Non-invasive biological indicators of the absence/presence or progress of the disease that could be used to support diagnosis and to evaluate the effectiveness of treatment are of utmost importance in Duchenne Muscular Dystrophy (DMD). This neuromuscular disorder affects male children, causing weakness and disability, whereas female relatives are at risk of being carriers of the disease. A biomarker with both high sensitivity and specificity for accurate prediction is preferred. Until now creatine kinase (CK) levels have been used for DMD diagnosis but these fail to assess disease progression. Herein we examined the potential applicability of serum levels of matrix metalloproteinase 9 (MMP-9) and matrix metalloproteinase 2 (MMP-2), tissue inhibitor of metalloproteinases 1 (TIMP-1), myostatin (GDF-8) and follistatin (FSTN) as non-invasive biomarkers to distinguish between DMD steroid naïve patients and healthy controls of similar age and also for carrier detection. Our data suggest that serum levels of MMP-9, GDF-8 and FSTN are useful to discriminate DMD from controls (p < 0.05), to correlate with some neuromuscular assessments for DMD, and also to differentiate between Becker muscular dystrophy (BMD) and Limb-girdle muscular dystrophy (LGMD) patients. In DMD individuals under steroid treatment, GDF-8 levels increased as FSTN levels decreased, resembling the proportions of these proteins in healthy controls and also the baseline ratio of patients without steroids. GDF-8 and FSTN serum levels were also useful for carrier detection (p < 0.05). Longitudinal studies with larger cohorts are necessary to confirm that these molecules correlate with disease progression. The biomarkers presented herein could potentially outperform CK levels for carrier detection and also harbor potential for monitoring disease progression.