This report complies with the Strengthening the Reporting of Observational Studies in Epidemiology – Molecular Epidemiology (STROBE-ME) guidelines.14 The study was conducted in a tertiary center for the treatment of patients with psoriasis at the University Hospital of Brasília, which is located in Brasília, Brazil. This center is responsible for a population of more than a million people, and patients are referred from secondary facilities.

From November 1st, 2022, to December 1st, 2023, the authors included consecutive patients attending the Psoriasis Ambulatory Department. To form the main analysis group, the authors included patients who were regularly followed up at the University Hospital of Brasília and who had a confirmed history of severe psoriasis vulgaris (International Classification of Diseases 10th Revision [ICD10 = L40.0]) (Psoriasis Area and Severity Index [PASI > 10] or Dermatology Life Quality Index [DLQI > 10]) and who subsequently achieved disease remission (PASI or DLQI < 5) for at least 1-year after the use of methotrexate or biologics. Consecutive psoriasis patients who were referred to this specialized dermatological center during the same period and who were not receiving systemic treatment were also included to form a comparator group. Twenty healthy controls matched for sex and age to psoriasis patients were also included (convenience sampling) for comparison. The authors excluded patients who had any other autoimmune conditions (except for psoriatic arthritis), pregnant women, or indigenous individuals. All mutation and gene expression tests were conducted on every patient.

After signing an informed consent form, all patients completed a comprehensive questionnaire aimed at gathering detailed clinical information. The information requested included age, sex, time since psoriasis diagnosis and previous medications used for the treatment of psoriasis. All patients were also examined by 2 board-certified dermatologists who collected data related to the type and severity of psoriasis (PASI and DLQI scores).

In the main analysis group, all psoriasis patients were regularly followed up for more than a year and had previously achieved a satisfactory clinical response to systemic treatment. Upon inclusion in the study, patients were divided into two main groups: 1) Those who developed a loss of systemic treatment response (defined as the recurrence of one or both of the following scores ‒ PASI > 10 or DLQI > 10 ‒ after previously effective treatment) and 2) Those who continued to have a sustained effective response to systemic treatment. Although the group of patients who did not receive any systemic treatment was also divided into patients with a PASI or DLQI > 10 or not, the term “loss of systemic treatment response” was not used for this group.

The authors collected 20 ml of whole blood from the study participants in 2 heparin tubes. Immediately after collection, PBMCs were isolated using a Ficoll Paque Plus (Merck KGaA, Darmstadt, Germany) gradient. The samples were stored in RNA Later (Thermo Fisher Scientific, Waltham, United States) at −80 °C until RNA extraction, which was performed up to 1-month after sample collection. Whole blood and plasma were also stored at −80 °C.

Total RNA from PBMCs was extracted using the mirVana™ PARIS™ RNA and Native Protein Purification Kit (Thermo Fisher Scientific, Waltham, United States). The purified RNA was quantified with a NanoDrop One/OneC UV–vis spectrophotometer (Thermo Fisher Scientific). Samples were treated with RNase-free DNase I (1 U/µl) (Thermo Fisher Scientific). Complementary DNA (cDNA) was generated using a High-Capacity cDNA Reverse Transcription Kit (Thermo Fisher Scientific) in a T100 Thermal Cycler (Bio-Rad, Hercules, United States) according to the manufacturer’s instructions.

Messenger RNA (mRNA) expression analysis was performed using TaqMan-based manufactured probes (Thermo Fisher Scientific) labeled with 5'Fluorescein Amidite (FAM) and 3′Minor Groove Binder (MGB) probes. The authors measured the expression of the TYK2, IL-12A, IL-12B, IL-23A, IL-23 receptor, IL-6, IL-6R, IL-17A and Tumor Necrosis Factor (TNF) genes using probes that spanned exon junctions. Despite the primary objective of this study being the evaluation of cytokines in the IL-23 axis and other activators of the TYK2 receptors, effector cytokines such as TNF and IL-17A were additionally measured to assess and compare the activation of the entire Th1 and Th17 stimulation pathway. Three target candidates, namely, Glyceraldehyde-3-Phosphate Dehydrogenase (GAPDH), eukaryotic 18S rRNA and beta actin, were screened for use as endogenous controls, and GAPDH was selected. For each assay, the genes analyzed are detailed in Supplementary Table 1.

