Melanoma is a diffuse neuroendocrine tumor originating from the neural crest that mainly occurs in the skin and mucosa. The global number of melanoma cases increased from 232,000 in 2012 to 351,880 in 2015, and 62,000 patients died from melanoma in 2015. As the most common cutaneous malignant tumour, melanoma has the characteristics of high malignancy levels, increasing morbidity and mortality rates, and extremely high treatment costs (1). Ethnicity, sun exposure, alcohol consumption, vitamin D deficiency, obesity and exposure to chemicals such as oil and pesticides have all been cited as causes of melanoma (2). With the continuing rise in morbidity, melanoma has become a challenging public health problem worldwide, especially in New Zealand, Australia, Norway, Sweden, and the Netherlands (3). Moreover, melanoma is projected to become the second most common cancer in the United States by 2040 (4). The global incidence of melanoma has increased observably in recent years; however, some emerging therapies, such as immune checkpoint inhibitors, have led to a significant decline in melanoma patient mortality rates (5).

As a nuclear protein expressed in proliferating mammalian cells, Ki-67 controls gene expression by organizing heterochromatin spatially (6). Immunohistochemical (IHC) staining is usually used to detect Ki-67, and the expression level of Ki-67 is related to cell proliferation activity, disease progression and cancer recurrence (7). Some studies have shown that Ki-67 is one of the prognostic indices of multiple solid tumors, such as nasopharyngeal carcinoma (8), stage I non-small cell lung cancer (9), gastrointestinal stromal tumour (10), and gliomas (11), resected triple-negative breast cancer (12), colorectal cancer (13), hepatocellular carcinoma (14), and thyroid cancer (15). Previous studies have shown an association between Ki-67 expression and melanoma patient prognosis, but the results have been contradictory (16–25). Some studies show that high Ki-67 expression is an indicator of worse prognosis (16–22, 25), while other studies suggest that high Ki-67 expression predicts favorable prognosis (23, 24). Therefore, we conducted this meta-analysis to accurately determine the prognostic and clinicopathological significance of Ki-67 in melanoma patients to optimize treatment strategies.

This meta-analysis was conducted on the basis of the Preferred Reporting Items for Systematic Reviews and Meta-Analyses statement (26).

Currently, the association between ki-67 expression levels and prognosis in melanoma patients is not clear. In this study, we integrated 10 clinical studies to determine the prognostic value of Ki-67 expression in melanoma patients. The results showed that Ki-67 expression was connected with different survival endpoints,including OS, PFS and RFS rate, suggesting that Ki-67 could be used as a valuable index in the prognostication of patients with melanoma. The pooled data also showed that high expression of Ki-67 was associated with melanoma thickness but not with sex, location, ulceration, or vascular invasion. In addition, subgroup analyses indicated that a high level of Ki-67 expression was related to poor OS outcomes in melanoma patients regardless of region, patients, median age of the patients, study type, cut-off of Ki-67% and length of follow-up. Taken together, this is the first study to reveal that high Ki-67 expression is associated with poor prognosis in melanoma patients by using meta-analysis approach.

As the best marker to evaluate cell proliferation, Ki-67 is expressed in all active stages of the cell cycle, including G1, S, G2 and mitosis (28). Therefore, the higher the expression level of KI-67, the faster the tumor growth and the larger the tumor volume will be, leading to the worse prognosis of patients. KI-67 has been proved to be an prognostic index for multiple solid tumors, for example, high ki-67 expression is closely associated with poor OS and DFS (Disease Free Survival) in lung adenocarcinoma (29). A meta-analysis including 8 studies showed that high ki-67 expression was associated with poor OS, PFS, and DMFS (distant metastasis-free survival) in patients with nasopharyngeal carcinoma (8). In hepatocellular carcinoma, high expression of KI-67 has been associated with poor DFS, RFS, and OS (14). Qiu et al. reported that Ki-67 overexpression was correlated with poor OS in patients with ovarian cancer (30). In our study, we revealed that the high expression of KI-67 was closely associated with poor OS in melanoma patients, which is consistent with the conclusion of other solid tumors. However, the reasons why Ki-67 is not related to PFS and RFS may be as follows. First, there are only two studies including PFS and RFS, so there is maybe some bias in the conclusion. Second, high expression of KI-67 may indeed be unrelated to poor PFS and RFS, but the current research data are insufficient, and we expect more data to confirm this conclusion in the future.

Previous studies have shown that the expression level of KI-67 is closely related to the tumor size, such as bladder cancer (31), hepatocellular carcinoma (14), and gastric cancer (32). In our study, we found the relationship between high expression of Ki-67 and melanoma thickness. Thus, the previous findings of the correlations of Ki-67 expression and tomor size in various solid cancers are in accordance with the present study.A recent study showed that melanoma tumor thickness is strongly associated with poor 5-year OS (33). Therefore, the high expression of KI-67 indicates the poor prognosis of melanoma patients, which may be related to tumor thickness. However, a study examining the relationship between ki-67 expression and patient prognosis in thick melanoma (≥4 mm) showed that KI-67 remains an index of poor prognosis in melanoma (19). Therefore, we believe that Ki-67 is still a marker of poor prognosis in melanoma patients even after removing the effect of tumor thickness. Furthermore, ki-67 expression was strongly associated with increased Breslow thickness, Clark level, ulceration, lymphovascular invasion, number of mitosis, and pT stage (34). However, pooled data of this study showed that high ki-67 expression was independent of gender, location, ulceration, or vascular invasion, which may account for the limited sample size of this study.

The advantages of our study are as follows. First, this is the first study to use a meta-analysis to demonstrate that Ki-67 overexpression is a predictor of poor prognosis in melanoma patients. Second, we did a through search to find the best fitting studies, and finally a total of 10 previours studies were included in our meta-analysis. Third, the expression level of KI-67 in the 10 eligible studies included was detected by IHC, which ensured the reliability of the results. Finally, only high quality English literature is included in this study to reduce errors and ensure the authenticity of research conclusions.

However, there are several limitations in our meta-analysis. First, both the number of studies and the total sample size were small. Second, the cut-off point for Ki-67 positivity was different among the included studies, which may have led to heterogeneity. Third, clinical data from Asian and African countries are scarce. Therefore, we need more data from other region groups to determine the influence of region on the study results. Forth, different types of melanomas behave differently, having different molecular signitures and Ki67 proliferation index may have different impacts in those types of melanomas. Finally, as with all meta analysis,it cannot correct some of the bias in the included original studies, and some of the studies which were included in this article were too small for statistical analysis. Therefore, we need more comprehensive designs and large-scale clinical trials for further investigation.

Despite some limitations, our meta-analysis conclusively indicates that Ki-67 overexpression is associated with worse OS rates in melanoma patients. Ki-67 can be used as an important reference marker when evaluating the survival outcomes and prognoses of melanoma patients. Therefore, our study can provide some reference for clinicians in the formulation of melanoma diagnosis and treatment plan, rational allocation of medical resources and preliminary judgment of patient prognosis. There are some shortcomings in our study; thus, we look forward to the completion of more prospective multicentric clinical studies with reasonable designs and on larger scales to verify and supplement our conclusions.

The original contributions presented in the study are included in the article/supplementary material. Further inquiries can be directed to the corresponding author.

QL, ZP, and LSe conceived and designed the study. QL and ZP performed the analysis, prepared the figures and tables and wrote the main manuscript. LFS reviewed and revised the manuscript. All authors contributed to the article and approved the submitted version.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.