Background: Neoadjuvant chemotherapy is currently mainly used for locally advanced breast cancer. However, there is currently no effective way to predict the prognosis of neoadjuvant chemotherapy patients.
Methods: All relevant data before and after chemotherapy were collected and evaluate the efficacy of chemotherapy through RECIST1.1 and pathology. Survival curves were generated by the Kaplan-Meier method and compared using the log-rank test. A nomogram was developed using a binary logisti regression model with a cross validation.
Results: A total of 158 patients were included in our study. The ORR was 64.6% (cCR, 3.8%; cPR, 60.8%). The pCR rates wer 12.0% for breast tumor and axillary nodes. The pCR rates were 17.1% for breast tumor and 30.4% for axillary nodes. Our study found that only differences in HER-2 status could predict different clinical efficacy, and clinical remission (CR+PR) was more likely to be achieved if HER-2 positive than negative (16.7% vs 7.5%, p=0.05). There was no significant difference in the clinical efficacy of neoadjuvant chemotherapy in age, clinical stage, and molecular classification of patients. In a single factor COX proportional hazard model study, it was found that ER positive relative to HR negative(HR 0.479,CI 95% 0.272-0.844, p=0.011), PR positive relative to PR negative (HR 0.423,CI 95% 0.235-0.763, p=0.004), HER-2 positive relative to HER-2 negative (HR 2.011, CI 95% 1.000-4.042, p=0.050), TNBC (HR 2.229, CI 95% 1.189-4.178, p=0.012), and HER-2 (HR 2.808, CI 95% 1.301-6.062, p=0.009) type are more effective than patients with Luminal type breast cancer 0btain a longer survival time, which is considered a protective factor. A nomogram was developed based on the clinical and statistically significant predictors. ROC curve for 3-year survial with area under the curve = 0.731 (95% confidence interval, 0.663-0.828); ROC curve for 5-year survial with area under the curve = 0.743 (95% confidence interval, 0.652-0.834); ROC curve for 10-year survial with area under the curve = 0.726 (95% confidence interval, 0.639-0.812).
Conclusion: In conclusion, the hormone receptor status before chemotherapy can predict the prognosis of patients in advance, and we successfully constructed nomogram, which can help us predict the OS of patients with NAC.