To evaluate the commutability between HER2 genomic DNA reference materials (RM) and clinical samples, ddPCR and qPCR were used to study the commutability of HER2 genomic DNA RM, which can provide traceable RM with commutability for clinical laboratory testing. 29 clinical samples were randomly measured among the 5 levels of RM by real time quantitative PCR (qPCR) and droplet digital PCR (ddPCR). The averaged copy number ratio of HER2 to RPPH1 was calculated. According to the guideline EP30 of American Society for clinical laboratory standardization (CLSI) and the guideline WS/t356-2011 for Guideline for Evaluation of Matrix Effects and Commutability of the People’s Republic of China, a regression curve was drawn with the result of ddPCR as abscissa and the result of qPCR as ordinate, and the commutability of RM was evaluated by Deming regression method. If the test results of the proposed RM fall within the 95% confidence interval of its predicted, it is considered that the analyzed RM is communtable; if it falls outside the range, it is considered that the analyzed RM is not communtable. In addition, the results were compared with the ratio of copy number concentration when CEP17 gene was used as reference gene. HER2/RPPH1 of the five RMs determined by ddPCR and qPCR were: 1.91, 1.86; 5.70, 4.45; 16.94, 12.21; 22.38, 17.19; 35.38, 28.84, respectively, which fall in the prediction range, indicating the five RMs are communtable. Moreover, this was confirmed by the ratio of HER2/CEP17 determined by the two methods. However, the regression coefficient of HER2/RPPH1 (R² = 0.97) was better than that of HER2/CEP17 (R² = 0.78) and even one group of clinical sample fell outside the 95% confident interval. We used FISH to detect the cell lines used to prepare HER2 genomic DNA RM. The results showed that in the same nucleus, part of chromosome 17 showed HER2 amplification in the short arm, and part of chromosome 17 showed no HER2 amplification. Additionally, CEP17 was amplified in some cells, which indicates it is not suitable to be used as reference gene as this will cause false negative results. This confirms the necessity of using RPPH1 gene as reference gene when diagnostic of HER2. In conclusion, the five RMs are communtable, which can be used for the method validation and quality control in analyzing of HER2 copy number variation in clinical laboratories.