Work

Development of an IL-6- and Th1/Treg-based nomogram for identifying low bone mass in patients with rheumatoid arthritis

Year: 2026

Type: article

Abstract: Objective: This study aimed to analyze clinical and immunological characteristics in patients with rheumatoid arthritis (RA) according to bone loss status, integrate traditional risk factors with immune-inflammatory indicators, and construct and validate an individualized nomogram for estimating the probability of concurrent BMD-defined low bone mass in patients with RA. Methods: A total of 442 patients with RA were enrolled and divided into normal bone mass (RA-N, n=219) and low bone mass (RA-LBM, n=223) groups based on bone mineral density values. Data on the clinical characteristics, laboratory parameters, peripheral blood lymphocyte subsets, and cytokine levels were collected for both groups. The Benjamini-Hochberg (BH) method was applied for multiple testing correction. Univariate logistic regression was used to screen relevant variables, and the random forest algorithm (mean decrease Gini importance ranking and out-of-bag error analysis) was employed to determine the optimal combination of variables. Multivariate logistic regression was performed to identify factors independently associated with low bone mass, based on which a nomogram for identifying concurrent low bone mass was constructed. The discriminative ability of the model was evaluated using the receiver operating characteristic curve, calibration was assessed using the calibration curve and Hosmer-Lemeshow test, and clinical net benefit was evaluated using decision curve analysis. Results: Patients in the RA-LBM group exhibited significantly higher values for age, disease duration, erythrocyte sedimentation rate, aspartate aminotransferase/alanine aminotransferase ratio, and alkaline phosphatase, while body mass index (BMI), red blood cell count, hemoglobin levels, albumin levels, and albumin/globulin ratio were significantly lower than those in the RA-N group (BH-adjusted P<0.05). Immunological analysis revealed that interleukin-6 (IL-6) levels were significantly elevated in the RA-LBM group (BH-adjusted P<0.001), and the T helper 1 cell/regulatory T cell (Th1/Treg) ratio was also significantly increased (BH-adjusted P = 0.016). Random forest analysis followed by multivariate logistic regression ultimately identified age, Th1/Treg ratio, BMI, IL-6, albumin, and uric acid levels as factors independently associated with low bone mass in patients with RA (BH-adjusted P<0.05). Model 1 constructed based on these six variables achieved an area under the curve (AUC) of 0.906 (95% CI: 0.877-0.935), which was significantly higher than that of model 2 containing only traditional risk factors (AUC = 0.861, P = 0.001). The AUCs of model 1 in the training and validation sets were 0.906 and 0.905, respectively. The Hosmer-Lemeshow test indicated good model calibration. Decision curve analysis showed that the model provided a higher net benefit than the treat-all and treat-none strategies across a range of threshold probabilities. Conclusion: This study integrates immune-inflammatory indicators, such as IL-6 and Th1/Treg ratio, with traditional risk factors to construct a nomogram for identifying concurrent BMD-defined low bone mass in patients with RA. The model showed favorable discrimination and calibration and may provide a useful tool for estimating the probability of concurrent low bone mass in this population.

Source: Frontiers in Immunology

DOI: 10.3389/fimmu.2026.1952321