Hepatocellular carcinoma (HCC) is one of the most common and deadly types of liver cancer, significantly contributing to cancer-related deaths globally (1). The outlook for HCC patients largely depends on how early the cancer is detected and the effectiveness of the treatments available (2). Among the various treatment options, transarterial chemoembolization (TACE) has become a key therapy for intermediate-stage HCC. TACE involves the local delivery of chemotherapy combined with embolization to cut off the tumor’s blood supply, enhancing drug retention and effectiveness (3). Despite its extensive use, responses to TACE vary greatly, with many patients experiencing poor outcomes (4). Therefore, finding predictive biomarkers to accurately predict the response to TACE is crucial (5). Recently, microRNAs (miRNAs) have attracted considerable interest as potential biomarkers for various cancers, including HCC (6, 7). miRNAs are small, non-coding RNA molecules, typically 18-25 nucleotides long, that regulate gene expression post-transcriptionally (8). They are vital in numerous cellular processes such as cell proliferation, differentiation, apoptosis, and metastasis. The dysregulation of miRNAs is linked to the development and progression of HCC, making them promising candidates for diagnostic, prognostic, and predictive biomarkers (9). Several studies have investigated the potential of miRNAs to predict the response to TACE in HCC patients (10). However, results have been inconsistent, with differences in miRNA profiles, sample sizes, and methodologies (10–13). This inconsistency highlights the need for a comprehensive review of the existing evidence to clarify the diagnostic performance of miRNAs in this context. This meta-analysis aims to systematically assess the diagnostic accuracy of miRNAs in predicting the response to TACE in HCC patients. By combining data from multiple studies, we aim to provide robust estimates of sensitivity, specificity, diagnostic odds ratio (DOR), and the area under the summary receiver operating characteristic (SROC) curve. We also intend to identify sources of variability and evaluate the impact of study design, miRNA profiling techniques, and other factors on diagnostic performance. The findings from this meta-analysis could significantly impact the clinical management of HCC. Demonstrating high diagnostic accuracy of miRNAs could lead to their incorporation into clinical practice, improving patient stratification and treatment outcomes. Additionally, understanding the limitations and sources of variability in current research could guide future studies toward more standardized and reliable approaches. In summary, this meta-analysis aims to provide a thorough and rigorous evaluation of the diagnostic performance of miRNAs in predicting the response to TACE in HCC patients. By doing so, it seeks to offer valuable insights that could inform clinical decision-making and enhance the prognosis for HCC patients undergoing TACE.

This meta-analysis was carried out in accordance with the 2020 guidelines specified by the PRISMA-DTA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses of Diagnostic Test Accuracy) statement (14).

TACE is a pivotal treatment for HCC, especially in patients with intermediate-stage disease who are ineligible for surgical removal or liver transplants. TACE functions by combining avascular necrosis and localized chemotherapy, which is administered directly through the hepatic artery that supplies blood to the tumor. This method effectively cuts off the tumor’s blood supply, causing tumor cell death, while simultaneously delivering high doses of chemotherapy to the cancerous area. TACE has demonstrated significant improvements in survival rates and symptom management, without substantially impairing liver function, assuming the patient’s liver function is reasonably preserved and there are no issues such as portal vein thrombosis or ascites. This procedure can be used alone or (19, 20) in conjunction with other treatments like radiation therapy, radiofrequency ablation (RFA), and systemic therapies like sorafenib for more advanced cases. The effectiveness of TACE hinges on careful patient selection and customized treatment plans to ensure the best outcomes and minimize potential risks (20).

Evaluating the response to TACE treatment is vital because it provides critical information on the effectiveness of the therapy in targeting liver tumors. Accurate assessment allows healthcare professionals to determine whether the treatment is successfully reducing tumor size, necrosis, and viability, which directly correlates with patient prognosis and survival rates. Moreover, evaluating treatment response aids in making informed decisions about subsequent therapeutic strategies, such as additional TACE sessions, alternative treatments, or supportive care. It also helps in identifying potential complications early, ensuring timely interventions to mitigate adverse effects and improve overall patient outcomes (19, 20).

To the best of our knowledge, this study is the first to evaluate the diagnostic performance of serum/plasma-derived biomarkers in predicting response to TACE therapy in HCC patients. Typically, therapeutic response in HCC is evaluated using imaging methods such as CT scans and MRI. While these imaging techniques are standard, the advent of artificial intelligence-based models (radiomics) has led to the development of models capable of predicting response to TACE in HCC cases, achieving an impressive diagnostic performance with an AUC of 0.93 (21).

However, these methods have significant limitations. The cost of advanced imaging and the infrastructure required for AI-based analysis can be prohibitively high, limiting accessibility in resource-limited settings. The accuracy of radiomics models is highly dependent on the quality of the imaging data, which can vary significantly across different machines and institutions. There is considerable variability in the interpretation of imaging results, leading to inconsistent outcomes. Radiomics models also require large, well-annotated datasets for training, which may not be available for all patient populations. Additionally, issues with reproducibility and generalizability of these models across diverse clinical settings pose further challenges (22–24).

This study, therefore, fills a crucial gap by exploring alternative, potentially more accessible, and less resource-intensive biomarkers that could complement or enhance current imaging-based diagnostic strategies. By evaluating serum/plasma-derived biomarkers, we aim to provide a more universally applicable method for predicting response to TACE therapy in HCC, which could lead to improved patient outcomes and more personalized treatment approaches.

