-Objectives and study: Human milk is the ideal food for neonatal nutrition and optimal growth. When the mother's own milk is unavailable or limited, pasteurized human milk from milk banks is preferentially administered instead of infant formula, especially for preterm hospitalized neonates. Holder pasteurization (62.5 °C, 30 min) is applied for sanitary reasons but alters human milk components such as enzymes and immunoglobulins. In vitro studies have also shown that pasteurization of human milk impacts its hydrolysis and disintegration in term newborns (De Oliveira et al., 2016). Our study aimed at investigating the impact of pasteurization of human milk on its gastric digestion in preterm infants. -Methods: In vivo study was conducted at Rennes Hospital on preterm infants (n=12) fed by a feeding nasogastric tube each three hours (NCT02112331). Over a six-day sequence, gastric aspirates were collected twice a day, before and after administration of raw or pasteurized human milk. Samples were collected at 35, 60 or 90 min after meal ingestion. Gastric volume and pH were measured. Structural changes of the digesta were evaluated by confocal microscopy and laser light scattering. In digesta, residual intact proteins left were followed by gel electrophoresis (SDS-PAGE) submitted to densitometry, and lipolysis degree was evaluated by gas and thin layer chromatography fitted to flame ionization detector. -Results: On average (±SD), infants were 27.5 ± 12.3 days old at the first day of the study. Birth weight was 1.4 ± 0.3 kg and gestational age 29.6 ± 1.0 weeks. Digestive kinetics presented high inter and intra-individual variabilities. Regarding proteins, results showed a rapid disappearance of intact caseins and lactoferrin, but a resistance of alpha-lactalbumin. The contribution of meal gastric emptying or hydrolysis in protein disappearance depended on the protein. Some pre-lipolysis was determined in milk before digestion and it was in overall significantly lower in pasteurized than in raw milk, likely due to the heat-denaturation of endogenous lipases. During gastric digestion the kinetics of lipolysis were not affected by pasteurization (p > 0.05). The lipolysis degree ranged from 6 to 20% at 90 min. This relatively limited extent of gastric lipolysis was observed for both raw and pasteurized human milk, and was illustrated by the microscopic observations: some native milk fat globule structure (hydrophobic core enveloped by an amphiphilic membrane) persisted through the gastric digestion for both raw and pasteurized human milk. Regarding the structure, pasteurization led to heat-induced protein aggregates in the soluble phase and at the interface of the human milk fat globule membrane, and also impacted the protein aggregation and emulsion disintegration during gastric digestion. -Conclusion: This study represents a unique and important dataset on the behavior of pasteurized versus raw human milk. The gastric digestion is a key step which can further modulate nutrient absorption and infant nutrition. Digestive hydrolysis may also impact on gut microbiota, a major contributor to the development of the intestinal and systemic immune systems in the neonatal period. Physiologic and metabolic consequences remain to be investigated.