Chastre et al. Critical Care (2022) 26:355 https://doi.org/10.1186/s13054-022-04204-9 RESEARCH Open Access Safety, efficacy, and pharmacokinetics of gremubamab (MEDI3902), an anti‑Pseudomonas aeruginosa bispecific human monoclonal antibody, in P. aeruginosa‑colonised, mechanically ventilated intensive care unit patients: a randomised controlled trial Jean Chastre1*†, Bruno François2, Marc Bourgeois3, Apostolos Komnos4, Ricard Ferrer5, Galia Rahav6, Nicolas De Schryver7, Alain Lepape8, Iftihar Koksal9, Charles‑Edouard Luyt1, Miguel Sánchez‑García10, Antoni Torres11, Philippe Eggimann12, Despoina Koulenti13,14, Thomas L. Holland15, Omar Ali16, Kathryn Shoemaker16,17, Pin Ren17, Julien Sauser18, Alexey Ruzin16, David E. Tabor16, Ahmad Akhgar19, Yuling Wu19, Yu Jiang19, Antonio DiGiandomenico16, Susan Colbert20, Drieke Vandamme21, Frank Coenjaerts22, Surbhi Malhotra‑Kumar23, Leen Timbermont23, Antonio Oliver24, Olivier Barraud25, Terramika Bellamy16, Marc Bonten22,26, Herman Goossens23, Colin Reisner17,27, Mark T. Esser16, Hasan S. Jafri16,28*† and The COMBACTE-MAGNET EVADE Study Group Abstract Background: Ventilator-associated pneumonia caused by Pseudomonas aeruginosa (PA) in hospitalised patients is associated with high mortality. The effectiveness of the bivalent, bispecific mAb MEDI3902 (gremubamab) in prevent‑ ing PA nosocomial pneumonia was assessed in PA-colonised mechanically ventilated subjects. Methods: EVADE (NCT02696902) was a phase 2, randomised, parallel-group, double-blind, placebo-controlled study in Europe, Turkey, Israel, and the USA. Subjects ≥ 18 years old, mechanically ventilated, tracheally colonised with PA, and without new-onset pneumonia, were randomised (1:1:1) to MEDI3902 500, 1500 mg (single intravenous dose), or † Jean Chastre and Hasan S. Jafri contributed equally *Correspondence: jean.chastre@gmail.com; hasan.jafri.md@gmail.com 1 Service de Médecine Intensive Réanimation, Institut de Cardiologie, Groupe Hospitalier Pitié‑Salpêtrière, Assistance Publique‑Hôpitaux de Paris, Sorbonne University, 47–83 Bd de l’Hôpital, 75651 Paris, France 16 Clinical Research and Development, Vaccines and Immune Therapies, BioPharmaceuticals R&D, AstraZeneca Biopharmaceuticals, One MedImmune Way, Gaithersburg, MD 20878, USA Full list of author information is available at the end of the article © The Author(s) 2022. Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://​creat​iveco​mmons.​org/​licen​ses/​by/4.​0/. The Creative Commons Public Domain Dedication waiver (http://​creat​iveco​ mmons.​org/​publi​cdoma​in/​zero/1.​0/) applies to the data made available in this article, unless otherwise stated in a credit line to the data. Chastre et al. Critical Care (2022) 26:355 Page 2 of 14 placebo. The primary efficacy endpoint was the incidence of nosocomial PA pneumonia through 21 days post-dose in MEDI3902 1500 mg versus placebo, determined by an independent adjudication committee. Results: Even if the initial sample size was not reached because of low recruitment, 188 subjects were randomised (MEDI3902 500/1500 mg: n = 16/87; placebo: n = 85) between 13 April 2016 and 17 October 2019. Out of these, 184 were dosed (MEDI3902 500/1500 mg: n = 16/85; placebo: n = 83), comprising the modified intent-to-treat set. Enrol‑ ment in the 500 mg arm was discontinued due to pharmacokinetic data demonstrating low MEDI3902 serum con‑ centrations. Subsequently, enrolled subjects were randomised (1:1) to MEDI3902 1500 mg or placebo. PA pneumonia was confirmed in 22.4% (n = 19/85) of MEDI3902 1500 mg recipients and in 18.1% (n = 15/83) of placebo recipients (relative risk reduction [RRR]: − 23.7%; 80% confidence interval [CI] − 83.8%, 16.8%; p = 0.49). At 21 days post-1500 mg dose, the mean (standard deviation) serum MEDI3902 concentration was 9.46 (7.91) μg/mL, with 80.6% (n = 58/72) subjects achieving concentrations > 1.7 μg/mL, a level associated with improved outcome in animal models. Treat‑ ment-emergent adverse event incidence was similar between groups. Conclusions: The bivalent, bispecific monoclonal antibody MEDI3902 (gremubamab) did not reduce PA nosocomial pneumonia incidence in PA-colonised mechanically ventilated subjects. Trial registration Registered on Clinicaltrials.gov (NCT02​696902) on 11th February 2016 and on EudraCT (2015-​001706-​ 34) on 7th March 2016. Keywords: Monoclonal antibody, Prevention, Pharmacokinetics, Pseudomonas aeruginosa ventilator-associated pneumonia, Safety Introduction 14]. A placebo-controlled phase 2a study highlighted Pseudomonas aeruginosa (PA) is a common cause of the potential for reducing PA pneumonia incidence in ventilator-associated pneumonia (VAP) in hospitalised colonised, mechanically ventilated intensive care unit patients [1–3], associated with mortality rates > 30% (ICU) patients dosed with a monovalent, monoclonal in patients with antibiotic-susceptible and > 44% with antibody against the PA PcrV protein [6]. multidrug-resistant (MDR) strains [4, 5]. Tracheobron- MEDI3902 (gremubamab) is a first-in-class bivalent, chial colonisation with PA increases the odds of devel- bispecific human immunoglobulin G1 kappa mono- oping VAP by around eightfold [2, 6], but no specific clonal antibody that selectively binds to the PA PcrV guidelines exist for managing these patients before protein and Psl exopolysaccharide involved in host cell pneumonia is diagnosed. Current strategies involve cytotoxicity and PA colonisation and tissue adherence, antibiotic treatment post-diagnosis [1, 7, 8]. Although respectively [15–17]. Prophylactic MEDI3902 admin- evidence suggests the earlier treatment may be ben- istration protected against lethal PA in animal models eficial [9], routine prophylactic antibiotics to prevent [15, 18], with a significant reduction in the expression VAP can contribute to PA resistance [8, 10, 11]. Con- of genes encoding key inflammatory cytokines in ani- sequently, Infectious Diseases Society of America and mals who received MEDI3902 versus control immuno- the American Thoracic Society guidelines recommend globulin G [18]. withholding antibiotic treatment in patients with sus- The use of rapid diagnostic techniques, such as real- pected VAP and endotracheal culture results below time PCR, in ICU settings can enable prompt identi- the diagnostic threshold for VAP (protected specimen fication of patients with bacterial colonisation of the brush [PSB] with < ­103 colony-forming units [CFU]/ lower respiratory tract before the onset of nosocomial mL, bronchoalveolar lavage [BAL] with < ­104 CFU/mL) pneumonia, bypassing the time required to obtain the [12]. No systemic agents are currently approved for the results of conventional microbiological cultures which pre-emptive treatment of ventilated patients with Pseu- can take 48–72 h. Accordingly, rapid identification of domonas airway colonisation to prevent PA pneumonia patients with respiratory PA colonisation could assist in [7, 13], highlighting an unmet need for effective, tar- the timely initiation of pre-emptive or curative thera- geted prevention. pies. A phase 1, placebo-controlled, dose escalation study Monoclonal antibodies are an attractive alternative to (NCT02255760) assessed an intravenous (IV) infusion of systemic antibiotics for the pre-emption of PA pneumo- MEDI3902 in 56 healthy adults [16]. MEDI3902 serum nias. Their benefits include enhanced specificity, longer concentrations through day 29 remained above the tar- half-life, and complementary mechanism of action to get therapeutic concentration of 5.3 μg/mL (based on a antibiotics and do not induce antibiotic resistance [7, Chastre et al. Critical Care (2022) 26:355 Page 3 of 14 murine model of PA pneumonia where mice were inoc- Subjects ulated with PA 5 × LD100, data on file) in subjects who Adults ≥ 18 years of age were eligible if they met the fol- received the highest doses (750, 1500, and 3000 mg), lowing key inclusion criteria: currently intubated and and dose-dependent increases in serum anticytotoxic mechanically ventilated and expected to remain so for and opsonophagocytic killing activities were observed at least 72 h; tracheal PA colonisation as assessed by [16]. MEDI3902 was well