis and/or other immunological diseases were selected and analysed from salivary samples of patients (n = 88). Treatment effectiveness and patient improvement was assessed clinically through Relative Psoriasis Area and Severity Index (PASI), also called ‘PASI response’, as well as absolute PASI. Associations between SNPs and PASI factors were assessed at 3 and 12 months for every treatment category of IL‐17, IL‐23, IL‐12&23 and TNF‐α inhibitors. Multivariate correlation analysis and Fisher's exact test were used to analyse the relationship between SNPs and therapy outcomes. Several SNPs located in the TLR2, TLR5, TIRAP, HLA‐C, IL12B, SLC12A8, TNFAIP3 and PGLYRP4 genes demonstrated association with increased short and long‐term therapy‐effectiveness rates. Most patients achieved values of PASI response ≥75 or absolute PASI<1, regardless of the biological treatment administered. In conclusion, we demonstrate a relationship between different SNPs and both short‐ and especially long‐term effectiveness of biological treatment in terms of PASI. These polymorphisms may be used as predictive markers of treatment response in patients with moderate‐to‐severe psoriasis, providing personalized treatment.


| INTRODUC TI ON
Psoriasis is a chronic systemic immune-mediated disease that affects 2%-3% of the general population. 1,2Prevalence and incidence vary by population and age.Among Europeans, psoriasis primarily impacts adults rather than children, with both sexes being equally affected. 3The majority of psoriasis cases, around 90%, are attributed to chronic plaque psoriasis, also referred to as psoriasis vulgaris.
Furthermore, in approximately 6%-42% of cases, psoriasis is accompanied by psoriatic arthritis throughout a person's life. 4Additionally, patients may experience other comorbidities like cardiovascular conditions, metabolic disorders, cancer or Crohn's disease, which significantly impact their quality of life. 5e cause of psoriasis remains uncertain, but it results from a combination of genetic, immunological and environmental factors that disrupt the normal functioning of epidermal cells, particularly the proliferation of keratinocytes and the development of epidermal hyperplasia.These alterations lead to the formation of thickened and inflamed scaly patches, causing discomfort and itchiness.
Among these factors, it has been established that genetic variability explains about 70% of disease susceptibility. 6er the past few decades, research on psoriasis has substantially advanced our understanding of the disease's pathophysiology, emphasizing the significance of the immune system and enabling the creation of secure and highly effective targeted treatments using monoclonal antibodies.Among these medications are anti-IL-17, anti-IL-12&23 and TNFα inhibitors, 7,8 as well as anti-IL-23 drugs, which represent the most recently approved therapies for moderate-to-severe psoriasis. 9ese biological agents hinder the interaction between interleukins and their dedicated receptors, thereby suppressing IL-dependent cellular signalling and the release of proinflammatory cytokines.
The pathophysiology of psoriasis heavily relies on the involvement of T cells and cytokines.Numerous studies have indicated elevated levels of IL-2, IL-6, IL-8, IL-12, IL-17, IL-19, IL-23, IFNγ and TNFα in psoriasis. 7,10This overexpression is believed to be responsible for the excessive proliferation of epidermal keratinocytes and the development, maintenance and recurrence of skin lesions.
Among these interleukins, IL-12 and IL-23 hold particular importance due to their impact on T cells. 7,10While IL-12 drives Th1 cell differentiation, IL-23 is critical for the proliferation of Th17 and Th22 cells.Th17 cells produce interleukins such as IL-17, IL-22 and TNFα.
