A Competing Nomogram to Predict Survival Outcomes in Invasive Micropapillary Breast Cancer

Background: Although it is widely accepted that invasive micropapillary carcinoma (IMPC) presents more aggressive behavior and has a higher aggressive behavior, the prognosis of IMPC compared with invasive ductal carcinoma (IDC) remains controversial. We conducted this study to explore gene expression profiles of IMPC and establish a competing nomogram that predicts the survival outcomes across these two groups of patients. Methods: Data from the Surveillance, Epidemiology, and End Results (SEER) database were reviewed. Propensity score matching (PSM) was used to adjust for potential baseline confounding between IMPC and IDC group. The Kaplan-Meier method was used to calculate the occurrence of overall mortality. The Gray method was used to estimate the rate of breast cancer specific death (BCSD). A competing regression model was used to evaluate factors associated with BCSD. A nomogram based on the competing risk regression model was established to predict individual outcomes. IMPC-specific gene expression profiles were explored using microarrays data from the Gene Expression Omnibus (GEO) database. Gene ontology (GO) and Kyoto Encyclopedia of Genes and Genomes pathway (KEGG) enrichment analyses were performed. Results: In this study, 330786 (99.62%) patients with IDC 1247 (0.38%) patients with IMPC were included. Patients with IMPC had more lymph node involvement and a larger tumor size compared with those with IDC. After PSM, many distributional differences were eliminated, showing that the IMPC and IDC group were more similar. Patients with IMPC had a favorable prognosis with statistical significance compared with patients with IDC (overall mortality HR = 0.68; 95%CI, 0.53-0.86; P = 0.002). Based on Gray method, patients with IMPC had a favorable prognosis with significant statistical significance compared with patients with IDC (BCSD SHR = 0.64; 95%CI, 0.47-0.88; P = 0.006). Multivariate analysis based on competing risk model demonstrated that IMPC was a favorable independent factor for BCSD. The nomogram could accurately predict BCSD with a high internal and external validated C-index (0.835, 0.818 respectively). A total of 53 upregulated differentially expressed genes (DEGs) and 40 downregulated DEGs of IMPC was identified. The GO analysis results showed that downregulated DEGs were significantly enriched in extracellular structure organization, extracellular matrix, cell-substrate adhesion junction. KEGG analysis of selective gene sets shows that downregulated DEGs significantly enriched for processes related to carbon metabolism, Rap1 signaling pathway. Conclusion: In the current study, IMPC accounted for 0.38% of the entire cohort. IMPC was found to be a favorable independent prognostic factor. The present study identified gene expression profiles and signal pathways of IMPC. The developed nomogram can help the oncologists to predict individual outcomes more accurately.