Background: Appropriate cancer treatment requires determination of the primary site of origin. The standard of care for diagnosing Cancer of unknown primary (CUP) based on the 2016 NCCN Guidelines is a comprehensive clinical work-up including biopsy and immunohistochemistry (IHC). However, IHC was able to identify a primary tumor in only 25% of CUP. Molecular tumor profiling (MTP) validation studies demonstrated sensitivities of 74-89% and specificities of 95-99%. Only a few studies with small sample sizes using MTP to diagnose the primary in CUP have been performed yielding variable results. This study aims to evaluate the diagnostic accuracy of MTP in determining the tissue of origin in adult patients with CUP. Methods: Literature search included articles published in any language indexed in MEDLINE and Cochrane. A Google Scholar search and a review of all published articles’ references were performed. Unpublished studies and abstracts from conference proceedings were also reviewed. Studies involving adult patients with CUP who underwent both MTP and clinical evaluation with IHC were identified. Articles were included if they had primary data sufficient to calculate both sensitivity and specificity. Methodological qualities of the included studies were evaluated using the QUADAS-2 tool. Data was analyzed using Review Manager 5.3 and MetaDiSc 1.4. Results: Seven studies with 549 patients were analyzed. MTP was found to have a pooled sensitivity of 89% (95% CI 0.85-0.92)and pooled specificity of 74% (95% CI 0.67-0.80)with pooled positive and negative likelihood ratios of 2.97 (95% CI 1.17-7.63) and 0.20 (95% CI 0.07-0.51), respectively. Area under the curve was measured to be 0.9107. Conclusions: This meta-analysis suggests that MTP could be useful in determining the tissue of origin in adult patients with CUP. With an area under the curve of 0.9107, it is an excellent diagnostic test. The results showed considerable heterogeneity as expected in meta-analyses of diagnostic test accuracy. In this case, it may be due to variability among the MTP assays used which may reflect different cutoffs.