Breast cancer remains the number one cancer threat to women's health worldwide.(16) HR + HER2- patients who are newly diagnosed without lymph node metastasis are generally considered to have a better prognosis. In addition, some studies suggest that patients with primary operable breast cancer and no signs of distant metastasis do not need perioperative metastatic screening program.(17) Notably, our study suggests that 1.3% of HR + HER2-N0 patients are newly diagnosed with DM, which has a poor prognosis and is prone to misdiagnosis in clinical practice. Patients with and without metastasis (stage I and IV) have significantly different treatment options and prognosis. Stage I patients can be preferentially treated with surgery to achieve radical effect. The treatment of stage IV patients is controversial. Some people advocated palliative surgery, while others considered stage IV patients have already lost the chance of surgery and should be treated with chemotherapy. (18)So our study may also help to avoid unnecessary mastectomy in stage IV patients.
As well, this suggests that regional lymph node metastasis do not necessarily precede distant metastases. With the widespread use of circulating tumor cell (CTC) analysis, many scholars believe that DM is caused by hematogenous spread of CTC rather than by lymphatic or direct intracavitary spread.(19) The underlying mechanism of distant metastasis of breast cancer needs to be further investigated.
In this study, we analyzed the differences in clinicopathological features between HR + HER2-N0 patients with and without DM and predicted DM by our nomogram. Race, T stage, tumor location, grade, and PR have been shown to be significantly associated with DM. Swati Sakhuja et al. found that black patients were more likely to have de novo metastasis than white patients, consistent with our study.(20) As T stage increases, the probability of de novo metastasis also increases significantly.(21, 22) Central location of the primary site and high pathological grade have also been shown to be significantly associated with increased risk of de novo metastasis.(23, 24) An interesting finding in this study was that PR negative was significantly associated with de novo metastasis in patients with HR + breast cancer. Studies have shown that Luminal B subtype are more likely to lose PR expression and up-regulate HER2 expression. Loss of PR expression is associated with breast cancer progression.(25) This confirms that Luminal B subtype is more aggressive than Luminal A subtype.
Our study found that elderly patients had worse prognosis than younger patients, which was also confirmed in the study of Han and Purushotham et al.(26, 27) Many age-related factors may play an important role in metastasis, including accumulation of DNA damage, immune responses, chronic inflammation, changes in hormone levels, etc.(28, 29) It may also be that younger patients are more likely to receive systematic treatment, while older patients are undertreated. In addition, black race,(2, 30) high pathological grade, ER, PR negative, and metastatic site are also important factors affecting BCSS in patients with N0M1, which have been shown to be important for predicting survival in recent years. (31, 32)
Whether surgery for primary tumors provides a survival benefit for patients with metastatic breast cancer remains controversial. Many studies, including our study, have demonstrated that surgical resection of primary tumors improves survival by reducing the overall tumor burden and the source of metastatic spread. (33–35) On the contrary, some studies suggest that surgery has no survival benefit for advanced patients. The survival benefit may be due to confounders such as timing of surgery, type of systemic treatment administered prior to surgery, and coding errors were investigated.(36–38) The results of large prospective trials are needed to confirm the benefits of surgery. We used these 7 factors to construct the nomogram, and the 1-year, 3-year and 5-year AUC values were 0.74, 0.77, and 0.78 in the development cohort, and 0.66, 0.73, and 0.80 in the validation cohort, which proved that our model had good predictive ability.
Our prediction model can screen out high-risk patients with distant metastasis by comprehensively evaluating the clinicopathological features of N0 patients. These patients should be given more aggressive imaging to prevent misdiagnosis. In addition, the prediction model of BCSS can be used to predict the 1 -, 3 - and 5-year survival rates of N0M1 patients, so as to provide guidance for clinical decision.
This study has some limitations. Firstly, this study is a retrospective study, which inevitably has selection bias and is not as convincing as a prospective study. Secondly, SEER database does not provide some potential predictors, such as gene mutations, Ki67, and more detailed treatment information, such as endocrine therapy and chemotherapy regimens, which may affect the comprehensiveness of the nomogram. Finally, despite internal validation of our model, there is a lack of external validation to further determine the accuracy of the model.