As one of the most effective treatments for cancer, chemotherapy is the basic treatment for breast cancer and the most important adjuvant treatment for patients after surgery. However, traditional clinicopathological indicators do not accurately predict patients’ risk of recurrence and therefore do not accurately screen patients who need chemotherapy and those who do not. This new study, published in the New England Journal of Medicine, shows that genetic testing can predict the need for chemotherapy in patients with early-stage breast cancer. The promising findings were led by Laura van’t Veer and colleagues at the Helen Diller Family Comprehensive Cancer Center. The initial aim of the study was to assess whether people at low risk of gene expression who were not receiving chemotherapy were at high clinical risk. In this randomized phase 3 study, they enrolled 6,693 women with early-stage breast cancer and identified their genomic risk (using markers for 70-gene) and clinical risk. Women at low clinical and genomic risk did not receive chemotherapy, while women at high clinical and genomic risk required chemotherapy. Patients whose clinical and genomic risks were not known were treated with chemotherapy. The results found that after 5 years, 94.7% of patients with a high clinical but low gene expression risk were alive without chemotherapy, which is only slightly less than 1.5 percentage points of the proportion of the population receiving chemotherapy in the same condition. In this study, all patients had received standard treatment after surgery, including hormone therapy and radiation therapy. This result suggests that the need for such patients to receive chemotherapy can be predicted by detecting markers of 70-gene. The study was reportedly conducted using the MammaPrint system, a high-throughput genetic testing system for breast cancer. Compared to traditional clinical parameters used to determine the risk of breast cancer recurrence, such as mass size, grade, patient age and lymph node status, which are currently commonly used in the United States, MammaPrint is said to help physicians clearly classify all patients into high- and low-risk subgroups, eliminating the uncertainty associated with ambiguous data given by other genomic testing methods. Since this study could significantly reduce the number of breast cancer patients requiring chemotherapy, a large number of patients could avoid the side effects and toxicity of chemotherapy. Therefore, it can further guide personalized medicine.