Reactions were performed in a QuantStudio 5 Thermocycler (Thermo Fisher Scientific). The reaction was initiated at 50 °C for 2 min, followed by 95 °C for 10 min (polymerase activation) and 40 PCR cycles of 95 °C for 15 s and 60 °C for 60 s. Reactions were performed in a final volume of 15 µl comprising 1× TaqMan™ Gene Expression Master Mix (Thermo Fisher Scientific, Waltham, United States), 1× TaqMan-based probes, 2 µl of cDNA sample from each patient and ultrapure water. All the samples were tested in triplicate. For normalization, the authors used a reference sample (calibrator) formed from a pool of total RNA extracted from the PBMCs of the 20 included healthy controls. No-template controls were also used for comparison.

The authors used the comparative Ct (ΔΔCt) method for data analysis. The individual results are reported as the Relative Quantification (RQ) of gene expression (fold change). Calculations were performed with Applied Biosystems™ Analysis Software (Thermo Fisher Scientific). Values of each expression considered inaccurate are suppressed by the analysis software.

The authors chose a robust manufactured assay that precisely detects the psoriasis-related protective Single-Nucleotide Polymorphism (SNP) rs12720356.19 This SNP was the only SNP in the TYK2 gene that was found to be related to psoriasis and that was reproduced in more than one scientific article, according to a previous systematic review of the literature and meta-analysis.19 The kit uses multiplex qPCR with [VIC/FAM] probes to detect the following mutation in the TYK2 gene: GTGCTCCACGTACTCTGTCACCATG[A/C]TATCTGTAAAGACACAGCTGCTCTG, located on chromosome 19:10359299 on build GRCh38. Transcripts of this region are similar to those of the target tested for TYK2 gene expression.

Genomic DNA from whole blood was extracted using the PureLink™ Genomic DNA Mini Kit (Thermo Fisher Scientific). The genotyping mixture included 1× TaqPath™ ProAmp™ Master Mix (Thermo Fisher Scientific), 1× TaqMan™ SNP Genotyping (Thermo Fisher Scientific, Waltham, United States) (CN: C_34042925_10, 4351379), 20 ng of genomic DNA and ultrapure water in a final volume of 25 µl. Reactions were performed in a QuantStudio 5 Thermocycler (Thermo Fisher Scientific). The cycling conditions comprised a preread step of 30 s at 60 °C, followed by an initial denaturation/enzyme activation step of 5 min at 95 °C; 40 cycles of denaturation (15 s at 95 °C) and annealing/extension (60 s at 60 °C); and a post read step of 30 s at 60 °C.

The plasma levels of TNF-α, IL-10, IL-1β, IL-8, IL-6 and IL-12p70 in the first 80 included patients were measured using the Human Inflammatory Cytokine Cytometric Bead Array (CBA)-I Kit (RUO) (Becton Dickinson, Franklin Lakes, USA) and a FACSVerse flow cytometer (Becton Dickinson).

For initial statistical analysis, the authors considered the PASI score the main outcome and divided patients into two groups: those with a PASI > 10 and those with a PASI ≤ 10. Another important outcome was the loss of systemic treatment response, defined by a PASI or DLQI score > 10.

The authors explored the differences in the RQs of gene expression and clinical characteristics. Subsequently, the authors applied hierarchical cluster techniques and heatmaps to find groups of patients who could represent a different endotypic profile of individuals experiencing loss of systemic treatment response. The assumptions for multivariate analysis were not met, and this analysis was determined to be unreliable.

The clinical relevance of gene expression changes was defined as a minimum 2× difference in gene expression (RQ) between groups.20,21 A p-value < 0.05 indicated statistical significance. Whenever statistical significance was met for RQ, the authors applied Benjamini-Hochberg False Discovery Rate (FDR) adjustment. Statistical analysis was performed using the program R version 4.1.2 (R Core Team [2021]). R: A language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, Austria. URL https://www.R-project.org/).

All patients were included after signing an informed consent form. This study complied with the Declaration of Helsinki and was approved by the Ethics Committee of the Faculty of Medicine of the University of Brasília, Brazil (CAAE: 68068323.3.1001.5558).