tolerated, supporting further PCR (GeneXpert System with PA Xpert test cartridge assessment in PA-colonised subjects at risk for devel- [research use only], Cepheid, Sunnyvale, CA, USA) no oping PA pneumonia. However, previous studies done more than 36 h and no diagnosis of new-onset pneumo- in ICU patients for assessing the potential usefulness of nia within 72 h before randomisation (patients with evi- monoclonal antibodies in preventing bacterial infections, dence of resolved pneumonia were eligible for inclusion); while showing some non-statistically significant trends in expected to survive for > 2 weeks and participate in the favour of antibodies in post hoc analyses, were based on study through 49 days post-dose. a limited number of patients and mostly negative, render- Key exclusion criteria were: acute confirmed or sus- ing difficult any conclusions [6, 14, 19]. Many factors that pected pseudomonal disease or active pulmonary dis- are not influenced by monoclonal antibodies can also ease; a Clinical Pulmonary Infection Score (CPIS) of at contribute to the development of pneumonia, including least 6 measured no more than 24 h before treatment; disease severity, underlying immune function and con- an Acute Physiology and Chronic Health Evaluation-II comitant therapies, warranting further studies before (APACHE-II) score of at least 25 or a Sequential Organ concluding that MEDI3902 represents a valuable com- Failure Assessment (SOFA) score of at least 12; and sys- plement to conventional measures for preventing lower temic or aerosolised colistin received for > 72 h within respiratory tract infections caused by PA. Here, we pre- 96 h before randomisation. The Additional file 1: Meth- sent the results of a single-dose, proof-of-concept study ods detail full eligibility criteria. of MEDI3902 for the pre-emptive treatment of PA noso- The modalities of the screening process were left to the comial pneumonia in PA-colonised, mechanically venti- discretion of the investigators. In a minority of centres, lated subjects in the ICU. mechanically ventilated patients were routinely screened for PA colonisation using conventional microbiological Methods cultures of endotracheal aspirates (ETA) once or twice a Study design week until death or weaning from mechanical ventilation, EVADE (Clinicaltrials.gov NCT02696902; EudraCT according to standard practice. When cultures grew posi- 2015-001706-34) was a phase 2, randomised, paral- tive for PA, informed consent was obtained and tracheal lel-group, double-blind, placebo-controlled study of colonisation was then confirmed using PCR. The other MEDI3902 in mechanically ventilated patients with PA centres did not use routine serial microbiological cultures lower respiratory tract colonisation, confirmed by a poly- for screening. Patients’ eligibility for study enrolment merase chain reaction (PCR)-based test on tracheal aspi- was regularly checked as long as they were expected to rates collected no more than 36 h before randomisation. remain on mechanical ventilation for at least 72 h and did All randomised subjects were positive for PA by PCR, not meet any exclusion criteria (see above), and informed while 154/184 (83.7%) had positive cultures, highlighting consent was obtained for using PCR on ETA. the greater sensitivity of PCR for detecting PA airway col- Routine use of VAP prevention bundles was highly onisation in ventilated patients. EVADE was performed recommended. To check whether these prevention bun- across 48 sites in 13 countries (Europe, Turkey, the USA, dles had been implemented correctly, sites were asked to and Israel; Additional file 1: Tables S1 and S2). report the VAP bundle application in the electronic case The study was conducted within the European public– report form for each enrolled patient (Table 1 and Addi- private partnership Combatting Bacterial Resistance In tional file 1: Table S3). Europe—Molecules Against Gram-Negative Infections (COMBACTE-MAGNET) consortium [7] in accordance Randomisation and masking with the ethical principles of the Declaration of Helsinki Per protocol, subjects were randomised (1:1:1) to a single and the International Council for Harmonization Guid- intravenous (IV) dose of MEDI3902 500 mg, 1500 mg, ance for Good Clinical Practice. The Antibiotic Resist- or placebo. Based on previous studies [19, 20] and phar- ance Leadership group also participated. Study-related macokinetic (PK) data [21] received after the start of the documents were reviewed and approved by the local study, a single MEDI3902 500 mg dose was not expected independent ethics committees or institutional review to maintain a target level of 1.7 μg/mL (derived from a boards. All subjects/legally acceptable representatives murine model of PA pneumonia where mice were inocu- provided written informed consent. lated with PA 1 × LD100, data on file) for 21 days, and Chastre et al. Critical Care (2022) 26:355 Page 4 of 14 enrolment in this arm was discontinued after 16 subjects end of infusion, and 8- and 24-h later and on days 4, 8, 15, were dosed. Interim PK confirmed MEDI3902 1500 mg 22, 29, and 50 of follow-up. maintained the target level in 80% of patients through day The Additional file 1: Methods include exploratory 21 [21]. Subsequently, following protocol and statistical endpoints and sample collection details. analysis plan (SAP) amendment, subjects were randomised (1:1) to MEDI3902 1500 mg or placebo (Additional file 1: Statistical analysis Fig. S1), stratified by geographical region and duration of Following protocol/SAP amendment, planned enrol- anti-PA antibiotic treatment within 96 h before randomi- ment was approximately 286 subjects randomised (1:1) sation (no antibiotic use, duration of no more than 72 h, to MEDI3902 1500 mg or placebo. Given the explora- duration > 72 h [except for systemic or aerosolised colis- tory nature of the study, power was calculated based tin; see exclusion criteria]). Subjects were followed until on Poisson regression with robust variance comparing the end of the study period (day 50). An interactive web MEDI3902 versus placebo groups (two-sided, α = 0.2), response system was used for randomisation to the treat- assuming a placebo group PA pneumonia incidence of ment group and assignment of blinded investigational 20%, a relative reduction of 50%, at least 80% power, and product kit numbers. To complete screening and ensure 20% adjustment for attrition. A relative reduction of 50% uniformity of inclusion criteria across sites, the eligibil- was considered clinically meaningful based on expert ity of all potential patients was confirmed by the Clinical advice and published data [6]. Coordinating Centre (Saint-Luc University Hospital, Brus- Recruitment was stopped early after 168 patients were sels, Belgium). MEDI3902 and placebo were administered included (MEDI3902 1500 mg: n = 85; placebo: n = 83) in a blinded fashion, and neither the subjects, their legal due to slow enrolment. representatives, nor the investigators and sponsor staff Efficacy and PK were assessed in the modified intent- involved in the treatment or clinical assessment of subjects to-treat (mITT) population (all subjects randomised and were aware of the treatment received. The investigational dosed, analysed by randomised treatment). Following products were handled by an unblinded investigational discontinued enrolment in the MEDI3902 500 mg arm product manager at each site. and protocol/SAP amendment, efficacy was assessed in MEDI3902 1500 mg versus placebo recipients. The Endpoints and assessments primary endpoint of nosocomial PA pneumonia was The primary efficacy endpoint was the incidence of noso- assessed by relative risk reduction (RRR), defined as comial PA pneumonia through 21 days post-dose in 1—relative risk, and its corresponding two-sided 80% MEDI3902 1500 mg recipients versus placebo as deter- confidence interval (CI), as estimated from a Poisson mined by a blinded independent Endpoint Adjudication regression model with robust variance and treatment Committee (EAC). The EAC included three experts in group as a covariate (two-sided, α = 0.2). A positive RRR intensive care medicine