1. IntroductionBreast cancer (BC) is one of the most common malignancies among women worldwide [1,2]. According to World Health Organization statistics in 2020, 2.3 million women were diagnosed with BC, with a total of 685,000 deaths worldwide. At the end of 2020, 7.8 million women had a diagnosis with the disease in the last 5 years. BC is the most prevalent type in the world [3]. The complex and heterogeneous BC complex, a heterogeneous disease that comprises multiple tumor groups, is associated with different histological patterns, clinical behaviors, and biological characteristics [4]. Triple-negative breast cancer (TNBC) stands out as a highly diverse group of cancers characterized by its aggressive metastatic nature, with poor survival and poor prognosis, representing 15% to 20% of all invasive breast cancers [5,6] and is associated with younger age as a risk factor for developing this subtype [7]. There are several risk factors for developing breast cancer, such as sex, advanced age at menarche [8], family history, reproductive history, number of children, breastfeeding time [9], genetic characteristics [10], race, ethnicity [11], dense breast tissue, certain benign breast conditions, previous chest radiations, radiation therapy [8], anthropometric factors, overweight or obesity [12], menopausal age, menopausal hormone therapy [13], alcohol intake [10], smoking, and physical inactivity [14].TNBC tumors have larger size, nuclear grade, and mitotic activity, generally have lymph node involvement, and are biologically more aggressive [15]. Studies have extensively documented the clinicopathological risk factors of the disease, such as the status of nodules, tumor size, grade, and the rate of monoclonal antibody marker of cell proliferation (ki-67) [16,17]. However, there are no adequate prognostic and predictive biomarkers for use in clinical practice in TNBC [18]. Research has advanced in discovering possible genetic biomarkers that can contribute to the diagnosis, prognosis, and treatment of the disease [19,20]. Currently, these studies involve the search for microRNAs (miRs) that have scientific relevance in their use as molecular biomarkers. miRs are a new class of endogenous, short-molecule, nonprotein-coding, single-stranded RNAs that are 19 to 25 nucleotides in length. Its function is is to modulate the gene expression in eukaryotes by post-transcriptional regulation. These miRs can act as post-transcriptional silencers, inhibiting the translation of target messenger RNAs [21,22,23,24]. With the advancement of research, there is much evidence that indicates the abnormal expression of miRs may be linked to the development and evolution of cancer [25]. Several of these studies have shown that miRs play significant roles in the prognosis of TNBC [26,27,28,29].Despite the extensive original publications depicting the different types of miRs and their possible roles as oncogenes and tumor suppressors, it is necessary to organize these published studies and analyze their results. However, to date, only one study has been carried out using a systematic review and meta-analysis, with 19 articles up to the year 2016 [25]. From 2017 to 2022, there was a significant increase in studies that evaluated the role of miRs in the survival of patients with TNBC, and the relevance of the results needs to be analyzed together. In this sense, this work aims to conduct a meta-analysis of studies on the expression of miRs in the prognosis of TNBC, providing a better understanding of the associations between specific miRs. 4. DiscussionBreast cancer is a heterogeneous disease with different molecular subtypes. TNBC is the second most common subtype of BC in women worldwide [75], accounting for 15–20% of all BC. In recent years, many reports have shown differential expression levels of miRs in BC [76,77]. However, in general, these studies evaluated specific miRs or combinations of multiple miRs for the diagnosis of BC, and the miR expression for each molecular subtype of CM remains poorly explored. Specifically, the TNBC subtype shows distinct histopathological characteristics and clinical behavior featuring an aggressive phenotype [78]. This subtype usually has early and high recurrence rates, resulting in poor survival, as well as the patients not receiving treatment-directed therapy [79]. The absence of targeted therapy is one of the obstacles in the treatment of TNBC, which may partially explain the worse prognosis in patients with TNBC than those with other subtypes of BC [5]. In this context, it becomes interesting to search for specific miRs for the diagnosis of TNBC to direct the treatment and improve the clinical outcome of this disease.Therefore, this systematic review and meta-analysis aimed to verify the role of miRs as predictive biomarkers that can be used in the prognosis of TNBC, thus comparing the effect of a specific miR among the study population. This work provide an up-to-date review of the topic with 43 articles between cohort and case–control studies that examined the role of miRs in the prognosis of TNBC. The pattern of several miRs that appear to be altered in TNBC can be observed, suggesting the possibility of being used as a prognostic tool in the disease [41,56]. For example, the reduced expression of miR-155 predicted poor overall survival in patients with TNBC, while the elevated levels of miR-21, miR-27a/b, miR-210, and miR-454 were associated with shorter overall survival [19,55,80]. Similarly, a decrease in the miR-374a/b/5p expression and an increase in miR-454 levels were correlated with lower disease-free survival [28,36]. Other miRs have been associated with chemoresistance during treatment [42,56,57].miR-155 appears to be a promising prognostic tool in many types of cancer, especially BC. Generally, miR-155 is considered an oncogenic miR that promotes tumor growth, angiogenesis, and BC aggressiveness [81]. However, in patients with TNBC, this miR has the opposite behavior; that is, the overexpression of miR-155 has a protective effect, increasing survival. This may be due to the various molecular mechanisms found in TNBC that may play a role in DNA damage pathways [42,44]. miR-21 was one of the first oncomiRs identified as overexpressed in breast carcinoma. Its function has been correlated with cell survival, proliferation, stimulating invasion, extravasation, and metastases [82], which has been associated with advanced clinical stage, lymph node metastasis, and poor prognosis of the patient [83]. The target of miR-21 may be pro-apoptotic phosphatase and the tensin homolog (PTEN), which promote tumor cell proliferation by inhibiting tumor cell apoptosis [41]. miR-27a/b is one of the significantly increased miRs in TNBC, and this high expression has been associated with poor overall patient survival [42]. miR-27a is an oncogenic microRNA, and its elevated expression is associated with poor overall survival in BC patients, which suggests that miR-27a may be an important biomarker in disease progression; high levels of miR-27a appear to be significantly correlated with tumor size, lymph node metastases, distant metastases, and poor prognosis in patients with BC [60]. The peroxisome proliferator-activated receptor (PPAR) and PTEN signaling in TNBC cells act as tumor suppressors, regulating the cell division cycle gene (CDC27), which is an essential component of the anaphase promoter complex (APC) important in the regulation of the mitotic checkpoint to ensure chromosomal integrity [50,84]. miR-374a has been reported as a “protective” gene in invasive breast cancer. In analyzing the mRNA profile of this miR, researchers observed that an important tumor suppressor inhibitor of cyclin-dependent kinase 2A (CDKN2A) is also a target of miR-374a. Furthermore, it has been associated with better survival in lung cancer due to this protective ability [36]. The in vitro study by Liu Y. et al. [51] showed that miR-374b-5p overexpression reduced tumor cell invasion, supporting the epidemiological findings that it is associated with better outcomes for TNBC [51]. However, a study by Son et al. [85] with in vitro and in vivo models indicated that a high expression of miR-374a-5p promotes tumor progression in TNBC. Therefore, miR-374a-5p may be a potential prognostic marker of TNBC [85].MiR-30a expression was significantly reduced in TNBC, and patients with a low miR-30a expression were associated with a higher histological grade and more lymph node metastases. Furthermore, researchers found that miR-30a curtailed the epithelial-mesenchymal transition (EMT) of TNBC cells. The overexpression of miR-30a increased the expression of the epithelial marker E-cadherin but decreased the expression of the mesenchymal markers N-cadherin and vimentin. In this study, it was shown that the overexpression of miR-30a significantly suppressed the invasion and migration of cancer cells [86]. A reduced expression of miR-34a/34b/34c is strongly associated with tumor progression and a worse prognosis. Furthermore, miR-34c was an independent risk factor for OS in patients with TNBC [73]. However, in a study with 39 patients, it was observed that high levels of miR-34b expression negatively correlated with DFS and OS of patients with TNBC [26]. The high expression of miR-301a was responsible for cell proliferation induced by the cancerous protein phosphatase 2A inhibitor (Cip2a) and cell invasion. MiR-301a levels were overexpressed in tumor tissues compared with adjacent tissues. Furthermore, the level of miR-301a in TNBC cells increased compared with other subtypes [87]. miR-301a level was positively correlated with size, depth of invasion, stage of TNM, and lymph node metastasis. The expression of MiR-301a was an independent prognostic factor for survival in patients with TNBC [68].miR-210 is highly expressed in TNBC, and its overexpression has been associated with a poor prognosis of the disease [88]. Consequently, a low expression of miR-210 has been shown to improve DFS and OS more than a high expression [89]. The miR-210 signal was detected in tumor cells, and in the tumor microenvironment, in a positive region for the pan-leukocyte marker—the common leukocyte antigen (CD45-LCA). miR-210 is hypoxia-regulated and is a target of hypoxia-inducible factor 1 alpha (HIF1-alpha). As hypoxia is an important feature of solid tumors, miR-210 has been shown to be overexpressed in several types of cancer, including breast, lung, head, and neck, pancreatic