Acute respiratory distress syndrome (ARDS) is an acutely induced respiratory failure characterised by the appearance of bilateral alveolar opacities on imaging and hypoxemia(1). Among strategies that have proved beneficial in terms of patient outcome, prone position (PP) is one associated with the greatest impact in terms of reduction in mortality, with PP sessions of 17h on average(2). The benefit of PP is to better homogenize the stress and strain applied to the lung by invasive ventilation. In a multicentric retrospective study involving COVID-19-related intubated ARDS patients, a strategy of extended PP, where PP was maintained for a median duration of 40 hours (h), was associated with increased survival compared to standard PP where each session lasted less than 24h(10). The main objective of this study is to investigate whether extended PP, defined as 40-hour PP sessions, can significantly reduce mortality of patients with invasively ventilated, severe ARDS, at day 28 after inclusion, compared to standard, 16-hour, PP sessions. The primary endpoint is all-cause mortality 28 days after inclusion. The study is a prospective, randomised, open-label, two-group parallel, multicenter trial comparing 40-hour extended PP to 16-hour standard PP in patients with moderate to severe ARDS. After inclusion, patients will be randomised either to standard PP, or extended PP. In the control arm, patients will be turned prone for 16-hour sessions. In the intervention arm, patients will be turned prone for 40h. In both groups, maintaining PP for up to an extra 8 hours will be permitted, allowing PP sessions to last 24 hours in the comparator group, and 48 hours in the intervention group. Inclusion criteria will be: severe ARDS, with ARDS defined according to the 2012 Berlin definition; and the severity defined as a PaO2/FiO2 ratio of \<150 mm Hg, with an FiO2 of ≥ 60% persisting for at least 4h after optimization of ventilatory setting. Exclusion criteria will be : PP during the same Intensive care unit ( ICU ) stay and before inclusion, ECMO before PP, arterio-venous ECMO, persistent PaO₂/FiO₂ ratios \<150 mm Hg, with FiO2 ≥ 60% on all arterial blood gases collected over a 24-hour period without prompting study inclusion, spine instability, intracranial pressure \> 20 mmHg, severe brain injury, hemodynamic instability deemed to contraindicate PP by the physician in charge.