Evaluation of vaccine immunogenicity is a key step in the clinical development of vaccines; the T-cell responses to vaccine candidates are typically assessed by intracellular cytokine staining using flow cytometry. Conventional approaches for analyzing T-cell responses do not take into consideration the relationship between the specific response and the background response and can compromise statistical performance, particularly in terms of the type-I error rate and statistical power. We propose a new modeling approach that considers all measured data and is more flexible than conventional methods, which yields more detailed results (vaccine effect on the non-stimulated response and non-stimulated response effect on stimulated response) and enables accurate interpretation of vaccine-induced response.