Machine Learning-Based Prediction of Treatment Response to Vitamin D3 in Adolescents and Young Adults with Inflammatory Bowel Disease: A Prospective, Open-Label, Alternation-Based Comparative Study.

Background

Chronic inflammatory bowel diseases (IBDs), specifically Crohn's disease and ulcerative colitis, present substantial clinical challenges characterized by a relapsing-remitting course. These conditions often lead to serious complications and significantly impair nutritional status, physical development, and overall quality of life.

Methods

This unicentric, prospective, open-label, alternation-based comparative study investigated the efficacy of vitamin D3 supplementation in adolescents with IBD. Patients aged 15-25 years were randomly assigned to receive either 1000 IU/day or 2000 IU/day of cholecalciferol for 6 months as an adjuvant to standard care. Eighty patients completed the study.

Results

By month 6, remission was achieved in 52.5% of patients receiving 1000 IU/day (Group A) and 87.5% of those receiving 2000 IU/day (Group B). Significant differences were observed between groups regarding disease activity (p = 0.007), C-reactive protein levels (30.17 ± 15.74 mg/L vs. 62.82 ± 37.14 mg/L; p < 0.001), and fecal calprotectin levels.

Conclusions

The machine learning models demonstrated consistent performance in assessing inflammatory indicators, with a weighted average recall exceeding 72% (Random Forest: 82.50%). No adverse effects were reported. Supplementation with 2000 IU/day of vitamin D3 for 6 months was associated with more favorable clinical and biochemical outcomes than 1000 IU/day in this single-center open-label study as an adjunct to standard treatment in adolescents and young adults with IBD. The results suggest that higher doses of vitamin D may be beneficial in achieving remission and reducing systemic inflammation.