Crohn’s disease (CD) is a chronic inflammatory gastrointestinal disease with recurrence, progression, and disability (1). CD was first described in 1932, and the incidence is rising now, but the pathogenesis is still poorly understood (2). Existing studies have shown that CD is related to genetic factors, triggered by environmental factors, and associated with immune disorders (i.e., imbalance of effector and regulatory T cells and cytokines, and migration and retention of leukocytes) (3). Patients with CD were divided into three groups by disease behavior (non-stenosis and non-penetrating, stenosis, and penetrating) according to the Montreal classification. Approximately 50% of patients with CD experienced progression from inflammation to fibrosis, which leads to intestinal stenosis and even bowel obstruction (4). Intestinal strictures in patients with CD often require surgery and seriously affect patients’ quality of life (4). The medications used to treat CD include mesalazine, methotrexate, thiopurines, and biologic therapies, such as antibodies to tumor necrosis factor alpha (TNF-α), interleukin (IL)-12/23, and integrin α4β7 (1, 5). Biologic therapies have benefited patients with CD, but there is still a therapeutic ceiling and some patients do not respond to some biologic therapies.

Ustekinumab, human monoclonal IL-12/23 p40 antibody, is a new biological agent and is approved for the treatment of patients with moderate-to-severe CD (6). Inflammatory changes in CD are related to an imbalance between Th1, Th17, and Treg cells. Moreover, IL-12 is responsible for the differentiation of naive T helper cells into Th1 cells and IL-23 is important for the proliferation of Th17 cells (7). Many studies have shown the efficacy of ustekinumab in patients with CD (8–10). Treatment response includes clinical response, biological response, and endoscopic healing. Clinical remission is a short-term goal, while endoscopic remission and mucosal healing are long-term goals (11). However, up to 30% of patients do not respond to initial treatment (12). Therefore, finding factors that can predict which patients will respond to ustekinumab and which will not is essential in expediting effective patient treatment.

Currently, no globally accepted predictive factors have been identified to determine if a patient will be a responder or non-responder with ustekinumab treatment. We used datasets from the public Gene Expression Omnibus (GEO) database to analyze gene alterations between the responders and non-responders with ustekinumab treatment. In addition, intestinal fibrosis seriously affects the quality of life of patients. Currently, there are no effective drugs for treating intestinal fibrosis. One study has shown that the intestinal wall thickness in patients with CD significantly improved after treatment with infliximab and ustekinumab; however, the shear wave velocity index only in the ustekinumab group significantly decreased after treatment (13). Shear wave velocity index is a measurement that can reflect the hardness and the fibrosis degree of a tissue. It is measured by intestinal ultrasound to evaluate the scissoring speed of a shear wave induced by an acoustic radiation force impulse (13). Does ustekinumab have an impact on fibrosis? We wanted to elucidate the impact of ustekinumab on fibrosis by analyzing publicly available datasets before and after treatment with ustekinumab.

CD is a progressive disease that can lead to the formation of complications such as abscesses and strictures. This situation seriously influences patient’s quality of life. Currently, the therapeutic medication for moderate–severe CD includes corticosteroids, immunosuppressants, and biologicals (1). The introduction of monoclonal antibodies against TNF-α (anti-TNFs) has significantly improved the treatment of CD. However, a considerable proportion of patients with CD either failed to respond or experienced a loss of response to these agents over time (27).

Ustekinumab is a new monoclonal antibody that binds with high affinity to the p40 subunit of human IL-12 and IL-23. IL-12 and IL-23 are produced by dendritic cells and macrophages (28). These two cytokines mainly affect Th1 and Th17 cells, which are important in the pathogenesis of CD. IL-12 and IL-23 can also affect NK cells, ILC1 cells, and ILC3 cells (28). Ustekinumab has been shown to be effective in the treatment of CD (8–10). However, up to 30% of patients with CD have no response to initial treatment (29). The immune response in the gut is complex and the prediction of response to treatment is helpful in the selection of therapy (30). However, there are limited studies on the predictors of treatment response to ustekinumab.

