Influence of biological treatments on intestinal microbiota of psoriasis patients

To the Editor: Psoriasis is a common, chronic inflammatory skin disease occurring worldwide and presenting at any age.^[1] Biologics are the most important therapeutic advances in psoriasis treatment.^[2] With the development of 16S rRNA sequencing technology, the association between psoriasis and intestinal flora has been gradually revealed. Some bacteria secrete short-chain fatty acids (SCFAs) with anti-inflammatory properties, serving as active microbial metabolites that regulate the function of immune cells in the intestine and other tissues.^[3] However, the impact of biological treatment on gut microbiota and related functional changes in psoriasis patients remains unclear. Our study aimed to explore the effects of biological treatments on the intestinal microbiota of psoriasis patients.

This research was reviewed and approved by the Ethic Committee of The First Affiliated Hospital of Chongqing Medical University (No. 2023-433). Patients have given written informed consent to publication of their case details. Inclusion criteria and exclusion criteria are shown in Supplementary Materials, http://links.lww.com/CM9/C83. Finally, a total of 181 plaque psoriasis vulgaris patients were enrolled, and their fecal samples were collected. Participants characteristics are presented in Supplementary Table 1, http://links.lww.com/CM9/C83.

Biologics included tumor necrosis factor (TNF)-α inhibitors (adalimumab), interleukin (IL)-17A inhibitors (secukinumab and ixekizumab), IL-12/23 inhibitors (ustekinumab), and IL-23 inhibitors (guselkumab). Age, sex, and body mass index (BMI) were matched in non-bio-treated and bio-treated group [Supplementary Table 1, http://links.lww.com/CM9/C83].

According to our findings, there were no significant differences in alpha-diversity between the bio-treated and non-bio-treated groups [Supplementary Figure 1A–E, http://links.lww.com/CM9/C83]. Furthermore, no significant statistical differences in terms of the Firmicutes/Bacteroidetes (F/B) ratio between these two groups [Supplementary Figure 1F, http://links.lww.com/CM9/C83]. No significant clustering was observed based on Bray-Curtis dissimilarity [Supplementary Figure 1G,H, http://links.lww.com/CM9/C83]. We found that the fecal microbiota in both groups appeared to be dominated by Firmicutes and Actinobacteriota at the phylum level, and by Blautia and Faecalibacterium at the genus level [Supplementary Figure 1I,J, http://links.lww.com/CM9/C83]. To detect the specific bacteria’s discrepancy, we analyzed the gut microbiota at genus level. Our results revealed a significantly higher relative abundance of Coprococcus (P = 0.0450) and Adlercreutzia (P = 0.0062), and a significantly lower relative abundance of Dialister (P = 0.0189), Veillonella (P = 0.0308), Eggerthella (P = 0.0102), and Erysipelatoclostridium (P = 0.0440) were observed in the bio-treated group [Supplementary Figure 2A, http://links.lww.com/CM9/C83]. We then conducted linear discriminant analysis effect size (LEfSe) comparison of these two groups. The structure and predominant bacteria of the microbiota were represented in a cladogram. The greatest difference in taxa from phylum to genus level was identified by the linear discriminant analysis (LDA) score. Coprococcus were enriched in fecal microbiota in the bio-treated group, while the top two in non-bio-treated group were Dialister and Veillonella at the genus level [Supplementary Figure 2B,C, http://links.lww.com/CM9/C83]. Heat map showed Enterococcus decreased in the bio-treated group [Supplementary Figure 2D,E, http://links.lww.com/CM9/C83]. The Kyoto Encyclopedia of Genes and Genomes (KEGG) pathways are shown in Supplementary Figure 3, http://links.lww.com/CM9/C83.