Therefore, activation of the Th17 pathway by IL-23 is closely linked to the inflammatory processes and autoimmune responses observed in psoriasis. 11In this regard, targeted immunomodulation through biological therapies has revolutionized the ability to manage this immune-mediated disorder.
On the other hand, large-scale genome-wide association studies (GWAS) have provided insights into the genetic underpinnings of psoriasis.These revelations have, in the last decades, paved the way for biological therapies (e.g.cyclosporine, methotrexate or acitretin) that surpass conventional systemic treatments in both efficacy and safety. 4Specifically, these new therapies have improved the clinical symptoms and reduced the risk of developing comorbidities.
In contrast to conventional systemic medications, biologic drugs selectively target specific components of the immune system, inhibiting the activity of T cells or cytokines such as tumour necrosis factor-alpha (TNFα), interleukins (IL) 17, IL-12 or IL-23. 12,13According to these molecular targets, TNFα or IL inhibitors have been developed for the treatment of psoriasis.However, patients show heterogeneity in their treatment outcomes, both short-and long-term, and some patients develop different degrees of toxicity. 14,15The variability in treatment response is influenced by the patient's genetic makeup, specifically the presence of genetic variations that can impact therapy effectiveness and safety.This heterogeneity may affect drug metabolism, the cellular environment or the drug's mechanism of action.Hence, the identification of genetic biomarkers as predictors of treatment response becomes crucial to optimize and provide personalized treatment options.This not only reduces the likelihood of adverse effects but also enhances the overall efficiency of therapies, benefiting health systems in the process.
This particular study aims to unveil the correlation between SNPs and both short-term and long-term clinical outcomes of psoriasis biological therapies.The findings from this study could significantly influence clinical decision-making, leading to the administration of more personalized biological treatments to patients, thereby ensuring enhanced efficacy and, subsequently, an improved quality of life for those affected.For this, Relative Psoriasis Area and Severity Index (PASI) also called 'PASI response' was used.All patients included in the study had an initial PASI score of 4 or higher (≥4), which is the specific PASI score that was considered for moderate-to-severe psoriasis.International guidelines consider as effective 'a reduction of a PASI between baseline and a specified treatment period by at least 75% (PASI75 response)'. 18SI75 has been the treatment goal for moderate-to-severe psoriasis in most clinical trials. 19However, with the implementation of more effective biological drugs for psoriasis treatment, some authors consider that PASI90 or PASI100 should be considered as the new assessment parameters. 19e present study considered PASI50, PASI75 and PASI90 at 3 and 12 months from treatment onset (Figure 1).Patients with PASI50 response were classified as non-responders; patients with PASI75 as responders; and patients with PASI90 response as super-responders. 19The latter group includes patients that present a fast and exceptional improvement with treatment.
In addition to PASI response, absolute PASI<3 and PASI<1 was also collected.Absolute PASI is the most frequent score used to quantify treatment effectiveness in routine clinical practice since it indicates the status of the disease.In general, there is a consensus that absolute PASI scores of ≤3 and ≤5 represent appropriate treatment responses. 20e study presented here was carried out in accordance with the Declaration of Helsinki.Moreover, the protocol of the study was approved by the Ethics Committee of the Castellon General University Hospital, and all patients gave written informed consent to participate.

