Patients who achieve a tumor pathologic complete response (pCR) after neoadjuvant chemotherapy (NAC) have better outcomes than patients with residual tumor. However, tumors still recur in the pCR patients. Therefore, we aim to explore factors associated with tumor recurrence in this patient population.
A total of 1,913 patients diagnosed with breast cancer between 1995 and 2020 and received NAC were included in this analysis. Clinicopathological data of the patients were retrospectively collected. We used Cox regression analysis to assess the associations of clinicopathological factors with patients’ outcome. Proteomic study of tumors was applied to identify differentially expressed proteins (DEPs) between tumors from the pCR patients with tumor recurrence and tumors from those without tumor recurrence. PPI network analysis of the corresponding genes of DEPs was used to identify the hub genes. The prognostic value of the corresponding genes of DEPs was evaluated using two online databases, Kaplan-Meier Plotter and bc-GenExMiner. The genes that were significantly associated with patients’ survival in both databases, as well as being identified as hub genes, were considered as potential prognostic markers for pCR patients. Publicly available data from Gene Expression Omnibus (GEO) was used to verify the prognostic value of the identified marker.
Among the 1,913 included patients, 420 had tumor pCR. The median follow-up for the pCR patients was 32.6 months (IQR, 16.3-55.5). Overall estimated 5-year risk of tumor recurrence for the pCR patients was 11%. Multivariable analysis showed that a higher pre-NAC clinical T stage and N stage were independent predictors for increased risk of tumor recurrence (hazard ratio [HR] 2.57, 95% confidence interval [CI] 1.01-6.51, P=0.047 for clinical T stage and HR 3.48, 95%CI 1.37-8.83, P=0.009 for clinical N stage). NAC regimens, the type of breast and axillary surgery, and adjuvant chemotherapy were not associated with tumor recurrence. Finally, aldehyde dehydrogenase (ALDH) 3A2 was identified by the proteomic study and was verified as a potential predictor for tumor recurrence in the pCR patients (with a median follow up of 3.78 years for dataset GSE32603 and 2.74 years for dataset GSE25066 from GEO, tumor recurrence rate: low versus high expression, 20.7% versus 4.5% [data from GSE32603]; 10.9% versus 0% [data from GSE25066]).
Clinical T stage, clinical N stage and tumor expression of ALDH3A2 were potential markers for predicting tumor recurrence in the pCR patients after NAC.