Subgroups (IL-12/23i, IL-17i, and IL-23i) of the bio-treated group were then analyzed. Results revealed that although the Sobs index had no significant differences at Operational Taxonomic Units (OTU) level, Chao index between the IL-12/23i group and IL-17i group displayed a significant difference in genus level (P = 0.0247) [Supplementary Figure 4A–E, http://links.lww.com/CM9/C83]. Beta-diversity analysis based on Principal Co-ordinates Analysis (PCoA) showed no significant differences among the three bio-treated groups. The F/B ratio of IL-12/23i, IL-17i, and IL-23i groups were decreased in turn, but there were no significant differences between the three groups [Supplementary Figure 4F, http://links.lww.com/CM9/C83]. The fecal microbiota in the three bio-treated groups appeared to be dominated by Firmicutes at phylum level, and by Blautia at genus level [Supplementary Figure 4G–J, http://links.lww.com/CM9/C83].

At genus level, the IL-12/23i group had a significantly higher relative abundance of Megamonas (P = 0.0408); the IL-17i group had a significantly higher relative abundance of Rothia (P = 0.0216) and Granulicatella (P = 0.0495); and IL-23i had a significantly higher relative abundance of Alistipes (P = 0.0361), Oscillibacter (P = 0.0451) and Colidextribacter (P = 0.0453) [Supplementary Figure 5A, http://links.lww.com/CM9/C83]. Megamonas was enriched in the fecal microbiota of the IL-12/23i group, Rothia in the IL-17i group, and Alistipes in the IL-23i group in genus level [Supplementary Figure 5B,C, http://links.lww.com/CM9/C83]. The heatmap indicated an increase in Sarcina in the IL-23i group and an increase in CAG-352 in the IL-12/23i group compared with the other two groups at the genus level [Supplementary Figure 5D,E, http://links.lww.com/CM9/C83].

To understand whether severity of skin lesions affects the changes in intestinal flora composition and abundance in psoriasis patients, we divided all patients into four different disease severity groups according to psoriasis area and severity index (PASI) score: mild (PASI <3; n = 9), moderate (3 ≤PASI <10; n = 36), severe (10 ≤PASI <30; n = 109), and extremely severe (PASI ≥30; n = 27). The relative abundance of Enterococcus (P = 0.0158) was higher in the mild and moderate groups, while the relative abundance of Anaerostipes (P = 0.0277) and Lachnospira (P = 0.0083) was higher in the severe and extremely severe groups [Figure 1A–C]. There was no significant difference in F/B values between the four groups [Figure 1D]. At the genus level, Enterococcus were enriched in the mild group, TM7x (a human oral Saccharibacteria isolate) in the moderate group, Anaerostipes in the severe group, and Lachnospira in the extremely severe group at the genus level [Supplementary Figure 6A,B, http://links.lww.com/CM9/C83]. The heatmap indicated Enterococcus increased in both the mild and moderate groups, while Akkermansia and Lactobacillus decreased in the mild group compared with the others [Figure 1E,F].

Figure 1.

(A–C) Comparison of bacterial abundance at the genus level among PASI <3 (n = 9), 3 ≤PASI <10 (n = 36), 10 ≤PASI <30 (n = 109), and PASI ≥30 (n = 27) groups. The Kruskal–Wallis H test was applied. ^*P-value <0.05, ^†P <0.01 represent significant differences. (D) F/B ratio among PASI <3, 3 ≤PASI <10, 10 ≤PASI <30, and PASI ≥30 groups. (E,F) Heat map based on the abundance ranks of the phylum and genus level. Red and blue indicate high and low abundance, respectively. LDA: Linear discriminant analysis; LefSe: Linear discriminant analysis effect size; ns: Not significant; PASI: Psoriasis area and severity index.

PASI improvement rate (PASI IR) was calculated according to the patients’ skin lesions before treatment with biological agents and at the sampling time. The patients were divided into four groups: PASI IR <50 (n = 7), 50 ≤PASI IR <70 (n = 11), 70 ≤PASI IR <90 (n = 32), and PASI IR ≥90 (n = 69). The relative abundance of Butyricicoccus (P = 0.0179) and Adlercreutzia (P = 0.0191) decreased in turn in four groups [Supplementary Figure 7A,B, http://links.lww.com/CM9/C83]. There was no significant difference in F/B values between the four groups [Supplementary Figure 7C, http://links.lww.com/CM9/C83]. On genus level, Adlercreutzia and Butyricicoccus were enriched in the PASI IR <50 group, Brevundimonas was enriched in the 50 ≤PASI IR <70 group, and NK4A214 group was enriched in the 70 ≤PASI IR <90 group [Supplementary Figure 7D,E, http://links.lww.com/CM9/C83]. The microbiota communities of the four different PASI improvement groups were compared based on the relative abundance ranks of the phylum and genus levels. Heatmap showed that Lactobacillus decreased in the PASI IR <50 group, Akkermansia decreased in both PASI IR <50 and PASI IR ≥90 groups, and increased in the 50 ≤PASI IR <70 group [Supplementary Figure 7F,G, http://links.lww.com/CM9/C83].

