The use of alternative interventions, such as gluten-free and casein-free (GFCF) diets, is frequent due to limited therapies for Autism Spectrum Disorder (ASD). Our aims were to determine the influence of a GFCF diet on behavior disorders in children and adolescents diagnosed with ASD and the potential association with urinary beta-casomorphin concentrations. Thirty-seven patients were recruited for this crossover trial. Each patient consumed a normal diet (including gluten and casein) for 6 months and a GFCF diet for another 6 months. The order of the intervention (beginning with normal diet or with GFCF diet) was assigned randomly. Patients were evaluated at three time-points (at the beginning of the study, after normal diet and after GFCF diet). Questionnaires regarding behavior and autism and dietary adherence were completed and urinary beta-casomorphin concentrations were determined at each time-point. No significant behavioral changes and no association with urinary beta-casomorphin concentrations were found after GFCF diet. A 6-month GFCF diet do not induce significant changes in behavioral symptoms of autism and urinary beta-casomorphin concentrations. Further studies with a long follow-up period similar to ours and including placebo and blinding elements are needed to identify better those respondents to GFCF diets.