| Selection of genetic markers
A total of 21 psoriasis-related SNPs previously described 21,22 were chosen for analysis based on (i) promising susceptibility in psoriasis (SNPs rs3213094 and rs2546890 in IL12B, rs11209026 in (rs11126740 in CTNNA2).Online platforms such as GeneCards, PathCards, SNPedia, National Center for Biotechnology Information (dbSNP) and GWAS Catalog were used to establish the relationship between genes and SNPs to biological pathways and diseases.

| Sample processing and genotyping
A sample of saliva was collected from each patient using a sterile buccal swab.Genomic DNA was isolated from saliva samples using the QIAamp DNA mini kit (Qiagen), following the manufacturer's recommendations; and afterward, samples were stored at −80°C.

SNP genotyping was conducted by the Spanish National
Genotyping Centre (CeGen-PRB2, Santiago de Compostela, Spain) as a contract service, as previously described. 23Briefly, SNP genotyping was achieved by using the iPLEX Gold MassARRAY technology, according to manufacturer's protocol (Sequenom).Assays were implemented in 384-well plates, together with a negative control and three Coriell samples for quality control.Genotyping accuracy was also assessed by adding three DNA duplicates per plate, producing 100% consistent replication outcomes.

| Statistical analysis
Data were analysed using the R software, version 4.2.3 (http:// www.R-proje ct.org).SNPs deviations and genotype frequencies were tested for all the bi-allelic SNP markers using a Hardy-Weinberg equilibrium (HWE) test.Only the SNPs which allele frequencies were in HWE were included.
The associations between the SNPs and the other variables were performed using the R-package SNPassoc program (version 2.1-0).
Every SNP was tested for all genetic models of inheritance considering a p-value <0.05 as statistically significant.All association tests were performed adjusting each SNP by covariates, sex, obesity, age at onset and years since diagnosis.
To assess differences among the variables and type of treatment groups, Pearson's chi-squared test and Fisher's exact test were used.
Statistical significance was set at p-value ≤0.05.
Unknown values were excluded at each specific analysis.

| Patient and treatment characteristics
A total of 88 patients (37 females and 51 males) diagnosed with plaque psoriasis were enrolled in this study.All recruited patients were of European Spanish descent.The average age of psoriasis onset was 26 years, and the mean duration from disease diagnosis to the present time was 23 years.A majority of patients had generalized lesions (54.12%), but others only showed lesions in specific areas of the body such as the head, elbows or limbs (45.88%).Approximately 72% of patients were overweight or obese according to OMS criteria (BMI > 25 or 30, respectively), 15% were smokers, and 50% had family history of psoriasis.Among the patients, 29% had additional comorbidities associated with the immune system, with psoriatic arthritis being particularly notable, concurrent with psoriasis in 23% of the cases (Table S1).
Regarding treatment, 79 patients (91%) had been previously treated with single or combined systemic treatment (e.g.methotrexate, cyclosporine, methotrexate/cyclosporine and acitretin); 36 patients (48%) had already undergone a first line of biological treatment for psoriasis; and the remaining 42 patients (52%) were naive regarding biological treatment.Patient and treatment characteristics are described in Table S1.

| Anti-IL-17
A total of 13 patients were treated with Secukinumab as first biological treatment option, and 6 as the second therapeutic option.
Only one patient was treated with Ixekizumab as a second biological treatment.At 12 months of treatment, 53% of patients achieved an absolute PASI<1 and 18% a PASI<3.If treatment response is analysed according to PASI response, 24% of patients were nonresponders, 6% were responders, and 56% were super-responders (Table 1).

| Anti-IL-23
Out of 14 patients, 13 received Guselkumab as their second biological treatment, while 2 out of 4 patients were administered Risankizumab as their second-line option.The 12-month effectiveness results indicated that 67% of patients achieved PASI<1, and 33% attained PASI<3.Remarkably, 82% of the total group were classified as super-responders (Table 1).

| Anti-IL-12 & IL-23
This group of patients was treated with Ustekinumab, with 54% of them receiving it as a second-line treatment.After 12 months of treatment, 48% of patients attained a PASI<1.The PASI response analysis revealed that 59% were classified as super-responders, 16% as responders, and 22% as non-responders (Table 1).
Among them, 62% received this as their initial treatment option.
In Table 2, we present the SNPs associated with improved clinical outcomes in relation to PASI<1 and PASI90 after 3 and 12 months of commencing treatment.These time points were selected to evaluate early and enduring treatment responses, respectively.
The rs2916205 SNP demonstrated a strong association with enduring clinical responses to IL-12&23 treatment, as evidenced by a highly significant p-value (***p ≤ 0.0001; Table 2).Additionally, rs5744174 exhibited a correlation with TNFα inhibitors and anti-IL-23 treatments, while rs610604 was related to TNFα inhibitors and anti-IL-17 treatments.Conversely, rs9304742 and rs1800629 were linked to early clinical responses in patients undergoing anti-IL-17 treatments.
Out of all 15 significant SNPs, a total of eight demonstrated significant associations with sustained clinical outcomes at the 12month mark.These include SNPs located in the TLR2, TLR5, IL12B, TNFAIP3, SLC12A8, HLA-C, TIRAP and PGLYRP4 genes.Detailed information about these SNPs, such as their functions, related diseases, associated drugs and involvement in cellular pathways, can be found in Table S2.Additionally, their contribution to psoriasis pathogenesis is discussed in the following section.