In patients with BMI ≥25, the relative abundances of Christensenellaceae_R-7 group (P = 0.0368) and Coprococcus (P = 0.0193) increased with biological treatment [Supplementary Figure 8A,E, http://links.lww.com/CM9/C83]. The relative abundance of Coprococcus (P = 0.0077) increased in male patients with biological treatment, but that of Enterococcus (P = 0.0292) decreased in female patients with biological treatment [Supplementary Figure 8B, F, http://links.lww.com/CM9/C83]. In patients with biological treatment and comorbidities, the relative abundance of Coprococcus (P = 0.0312) was increased, while that of Enterococcus (P = 0.0008) was decreased [Supplementary Figure 8C, G, http://links.lww.com/CM9/C83]. Smoking patients with biological treatment had an increased relative abundance of Coprococcus (P = 0.0420) [Supplementary Figure 8D, http://links.lww.com/CM9/C83]. Moreover, the relative abundance of Prevotella (P = 0.0291) decreased in bio-treated patients <40 years old [Supplementary Figure 8H, http://links.lww.com/CM9/C83]. We found that patients with IL-12/23i treatment for less than 6 months showed the most significant increase in the relative abundance of Ruminococcus (P = 0.0443), and patients with IL-23i treatment for >6 months showed the most significant increase in the relative abundance of Butyricimonas (P = 0.0469) [Supplementary Figure 8I,J, http://links.lww.com/CM9/C83]. The Redundancy analysis/Canonical Correlation Analysis (RDA/CAA) environmental factor analysis are shown in Supplementary Figure 9, http://links.lww.com/CM9/C83. The 16s rRNA sequencing data of this article were deposited in GenBank (https://www.ncbi.nlm.nih.gov/bioproject/PRJNA1014805).

Adlercreutzia is an anti-inflammatory bacterium, and Butyricicoccus is a butyrogenic bacterium. SCFAs play an anti-inflammatory role by down-regulating inflammatory cytokines such as IL-6 and IL-8.^[4] In this study, as the PASI IR score increases, the relative abundance of Adlercreutzia and Butyricicoccus decreases in turn. The presence of butyrogenic bacteria seems to correlate with the intensity of the inflammatory response. A stronger inflammatory response in psoriasis seems linked to a higher abundance of anti-inflammatory gut bacteria, suggesting that gut microbiota could indicate the severity of inflammation.

Our findings revealed that the gut microbiota of psoriasis patients with different lesion severity had different characteristics, and biological treatment changed the composition and structure of intestinal flora. There was a certain correlation between PASI IR and butyrogenic bacteria in psoriasis patients. Butyrogenic bacteria produce SCFAs with anti-inflammatory properties and modulate immune cell function in the intestine and other tissues. They might play a role in the intestinal immunoinflammatory response of psoriasis and could serve as a biomarker for its severity. However, further research into their mechanisms with larger sample sizes is necessary to strengthen the findings. In the course of psoriasis biological treatment, adjusting diet and lifestyle habits and paying attention to the increase and decrease of beneficial biomarkers of intestinal flora can contribute to disease prognosis and individualized treatment.

Acknowledgment

The authors are grateful to Ying Chen for her kindly help in this study.

Funding

This work was supported by grants from the National Natural Science Foundation of China (No. NSFC 82103733) and Science and Technology Research Program of Chongqing Municipal Education Commission (No. KJQN202100412).

Conflicts of interest

None.