Background: In-clinic application of Randomized Controlled Trial  RCT  results is complex. There is variability in conclusions of RCTs studying similar therapies and it is difficult to estimate their external validity. Individual patient data meta-analyses attempt to address these issues ith aggregate and subgroup results, but are only possible many years later. We have developed Machine Learning  ML  based tools to predict hether results of any RCT ill agree ith that of subsequent meta-analyses, and provide real-time ranking of RCTs based on the predicted agreement. Methods: The Grade Index algorithm considers quality of an RCT based on methodology features such as sample size, endpoint, number of years of follo-up, etc. The Applicability Index algorithm considers similarity of an index patient to a trial population based on features such as age, tumor size, etc. The algorithms predict agreement on direction of effect  experimental arm better or equal to control  beteen any RCT and meta-analyses aggregate and subgroup results, respectively. Primary endpoint is percent accuracy of predicted agreement of an RCT ith recently published EBCTCG  Early Breast Cancer Trialists Cooperative Group  meta-analyses. Results: A comprehensive sample of 210 published early breast cancer RCTs conducted beteen 1976 and 2015  total sample size of 159,488 patients  as curated from MEDLINE. Nine features pertaining to the quality of an RCT and nine other features describing disease and patient information from the demographic data tables ere selected. Over 100 state of the art ML algorithms such as Support Vector Machines  SVM  and Random Forest ere systematically evaluated. The inning Grade Index algorithm and the Applicability Index algorithms achieved accuracies of 98 percent and 95 percent, respectively. Conclusions: Quality related features as ell as demographic data of an RCT can be used by the Grade Index and Applicability Index algorithms, respectively, to rank agreement ith subsequent meta-analyses. These ranks are used in a clinically validated informatics system to generate real time evidence based treatment recommendations for breast cancer patients.,J Clin Oncol 34, 2016  suppl; abstr e18165 ,Publication Only Health Services Research and Quality of Care