cancer, or glioblastoma [88]. In addition to being correlated with lymph node metastases, clinical staging, and differentiation, it may act as a promising prognostic tool for breast cancer [89]. miR-454 promoted the proliferation of TNBC and collaborated in the migration and invasion of TNBC cells. Furthermore, it improved the survival of TNBC cells after ionizing radiation. miR-454 inhibited radiation-induced apoptosis in TNBC cells by regulating the expression of caspase 3/7 and the antiapoptotic protein BCL-2. The levels of phosphatase tensin homolog (PTEN) and pAKT in TNBC cells were found to change after the overexpression of miR-454 [90]. A case–control study found that miR-454 upregulated TNBC tissues and cell lines. Therefore, the fall in miR-454 inhibited the progression of TNBC by suppressing cell proliferation and EMT and inducing cell apoptosis, indicating a promising treatment strategy for patients with TNBC [91].The rate of miR-146a in TNBC tissues was significantly higher than its rates in other BC subtypes and is related to larger tumor size and histological stage. In this study, patients with a high miR-146a expression were found to have a lower survival rate than individuals with a low expression. miR-146a is highly expressed in TNBC, leading to a worse prognosis, making miR-146a expression an independent factor that affects the prognosis of patients [66]. However, in another study, the results suggested that the high expression of miR-146a improves overall survival in patients with TNBC and BRCA1 deficiency [72]. miR-128 has been investigated in a series of in vitro and in vivo experiments to investigate the function and mechanism in the development of invasive ductal breast cancer. As a result of this study, we found that a low miR-128 expression was correlated with lower OS and DFS in TNBC, as well as in other subtypes. In addition, miR-128 was able to inhibit glucose metabolism, mitochondrial respiration, and TNBC cell proliferation. These effects were consistent with the direct inhibition of the insulin receptor miR-128 and the insulin receptor substrate 1 [71]. MiR-181 expression in TNBC tissue samples was found to increase compared with nontumor tissues. A high expression of miR-181a has been associated with a higher tumor grade, lymph node metastasis, and chemoresistance [10]. Furthermore, miR-181a expression was significantly correlated with worse OS in TNBC patients with or without chemotherapy [29]. Multivariate analyses revealed that miR181a was an independent prognostic factor of OS in TNBC [10]. The expression level of miR-493 is a significant prognostic factor in breast cancer. During the Kaplan–Meier analysis, one study observed that patients with a high miR-493 expression had better disease-free survival than patients with a low expression, soon after adjusting for clinical pathological factors common in breast cancer. Further subtype analyses revealed that miR-493 expression levels were only significantly seen in patients with TNBC [70]. The decreased expression of miR-16 was associated with the worsening of OS and DMFS in patients with TNBC, as shown by studies by Cascione et al. [36] and De Rinaldis et al. [40]. Therefore, the overexpression of miR-16 decreases cell growth and proliferation, inducing apoptosis in human cancer. However, Usmani et al. [92], using a sample of 194 participants, found a high expression of this miR in all stages of BC, especially in relation to patients with TNBC who showed higher expression levels of this miR, concluding that miR-16 is an important oncogene and has high functionality for prognostic biomarker and disease diagnosis [92].

In summary, the results of this systematic review demonstrated that the elevated levels of miR-21, miR-27a/b, miR-210, miR-181a, miR-301a/b, and miR-454 were associated with a reduction in overall survival time. Likewise, a higher miR-454 expression was associated with shorter DFS. However, the lower expression levels of miR-128, miR-34 a/b/c, miR-30a/c/e, and miR-16 miR-155, miR-374a/b/5p, and miR-493 were associated with the worsening of OS, DMFS, DFS, and RFS in patients with TNBC. In the same sense, statistical analyses corroborate the above results, except for the effect on miR-301a/b and its high expression concerning OS, which did not obtain a significant result. Some limitations can be observed in this work; for example, there are varieties of miRs found in the studies compared with studies that portray the same specific miR. Another example found was the marked heterogeneity in some analyses, which was certainly due to differences in sample size, cohort or case–control demographics, patient characteristics, prognostic factors related to tumor size, grade, differentiation, stage, the lymph node involvement axillary, and high-risk genetic mutations. In addition, various survival assessment models were also used, such as OS, DMFS, DFS, and RFS, the use of different miR detection methods and cut-off values for expression levels, sample types, the methods of preparing these samples, schedule, duration of follow-up, and HR extraction methods. Thus, we suggest that further studies should be conducted to evaluate specific miRs as promising tools in the prognosis of TNBC.