Fig. 1: Trial flow chart.

Patients (N = 267) were enrolled in the PRE-GVHD trial after informed consent and 259 were eligible. The screening phase was from day 0 to day +3 and the first samples were collected and analyzed on day +7 (±3). Upon aGvHD_MS17 CF (CF > 0.1) positivity, patients were randomized to receive either prednisolone or placebo for 5 days followed by 19 days of tapering. Primary endpoint was aGvHD grades II–IV. Without clinical manifestation of aGvHD, the medication was be tapered on days 6–19 after initiation of the therapy. Patients with aGvHD_MS17 negative samples were continually monitored until either a sample was positive for the aGvHD_MS17 CF, when they were randomized, or until clinical manifestation of aGvHD. Upon clinical manifestation of aGvHD, the patients were treated with standard therapy for aGvHD and counted as “pattern failure.” Eight patients were excluded due to missing urine samples and proteomic tests. The observation group consisted of 167 patients. Patients (N = 92; ITT population) were randomized to either the placebo (N = 48) or prednisolone (N = 44) arms. Fife patients (N = 3 placebo and N = 2 prednisolone) did not receive the study medication and were excluded from the safety group (N = 87) and safety analyses. The per-protocol population (N = 84) excluded another three patients from the placebo group who either received the study medication for <3 days (N = 2) or had no positive proteomic pattern test (N = 1).