BACKGROUND: Critically ill extremely preterm infants fed human milk are often underrepresented in neonatal nutrition trials aimed to determine the effects of enteral protein supplementation on body composition outcomes. METHODS: Masked randomized trial in which 56 extremely preterm infants 25-28 weeks of gestation were randomized to receive either fortified milk enriched with a fixed amount of extensively hydrolyzed protein (high protein group) or fortified milk without additional protein (standard protein group). RESULTS: Baseline characteristics were similar between groups. In a longitudinal analysis, the mean percent body fat (%BF) at 30-32 weeks of postmenstrual age (PMA), 36 weeks PMA, and 3 months of corrected age (CA) did not differ between groups (17 ± 3 vs. 15 ± 4; p = 0.09). The high protein group had higher weight (-0.1 ± 1.2 vs. -0.8 ± 1.3; p = 0.03) and length (-0.8 ± 1.3 vs. -1.5 ± 1.3; p = 0.02) z scores from birth to 3 months CA. The high protein group also had higher fat-free mass (FFM) z scores at 36 weeks PMA (-0.9 ± 1.1 vs. -1.5 ± 1.1; p = 0.04). CONCLUSIONS: Increased enteral intake of protein increased FFM accretion, weight, and length in extremely preterm infants receiving protein-enriched, fortified human milk. IMPACT: Extremely preterm infants are at high risk of developing postnatal growth failure, particularly when they have low fat-free mass gains. Protein supplementation increases fat-free mass accretion in infants, but several neonatal nutrition trials aimed to determine the effects of enteral protein supplementation on body composition outcomes have systematically excluded critically ill extremely preterm infants fed human milk exclusively. In extremely preterm infants fed fortified human milk, higher enteral protein intake increases fat-free mass accretion and promotes growth without causing excessive body fat accretion.