Growth pattern can be used as a new characteristic to predict malignancy in breast cancer

BACKGROUND: To date, anatomic tumor length is a key criterion for cancer staging and can be used to evaluate the effectiveness of therapies. This article describes growth pattern that can be used as a new characteristic to represent disease burden and tumor features and predict lymphatic metastasis. METHODS: Patients with breast cancer were included in this 10-year (1999–2008) hospital-based multicenter retrospective study. The pathologic length/height ratio was used to illustrate the correlation between tumor features, behaviors and treatments in breast malignancies. The most appropriate ratio was chosen based on the comprehensive evaluation of p value and changing trend of each characteristic. RESULTS: The sample consisted of 4211 women diagnosed with breast cancer. Among them, 2037 patients with complete pathologic length, width and height information were included in the final analysis. There were 2.34 ± 4.77 metastatic lymph nodes for spheroid tumors and 3.21 ± 5.82 for ellipsoid tumors when the cutoff point was 2. In addition, the proportion of ellipsoidal tumors gradually increased from 54.36 to 56.67% in the upper outer quadrant (UOQ) and from 6.7 to 9.03% in the central region with an increase in the cutoff point. The proportion of ER + PR + ellipsoid tumors significantly decreased from 50.1 to 45.35% and that of ER–PR ellipsoid tumors significantly increased from 32.73 to 36.24% with an increase in the cutoff point. Additionally, the best length/weight ratio to distinguish spheroid and ellipsoid tumors was 2. CONCLUSION: This study described for the first time how growth pattern is correlated with tumor malignancy and how it influences the selection of therapeutic strategies for patients. ELECTRONIC SUPPLEMENTARY MATERIAL: The online version of this article (10.1007/s12282-019-01041-7) contains supplementary material, which is available to authorized users.