Survival analysis and nomogram for early-stage occult breast cancer with positive lymph nodes based on the SEER database
Xiaochu Man,Han Xu,Hongxia Wang,Jiuda Zhao,Xuesong Chen,Sha Yin,Qiaorui Tan,Jie Huang,Shujuan Sun,Dongdong Zhou,Fangchao Zheng,Tong Liu,Lihua Song,Baoxuan Zhang,Na Li,Huihui Li
DOI: https://doi.org/10.21037/atm-22-5701
IF: 3.616
2022-12-27
Annals of Translational Medicine
Abstract:Background: Occult breast cancer (OBC) is a rare type of breast cancer, which accounts for 0.3-1.0% of all breast cancers. However, the treatment of OBC remains controversial, especially the local treatment. We aimed to analyze the impact of different treatment and N stage on survival in early-stage OBC patients, and construct a nomogram to predict the prognosis.Methods: The data of patients with early-stage breast cancer were obtained from 17 registries in the Surveillance, Epidemiology, and End Results (SEER) database. Patient characteristics and breast cancer-specific survival (BCSS) were compared among the groups. Cox proportional risk models were used for both the univariate and multivariate analyses. Variables with a P value <0.07 in the univariate analysis were included in the multivariate analysis. The independent prognostic factors were included in the nomogram and validated internally.Results: A total of 492 early-stage OBC patients were randomized at a 7:3 ratio into the training cohort (n=348) and the testing cohort (n=144). N2+ stage patients had a worse prognosis than N1 stage patients (P=0.0051). Triple-negative breast cancer (TNBC) patients had the worst prognosis. Early-stage OBC patients benefited from surgery (P=0.0093) and radiotherapy (P=0.0102), but not chemotherapy (P=0.4030). An analysis of OBC patients with different N stages showed that in terms of treatment, N1 stage patients benefited from surgery (P=0.023), but did not benefit significantly from radiotherapy (P=0.0793), whereas N2+ stage patients benefited from radiotherapy (P=0.0098), but the benefit from surgery was not significant (P=0.1005). In the multivariate analysis, N stage, surgery, and radiotherapy remained statistically significant. Based on the results of the multivariate analysis, we constructed a nomogram for estimating the 3- and 5-year BCSS of OBC patients. The concordance index and the calibration plots show that our nomogram had sufficient accuracy and good coordination.Conclusions: N stage, surgery, and radiotherapy were identified as independent prognostic factors for OBC. We successfully constructed a nomogram using these independent risk factors and demonstrated that it could help predict the 3- and 5-year BCSS of OBC patients. Further data analyses need to be conducted to revise the treatment of early-stage OBC.
oncology,medicine, research & experimental