MOTIVATION: High-dimensional mass cytometry (CyTOF), which provides both cellular signatures and inter-cluster interactions like the antagonism between immune activation and suppression, and the pro-inflammatory synergy, sheds light on the cellular and molecular basis of disease pathogenesis. However, revealing the aberrance of inter-cluster communication networks in CyTOF datasets remains a significant challenge. RESULTS: Here, we developed Sample Classification and direct Association Network among Cell clusters (SCANCell) that quantifies the direct association (DA) network of cell clusters. SCANCell was applied to profile inter-cluster interaction patterns of a well-recruited systemic lupus erythematosus (SLE) cohort, including 8 healthy controls, 10 active SLE patients (APs) and 8 remission SLE patients (RPs). SCANCell identified decreased inter-cluster interactions of CD8+ T cells in APs compared with RPs, and enhanced DA of CD8+ T cells after stimulation with immunostimulatory cytokine interleukin-2 in vitro. These discoveries prove that SCANCell can uncover pathology- and drug stimulation-associated inter-cluster interactions, which potentially benefits understanding of pathogenesis and novel therapeutic strategies. AVAILABILITY AND IMPLEMENTATION: The main processing scripts of SCNACell are available at https://github.com/Lxc417/SCANCell. Other codes for the following data statistics are available from the corresponding author upon request. SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.