Related Experiment Video
Updated: Nov 2, 2025
Constructing and Visualizing Models using Mime-based Machine-learning Framework
Published on: July 22, 2025
[The Prognosis and Predictive Value of Complement C3 in Patients with Multiple Myeloma]
Yi-Fei Chen1, Yi-Long Lu2, Ying Zhu1
1Department of Hematology, Jiangdu People's Hospital of Yangzhou, Yangzhou 225000, Jiangsu Province, China.
Objective:
To investigate the effect of complement C3 on the prognosis of patients with multiple myeloma (MM), and to establish a predictive model to evaluate the overall survival.
Methods:
Eighty newly diagnosed MM patients were enrolled, and clinical characteristics, such as sex, age, platelet count, white blood cell count, ISS stage, FISH, levels of kappa and lammda chain, complement C3 and C4 were retrospectively analyzed. Cox regression model was used for univariate and multivariate analysis about risk factors that affecting the prognosis of the MM patients. A nomogram based on C3 level was established for predicting the prognosis of MM patients.
Results:
The average age of the MM patients was 63.15±10.41, including 36 males and 44 females. The median overall survival (OS) was 36.3 months, and the median progression-free survival (PFS) was 35.2 months, the 3-year OS rate and PFS rate of the MM patients were 67.5% and 52.5%, respectively. The variants selected by univariate analysis were put into multivariate regression model, the result showed that C3 level ≥0.7 U/L and PLT count <100×109/L were the independent risk factors for OS. Nomogram based on C3 level, PLT count as well as β2-protein level showed an excellent accuracy in estimating prognosis of MM (C-index: 0.775).
Conclusion:
Patients with C3 level≥0.7 U/L or PLT count <100×109/L show poor prognosis. Nomogram based on the two variants can estimate overall survival of MM patients and provide suggestions to clinical decision.