The findings from this meta-analysis underscore the significant potential of miRNAs as predictive biomarkers for assessing the response to TACE in HCC patients. The pooled sensitivity and specificity of 0.79 and 0.82, respectively, along with an AUC of 0.85, demonstrate that miRNAs possess excellent diagnostic accuracy. These results indicate that miRNAs could serve as reliable indicators for predicting which patients are likely to respond favorably to TACE, thereby enhancing the ability to personalize treatment strategies and improve patient outcomes.

One of the key strengths of this study is the comprehensive analysis of miRNA diagnostic performance across multiple studies, encompassing diverse geographical locations and varied patient populations. The inclusion of studies from different countries, such as Egypt, China, and Italy, highlights the broad interest and applicability of miRNA research in HCC. This geographical diversity also reinforces the robustness of the findings, suggesting that miRNAs could be universally effective biomarkers, irrespective of population-specific genetic and environmental factors.

Among the analyzed microRNAs, miR-210 and miR-373 demonstrated elevated expression, functioning as oncogenes in HCC patients. Notably, their expression significantly diminished after undergoing TACE therapy. Conversely, certain other microRNAs displayed a reverse pattern of expression: MiR-373 acts as an oncogenic miRNA in hepatocellular carcinoma by promoting cell proliferation and the G1/S cell cycle transition through downregulation of the tumor suppressor PPP6C (25). Similarly, miR-210 promotes HCC progression by enhancing autophagy in M2-polarized macrophages through inhibition of the PI3K/AKT/mTOR signaling pathway, thereby fostering tumor cell proliferation, invasion, and immune evasion in the tumor microenvironment (26). In addition, miR-210 is a specific biomarker for differentiating HCC from other metastatic lesions in the liver (27). Therefore, their downregulation following TACE treatment might be an indicator of the treatment efficacy (12, 15). In contrast, substantial evidence highlights the tumor-suppressive role of miR-1271 in HCC, demonstrating its involvement in inhibiting cell proliferation (28), enhancing radiosensitivity (29), and suppressing epithelial-to-mesenchymal transition (30). Similarly, miR-214 is downregulated in HCC patients, and restoring its expression has been shown to inhibit cell proliferation (31), migration (32), metabolism (32), angiogenesis (33), and the expression of β-catenin protein (34), a key factor in HCC progression (35). In addition, miR-133b is downregulated in HCC patients (36), leading to reduced growth factor levels in connective tissues (37). Its expression suppresses cell proliferation (36), migration (38), and invasion (38), is positively associated with better patient prognosis (39), and enhances sensitivity to cisplatin treatment (40). More importantly, researchers have discovered that encapsulating miR-335 in exosomes enhances its stability, bioavailability, and therapeutic efficacy, overcoming resistance mechanisms in HCC (41).

The subgroup analyses provided further insights into factors affecting the diagnostic performance of miRNAs. Significant differences were observed in diagnostic accuracy based on response criteria and geographical location. For instance, studies using the mRECIST criteria showed moderate heterogeneity, while those using RECIST criteria exhibited higher diagnostic accuracy. This suggests that standardized response evaluation methods are crucial for accurately assessing the effectiveness of miRNAs as predictive biomarkers. Additionally, the higher diagnostic performance observed in studies from Egypt suggests potential regional differences in miRNA expression or detection methodologies that warrant further investigation.

Differences in RNA extraction methods and endogenous controls can introduce variability in the results and cause heterogeneity. This is particularly relevant in the context of small, non-coding RNA molecules like miRNAs, where consistency in methodology is crucial for reliable quantification. The included studies employed various RNA extraction methods. While methodological differences exist, the extraction methods used across the studies are validated techniques widely accepted in the field. Additionally, studies were included based on stringent criteria that ensured reliable miRNA quantification and diagnostic reporting. Furthermore, the extraction methods varied among the studies, with only two utilizing the miRNeasy kit. We conducted a new subgroup analysis focusing on these two studies to address this. Our findings indicate that RNA extraction methods might contribute to interstudy heterogeneity. However, since both studies were conducted in Egypt, reaching a definitive conclusion remains challenging. Lastly, we performed a sensitivity analysis to evaluate the robustness of the results. We observed that if one study is removed one by one (leave one out analysis), the overall pooled results for sensitivity, specificity, and DOR do not change significantly. These findings suggest that miRNAs demonstrate promising diagnostic potential in predicting the response to TACE in HCC patients. However, further validation with larger and more diverse patient cohorts is essential before their incorporation into routine clinical practice.

The clinical implications of these findings are substantial. Incorporating miRNA profiling into routine clinical practice could enhance the stratification of HCC patients, allowing clinicians to identify those who are more likely to benefit from TACE. This targeted approach could lead to better treatment outcomes, reduced side effects, and more efficient use of healthcare resources. Furthermore, understanding the limitations and sources of variability in current miRNA research could guide future studies towards more standardized and reliable approaches, ultimately improving the reliability of miRNAs as predictive biomarkers.

In conclusion, this meta-analysis provides compelling evidence for the diagnostic potential of miRNAs in predicting the response to TACE in HCC patients. The high sensitivity and specificity, along with the robust AUC, highlight the promise of miRNAs as valuable predictive biomarkers. Implementing miRNA profiling in clinical practice could revolutionize the management of HCC, leading to more personalized and effective treatment strategies. Further research is needed to address the existing limitations and enhance the reliability of miRNAs as predictive tools in HCC therapy.