and two radiologists that used denoted less PA pneumonia in the MEDI3902 1500 mg prespecified, stringent and non-subjective criteria agreed group compared to placebo, and a negative RRR had upon by both the US Food and Drug Administration and more PA pneumonia in the MEDI3902 group. Patients the European Medicines Agency. Subjects must have met with mixed culture results that included PA were counted radiological (new or worsening infiltrate consistent with towards the primary endpoint. Early discontinuation pneumonia on chest X-rays), clinical, and microbiological due to death from underlying disease was expected as criteria concurrently to be diagnosed with PA pneumonia the main cause of missing data. If no PA pneumonia (Additional file 1: Methods). In subjects with suspected occurred before discontinuation, the subject was consid- or confirmed pneumonia, tracheobronchitis, or bacterae- ered as having no PA pneumonia infection in the primary mia, blood and respiratory specimens were collected, and efficacy analysis. No other imputation was applied to this chest X-rays were performed as clinically indicated, until analysis. Time to diagnosis of PA pneumonia, as judged clinical resolution. Primary safety endpoints included by the EAC, was estimated by use of the Kaplan–Meier the incidence of treatment-emergent adverse events method. Safety was assessed in the as-treated popula- (TEAEs), serious AEs (SAEs), and AEs of special interest tion (all subjects randomised and dosed, analysed by assessed through 49 days post-dose. treatment received). Safety and PK analyses included Secondary endpoints were MEDI3902 serum/ETA the 16 subjects who received MEDI3902 500 mg. Data PK parameters and serum anti-drug antibody responses were summarised descriptively, with no multiplicity through 49 days post-dose. Blood samples were collected adjustments. immediately before MEDI3902 dosing on day 1, at the Chastre et al. Critical Care (2022) 26:355 Page 5 of 14 Table 1 Demographics and baseline characteristics (mITT) MEDI3902 500 mg MEDI3902 1500 mg Placebo (N = 83) Total (N = 184) (N = 16) (N = 85) Age, years; mean (SD) 62.7 (9.3) 60.3 (15.2) 64.1 (12.9) 62.2 (13.8) Age < 65 years; n (%) 7 (43.8) 42 (49.4) 39 (47.0) 88 (47.8) Sex, male; n (%) 10 (62.5) 54 (63.5) 62 (74.7) 126 (68.5) Race, n (%) Asian 0 (0.0) 0 (0.0) 1 (1.2) 1 (0.5) Black or African American 0 (0.0) 2 (2.4) 4 (4.8) 6 (3.3) Native Hawaiian or Other Pacific Islander 0 (0.0) 0 (0.0) 1 (1.2) 1 (0.5) White 16 (100.0) 81 (95.3) 75 (90.4) 172 (93.5) Other 0 (0.0) 2 (2.4) 2 (2.4) 4 (2.2) Weight, kg; mean (SD) 82.5 (25.2) 78.8 (19.5) 84.4 (21.0) 81.6 (20.7) Height, cm; mean (SD) 167.9 (10.0) 169.1 (9.6) 171.1 (10.0) 169.9 (9.8) BMI, kg/m2; mean (SD) 29.5 (9.4) 27.5 (6.4) 29.0 (7.7) 28.4 (7.3) BMI ≤ 30 kg/m2; n (%) 11 (68.8) 60 (70.6) 54 (65.1) 125 (67.9) Clinical severity scores at baseline; mean (SD) APACHE-II 16.9 (2.9)* 15.3 (5.4)† 15.5 (5.2)‡ 15.5 (5.1)§ SOFA 4.5 (2.4)* 4.4 (2.7)† 4.0 (2.1) 4.2 (2.4)¶ CPIS 3.5 (1.5)* 3.0 (1.5)# 3.2 (1.5)# 3.1 (1.5)** Duration of mechanical ventilation; days, mean (SD) 19.5 (15.5) 25.2 (27.6) 31.1 (28.4)†† – Previous PA infections ≤ 3 months before randomisation; n (%) Yes 7 (43.8) 25 (29.4) 31 (37.8) 63 (34.4) No 9 (56.3) 60 (70.6) 51 (62.2) 120 (65.6) Missing 0 0 1 1 Use of antibiotics in the 3 months before randomisation; n (%) Yes 14 (93.3) 59 (74.7) 63 (81.8) 136 (79.5) No 1 (6.7) 20 (25.3) 14 (18.2) 35 (20.5) Missing 1 6 6 13 CLSI susceptibility at baseline;‡‡ n (%) Any culture result, n 16 83 82 181 P. aeruginosa positive 15 (93.8) 72 (86.7) 67 (81.7) 154 (85.1)   Non-MDR 6 (37.5) 37 (44.6) 25 (30.5) 68 (37.6)   ­MDR§§ 4 (25.0) 12 (14.5) 16 (19.5) 32 (17.7)   ­XDR§§ 4 (25.0) 14 (16.9) 21 (25.6) 39 (21.5)   ­PDR§§ 0 (0.0) 0 (0.0) 0 (0.0) 0 (0.0)   ­Unknown¶¶ 1 (6.3) 9 (10.8) 5 (6.1) 15 (8.3) P. aeruginosa ­negative## 1 (6.3) 11 (13.3) 15 (18.3) 27 (14.9) MEDI3902 500 mg MEDI3902 1500 mg Placebo (N = 83) MEDI3902 total (N = 101) (N = 16) (N = 85) P. aeruginosa PCR CT value; mean (SD) 28.3 (3.7) 28.5 (5.7) 29.2 (6.2) 28.5 (5.4) White blood cell count, ­103/μL; mean (SD) 11.6 (4.4) 13.2 (6.3) 11.4 (6.1) 12.9 (6.1) Absolute Neutrophil count, ­103/μL; mean (SD) 9.1 (4.1) 10.6 (6.0)*** 8.4 (4.3)††† 10.4 (5.7)‡‡‡ Procalcitonin, μg/L; mean (SD) 4.8 (15.5)§§§ 1.0 (2.4)¶¶¶ 0.61 (1.3)### 1.5 (6.1)**** CRP, mg/dL; mean (SD) 7.9 (6.3) 14.7 (37.9)†††† 15.1 (34.3)¶¶¶ 13.5 (34.6)‡‡‡‡ † Ventilator-associated pneumonia ­prevention All 5 measures used, n (%) 8 (50) 36 (42.4) 31 (37.3) 44 (43.6) APACHE-II Acute Physiology and Chronic Health Evaluation-II, BMI body mass index, CLSI Clinical and Laboratory Standards Institute, CPIS Clinical Pulmonary Infection Score, CRP C-reactive protein, MDR multidrug-resistant, mITT modified intent-to-treat population, PCR CT polymerase chain reaction cycle threshold, PDR pan-drug-resistant, SD standard deviation, SOFA Sequential Organ Failure Assessment, XDR extensively drug resistant. †Preventive ventilator-associated pneumonia measures (bundles) were elevation of the head of the bed, daily sedation vacations and extubation readiness assessment, peptic ulcer disease prophylaxis, deep vein thrombosis prophylaxis, and daily oral care with chlorhexidine * n = 15; †n = 82; ‡n = 82; §n = 179; ¶n = 180 #n = 81; **n = 177; ††n = 82; ‡‡minimum inhibitory concentrations were determined by CLSI broth microdilution at a centralised laboratory; §§MDR, PDR, and XDR as defined by;[29] ¶¶subjects with PA-positive culture results but missing minimum inhibitory concentration records; ##all randomised subjects were positive for PA by PCR within 36 h before randomisation, but not all had positive cultures; ***n = 83; †††n = 81; ‡‡‡n = 99; §§§n = 12; ¶¶¶n = 77; ### n = 74; ****n = 89; ††††n = 76; ‡‡‡‡n = 92 Chastre et al. Critical Care (2022) 26:355 Page 6 of 14 Post hoc analyses p = 0.49) (Table 2). The primary endpoint was not met. An adjusted post hoc analysis of the primary efficacy Similar results were observed regarding exploratory endpoint was done to address possible selection bias due efficacy endpoints, including all-cause pneumonia, all- to an imbalance in baseline covariates including ECMO, cause pneumonia or death, antibiotic usage and dura- PA-positive cultures, absolute neutrophil counts (ANC) tion of mechanical ventilation (Table 2 and Additional and procalcitonin (PCT) levels. The impact of key base- file 1: Results and Table S4). Specifically, when patients line covariates on the RRR of PA pneumonia and all- who died before study completion without PA pneumo- cause mortality in MEDI3902 1500 mg recipients versus nia were classified as failure (and not as success as done placebo, and on MEDI3902 serum PK, was also assessed in the primary analysis), the incidence of all-cause pneu- (Additional file 1: Methods). Efficacy data were calcu- monia or death was 30/85 (35.3%) in MEDI3902 1500 mg lated using PCT and ANC quartiles. The groups with recipients and 22/83 (26.5%) in placebo recipients; RRR high and statistically significant RRR with baseline levels − 33.3% (80% CI − 79.8%, 1.4%). The results of the time- of ≤ 0.55 μg/L for procalcitonin and ≤ 8.17 × ­103 cells/μL to-event analysis are shown in the Additional file (Addi- for ANC correspond to combined quartiles 1–3 for proc- tional file 1: Fig. S2), with most PA pneumonia occurring alcitonin and combined quartiles 1–2 for ANC. 15 days after randomisation in both groups. No signifi- cant difference in the incidence of other serious PA infec- Role of the funding source tions (tracheobronchitis, bacteraemia, intraabdominal The study sponsor was involved in study design, data col- or deep skin and soft tissue infections) was observed lection, data analysis, and data interpretation, with input between the placebo group and the MEDI3902 1500 mg from the authors, and in the writing of the report. group (Additional file 1: Table S5). To make sure that decisions made by the EAC were in agreement with those Results made by the physicians at the bedside, the number of sub- Subjects jects with an adverse event of PA pneumonia, not con- Subjects were randomised between 13 April 2016 and 17 firmed as PA pneumonia by