Investigating the quality of the evidence provided by empirical studies on women with breast cancer treated with mistletoe using GRADE

Background: With health costs rising and economic crisis affecting many countries, solutions that offer advantages for patients and society are most wanted, where Complementary and Alternative Medicine (CAM) appears to be calling the attention of politicians and governments. Among unconventional cancer therapies, mistletoe is one of the most frequently used, especially in Central Europe. Many reviews have addressed its safety, and effectiveness, with controversial results, mainly because conventional reviews have mostly included evidence coming from randomized controlled studies (RCTs). This review aims to use the GRADE (Grading of Recommendation, Assessment, Development, and Evaluation) approach to assess the quality of the evidence on mistletoe and breast cancer and to evaluate whether the GRADE method is more suitable for appraising evidence coming from CAM studies because of the innovations it proposes. Methods: The databases PubMed/Medline, the Cochrane Library, SciVerse Scopus, Embase and CAMbase were searched for studies on mistletoe and breast cancer. The selected studies were grouped under six outcomes: survival, tumour progression, quality of life, immunological response, neutropenia, and safety. Quality was assessed regarding study limitations, imprecision, inconsistency, indirectness and publication bias, following GRADE guidelines, using standardized extraction sheets. Results: Eleven studies (4 RCTs and 7 observational studies) met the inclusion/exclusion criteria and were included in the analysis. The complete assessment of initial and final grading of the quality of the evidence, the descriptions of the assessment and calculations of absolute and relative effects were only possible for safety, due to lack of data for the other outcomes. The quality of the evidence for safety, coming from 2 RCTs, was assessed as good and evidence from observational studies on survival and tumour response was graded as very low. RCTs on quality of life and immunological response were not assessable due to incompatible outcome measurements, which also happened with the studies on neutropenia. Conclusion: The GRADE method proved to be an organized and transparent way of assessing quality of evidence. It treats many aspects differently from previous methods, for instance, considering evidence from observational studies and assessing quality per outcome. This makes it a good option for assessing evidence on CAM, often consisting of more observational studies than RCTs. The quality of the evidence on mistletoe shown by the GRADE approach was at least in line with evidence from other reviews. With the improvement of the quality of the observational studies on mistletoe, assessing this intervention with GRADE in the future might upgrade the quality of its evidence, providing better evidence for its recommendation.