| DISCUSS ION
Research on psoriasis has significantly advanced our understanding of its pathophysiology, highlighting the role of the immune system and leading to the development of secure and effective targeted   treatments using monoclonal antibodies.These include anti-IL-17, anti-IL-12&23, TNFα inhibitors 7,8 and the recently approved anti-IL-23 drugs for moderate-to-severe psoriasis. 9These biological agents suppress proinflammatory cytokines by blocking interleukins and their receptors.
In this study, the best efficacy rates of treatment were achieved with the anti-IL-23 treatment.Approximately 67% and 100% of patients attained a PASI<1 and PASI<3 respectively, and 82% were classified as super-responders (PASI≥90).On the other hand, patients who received anti-IL-17 or anti-IL-12&23 treatments exhibited comparable therapeutic effectiveness results.Around 50% of patients showed PASI values <1 and more than 70% of patients achieved a PASI<3.PASI75 and PASI90 were reached for 62% and 59% of patients on anti-IL-17 treatment, and 75% and 59% on anti-IL-12&23, respectively.Finally, a representative percentage of patients treated with TNFα inhibitors attained a PASI<3 (79%), 77% achieved PASI75, and 46% reached PASI90 after 12 months of treatment.
These results are largely in agreement with data provided by various clinical trials.For example, at Week 50, the reported PASI90 response in studies RESTORE and REVEAL was 45% for Infliximab and 50% for Adalimumab at 3 years. 24,25Regarding Ustekinumab, the PHOENIX 1 study reported a 5-year PASI90 response rate of 39.7%. 26At Week 52, around 60% of the patients treated with Secukinumab achieved a PASI90 response 27 ; and finally, phase III trials of Guselkumab, VOYAGE 1 and 2, provided good efficacy and safety profiles of this drug.Specifically, at 48 weeks, PASI90 response rate was achieved by 76.3% of patients. 28,29In addition, studies such as ECLIPSE demonstrated clinical superiority of Guselkumab vs Secukinumab in long-term efficacy based on PASI90, 30 and others like the NAVIGATE showed Guselkumab superiority vs Ustekinumab. 31Ultimately, several studies have demonstrated the safety and efficacy of Risankizumab, as well as the impressive response rates at Week 52 in terms of PASI90 and PASI100 (85.5% and 60% of patients, respectively. 32,33These clinical outcomes reveal the durable treatment efficacy and the benefits that these biological drugs have provided to patients with moderateto-severe psoriasis. However, patients present biological treatment response heterogeneity due to genetic variability.Consequently, new pharmacogenetic biomarkers linked to treatment effectiveness are required to carry out personalized medicine.Our study was designed to identify associations between multiple SNPs and biological treatment response in patients with moderate-to-severe psoriasis.The performed analyses showed very significant statistical associations between several SNPs and PASI<1 and PASI90 values (as measures of good treatment effectiveness) at 3 and 12 months after treatment onset.Among them, those related to TLR2, TLR5, IL12B, TNFAIP3, SLC12A8, HLA-C, TIRAP and PGLYRP4 genes were the most significant in the maintenance of long-lasting clinical responses in psoriatic patients treated with biological agents.
There are no described functional studies in the scientific literature for any of the 15 significant SNPs reported in this work.The toll-like receptors (TLRs) are crucial for recognizing pathogenassociated molecular patterns (PAMPs) and activating inflammatory responses. 34Psoriatic plaques have elevated TLR levels (TLR1, 2, 4, 5 and 9), linked to disease exacerbation.TLR2 induces proinflammatory genes in psoriatic keratinocytes.Therefore, TLRs antagonism is used for the treatment of diseases such as psoriasis. 35In fact, previous studies have reported how Adalimumab modulates TLR expression and improves clinical outcomes. 