MicroRNA-17 family as novel biomarkers for cancer diagnosis: a meta-analysis based on 19 articles.

Cancer remains as the leading cause of death all over the world due to the lack of efficient diagnostic techniques and therapeutic methods. Many studies have reported the potential diagnostic value of microRNA-17 (miRNA-17, miR-17) family members as biomarkers for cancer detection. However, inconsistent results were revealed from a wide range of studies. As a result of this, a meta-analysis based on 19 studies was conducted to assess the diagnostic performance of miR-17 family for cancer detection. A total of 1772 patients with certain types of cancer and 1320 healthy controls were involved in these studies. The overall diagnostic accuracy was measured by the following: sensitivity, 0.67 (95 % confidence interval (CI) 0.60-0.74); specificity, 0.83 (95 % CI 0.74-0.85); positive likelihood ratio (PLR), 3.9 (95 % CI 2.6-5.9); negative likelihood ratio (NLR), 0.40 (95 % CI 0.34-0.48); and diagnostic odds ratio (DOR), 10 (95 % CI 6-16), respectively. Additionally, the pooled area under the summary receiver operator characteristic (SROC) curve (area under the curve (AUC)) was 0.79 (95 % CI 0.75-0.82), indicating a relatively low accuracy of miR-17 family as biomarkers for cancer detection. Subgroup analysis further showed that miR-17 family had more reliable performance in cancer diagnosis for Asian than that for Caucasian. Moreover, multiple miRNAs containing miR-17, -20a/b, and -93 reflected higher diagnostic accuracy than both miR-106a/b (single miRNA) and the overall miR-17 family assay. Therefore, appropriate combinations of miR-17 family may be used as non-invasive screening biomarkers for cancer, and it is necessary to carry out a large-scale population-based study to further assess the potential diagnostic value of miR-17 family.