We downloaded three GEO datasets of ustekinumab treatment in CD. Firstly, we analyzed the DEGs in blood samples (GSE207465), ileal samples (GSE112366), and rectal samples (GSE207022), comparing ustekinumab responders and non-responders. Responder prediction analysis was performed on samples at week 0 (prior to drug administration). Response status was identified at week 8 post-ustekinumab treatment for ileal samples by clinical response (a decrease of 100 points from their baseline value or a value <150 by CDAI) and rectal samples by endoscopy examination of mucous membrane recovery. MUC1, DUOX2, LCN2, and PDZK1IP1 were identified as hub genes in the difference between ustekinumab responders and non-responders in ileal samples, with all genes having significantly lower expression in responders. Rectal samples were then used to confirm our findings in the ileal samples. We observed that MUC1, LCN2, and PDZK1IP1 were also lower in the response group compared with the non-response group. The RF prediction model based on these three genes and AUC–ROC analysis was then performed and indicated that MUC1, LCN2, and PDZK1IP1 have a good performance in predicting response to ustekinumab. These results suggested that MUC1, LCN2, and PDZK1IP1 may be involved in treatment response to ustekinumab and could serve as a predictive test to identify which patients will respond to ustekinumab treatment, saving patients’ time and money, and preventing patients from suffering.

MUC1 is a transmembrane mucin glycoprotein that is expressed on the apical surface of mucosal epithelial cells and hematopoietic cells (31). MUC1 expression can be modulated by inflammatory cytokines such as interferon γ (IFN-γ) and TNF-α and is related to the NF-κB pathway (32, 33). MUC1 has a protective effect on the epithelium; however, recurrent inflammation has been shown to increase the level of the hypo-glycosylated form of MUC1 (32). This form of MUC1 can increase chemotaxis of innate inflammatory cells, driving increased immune cell recruitment and inflammation. Anti-TNF-α therapy might be effective in lowering the expression of pro-inflammatory MUC1 (32). The study also shows that continued high-level expression of MUC1 may be an early biomarker for resistance to anti-TNF-α therapy (34). In addition, increased levels of MUC1 have been shown to be associated with more severe endoscopic recurrence scores (34). In our study, we found that the patients who respond to ustekinumab treatment had lower expression of MUC1 in both the ileum and rectum. We postulate that low MUC1 levels may indicate lower baseline inflammation and less recurrence, which may account for the susceptibility to treatment response in this subset of patients and may serve as a potential route for future investigations.

LCN2, also called neutrophil gelatinase–associated lipocalin (NGAL), is a potent bacteriostatic glycoprotein stored in neutrophil granules and released at sites of inflammation (35). Previous research also observed that LCN2 may be regarded as a disease activity marker of ulcerative colitis (36). The cytokines IL-17A and IL-22 are secreted by Th17 cells and induce the activation of transcription factor NF-κB, which is required for LCN2 transcription. Earlier studies have shown that LCN2 is significantly elevated in CD (37, 38). In our study, we saw that the expression of LCN2 was lower in patients who responded to ustekinumab treatment and may predict the treatment response to ustekinumab. The possible mechanism of LCN2 expression affecting ustekinumab treatment may be related to the effect of ustekinumab on Th17 cells (39).

PDZK1IP1 is found to be over-expressed in patients with CD compared to healthy patients (40). Royce et al. demonstrated a strong correlation between PDZK1IP1 expression and inflammatory cytokines, including TNF-α, IFN-γ, IL-6, and IFN-β. PDZK1IP1 is also related to the NF-κB and STAT3 pathways (41). PDZK1IP1 stimulates the sodium-dependent uptake of mannose and glucose through the regulation of the sodium-glucose linked transporter (SGLT) family (41). Moreover, PDZK1IP1 is known to induce the differentiation of monocytes to dendritic cells and regulates the immune microenvironment (42). Thus, lower expression of PDZK1IP1 in ustekinumab treatment responders than non-responders in our study may be associated with different immune microenvironments.