the EAC, but who received October 2019; the study was completed on 4 December anti-PA antibiotics, was tabulated by the treatment 2019. Of the 1023 subjects screened, 835 (81.6%) could group (Additional file 1: Table S6). Clinicians detected not be included in the study (Fig. 1). Overall, 184 ran- and treated with antibiotics only three and 0 additional domised subjects received MEDI3902 1500 mg (n = 85), events of potential PA pneumonia in placebo recipients 500 mg (n = 16), or placebo (n = 83) (Fig. 1); 134/184 and MEDI3902 1500 mg recipients, respectively. (72.8%) subjects completed the study, including 59 Additional post hoc analyses were done on baseline (69.4%) MEDI3902 1500 mg, 12 (75.0%) 500 mg, and 63 covariates (Additional file 1: Fig. S3). PCT and ANC had (75.9%) placebo recipients. The most frequent reason for the greatest effect on RRR in PA pneumonia. At baseline, discontinuation was death in three (18.8%), 24 (28.2%), 115 patients (68% of the study population) had PCT lev- and 19 (22.9%) subjects in the 500 mg, 1500 mg and pla- els below 0.55 μg/L. In that group of patients with a low cebo groups, respectively. PCT level, the incidence of PA pneumonia was 23.7% in Baseline demographics, clinical severity scores, and the placebo and 12.5% in MEDI3902 1500 mg groups, Clinical and Laboratory Standards Institute (CLSI) PA corresponding to an 47.3% RRR (80% CI: 6.1% to 69.9%; antibiotic susceptibility results were generally similar p = 0.135). Similar results were observed in the subgroup between groups (Table 1). However, there was a numeri- of 83 patients (49% of the study population) with an cally higher percentage of non-MDR PA-positive cul- ANC less than 8.17 × 103 cells/µL: PA pneumonia inci- tures in the MEDI3902 1500 (44.6%) mg group versus dence was 17% and 2.8% in the placebo and MEDI3902 placebo (30.5%). Subjects in the MEDI3902 arms also 1500 mg group, respectively, corresponding to a 83.6% had higher ANC (13.2 vs. 11.4 × 103 cells per microliter RRR (80% CI 39.5%, 95.5%; P = 0.038) (Additional file 1: and PCT levels (1.0 vs. 0.61 μg/L) at baseline versus pla- Tables S7A and S7B). In subjects with ANC ≤ 8.17 × 103/ cebo (Table 1). All five components of the VAP preven- µL, MEDI3902 treatment was associated with 56.5% RRR tion bundles were implemented in 42.4% and 37.3% of in all-cause mortality (80% CI 11.9%, 79.4%; P = 0.110). MEDI3902 and placebo recipients, respectively (Table 1 Conversely, RRRs were negative in patients with PCT lev- and Additional file 1: Table S3). els above 0.55 μg/L and those with ANC above 8.17 × 103 cells/µL, denoting more PA pneumonia in the MEDI3902 Efficacy group than in the placebo group (Additional file 1: Tables The PA pneumonia incidence was 19/85 (22.4%) in S7A and S7B). MEDI3902 1500 mg recipients and 15/83 (18.1%) in pla- Healthcare resource utilisation through 21 days post- cebo recipients; RRR − 23.7% (80% CI − 83.8%, 16.8%; dose for MEDI3902 1500 mg and placebo subjects Chastre et al. Critical Care (2022) 26:355 Page 7 of 14 N=1023 N=835 Subjects screened* Screen failures Reasons: Did not meet incl/excl criteria (n=809) Withdrawal of consent (n=15) N=188 Lost to follow-up (n=1) Subjects randomised Other (n=10) N=85 N=16 N=87 Placebo MEDI3902 500 mg MEDI3902 1500 mg N=2 N=2 Not treated Not treated N=83 N=16 N=85 Treated Treated Treated N=63 N=20 N=12 N=4 N=59 N=26 Completed study Early discontinuation Completed study Early discontinuation Completed study Early discontinuation Reasons: Reasons: Reasons: Death (n=19) Death (n=3) Death (n=24) Other (n=1) Lost to follow-up (n=1) Lost to follow-up (n=1) Withdrawal of consent (n=1) Fig. 1 Subject disposition. *Subjects signed informed consent Table 2 MEDI3902 phase II EVADE: efficacy through study day 22 (mITT) MEDI3902 MEDI3902 Placebo (N = 83) RRR​a 80% ­CIa P ­valuea 500 mg (N = 16) 1500 mg (N = 85) Primary endpoint P. aeruginosa Pneumonia 2 (12.5%) 19 (22.4%) 15 (18.1%) − 23.7% − 83.8 to 16.8% 0.491 Exploratory endpoints of interest (FDA) ­ neumoniab All-cause p 3 (18.8%) 25 (29.4%) 17 (20.5%) − 43.6% − 104 to -1.1% 0.186 ­ eathc All-cause pneumonia or d 4 (25.0%) 30 (35.3%) 22 (26.5%) − 33.2% − 79.8 to 1.4% 0.222 RRR: relative risk reduction (a negative RRR denotes more PA pneumonia in the MEDI3902 1500 mg group compared to placebo); mITT = modified intent-to-treat population All pneumonias were determined by an adjudication committee a Relative risk reduction (MEDI3902 1500 mg vs. placebo), 80% confidence interval (CI), and p value based on Poisson regression with robust variance b All-cause pneumonia: P. aeruginosa pneumonia or non-P. aeruginosa pneumonia occurring through study day 22 c All-cause pneumonia or death: P. aeruginosa pneumonia, non-P. aeruginosa pneumonia, or death occurring through study day 22 with PA pneumonia were similar, including duration of placebo patients with PA pneumonia (Additional file 1: mechanical ventilation (17.5 vs. 19.7 days, respectively), Tables S8 and S10). The results of other exploratory anal- systemic antibiotic use (12.3 vs. 14.4 days) and duration of yses are described in the Additional file 1: Results, Tables ICU stay (20.6 vs. 22.0 days) (Additional file 1: Table S4). S4 and S8-11 and Fig. S4. Severity scores on day 1 of pneumonia onset and day of resolution, as well as changes in cellular and protein Pharmacokinetics markers of inflammation, also did not demonstrate any Mean serum concentration–time profiles of MEDI3902 significant differences between MEDI3902 patients and are shown in Figs. 2a and 3a, b. At 21 days following a Chastre et al. Critical Care (2022) 26:355 Page 8 of 14 1500 mg infusion, the mean (standard deviation [SD]) Discussion serum MEDI3902 concentration was 9.46 (7.91) μg/mL, The primary efficacy endpoint of reduction in PA pneu- with 80.6% (n = 58/72) subjects achieving concentra- monia incidence, with a relative reduction of 50% con- tions > 1.7 μg/mL, the target level (Table 3). In ETA, the sidered clinically meaningful based on expert advice and geometric mean (SD) MEDI3902 concentration was 8.73 published data [6], was not achieved. However, the cer- (± 11) ng/mL at day 21 (Fig. 2b and Additional file 1: Fig. tainty of our overall study findings may be limited given S5); however, significant correlations were not observed that the study was interrupted prematurely before the between MEDI3902 concentrations in serum and ETA target number of patients was reached due to slow enrol- (R2 = 0.000081; data not shown). MEDI3902 PK param- ment, and thus, the analyses may be underpowered. A eters in serum and ETA are summarised in Table 3. single IV dose of MEDI3902 1500 mg provided PK serum Mean clearance from serum was 1.27 L/day, with a mean exposure above the target level of 1.7 µg/mL for at least half-life of 5.65 days following a 1500 mg infusion. On 21 days in most subjects, with MEDI3902 concentra- day 2, the median MEDI3902 ETA/serum PK ratio was tions detectable in the ETA. The incidence of persistent 0.06%, increasing to 0.1% on day 4. A positive exposure– serum anti-drug antibodies was low, occurring in < 5% response relationship was observed after a 1500 mg IV of MEDI3902 1500 mg recipients. A positive exposure– infusion, with numerically higher serum MEDI3902 con- response relationship was observed for the MEDI3902 centrations observed in subjects who did not develop PA 1500 mg group, with a greater MEDI3902 area under pneumonia versus those who did (Fig. 3a, b and Addi- the concentration–time curve from time zero to 21 days tional file 1: Fig. S6). post-dose associated with a lower probability of PA pneumonia. Serum anti‑drug antibodies For mechanically ventilated critically ill subjects, high Serum anti-drug antibodies were detected at baseline in baseline inflammatory status or cachexia increases pro- 2/85 (2.4%) subjects in the MEDI3902 1500 mg group, tein catabolism and volume distribution of many drugs 2/16 (12.5%) in the 500 mg group, and 3/83 (3.6%) in the [22], potentially lowering MEDI3902 exposure and placebo group. The number of subjects with persistent increasing PA pneumonia susceptibility simultaneously. serum anti-drug antibodies (defined as a positive test