The authors enrolled 198 participants in this study, including 162 patients with psoriasis who had previously achieved disease remission using systemic medications, 16 patients with psoriasis who were not receiving any systemic treatment, and 20 healthy controls. Twenty-seven psoriasis patients exhibited a PASI score > 10 (the main dependent variable), and another 32 patients exhibited DLQI scores > 10. The data in Supplementary Table 2 show that sex, age and disease duration were not different between patients with PASI scores > 10 and those with PASI scores ≤ 10.

Regarding the medication used by each patient at recruitment, 21 were using methotrexate, 83 were using anti-TNF biologics (5 etanercept, 7 infliximab, and 71 adalimumab), 17 were using anti-IL-12/23 biologics (ustekinumab), 7 were using anti-IL-23 biologics (5 risankizumab and 2 guselkumab), 34 were using anti-IL-17 biologics (secukinumab) and 16 were not using any systemic treatment. The authors did not detect a PASI > 10 in patients treated with anti-IL-23 biologics (n = 7) (Supplementary Table 3).

The percentage of patients who experienced a loss of systemic treatment response (PASI or DLQI score > 10) did not differ among those receiving different systemic treatments (p > 0.05). The number of patients with a PASI or DLQI > 10 was greater among those who did not receive systemic treatment than among patients who received anti-TNF biologics, anti-IL-17 biologics, anti-IL12/23 biologics or anti-IL-23 biologics (p < 0.05) (Supplementary Table 3).

The authors found a positive correlation between TYK2 gene expression RQ and IL-12A, IL-23A or IL-23R expression (p < 0.05) (Supplementary Fig. 1). Compared with that in the PBMCs of healthy controls, the RQ gene expression of most mediators was similar or reduced in the PBMCs of psoriasis patients. Similar results were found via CBA analysis (Supplementary Tables 4–8).

TYK2 gene expression in PBMCs was upregulated in patients with a PASI > 10 (Table 1). FDR correction revealed that, in isolation, TYK2 RQ gene expression could not differentiate patients with a PASI greater than 10 from those with a PASI less than or equal to 10. This result is probably due to the very wide range of RQ gene expression (Supplementary Fig. 2). Additionally, the authors did not find any relationship between TYK2 gene expression and other demographic or clinical characteristics (Supplementary Table 9).

For our hierarchical clustering analysis, the authors included only patients who had RQ values less than 2 standard deviations apart from all mediators tested (Fig. 1). The authors included 55 patients in this analysis (Tables 2 and 3). The authors identified a group of 19 patients with increased expression of mediators of the TYK2 signaling pathway. IL-12A, IL-12B, IL-23A, IL-23R, and IL-6 were upregulated in this group (Table 2). This group of patients also presented higher PASI (p = 0.021) and DLQI (p = 0.034) scores than those with lower TYK2 mediator expression (Table 3). Patients in this group who exhibited increased expression of mediators of the TYK2 signaling pathway and also loss of systemic treatment response were using methotrexate, anti-TNF agents and anti-IL-17 biologics but not anti-IL-12/23 or anti-IL-23 biologics (Supplementary Table 10). The remaining patients did not exhibit a gene expression pattern in PBMCs that could distinguish them from other patients (n = 36) (Fig. 1).

Fig. 1.

The image displays a heatmap of the gene expression of mediators that belong to the TYK2 signaling pathway. The data are derived from 55 psoriasis patients who had all the necessary measurements taken in their peripheral mononuclear cells and for whom the replicate results were within 2 standard deviations. This graphic was generated using the program R version 4.1.2 (R Core Team (2021)). R: A language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, Austria. URL https://www.R-project.org/). TYK2, Tyrosine Kinase-2; IL, Interleukin; R, Receptor.

Three patients (1.69%) were heterozygous for the rs12720356 SNP identified in the TYK2 gene (Supplementary Fig. 3). This mutation is also known as the TYK2 I684S variant.22 Of the 3 patients carrying the TYK2 I684S variant, 2 were females, 2 had associated nail psoriasis, and 1 had palmoplantar lesions. All 3 patients had a PASI value < 10. Because the frequency of this SNP was low, it was not possible to relate these findings to the measured gene expression. No mutations were found in healthy controls.