Current studies have shown that the neutrophil–lymphocyte ratio (NLR), which can reflect the status of inflammation in a disease, can predict the loss of response of infliximab therapy (43). The results in our study also showed that the ileum of responders contained more Tregs than non-responders. Patients had less neutrophils, less M1 macrophages, and more Tregs in the ileum after ustekinumab treatment. In addition, MUC1, LCN2, and PDZK1IP1 were mainly expressed in neutrophils and lymphocytes. In the GO analysis, these hub genes we identified were related to DNA damage, ferroptosis, and oxidative stress, which is related to inflammation, potentially indicating how the hub genes mediate the therapeutic response. Nowadays, the target of CD treatment has shifted from clinical response to mucous healing (11). MUC1, LCN2, and PDZK1IP1 had a good performance in predicting ustekinumab responders not only for clinical response (GSE112366) but also in mucous healing (GSE 207022).

We wanted to figure out the relationship between the expression changes in the ileum, rectum, and blood. We found that ITGA4 is changed both in the blood and in intestinal tissues. IL18 is lower in blood while IL18R1 is lower in intestinal tissues. ITGA4 is a well-known α4 integrin that is expressed on lymphocytes and is related to lymphocyte trafficking into the intestine (44). In our study, we analyzed the relative gene expression of ITGA4 and IL18R1 in immune cell populations. ITGA4 is related to T cells in the gut and neutrophils in the blood. A study showed that ITGA4 could also be expressed on the surface of neutrophils (45). Neutrophils and lymphocytes play an important role in the treatment response of ustekinumab (46). IL‐18, a member of the IL‐1 family, is similar to IL‐1β (47). In homeostatic conditions, IL‐18 is protective, but in pathological states, it is involved not only in the activation of Th1 and NK cells, but also in the activation of Th2, IL‐17‐producing γδT cells, and macrophages (47). In addition, it is associated with the TLR-MyD88 and NF-κB pathways. Further studies should be done to investigate the expression of IL-18 in patients with ustekinumab.

Intestinal fibrosis is an important complication of CD (4), the pathogenesis of the fibrosis is still unclear, and it may be related to some immune cells, such as Th17 cells (4). It is well known that ustekinumab can affect Th17 and Th1 cells, but there are few studies on whether ustekinumab has an anti-fibrotic effect. We wanted to explore the effect of ustekinumab on inflammation and fibrosis. In an inflammation-related analysis, we found that the IL-17 signaling pathway, NF-kappa B signaling pathway, cytokine–cytokine receptor interaction, Toll-like receptor signaling pathway, and TNF signaling pathway were changed after treatment with ustekinumab, as well as some inflammatory genes like IL1B, CXCL1, GPR109B, IL8, MMP1, MMP3, and S100A8. Some hub genes are associated with the NF-κB pathway, as we mentioned previously; thus, the NF-κB pathway may play an important role in the treatment response of ustekinumab. In a fibrosis-related analysis, we found that pathways like the ECM–receptor interaction (p = 0.045) and the PI3K-AKT pathway (p = 0.024), which were associated with fibrosis, were changed after treatment. The fibrosis module genes were investigated and we found that an extracellular matrix component, COL4A1, was decreased after treatment, and other fibrosis-related genes (TUBB6, IFITM2, SERPING1, DRAM1, NAMPT, MMP1, ZEB2, ICAM1, PFKFB3, and ACTA2) were also decreased after ustekinumab treatment. Ustekinumab may have an effect on alleviating fibrosis.

The limitation of our study is that the expression of hub genes was not validated in new patients but will be verified in the future. Moreover, our analysis of the effect of ustekinumab on fibrosis is only at the genetic level and will need to be further validated by cohort studies in the future. However, we hope that our study can provide a basis for further research.

In conclusion, MUC1, LCN2, and PDZK1IP1 are the hub genes in gut associated with ustekinumab response. The changes in the expression of ITGA4 and IL18/IL18R1 both in blood and in the gut might play an important role in the response to ustekinumab. Ustekinumab may have an impact on fibrosis.

The original contributions presented in the study are included in the article/Supplementary Material. Further inquiries can be directed to the corresponding author.

YX: Formal Analysis, Writing – original draft. SW: Visualization, Writing – original draft. ZY: Software, Validation, Writing – original draft. HZ: Writing – review & editing.