Cancer stem cells (CSC) have been investigated as prognostic markers in oral squamous cell carcinoma (OSCC). However, an assessment of the reporting quality of these studies has not been performed yet. The aim of this study was to describe the reporting quality of prognostic studies involving CSCs and OSCC, focusing mainly on the immunohistochemical reproducibility. By means a systematic review, 34 articles were selected. Analyses of both general reporting quality and immunohistochemistry technique were performed by using checklists for multiple aspects related to study reproducibility. A total of 21 different CSC markers were cited in the selected studies, evaluated by means of a wide range of antibodies, most of them (40.3%) without clone description. Discrepancies in intracellular immunolabeling were noted for some markers. The mean global score for general quality assessment revealed limits in the quality of the articles. The main problems were related to lack of report on OSCC characteristics and treatment, sample size rationale, and sensitivity analysis or internal validation of the markers. Although there was a high frequency of studies having “good or very good” score for immunohistochemistry reproducibility, the frequency of articles with “poor or very poor” score for individual items was expressive, mainly for description of immunolabeling analysis (38.2% of the studies were poorly described). In conclusion, although there is a significant range of CSC markers with promising results for prognosis of OSCC, the inadequate reporting of important sections in the published studies, including immunohistochemistry technique, may limit the quality of the investigation.