Effect of Narrowband Ultraviolet B Phototherapy on Systemic Metabolome in Korean Patients With Psoriasis

Dear Editor,

Psoriasis is a chronic skin disease that increases the risk of developing psoriatic arthritis, cardiovascular disease, and other metabolic disorders. Therefore, understanding the systemic impacts of psoriasis treatments is crucial beyond merely addressing skin lesions. Although phototherapy, particularly narrowband ultraviolet B (nbUVB), is recognised as a safe and effective treatment option for psoriasis, its mechanisms and effects on systemic metabolites are not fully understood. To investigate these potential systemic effects, we conducted an untargeted metabolomic analysis of plasma samples from adult patients with psoriasis who underwent nbUVB phototherapy at least twice a week for a minimum of 8 weeks.

In this single-centre, before-and-after study, 23 patients with moderate-to-severe psoriasis were treated with nbUVB phototherapy at Seoul National University Hospital between 2015 and 2018 (Table S1). Treatments were administered using TL-01 lamps in a UV 7001K cabin (Waldmann, Villingen-Schwenningen, Germany) twice or thrice a week, starting at 400 mJ/cm² and increasing by 10%–20% of the previous dose, with adjustments based on side effects such as erythema, irritation, and itching. The mean duration of phototherapy, mean session count, and cumulative nbUVB dose were 12.0 weeks, 25.0, and 35 423.9 mJ/cm², respectively. Topical corticosteroids of moderate potency or higher were allowed as needed, whereas no concurrent systemic treatment was administered. For each patient, the psoriasis area and severity index (PASI) score and plasma samples were collected before treatment, as well as on the day of the last UV exposure or the following day. Metabolomic profiling of the plasma samples was performed using an Orbitrap Exploris 120 mass spectrometer, equipped with a Vanquish Flex ultra-high-performance liquid chromatography system (Thermo Fisher Scientific; Waltham, MA, USA).

Following nbUVB phototherapy, the mean PASI score of the patient cohort significantly decreased from 10.6 to 3.3 (p < 0.001, Table S1). Changes in the overall levels of plasma metabolome, including 150 endogenous metabolites annotated with level 2 confidence according to the Metabolomics Standards Initiative, between pre- and post-treatment were subtle (Figure 1A). Five metabolites were found to be significantly changed, which may reflect the mechanism of action of nbUVB phototherapy (Figure 1B). Levels of 4-hydroxy docosahexaenoic acid, an anti-inflammatory metabolite derived from docosahexaenoic acid, decreased after treatment, indicating reduced inflammation [1]. Levels of creatine, which has been recognised as an antioxidant capable of preventing UVB-induced keratinocyte apoptosis, increased after treatment [2]. Carnitine, previously reported to be associated with the integral membrane protein CD147 highly expressed in psoriatic skin lesions, showed a similar increasing trend after UVB treatment [3]. These findings indicate the systemic effect of phototherapy and the inherent self-protection mechanism of the body against UVB damage.

To investigate whether patients' responses to nbUVB therapy could be predicted before treatment, we compared baseline metabolomic profiles between the good and poor responders. Good responders were defined as those exhibiting a decrease in PASI score of more than 75%, while poor responders were those who did not meet this criterion. No statistically significant differences were observed between the two groups in terms of age, sex, body mass index, disease duration, scalp or nail involvement, and nbUVB treatment (Table S1). After adjusting for sex, age, and PASI score, we found significant differences in five metabolites between good and poor responders, including gamma-glutamylglutamine (gamma-Glu-Gln), dodecylbenzenesulfonic acid (DBSA), phosphatidylethanolamine P-18:0/20:4 (PE P-18:0/20:4), glutamate, and proline (Figure 1C). The average area under the receiver operating characteristic curve for predicting response based on these five metabolites with a linear support vector machine reached 0.94 (Figure 1D). The inhibition of amino acid accumulation, such as glutamate and proline, has been recognised as beneficial for patients with psoriasis due to potential adverse metabolic risks [4]. Elevated levels of the key intracellular antioxidant marker gamma-Glu-Gln in good responders suggest that higher antioxidant capacity may enhance the response to phototherapy [5]. These results imply the potential significance of specific amino acids in the nbUVB phototherapy of psoriasis. DBSA, a commonly used surfactant in various dermatological and cosmetic products, may impact the barrier function and lipid metabolism of the skin. This effect is further supported by the observed difference in PE P-18:0/20:4 level, which is associated with inflammatory responses and immune regulation [5].

Phototherapy is believed to improve psoriasis by inducing immune changes in the skin and regulating metabolic markers [6]. However, most studies have focused on the roles of cytokines or specific metabolism, with limited examinations of the comprehensive changes in blood metabolites [6, 7]. Although the sample size was small, this study demonstrated that nbUVB phototherapy is safe for several months at the systemic level of the metabolome. Additionally, we identified several metabolites as potential biomarkers for predicting the response to nbUVB therapy, indicating certain clinical application values.

Yufei Li: conceptualization, data curation, formal analysis; methodology; visualisation; writing – original draft. Yujin Lee: conceptualization; supervision; writing – review and editing. Howard Lee: conceptualization; resources. Seong-Jin Jo: conceptualization; writing – review and editing. Joo-Youn Cho: conceptualization; supervision; writing – review and editing.

The study was approved by the institutional review boards of Seoul National University Hospital (IRB No. H-1711-057-899).

The patients in this manuscript provided written informed consent to the publication of their case details.

The authors declare no conflicts of interest.