of A higher MEDI3902 dose and/or direct administration at least 2 or the last post-baseline assessment) was simi- into the tracheobronchial tree by aerosolisation may be lar between the MEDI3902 1500 mg (n = 4/85; 4.9%) and considered for the most seriously ill subjects to achieve placebo (n = 4/83; 5.1%) groups. No differences were protection, especially since MEDI3902 ETA/serum PK observed in MEDI3902 PK or safety profiles in subjects ratio was low (0.1% on day 4). with and without post-baseline serum anti-drug antibod- An imbalance in baseline inflammation was reflected ies responses. in PCT and ANC levels. Subjects with lower baseline PCT or ANC had slightly higher mean PK levels and may Safety exhibit a greater MEDI3902 treatment response versus The incidence of TEAEs, SAEs, and TEAEs of > grade 3 placebo against PA pneumonia. These results provide severity reported through 49 days post-dose was simi- important lessons on pathogen-specific PA pneumo- lar between MEDI3902 1500 mg and placebo groups nia complexity in ICU trials with potential implications (Table 4). TEAEs related to treatment were reported for future study design. Whereas APACHE-II scores in three subjects receiving MEDI3902 1500 mg (sinus have been used as eligibility criteria to avoid enrolling tachycardia [n = 1, grade 1, resolved the same day] and the sickest patients [23], clinicians might also consider infusion-related reactions [n = 2, grade 1 and 3, resolved baseline levels of certain biomarkers as exclusion crite- during the study]) and in one placebo recipient (acute ria and/or for stratifying patients by severity at randomi- kidney injury, grade 4). No treatment-related SAEs were sation, particularly in pre-emptive treatment studies reported. of colonised patients. While PCT and ANC levels may Overall, 46 deaths were reported until 49 days not correlate directly with PA pneumonia, these mark- post-dose. A slightly higher number of deaths in the ers may identify patients with higher bacterial load and MEDI3902 1500 mg group (n = 24/85; 28.2%) versus pla- higher inflammatory status, and therefore a higher risk cebo (n = 19/83; 22.9%; Table 4) was not statistically sig- of pneumonia. [24–26]. Conversely, patients with higher nificant (RRR [80% CI] − 23.3% [− 73.3, 12.2%]; p = 0.43). levels of these biomarkers (regardless of apparent path- No deaths were considered related to MEDI3902, as ogen levels) may be too sick to benefit from treatment, adjudicated by the blinded investigators and independent or may have progressed too far in the development of unblinded DSMB. symptomatic pneumonia [27]. Whether MEDI3902 Chastre et al. Critical Care (2022) 26:355 Page 9 of 14 Fig. 2 Concentration–time profile following a single IV dose of MEDI3902 500 mg and 1500 mg in serum (a) and endotracheal aspirate (b) PK geometric mean profile. ETA = endotracheal aspirate. IV = intravenous. LLOQ = lower limit of quantification. PK = pharmacokinetic. SE = standard error could have increased the rate of PA pneumonia in no clinical benefit was observed in patients who were patients with high inflammatory status in the present seropositive at baseline, suggesting the possibility of a study remains highly speculative. However, some poten- conflict between the subjects’ own immune defences tial deleterious effects of monoclonal antibodies have and the monoclonal antibodies [28]. Further research is already been reported in other studies, including a large required to understand the role of these surrogate bio- randomised trial having assessed the benefit of a com- markers for the inflammatory response in risk stratifica- bination of casirivimab and imdevimab given together tion and treatment decisions. in patients admitted to the hospital with COVID-19. The safety profile of MEDI3902 was acceptable, with a Although the monoclonal combination improved sur- similar incidence of TEAEs and SAEs in the MEDI3902 vival and other clinical outcomes in patients who did not 1500 mg and placebo groups. TEAEs were generally have detectable anti-SARS-CoV-2 antibodies (i.e. had reflective of the critically ill patient population and the not yet mounted their own humoral immune response), numerical imbalance in deaths across treatment groups Chastre et al. Critical Care (2022) 26:355 Page 10 of 14 Fig. 3 a Mean (SE) serum MEDI3902 concentrations in subjects with or without PA pneumonia. LLOQ = lower limit of quantification. SE = standard error. b MEDI3902 area under the curve from 0 to 21 days and concentrations obtained after a single IV dose of 1500 mg in patients with and without PA pneumonia. Cmax = maximal observed concentration; ­C21 = concentration 21 days post-dosing; AUC​0to21 = Area under the curve from 0 to 21 days post-dose was expected due to the observed imbalance in baseline than expected, which resulted in an average of 4 patients disease characteristics. randomised per site. Other major contributing factors included the unusually complex screening process and Limitations the difficulty for the attending clinicians to distinguish This study has several limitations. First, the trial did not patients who were actually infected with PA from those achieve its planned sample size, with the recruitment only colonised, often resulting in the immediate admin- being stopped early due to low enrolment and thus was istration of new antibiotics and making patients ineli- underpowered to detect small but clinically important gible for randomisation. Second, the investigators may treatment effects in the entire study population, as well not have screened some eligible patients either because as in specific subgroups of patients. The main contrib- of the absence of consent or difficulties in identifying uting factor of low recruitment was that the proportion patients potentially colonised by PA. Since this num- of patients meeting all the eligibility criteria was lower ber was not recorded, the extent of this bias cannot be Chastre et al. Critical Care (2022) 26:355 Page 11 of 14 Table 3 MEDI3902 PK parameters PK parameter MEDI3902 500 mg (N = 16) MEDI3902 1500 mg (N = 85) n Mean (SD) n Mean (SD) Serum Cmax, µg/mL 16 87.6 (23.9) 84 299 (94.1) C21, µg/mL 14 2.56 (2.23) 72 9.46 (7.91) AUC​0–21, day*µg/mL 16 418 (126) 81 1410 (599) AUC​∞, day*µg/mL 16 440 (135) 81 1510 (675) CL, L/day 16 1.31 (0.64) 81 1.27 (0.86) t½, day 16 6.56 (4.03) 81 5.65 (2.69) n/N % n/N % Subjects with serum MEDI3902 levels > target level of 7/14 50.0% 58/72 80.6% 1.7 µg/mL on day 21 Endotracheal aspirate† Cmax, µg/mL – – n/a 0.083 AUC​0–21, day*µg/mL – – n/a 0.704 AUC​∞, day*µg/mL – – n/a 0.784 t½, day – – n/a 6.3 AUC​0–21 = the area under the concentration–time curve from time zero to 21 days post-dose. AUC​∞ = the area under the concentration–time curve from zero to infinity. CL = clearance. Cmax = maximum observed concentration. C21 = concentration 21 days post-dose. IV = intravenous. n/a = not applicable. SD = standard deviation. t½ = half-life † Due to the large variability in measured endotracheal aspirate concentrations and limited samples per subject, it was not plausible to calculate endotracheal aspirate PK parameter at an individual level. Therefore, endotracheal aspirate PK parameters were calculated based on geometric mean concentration–time profile Table 4 Overall summary of TEAEs through 49 days post-dose Subjects,* n (%) with ≥ 1 event MEDI3902 500 mg (N = 16) MEDI3902 1500 mg (N = 85) Placebo (N = 83) TEAE 15 (93.8) 84 (98.8) 81 (97.6) Treatment-related TEAE 0 (0.0) 3 (3.5) 1 (1.2) TEAE of ≥ grade 3 ­severity† 12 (75.0) 60 (70.6) 54 (65.1) Death (grade 5 severity†) 3 (18.8) 24 (28.2) 19 (22.9) Serious‡ TEAE 4 (25.0) 38 (44.7) 35 (42.2) Serious‡ and/or ≥ grade 3 ­severity† TEAE 12 (75.0) 60 (70.6) 55 (66.3) Treatment-related ­serious‡ TEAE 0 (0.0) 0 (0.0) 1 (1.2) TEAE leading to discontinuation of treatment 0 (0.0) 0 (0.0) 0 (0.0) AESI§ 0 (0.0) 2 (2.4) 1 (1.2) Treatment-related ­AESI§ 0 (0.0) 2 (2.4) 0 (0.0) AESI§ of ≥ grade 3 s­ everity† 0 (0.0) 1 (1.2) 1 (1.2) AE adverse event, AESI adverse event of special interest, TEAE treatment-emergent adverse event * Subjects are counted once for each category regardless of the number of