Astragali Radix-Curcumae Rhizoma is a classic drug pair mainly used for the treatment of digestive tract-related inflammation and tumors, but the ratio is not fixed in clinical practice. In order to study whether the anti-tumor effect of the drug pair is diffe-rent under different ratios, orthotopic transplantation model of colon cancer was established in mice. Then the principal component analysis(PCA) and cluster analysis(CA) were used to explore the effect of different ratios of the drug pair on the tumor growth and metastasis, and select the optimal ratio of Astragali Radix-Curcumae Rhizoma for anti-colon cancer effect. After administration for 15 days, the body weight of colon cancer mice with the tumor removed, the tumor volume and the number of liver metastases were mea-sured; the pathological changes of tumor tissue and liver tissue were observed by HE staining. At the same time, Western blot method was used to detect the protein expression level of tumor growth-related indicators in tumor tissue(Ki67, HBP1, AFP) and tumor metastasis-related indicators in liver tissue(β-catenin, E-cadherin, vimentin, p53) of the tumor-bearing mice. Subsequently, PCA and CA were used to select the optimal ratio of Astragali Radix-Curcumae Rhizoma for anti-colon cancer effect. The experimental results showed that different ratios of Astragali Radix-Curcumae Rhizoma inhibited tumor growth and metastasis to varying degrees. The ratio at 1∶1 of Astragali Radix-Curcumae Rhizoma had the best inhibitory effect on tumor growth, and the 2∶1 ratio group had the best effect on inhibiting liver metastasis and improving weighed loss. Astragali Radix-Curcumae Rhizoma significantly up-regulated the protein expression of HBP1 in tumor tissue of colon cancer mice, and significantly down-regulated the protein expression of Ki67 and AFP in tumor tissue; meanwhile, Astragali Radix-Curcumae Rhizoma significantly up-regulated the protein expression of E-cadherin in liver tissue of colon cancer mice, and significantly reduced the protein expression of β-catenin, vimentin and p53 in liver tissue. PCA results showed that the first three groups in the Astragali Radix-Curcumae Rhizoma compatibility group that were closer to the sham operation group were in the order of 2∶1, 1∶1 and 3∶2, among which the center distance of the 2∶1 group was the shortest from the sham operation group, indicating that the ratio 2∶1 of Astragali Radix-Curcumae Rhizoma had the best intervention effect on colon cancer in mice, consistent with the commonly used clinical proportion. CA results showed that 11 groups of colon cancer mice were classified into 3 categories: Astragali Radix-Curcumae Rhizoma compatibility group, sham operation group and model group, which was consistent with the theory. The results of this study provide a basis for more effective clinical application of Astragali Radix-Curcumae Rhizoma in the treatment of colon cancer, and provide new ideas for the development of classic drug pairs.