Related Experiment Video
Updated: Jul 7, 2026
Predictive Immune Modeling of Solid Tumors
Published on: February 25, 2020
[Preliminary study of multivariable model in predicting response to immunosuppressive therapy in patients with
Yun Chen1, Cong-Gao Xu, Nong-Jian Guo
1Medical College of Shandong University, Jinan 250012, China.
Objective:
To evaluate the potential usefulness of a multivariable model in predicting the response to immunosuppressive therapy (IST) in patients with aplastic anemia (AA), and its application to the clinical practice.
Methods:
PB T cells subpopulation and BM T cells intracellular IFN-gamma and IL-4 were serially analyzed by flow cytometry (FCM) before and during treatment. HLA-DRB1 * 1501 phenotype was analyzed by PCR-SSP. The predictive potentials of different parameter combinations for clinical responsiveness were statistically assessed.
Results:
In all evaluated parameters, CD8+ cell intracellular IFN-gamma had the relatively best diagnostic value with sensitivity and specificity of 94.3% and 62.5%, and positive and negative predictive value of 84.6% and 83.3% respectively. Positive CD8+ cell intracellular IFN-gamma plus Tc1/Tc2 < 50 could increase the positive predictive value to 92.3%. A multivariable model consisting of absolute neutrophil count (ANC), BM T cell intracellular IFN-gamma, Tc1/Tc2 ratio and HLA-DRB * 1501 phenotype of the patients was finally established.
Conclusion:
The multivariable model is superior to each of the single parameters in terms of predictive power of IST therapeutic outcome, and its higher accuracy and the clinical application make it potentially useful in practice.