events; †grade 3, severe; grade 4, life-threatening; grade 5, fatal; ‡serious adverse event criteria: death, life-threatening, required inpatient hospitalisation, prolongation of existing hospitalisation, persistent or significant disability/incapacity, important medical event, congenital anomaly/disorder (in the offspring of the subject); §AESI defined as targeted AEs of hepatic function abnormalities, hypersensitivity reactions (including anaphylaxis), infusion-related reactions, and immune complex disease (including vasculitis, endocarditis, neuritis, and glomerulonephritis) ascertained. Third, the patient population was primarily than just colonisation) could not be ruled out at the time Western European, thus reflecting the MDR profile from of treatment. Fifth, misclassification of pneumonia is also one region and the absence of strict guidelines for anti- possible due to difficulty in diagnosis, particularly once biotic treatment resulted in heterogeneity in antibiotic new antibiotics were prescribed. However, the trial was use. Fourth, colonisation, as determined by PCR, may not double-blinded and a strict protocol was applied to diag- coincide with a positive PA culture and, since colonisa- nose PA pneumonia. Sixth, MEDI3902 endotracheal aspi- tion assays were not quantitative, active infection (rather rate concentrations were rather low questioning whether Chastre et al. Critical Care (2022) 26:355 Page 12 of 14 adequate epithelial lining fluid levels of the monoclonal Czech Republic; Tomas Vymazal: Fakultni nemocnice v Motole, Czech Republic; Martin Novacek: Oblastni nemocnice Kolin, a.s., Czech Republic; antibody were present in these patients to prevent infec- Bruno François: Centre Hospitalier et Universitaire de Limoges, France; Djillali tion. Efficacy may have increased with higher antibody Annane: APHP Raymond-Poincaré de Garches, France; Jean Chastre: Groupe doses, which were not assessed in this study. Further- Hospitalier Pitié Salpétrière, France; Jean-Paul Mira: APHP Cochin, France; Bertrand Souweine: Centre Hospitalier Universitaire de Clermont Ferrand, more, results derived from post hoc analyses should be France; Pierre-François Dequin: CHRU de Tours, France; Ferhat Meziani: Nouvel interpreted with caution. Hôpital Civil Strasbourg, France; François Stephan: Centre Chirurgical Marie Lannelongue, France; Saadalla Nseir: CHRU Lille, France; Sebastien Gibot: CHRU Nancy, France; Carole Schwebel: Hôpital Albert Michallon La Tronche, France; Conclusions Alain Lepape: Centre Hospitalier Lyon Sud, France; Gaetan Plantefeve: Centre Hospitalier Victor Dupouy, France; Jean-Luc Diehl: APHP Hôpital Européen Among ICU patients requiring prolonged mechanical Georges-Pompidou, France; Christian Richard: APHP Hôpital de Bicêtre, France; ventilation and colonised by PA as detected by PCR, a Christian Lamer: Institut Mutualiste Montsouris, France; Kada Klouche: Centre single intravenous dose of 1500 mg of the bivalent, bispe- Hospitalier Universitaire de Montpellier/Lapeyronie hospital, France; Samir Jaber: Centre Hospitalier Universitaire de Montpellier/Hôpital St Eloi, France; cific monoclonal antibody MEDI3902 (gremubamab) Epaminondas Zakynthinos: University Hospital of Larissa, Greece; Georgios was safe, with a number of AEs similar to placebo, but Filntisis: Agioi Anargyroi Cancer Hospital, Greece; Apostolos Komnos: General it did not reduce PA nosocomial pneumonia incidence. Hospital of Larissa, Greece; Spyros Zakynthinos: Evangelismos General Hospital of Athens, Greece; Antonia Koutsoukou: Sotiria Chest Hospital of Athens, Whether MEDI3902 may help prevent PA pneumonia, Greece; Georgios Saroglou: Metropolitan Hospital, Greece; Charikleia Nikolaou: with higher doses or in more specific patient populations, Konstantopouleion General Hospital of Athens, Greece; Glykeria Vlachogianni: would require additional studies. Agios Dimitrios General Hospital of Thessaloniki, Greece; Ioannis Pnevmatikos: University Hospital of Alexandroupolis, Greece; Konstantinos Mandragos: General Hospital of Athens Korgialenio Benakio Greek Red Cross, Greece; Ildiko Kremer: Pest Megyei Flór Ferenc Kórház, Hungary; Zsolt Dezso Rozgonyi: Abbreviations Orszagos Koranyi TBC es Pulmonologiai Intezet, Hungary; Zsuzsa Marjanek: AE: Adverse event; ANC: Absolute neutrophil count; APACHE-II: Acute Jávorszky Ödön Kórház, Hungary; Ignacio Martin-Loeches: St James University Physiology and Chronic Health Evaluation-II; BAL: Bronchoalveolar lavage; CI: Hospital, Ireland; Pierre Singer: Rabin Medical Center, Israel; Vernon Van Confidence interval; CPIS: Clinical pulmonary infection score; EAC: Endpoint Heerden: Hadassah Medical Center, Israel; Yehuda Carmeli: Tel Aviv Sourasky Adjudication Committee; ETA: Endotracheal aspirate; ICU: Intensive care unit; Medical Center, Israel; Galia Rahav: Chaim Sheba Medical Center, Israel; Pedro IV: Intravenous; MDR: Multidrug-resistant; mITT: Modified intent-to-treat; PA: Povoa: Centro Hospital de Lisboa Ocidental – Hospital São Francisco Xavier, Pseudomonas aeruginosa; PCR: Polymerase chain reaction; PCT: Procalcitonin; Portugal; Antonio Alvarez Seoane: Centro Hospitalar Lisboa Norte, E.P.E. – Hos‑ PK: Pharmacokinetic; RRR​: Relative risk reduction; SAE: Serious adverse event; pital de Santa Maria, Portugal; Pedro Moura: Unidade Local de Saúde do Alto SAP: Statistical analysis plan; SOFA: Sequential Organ Failure Assessment; TEAE: Minho, EPE, Portugal; Filipe Gonzalez: Hospital Garcia de Orta, Portugal; Paula Treatment-emergent adverse event; VAP: Ventilator-associated pneumonia. Ramirez: Hospital Universitari i Politecnic La Fe de Valencia, Spain; Antonio Tor‑ res Marti: Hospital Clinic de Barcelona, Spain; Miguel Sánchez-García: Hospital Supplementary Information Clínico San Carlos Madrid, Spain; Ricard Ferrer Roca: Hospital Universitario Vall d’Hebron Barcelona, Spain; Lorena Oteiza: Hospital Universitario de Getafe The online version contains supplementary material available at https://​doi.​ Madrid, Spain; Dolores Escudero: Hospital Universitario Central de Asturias org/​10.​1186/​s13054-​022-​04204-9. Oviedo, Spain; Enrique Piacentini: Hospital Mutua de Terrassa Barcelona, Spain; Paula Vera: Hospital de La Santa Creu i Sant Pau, Spain; Luis Tamayo: Hospital Additional file 1: Supplementary Methods, Results, Tables and Figures. Universitario del Rio Hortega, Spain; Miguel Angel Gonzalez Gallego: Hospital Universitario Infanta Sofia, Spain; Borja Suberviola Canas: Hospital Universitario Marqués de Valdecilla, Spain; Iglesias Figueira: Hospital Universitario La Paz Acknowledgements – PPDS, Spain; Rafael Leon: Hospital General Universitario Reina Sofia, Spain; This research project receives support from the Innovative Medicines Initiative Volkan Korten: Marmara University Research and Training Hospital, Turkey; Joint Undertaking under grant agreement no 115737, with resources compris‑ Iftihar Koksal: Karadeniz Technical University Faculty of Medicine, Turkey; ing financial contributions from the European Union Seventh Framework Murat Akova: Hacettepe Universitesi Tip Fakultesi Hastanesi, Turkey; Duncan Programme (FP7/2007–2013) and EFPIA companies in-kind contribution. Wyncoll: St Thomas’ Hospital, UK; Tony Whitehouse: Queen Elizabeth Hospital, Research reported in this publication was supported by the National Institute UK; Phil Hopkins: King’s College Hospital, UK; Malcolm Sim: Southern General Of Allergy And Infectious Diseases of the National Institutes of Health under Hospital, UK; Yoav Golan: Tufts University Medical Center, UK; Marcus Zervos: award number UM1AI104681. The content is solely the responsibility of the Henry Ford Health Sys. Detroit, USA; Jose Vazquez: Georgia Regents Medical authors and does not necessarily represent the official views of the National Center-Augusta, USA; Kartikeya Cherabuddi: University of Florida, USA; George Institutes of Health. Medical writing support, under the direction of the Smulian: Univ. of Cincinnati, USA; Nadine Rouphael: Emory University Atlanta, authors, was provided by Claire Cairney, Ph.D. and Jennifer Stewart, PhD, MBA, USA; James Welker: Anne Arundel Health, USA; Mathew Sims: Beaumont CMC