The incidence of malignant melanoma is growing rapidly. Approximately half of the cases are BRAF mutated, making treatment with kinase inhibitors a (MEK and BRAF inhibitors) preferred choice in the advanced setting. The vast majority of these patients will benefit from the treatment. It is therefore of vital importance that the BRAF analysis is reliable and reflects the true nature of the tumour. Intraindividual tumour BRAF heterogeneity may exist, and changes of BRAF status over time might occur. We reviewed the literature by searching the PubMed database and 630 potentially relevant studies were identified. Thereafter, studies that investigated intralesional heterogeneity only, studies with ≤10 patients and studies that did not include adequate data to calculate discrepancy rates were excluded. Twenty-two studies met our inclusion criteria and were included in the meta-analysis. The pooled discrepancy rate between primary and metastatic lesions was 13.4% (95% confidence interval [CI]: 9.2-18.2%) while it was 7.3% (95% CI: 3.3-12.6) between two metastatic lesions. The number of patients whose tumoural BRAF status was changed from mutation to wild type and from wild type to mutation, respectively, was comparable. We conclude that a clinically meaningful discrepancy rate in BRAF status both between primary-metastatic and metastatic-metastatic melanoma lesions exists. Our results support the polyclonal model of melanomas in which subclones with different BRAF status co-exist in the same melanoma lesion. In addition, the results indicate a need for biopsy of a metastatic lesion for subsequent BRAF analysis when treatment with kinase inhibitors is considered.