With the rapid expansion of targeted treatment options for UC, one of the primary goals of UC management is to achieve precision therapy. Predicting the response to biologics, such as VDZ, has become a major focus of current research. In this study, we retrospectively analyzed the clinical data and colonic MMED of UC patients treated with VDZ at a tertiary care hospital to identify predictive factors for VDZ efficacy. The results demonstrated that higher MMED level was independent risk factor for treatment failure after 14 weeks of VDZ therapy. The combination of MMED and CRP proved to be highly accurate in predicting the therapeutic response to VDZ in UC patients.

In our study, 84 patients with moderate-to-severe UC were included, and the clinical response rate to VDZ was 69.05%. Our findings are consistent with a real-world study conducted in China involving 64 patients with moderate-to-severe UC treated with VDZ, which reported that 73.4% of patients achieved a clinical response after 14 weeks of treatment [7].

Eosinophils are a type of leukocyte that constitutes a component of the innate immune system and are primarily found in mucosal tissues, particularly in the gastrointestinal tract. These cells are highly sensitive to their environment and can respond to pathogen-associated molecular patterns (PAMPs) and damage-associated molecular patterns (DAMPs) during the immune response. This suggests that eosinophils play a crucial role in responding to pathogens and tissue damage resulting from focal infections. Activated eosinophils release a variety of biological substances, including granule proteins, cytokines, chemokines, enzymes, and growth factors, which mediate the diverse biological activities of eosinophils in infection and inflammation [20]. In patients with active UC , a marked increase in the number of eosinophils in the colorectal mucosa has been observed [21]. The Nancy scoring system, which is used to assess the degree of histologic inflammation in UC, includes a description of eosinophils [22]. Furthermore, studies have demonstrated that an increase in mucosal neutrophils, eosinophils, ulcers, and plasma cells is predictive of UC recurrence [23, 24]. Additionally, eosinophils are implicated in the chronic inflammation and fibrosis of the intestinal tract [25]. All of these findings suggest that tissue eosinophils are closely associated with the development and activity of UC. Histological evaluation is increasingly being utilized to predict the risk of recurrence in patients in clinical remission and has become a therapeutic target in various histologic grading systems in several studies [26, 27]. The results of the present study indicated that higher mucosal mean eosinophil density was negatively correlated with moderate-to-severe UC treated with VDZ, independent of peripheral blood eosinophil levels, which is consistent with findings from similar previous studies [12].

CRP is an acute time-phase response protein synthesized by the liver, which is elevated in inflammatory infections, traumas, stress reactions, fever, etc., and enhances phagocyte phagocytosis and activation of complement, etc., during inflammatory reactions, and is one of the commonly used indicators to assess the inflammatory state of the body in clinical practice, and one of the most important biomarkers to assess the inflammatory activity of UC. CRP is one of the most important biomarkers for assessing inflammatory activity in UC and is often used to monitor UC disease activity. CRP testing during treatment predicts the patient’s response to treatment and risk of complications (e.g., hospitalization and surgery) [28]. A population-based study conducted by the IBSEN Study Group demonstrated that in patients with ulcerative colitis, a CRP level above 10 mg/L at 1-year predicted an increased risk of surgery over the following 4 years. In patients with total colonic ulcerative colitis, a CRP level above 23 mg/L at diagnosis predicted an increased risk of colectomy during the first 5 years of follow-up. ALB, which is also synthesized by the liver, is widely used as an indicator of nutritional status and to monitor liver function, and the rate of synthesis is directly affected by the severity of acute infections [29]. Low levels of ALB interfere with immune mechanisms such as humoral and cellular immunity and phagocytosis. phagocytosis. A study showed that lower levels of ALB as well as higher levels of CRP reduced the likelihood of endoscopic remission in IFX-treated patients with IBD [30]. CRP and ALB, referred to as positive and negative acute-phase reactants, respectively, have been recognized in septic patients as a novel inflammation-based score, which, compared with CRP or albumin alone, provides more useful inflammatory status Information [31]. Other studies have shown that the CRP/ALB ratio is strongly correlated with IBD disease activity and is a highly valuable biomarker for assessing disease activity [32, 33]. The present study showed that higher serum CRP and higher CRP/ALB ratio were negatively correlated with the efficacy of VDZ in the treatment of UC, which may be due to the fact that high serum CRP levels and low albumin levels indicate a more severe disease state and poorer nutritional status, which further indicates a more refractory disease state, and thus may be the reason for the poor efficacy of VDZ.

Previous studies in the treatment of moderate-to-severe ulcerative colitis with VDZ have shown that prior treatment with TNF antagonists affects the efficacy of VDZ [34], but this was not observed in the present study, and it was considered that there may have been fewer patients with prior use of TNF antagonists in the present study, which did not reflect a statistical difference.

In order to further promote individualized treatment, save treatment costs, improve the efficiency of VDZ in the treatment of UC, and assist doctors and patients to jointly develop treatment strategies, we produced a prediction model for the 14-week efficacy of VDZ in the treatment of patients with moderate-to-severe UC and plotted columnar line graphs for clinical reference. Logistic regression analysis was used in this study, which was mainly used to identify risk factors for VDZ treatment failure and to predict and assess VDZ efficacy. This model is effective in controlling confounding factors when quantitatively analyzing the relationship between study variables and study outcomes. The model was internally validated using ten-fold cross-validation. By evaluating the adjusted consistency index (C-index) and constructing calibration curves, we found that the model performed well in terms of differentiation and calibration. In addition, the results of the Hosmer–Lemes how test indicated that the model was well fitted.

This study offers several strengths: (1) The analysis integrates clinical indicators, laboratory tests, and pathological indicators, all of which are objective test results and scales, ensuring high accuracy and practicality. Furthermore, eosinophil density in colonic tissues has been verified for inter-observer consistency [18]; (2) Currently, there are few studies on predicting the efficacy of biologics (including VDZ) for Chinese UC populations. This study is the first to establish an efficacy prediction model for VDZ in Chinese UC patients, complete with a clinical reference nomogram; (3) To ensure accuracy, all patients who used steroids within 4 weeks prior to biopsy were excluded to avoid potential bias from steroid’s effects on eosinophil counts in intestinal tissues. However, the study has certain limitations. First, the sample size is small, as VDZ has only recently been introduced in the Chinese market, with health insurance coverage available for just a year. This has limited the patient pool, with data derived from a single center and stringent inclusion criteria. Second, although the established prediction model demonstrates good differentiation and calibration, it has only undergone internal validation due to the limited sample size and lacks external validation. This limitation impacts the model’s reliability and generalizability. A multicenter study with a larger sample size would provide a solid data foundation to further validate and improve the model, ultimately enhancing its predictive accuracy and stability. Third, in clinical practice, most biopsies are typically performed during routine monitoring. It may be acknowledged that the 6-month biopsy collection window for eosinophils may introduce potential confounding factors due to concomitant treatments. To address this limitation, future prospective studies we will optimize the timing of biopsies, restricting the biopsy collection period to within 2 weeks prior to treatment.