36Our data also showed high significant associations between both TLR2 rs11938228 and TLR5 (OR = 9.42, p = 0.0051). 37In our study, TIRAP rs8177374 was significantly associated with PASI<1 at 3 and 12 months in patients treated with anti-TNFα (p < 0.0001).However, previous studies did not find significant associations between and TIRAP rs8177374 and TNFα therapy response. 37reover, genes as TNFα-induced protein 3 (TNFAIP3) also have been linked to psoriasis through their protective role against inflammatory responses by inhibiting the NFκβ pathway. 38Deregulated NF-κB activation is related to several inflammatory diseases, including psoriasis.Some studies have reported that TNFα inhibitors improve clinical outcomes and quality of life of psoriasis patients. 39,40ese data are in agreement with our results, in which the rs610604 derived allele variant, probably acting as a TNFAIP3 enhancer, is strongly associated with a PASI<1 at 12 months in patients treated with anti-TNFα drugs.
IL-12B is a proinflammatory cytokine that induces the Th1 pathway, related to the maintenance of Th1 memory cells and in turn associated with the IL-23A to IL-23 process.Thus, IL-12B plays an important role in the development of psoriasis. 41The rs2546890 polymorphism in the IL12B gene has been previously related to psoriasis susceptibility, to the development of psoriatic arthritis in psoriasis patients, and to TNFα treatment response. 13,22Our studies linked this SNP with good clinical responses to TNFα therapy (PASI<1 and PASI90), both in the short and the long term.
The SLC12A8 gene codes for an ion transporter related to the regulation of keratinocyte proliferation, closely implicated in psoriasis physiopathology.Previous studies have shown relationships between SLC12A8 rs651630 polymorphism and psoriasis risk. 42ecifically, the derived A allele of rs651630 was linked to a higher risk of paradoxical psoriasis in patients treated with TNFα inhibitors (p = 0.011). 13Mirroring those results, our study finds a strong association between the ancestral G allele of rs651630 and a good clinical response (PASI<1 and PASI90) at 3 and 12 months in patients treated with anti-TNFα and anti-IL-17.
Human leukocyte antigens (HLA) play a crucial role in recognizing exogenous proteins that trigger immune responses. 43In our study, a significant statistical association was identified between the HLA-C rs12191877 polymorphism and a PASI<1 and PASI90 values at 3 and 12 months, only for patients treated with anti-TNFα.
Therefore, this polymorphism may be used as a prediction marker of good short-term and long-term response to anti-TNFα drugs.
These data align with prior knowledge regarding the significance of this SNP in psoriasis and corroborates earlier investigations. 44nally, our study revealed a strong relationship between the SNP rs2916205 of the PGLYRP4 gene and the long-term efficacy of anti-IL-12&23 treatment psoriatic patients.significance of this polymorphism in psoriasis and its role on the response to anti-TNF therapy has also been documented in previous research. 44This gene encodes a peptidoglycan recognition protein involved in immune responses that bind to bacterial peptidoglycan.Consequently, an improved T-cell reaction to the bacterial antigen in psoriatic skin arises from this recognition. 45As a result, PGLYRP4 has been suggested as a compelling candidate gene for psoriasis.
Nonetheless, this work has limitations.While the findings are encouraging, additional studies are necessary to corroborate these indicators in more extensive patient cohorts, especially with a higher number of patients per treatment category.It is also worth noting that this research did not alter the standard clinical procedure (it is an observational study), thus lacking randomization of patients or establishing a control group.Lastly, it is important to emphasize that currently there is no genetic test available in the clinical practice to determine treatment response based on a subset of SNPs.