Connect, McCann Health Medical Communications, and was funded by Hospital Royal Oaks, USA; David Van Duin: UNC Chapel Hill, USA; Todd McCa‑ AstraZeneca, Gaithersburg, USA. The additional writing support was provided rthy: Univ. of Alabama, Birmingham, USA; Christopher Polk: Carolina Medical by Sarah Legrand Demai, MA, Inserm CIC 1435, CHU Limoges, France. Center/Atrium Health, USA. The COMBACTE-MAGNET EVADE Study Group: Michael Joannidis: Medical University of Innsbruck, Austria; Walter Klimscha: SMZOST Donauspital Wien, Author contributions Austria; Elisabeth De Waele: UZ Brussel, Belgium; Nicolas De Schryver: Clinique JC, BF, MS-G, AT, PE, DK, TLH, OA, KS, PR, AR, YW, AD, SC, DV, TB, MTE, and HSJ, Saint-Pierre, Belgium; Jacques Devriendt: CHU Brugmann, Belgium; Vincent conceptualised and designed the study. JC, BF, MBou, AK, RF, GR, NDS, AL, IK, Huberlant: Centre Hospitalier Jolimont-Lobbes, Belgium; Pieter Depuydt: C-EL, MS-G, AT, PE, OA, DET, AA, YW, AD, DV, FC, SM-K, LT, AO, and OB, collected University Hospital Gent, Belgium; Marc Bourgeois: AZ Sint-Jan AV, Belgium; the data. JC, BF, MS-G, AT, PE, DK, TLH, OA, KS, PR, JS, AR, DET, YJ, AD, MBon, HG, Sam Van Boxstael: Ziekenhuis Oost-Limburg, Belgium; Mladen Peric: Klinichki CR, MTE, and HSJ, analysed and interpreted the data. JC, BF, KS, PR, AR, and HSJ, Bolnicki Centar Zagreb, Croatia; Jasminka Kopic: General Hospital Dr Josip verified the data. JC, BF, AR, MTE, and HSJ, drafted the manuscript. All authors Bencevic, Croatia; Michal Hanauer: Krajska zdravotni, a.s. – Nemocnice Decin, had full access to study data, contributed to the development of the manu‑ o.z., Czech Republic; Tomas Hruby: Krajska zdravotni, a.s. – Nemocnice Teplice, script with medical writing support funded by the sponsor, critically reviewed o.z., Czech Republic; Vladimir Sramek: Fakultni nemocnice u sv. Anny v Brne, and revised the manuscript. AR and MTE accessed and verified the data. JC Czech Republic; Petr Svoboda: Nemocnice Kyjov, prispevkova organizace, Chastre et al. Critical Care (2022) 26:355 Page 13 of 14 and HSJ had final responsibility for the decision to submit for publication. All Polyvalente and Inserm CIC 1435 & UMR 1092, CHU, Limoges, France. 3 AZ Sint- authors read and approved the final manuscript. Jan Brugge-Oostende AV, Brugge, Belgium. 4 General Hospital of Larissa, Lar‑ issa, Greece. 5 SODIR‑VHIR Research Group, Hospital Universitari Vall d’Hebron, Funding Barcelona, Spain. 6 Chaim Sheba Medical Center, Sackler School of Medicine, This research project receives support from the Innovative Medicines Initiative Tel Aviv University, Tel Aviv, Israel. 7 Clinique Saint-Pierre, Ottignies, Belgium. 8 Joint Undertaking under grant agreement n° 115737 resources of which are Hospices Civils de Lyon Hôpital Lyon Sud, Lyon, France. 9 Faculty of Medicine, composed of financial contribution from the European Union Seventh Frame‑ Trabzon and Acibadem University Faculty of Medicine, Karadeniz Technical work Programme (FP7/2007‐2013) and EFPIA companies in-kind contribution. University, Istanbul, Turkey. 10 Critical Care Department, Hospital Clínico San Other support from the National Institute of Allergy And Infectious Diseases of Carlos, Universidad Complutense, Madrid, Spain. 11 Servei de Pneumologia, the National Institutes of Health, AstraZeneca. Hospital Clinic, University of Barcelona, IDIBAPS, CIBERES, ICREA, Barcelona, Spain. 12 Department of Locomotor Apparatus, Centre Hospitalier Universitaire Availability of data and materials Vaudois CHUV, Lausanne, Switzerland. 13 The University of Queensland Centre Data underlying the findings described in this manuscript can be obtained in for Clinical Research, Faculty of Medicine, The University of Queensland, Bris‑ accordance with AstraZeneca’s data sharing policy. bane, QLD, Australia. 14 2nd Critical Care Department, Attikon University Hos‑ pital, National and Kapodistrian, University of Athens, Athens, Greece. 15 Duke Clinical Research Institute, Durham, NC, USA. 16 Clinical Research and Develop‑ Declarations ment, Vaccines and Immune Therapies, BioPharmaceuticals R&D, AstraZen‑ eca Biopharmaceuticals, One MedImmune Way, Gaithersburg, MD 20878, Ethics approval and consent to participate USA. 17 Respiratory and Immunology, BioPharmaceuticals R&D, AstraZeneca, The study was conducted within the European public–private partnership Gaithersburg, MD, USA. 18 Infection Control Program, Faculty of Medicine, Combatting Bacterial Resistance In Europe—Molecules Against Gram-Nega‑ Geneva University Hospitals, Geneva, Switzerland. 19 Clinical Pharmacology tive Infections (COMBACTE-MAGNET) consortium in accordance with the ethi‑ and Quantitative Pharmacology, Clinical Pharmacology and Safety Sciences, cal principles of the Declaration of Helsinki and the International Council for BioPharmaceuticals R&D, AstraZeneca, Gaithersburg, MD, USA. 20 Oncology, US Harmonization Guidance for Good Clinical Practice. Study-related documents SM&M, AstraZeneca, Wilmington, NC, USA. 21 Inserm CIC 1435, CHU, Limoges, were reviewed and approved by the local independent ethics committees France. 22 Department of Medical Microbiology, University Medical Center or institutional review boards. All subjects/legally acceptable representatives Utrecht, Utrecht University, Utrecht, The Netherlands. 23 Laboratory of Medical provided written informed consent. Microbiology, Vaccine and Infectious Disease Institute, University of Antwerp, Antwerp, Belgium. 24 Servicio de Microbiología y Unidad de Investigación, Consent for publication Hospital Universitari Son Espases, Institut d’Investigació Sanitaria Illes Balears, Not applicable. Palma, Spain. 25 INSERM U1092, Centre Hospitalier Universitaire de Limoges, Université Limoges, Limoges, France. 26 Julius Center for Health Science Competing interests and Primary Care, University Medical Center Utrecht, Utrecht University, Jean Chastre received personal fees during the conduct of the study from Utrecht, The Netherlands. 27 DevPro Biopharma, Basking Ridge, NJ, USA. COMBACTE-MAGNET, personal fees from outside of the submitted work 28 Late‑Stage Development, Respiratory and Immunology, BioPharmaceuticals from Aridis, Bayer, Inotrem, Shionogi, and Tigenix/Takeda, and grants from R&D, AstraZeneca, Gaithersburg, MD, USA. AstraZeneca/Medimmune. Bruno François consulted for AM-Pharma, Aridis, Enlivex, GSK, and Inotrem and was part of an adjudication committee for Received: 12 September 2022 Accepted: 11 October 2022 Takeda. Ricard Ferrer received fees for conferences from BD, Grifols, MSD, and Pfizer. Alain Lepape received personal fees outside of the study from Fresenius and Tigenix/Takeda (DSMD committee). Iftihar Koksal has served on advisory boards for Abbvie, MSD, Pfizer, Gilead, GSK, and Roche; lectures: Abbvie, Gilead, Pfizer, and MSD. Charles-Edouard Luyt received fees from Bayer References Healthcare, ThermoFisher Brahms, Biomérieux, Faron, Carmat, Aerogen, and 1. Ramírez-Estrada S, Borgatta B, Rello J. Pseudomonas aeruginosa ventilator- Merck Sharp & Dohme outside of the submitted work. Miguel Sánchez-García associated pneumonia management. Infect Drug Resist. 2016;9:7–18. received funding for speaker fees from Biotest AG, Pfizer, Merck, Sharp & 2. Kollef MH, Chastre J, Fagon JY, Francois B, Niederman MS, Rello J, et al. Dohme, AstraZeneca, Orion, and Cepheid; for consulting fees from Bayer, Glax‑ Global prospective epidemiologic and surveillance study of ventilator- oSmithKline, Pfizer, and Masimo; for research grants from the European Union, associated pneumonia due to Pseudomonas aeruginosa. Crit Care Med. 7th Framework Programme IMI, H2020; was part of an adjudication committee 2014;42(10):2178–87. for Takeda. Antoni Torres has served on advisory boards for Pfizer, MSD, Bio‑ 3. Jones RN. Microbial etiologies of hospital-acquired bacterial pneu‑ mérieux, Menarini, Chiesi, and Jansen; lectures: Pfizer and MSD; grants: Bayer, monia and ventilator-associated bacterial pneumonia. Clin Infect Dis. AstraZeneca, and Cardeas. Despoina Koulenti was part of the adjudication 2010;51(Suppl. 