Current legislation (IDEA, 2004; NCLB, 2001) mandates all students, including students with disabilities, demonstrate progress toward the same standards. However, students continue to struggle with attainment of statewide academic standards as measured by high-stakes assessment. The purpose of the current study was to examine the degree that Dynamic Indicators of Basic Early Literacy Skills Oral Reading Fluency (DIBELS ORF) and Maze Curriculum-Based Measures (AIMSweb Maze-CBM) predict standard scores on the North Carolina End-of-Grade (EOG) Assessment of Reading Comprehension. The study also investigated differences in the relationship as a function of grade, examined the accuracy of established cutoff scores, and determined optimal cut scores. Participants included 336 students in third, fourth, and fifth grades. Results of the study were consistent with previous research, indicating the significance of fluency measures for determining the likelihood of proficiency on high-stakes assessments. Findings indicated ORF and Maze measures significantly predicted proficiency, with ORF accounting for the most variance in EOG scores. Receiver Operating Characteristic (ROC) Curves revealed statistically significant Area Under the Curve (AUC) values for ORF and Maze. Sensitivity levels were adequate for recommended cutoff values; specificity levels were less than adequate. Optimal cutoff scores to maximize sensitivity and specificity yielded slightly different cutoff points for ORF and Maze. Implications for practice, limitations, and suggestions for future research are provided.