| CON CLUS ION
In conclusion, we have identified several genetic polymorphisms as predictive markers of psoriasis treatment efficacy in both the short and long run.These types of studies are necessary, since patients exhibit diversity in their response to biological treatments due to genetic differences.As a result, there is a need for novel pharmacogenetic markers associated with treatment efficacy to facilitate personalized medicine.
Although these findings need validation in a larger cohort of patients, their application in clinical settings could enhance treatment optimization, achieve higher response rates, reduce side effects and costs and ultimately improve the quality of life for patients.

2 | ME THODS 2 . 1 |
Patient selection and study design A total of 88 patients (37 females and 51 males) diagnosed with psoriasis were recruited from the Department of Dermatology of the Castellon General University Hospital, La Plana Hospital of Vila-real and Valencia General University Hospital (Spain).Inclusion criteria for patient selection were as follows: (i) patients older than 18 years diagnosed with moderate-to-severe plaque psoriasis according to the Spanish Academy of Dermatology and Venereology Psoriasis Working Group consensus document criteria 17 ; (ii) patients undergoing biological treatment for psoriasis; and (iii) patients under follow-up and monitoring for over a year (Figure 1).At diagnosis, a saliva sample was collected from each patient to perform genetic analyses.Clinical data of patients were obtained from clinical records at 3 and 12 months from the start of treatment.Patients included in the study were classified into four groups according to the biological treatment administered: Group 1 included anti-IL-17 drugs such as Secukinumab or Ixekizumab; Group 2 included anti-IL-23 drugs such as Guselkumab and Risankizumab; Group 3 included anti-IL-23 and 12 drugs such as Ustekinumab; and Group 4 included TNFα inhibitors such as Infliximab, Adalimumab, Etanercept or Certolizumab (Figure 1).Treatment effectiveness was evaluated by the reduction or complete remission of the skin lesions.

PGLYRP4
and rs651630 in SLC12A8); (ii) relation to inflammatory or immune diseases such as rheumatoid arthritis, psoriatic arthritis and Crohn's disease among others (rs6908425 in CDKAL1, rs1801274 in FCGR2A, rs96844 in ANKRD55, rs11938228 in TLR2, rs5744174 in TLR5, rs1799724 in TNF, rs1143623 in IL1B, rs8177374 in TIRAP and rs1061622 in TNFRSF1B); and (iii) cell adhesion and migration F I G U R E 1 Study design.

TA B L E 1
Clinical response of patients to treatments based on PASI.TA B L E 2 P-values of the significant SNPs associated with clinical responses to different psoriasis treatments at 3 and 12 months.PASI90 anti-IL-23 treatment column is missing because significant data were not observed at PASI90.Genes of SNPs related to best and durable clinical responses are indicated in bold.Only significant p-values are shown.
However, only three of them code for exonic variants, all missense: rs1801274 (p.H167R amino acid change) in the FCGR2A gene, rs5744174 (p.F616L change) in TLR5 and rs8177374 (p.S180L change) in TIRAP.Their effect could indeed be due to the change in the protein they code for.In addition, five of the SNPs lie in regulatory sequences upstream of their genes: rs2546890 in the IL12B gene, rs1800629 and rs1799724 in TNF, rs1143623 in IL1B and rs96844 in ANKRD55.Out of all five variants, only the latter one (rs96844) is located within a described gene enhancer.Their effect could be conveyed by up-or down-regulating the expression of the gene.Finally, the rest of the SNPs (seven in total) are intronic variants that might be involved in alternative splicing or can influence gene expression in several different ways.
rs5744174 polymorphisms, and PASI90 and PASI<1 values, specifically for the TNFα and IL-17 treatments.These data suggest a close association between these genetic variants and clinical outcomes achieved after treatment.Related to TLR is the TIRAP gene, which encodes a protein that interferes with the TLR2 and TLR4 signalling cascade.Previous studies have shown that the TIRAP rs8177374 polymorphism modulates the immune response through reducing TLR2 signal transduction, and it was associated with UTK therapy (PASI75 at 12 weeks)