RESUMO
Background: Tumor microenvironment (TME) status is closely related to breast cancer (BC) prognosis and systemic therapeutic effects. However, to date studies have not considered the interactions of immune and stromal cells at the gene expression level in BC as a whole. Herein, we constructed a predictive model, for adjuvant decision-making, by mining TME molecular expression information related to BC patient prognosis and drug treatment sensitivity. Methods: Clinical information and gene expression profiles were extracted from The Cancer Genome Atlas (TCGA), with patients divided into high- and low-score groups according to immune/stromal scores. TME-related prognostic genes were identified using Kaplan-Meier analysis, functional enrichment analysis, and protein-protein interaction (PPI) networks, and validated in the Gene Expression Omnibus (GEO) database. Least absolute shrinkage and selection operator (LASSO) Cox regression analysis was used to construct and verify a prognostic model based on TME-related genes. In addition, the patients' response to chemotherapy and immunotherapy was assessed by survival outcome and immunohistochemistry (IPS). Immunohistochemistry (IHC) staining laid a solid foundation for exploring the value of novel therapeutic target genes. Results: By dividing patients into low- and high-risk groups, a significant distinction in overall survival was found (p < 0.05). The risk model was independent of multiple clinicopathological parameters and accurately predicted prognosis in BC patients (p < 0.05). The nomogram-integrated risk score had high prediction accuracy and applicability, when compared with simple clinicopathological features. As predicted by the risk model, regardless of the chemotherapy regimen, the survival advantage of the low-risk group was evident in those patients receiving chemotherapy (p < 0.05). However, in patients receiving anthracycline (A) therapy, outcomes were not significantly different when compared with those receiving no-A therapy (p = 0.24), suggesting these patients may omit from A-containing adjuvant chemotherapy. Our risk model also effectively predicted tumor mutation burden (TMB) and immunotherapy efficacy in BC patients (p < 0.05). Conclusion: The prognostic score model based on TME-related genes effectively predicted prognosis and chemotherapy effects in BC patients. The model provides a theoretical basis for novel driver-gene discover in BC and guides the decision-making for the adjuvant treatment of early breast cancer (eBC).
RESUMO
In the NCCN guidelines version 1.2019, aromatase inhibitors (AIs) or tamoxifen (TAM) for 5 years plus ovarian function suppression (OFS) were recommended for premenopausal breast cancer patient who has higher risk of recurrence. The meta-analysis established a comparison of the curative effect of two adjuvant endocrine therapies. In order to obtain randomized controlled trials (RCTs) related to this metaanalysis, PubMed and Embase database were searched systematically in English during May 2019. Two reviewers screened the articles and extracted data based on the criteria recommended by the Cochrane collaboration for evaluating evidence in RCTs. The first outcome was disease-free survival (DFS). Overall survival (OS) was the other endpoint. Hazard ratios (HRs) with 95% confidence intervals (CIs) were pooled utilizing fixed-effect model. The heterogeneity of this study has been described by Cochran's Q and the I2 statistics. Three RCTs which involved 7,203 premenopausal women with breast cancer were available in this meta-analysis. Pooled HRs showed that there was not difference between AIs plus OFS and TAM plus OFS in DFS (HR =0.87, 95% CI: 0.66-1.14, P=0.30). No statistical differences were found in OS between the two adjuvant therapies (HR =1.22, 95% CI: 0.75-1.99, P=0.43). Based on the included studies, there were no statistical differences between AIs plus OFS and TAM plus OFS in DFS and OS.
