OBJECTIVE： To provide reference for designing new CXCR4 receptor inhibitor. METHODS： The data of structure and active value （IC50） of 38 CXCR4 receptor inhibitors were collected from the literatures. Using Sybyl-X 2.0 software， 3D-QSAR model of CXCR4 receptor inhibitors was established with Topomer CoMFA. The external predictive power of the model was validated with test set for 18 samples. The relationship of R-group substituent group with activity was analyzed according to stereoscopic and electrostatic fields of inhibitors and CXCR4 receptors. Sybyl-X 2.0 software was used for molecular docking， and the key interaction between inhibitor and CXCR4 receptor were analyzed. RESULTS： The cross validation coefficients Q2 of Topomer CoMFA model was 0.735. The fitting validation coefficient R2 was 0.959， and Fisher validation value F was 416.4. After molecular splitting and three dimensional equipotential map analysis， the addition of big volume and positive charge group at the end of R1 group was beneficial to the improvement of the activity of CXCR4 receptor inhibitors. The introduction of substituent group with large volume and negative charge around R2 group was beneficial to the improvement of its activity. The introduction of substituent group with small volume and negative charge around R3 group was beneficial to the improvement of its activity. Molecule docking results showed the inhibitors interaction between CXCR4 receptor inhibitors and key amino acid residues as Asp97， Glu288 and Trp94. CONCLUSIONS： Established Topomer CoMFA model shows good predictability and statistical stability. CXCR4 receptor inhibitors have strong hydrogen bonding role with receptor target protein. 3D-QSAR and molecular docking study can provide reference for the design， reconstruction and R&D of CXCR4 receptor inhibitor molecule.