1):S81–7. committee of EVADE. Thomas L. Holland consulted for Basilea Pharmaceutica, 4. Micek ST, Wunderink RG, Kollef MH, Chen C, Rello J, Chastre J, et al. An Genentech, and Motif Bio and took part in a Scientific Advisory Board for Motif international multicenter retrospective study of Pseudomonas aeruginosa Bio. Antonio Oliver received grants from and participated as a speaker and in nosocomial pneumonia: impact of multidrug resistance. Crit Care. advisory boards for Pfizer, MSD, and Shionogi. Olivier Barraud received speaker 2015;19(1):219. fees and/or travel grants from MSD, Pfizer, Roche, and Sanofi and has been a 5. Tumbarello M, De Pascale G, Trecarichi EM, Spanu T, Antonicelli F, Maviglia consultant to bioMérieux and Mylan. Herman Goossens received grants from R, et al. Clinical outcomes of Pseudomonas aeruginosa pneumonia in European Union IMI Grant (in collaboration with Novartis). Omar Ali, Ahmad intensive care unit patients. Intensive Care Med. 2013;39(4):682–92. Akhgar, Pin Ren, Terramika Bellamy, Colin Reisner, Alexey Ruzin and Hasan S. 6. François B, Luyt CE, Dugard A, Wolff M, Diehl JL, Jaber S, et al. Safety and Jafri were employees of AstraZeneca during the conduct of the study and pharmacokinetics of an anti-PcrV PEGylated monoclonal antibody frag‑ hold shares in the company. Kathryn Shoemaker, David E. Tabor, Yuling Wu, Yu ment in mechanically ventilated patients colonized with Pseudomonas Jiang, Antonio DiGiandomenico, Susan Colbert and Mark Esser are employees aeruginosa: a randomized, double-blind, placebo-controlled trial. Crit of AstraZeneca and hold shares in the company. Marc Bourgeois, Aposto‑ Care Med. 2012;40(8):2320–6. los Komnos, Galia Rahav, Nicolas De Schryver, Drieke Vandamme, Surbhi 7. François B, Chastre J, Eggiman P, Laterre PF, Torres A, Sanchez M, et al. The Malhotra-Kumar, Philippe Eggimann, Julien Sauser Frank Coenjaerts, Leen SAATELLITE and EVADE clinical studies within the COMBACTE consor‑ Timbermont, and Marc Bonten have no conflicts of interest to disclose. tium: a public-private collaborative effort in designing and performing clinical trials for novel antibacterial drugs to prevent nosocomial pneu‑ Author details 1 monia. Clin Infect Dis. 2016;63(Suppl. 2):S46-51. Service de Médecine Intensive Réanimation, Institut de Cardiologie, 8. Torres A, Niederman MS, Chastre J, Ewig S, Fernandez-Vandellos P, Han‑ Groupe Hospitalier Pitié‑Salpêtrière, Assistance Publique‑Hôpitaux de Paris, berger H, et al. International ERS/ESICM/ESCMID/ALAT guidelines for the Sorbonne University, 47–83 Bd de l’Hôpital, 75651 Paris, France. 2 Réanimation Chastre et al. Critical Care (2022) 26:355 Page 14 of 14 management of hospital-acquired pneumonia and ventilator-associated 26. Huang Z, Fu Z, Huang W, Huang K. Prognostic value of neutrophil- pneumonia: guidelines for the management of hospital-acquired pneu‑ to-lymphocyte ratio in sepsis: a meta-analysis. Am J Emerg Med. monia (HAP)/ventilator-associated pneumonia (VAP) of the European 2020;38(3):641–7. Respiratory Society (ERS), European Society of Intensive Care Medicine 27. Melsen WG, Rovers MM, Groenwold RHH, Bergmans DC, Camus C, Bauer (ESICM), European Society of Clinical Microbiology and Infectious TT, et al. Attributable mortality of ventilator-associated pneumonia: a Diseases (ESCMID) and Asociación Latinoamericana del Tórax (ALAT). Eur meta-analysis of individual patient data from randomised prevention Respir J. 2017;50(3):1700582. studies. Lancet Infect Dis. 2013;13(8):665–71. 9. Ferrer R, Martin-Loeches I, Phillips G, Osborn TM, Townsend S, Dellinger 28. Group RC. Casirivimab and imdevimab in patients admitted to hospital RP, et al. Empiric antibiotic treatment reduces mortality in severe sepsis with COVID-19 (RECOVERY): a randomised, controlled, open-label, plat‑ and septic shock from the first hour: results from a guideline-based form trial. Lancet (Lond Engl). 2022;399(10325):665–76. performance improvement program. Crit Care Med. 2014;42(8):1749–55. 29. Magiorakos AP, Srinivasan A, Carey RB, Carmeli Y, Falagas ME, Giske CG, 10. American Thoracic Society, Infectious Diseases Society of America. Guide‑ et al. Multidrug-resistant, extensively drug-resistant and pandrug-resist‑ lines for the management of adults with hospital-acquired, ventilator- ant bacteria: an international expert proposal for interim standard defini‑ associated, and healthcare-associated pneumonia. Am J Respir Crit Care tions for acquired resistance. Clin Microbiol Infect. 2012;18(3):268–81. Med. 2005;171(4):388–416. 11. Pang Z, Raudonis R, Glick BR, Lin TJ, Cheng Z. Antibiotic resistance in Pseu- domonas aeruginosa: mechanisms and alternative therapeutic strategies. Publisher’s Note Biotechnol Adv. 2019;37(1):177–92. Springer Nature remains neutral with regard to jurisdictional claims in pub‑ 12. Kalil AC, Metersky ML, Klompas M, Muscedere J, Sweeney DA, Palmer lished maps and institutional affiliations. LB, et al. Management of adults with hospital-acquired and ventilator- associated pneumonia: 2016 clinical practice guidelines by the Infectious Diseases Society of America and the American Thoracic Society. Clin Infect Dis. 2016;63(5):e61–111. 13. Tümmler B. Emerging therapies against infections with Pseudomonas aeruginosa. F1000Res. 2019;8(F1000 Faculty Rev-1371):1371. 14. Que YA, Lazar H, Wolff M, François B, Laterre PF, Mercier E, et al. Assess‑ ment of panobacumab as adjunctive immunotherapy for the treatment of nosocomial Pseudomonas aeruginosa pneumonia. Eur J Clin Microbiol Infect Dis. 2014;33(10):1861–7. 15. DiGiandomenico A, Keller AE, Gao C, Rainey GJ, Warrener P, Camara MM, et al. A multifunctional bispecific antibody protects against Pseudomonas aeruginosa. Sci Transl Med. 2014;6(262):262ra155. 16. Ali SO, Yu XQ, Robbie GJ, Wu Y, Shoemaker K, Yu L, et al. Phase 1 study of MEDI3902, an investigational anti-Pseudomonas aeruginosa PcrV and Psl bispecific human monoclonal antibody, in healthy adults. Clin Microbiol Infect. 2019;25(5):629.e1–.e6. 17. Le HN, Tran VG, Vu TTT, Gras E, Le VTM, Pinheiro MG, et al. Treatment efficacy of MEDI3902 in Pseudomonas aeruginosa bloodstream infection and acute pneumonia rabbit models. Antimicrob Agents Chemother. 2019;63(8):e00710-e719. 18. Le HN, Quetz JS, Tran VG, Le VTM, Aguiar-Alves F, Pinheiro MG, et al. MEDI3902 correlates of protection against severe Pseudomonas aerugi- nosa pneumonia in a rabbit acute pneumonia model. Antimicrob Agents Chemother. 2018;62(5):e02565-e2617. 19. Francois B, Jafri HS, Chastre J, Sanchez-Garcia M, Eggimann P, Dequin PF, et al. Efficacy and safety of suvratoxumab for prevention of Staphylococ- cus aureus ventilator-associated pneumonia (SAATELLITE): a multicentre, randomised, double-blind, placebo-controlled, parallel-group, phase 2 pilot trial. Lancet Infect Dis. 2021;21(9):1313–23. 20. Yu XQ, Robbie GJ, Wu Y, Esser MT, Jensen K, Schwartz HI, et al. Safety, tolerability, and pharmacokinetics of MEDI4893, an investigational, extended-half-life, anti-staphylococcus aureus alpha-toxin human monoclonal antibody. Healthy Adults Antimicrob Agents Chemother. 2017;61(1):e01020-e1116. 21. Guo X, François B, Bourgeois M, Komnos A, Rozgonyi ZD, Ali SA, et al. Interim pharmacokinetic analysis from the EVADE phase-2 clinical trial of MEDI3902, a bispecific monoclonal antibody against PcrV and Psl of Pseudomonas aeruginosa. In: 2018 E, editor. 28th European congress of clinical microbiology and infectious disease; 21–24 April 2018; Madrid, Spain: ECCMID; 2018. p. P2213. 22. O’Leary-Kelley C, Bawel-Brinkley K. Nutrition support protocols: enhanc‑ ing delivery of enteral nutrition. Crit Care Nurse. 2017;37(2):e15–23. 23. Zeng J, Wang CT, Zhang FS, Qi F, Wang SF, Ma S, et al. Effect of probiot‑ ics on the incidence of ventilator-associated pneumonia in critically ill patients: a randomized controlled multicenter trial. Int Care Med. 2016;42(6):1018–28. 24. Marik PE, Stephenson E. The ability of procalcitonin, lactate, white blood cell count and neutrophil-lymphocyte count ratio to predict blood stream infection. Analysis of a large database. J Crit Care. 2020;60:135–9. 25. Gregoriano C, Heilmann E, Molitor A, Schuetz P. Role of procalcitonin use in the management of sepsis. J Thorac Dis. 2020;12(Suppl 1):S5–15.