Machine learning methods and harmonized datasets improve immunogenic neoantigen prediction.

The accurate selection of neoantigens that bind to class I human leukocyte antigen (HLA) and are recognized by autologous T cells is a crucial step in many cancer immunotherapy pipelines. We reprocessed whole-exome sequencing and RNA sequencing (RNA-seq) data from 120 cancer patients from two external large-scale neoantigen immunogenicity screening assays combined with an in-house dataset of 11 patients and identified 46,017 somatic single-nucleotide variant mutations and 1,781,445 neo-peptides, of which 212 mutations and 178 neo-peptides were immunogenic. Beyond features commonly used for neoantigen prioritization, factors such as the location of neo-peptides within protein HLA presentation hotspots, binding promiscuity, and the role of the mutated gene in oncogenicity were predictive for immunogenicity. The classifiers accurately predicted neoantigen immunogenicity across datasets and improved their ranking by up to 30%. Besides insights into machine learning methods for neoantigen ranking, we have provided homogenized datasets valuable for developing and benchmarking companion algorithms for neoantigen-based immunotherapies.

Dynamic readers for 5-(hydroxy)methylcytosine and its oxidized derivatives.

Tet proteins oxidize 5-methylcytosine (mC) to generate 5-hydroxymethyl (hmC), 5-formyl (fC), and 5-carboxylcytosine (caC). The exact function of these oxidative cytosine bases remains elusive. We applied quantitative mass-spectrometry-based proteomics to identify readers for mC and hmC in mouse embryonic stem cells (mESC), neuronal progenitor cells (NPC), and adult mouse brain tissue. Readers for these modifications are only partially overlapping, and some readers, such as Rfx proteins, display strong specificity. Interactions are dynamic during differentiation, as for example evidenced by the mESC-specific binding of Klf4 to mC and the NPC-specific binding of Uhrf2 to hmC, suggesting specific biological roles for mC and hmC. Oxidized derivatives of mC recruit distinct transcription regulators as well as a large number of DNA repair proteins in mouse ES cells, implicating the DNA damage response as a major player in active DNA demethylation.

Demonstration of fault-tolerant universal quantum gate operations.

Quantum computers can be protected from noise by encoding the logical quantum information redundantly into multiple qubits using error-correcting codes1,2. When manipulating the logical quantum states, it is imperative that errors caused by imperfect operations do not spread uncontrollably through the quantum register. This requires that all operations on the quantum register obey a fault-tolerant circuit design3-5, which, in general, increases the complexity of the implementation. Here we demonstrate a fault-tolerant universal set of gates on two logical qubits in a trapped-ion quantum computer. In particular, we make use of the recently introduced paradigm of flag fault tolerance, where the absence or presence of dangerous errors is heralded by the use of auxiliary flag qubits6-10. We perform a logical two-qubit controlled-NOT gate between two instances of the seven-qubit colour code11,12, and fault-tolerantly prepare a logical magic state8,13. We then realize a fault-tolerant logical T gate by injecting the magic state by teleportation from one logical qubit onto the other14. We observe the hallmark feature of fault tolerance-a superior performance compared with a non-fault-tolerant implementation. In combination with recently demonstrated repeated quantum error-correction cycles15,16, these results provide a route towards error-corrected universal quantum computation.

Realizing repeated quantum error correction in a distance-three surface code.

Quantum computers hold the promise of solving computational problems that are intractable using conventional methods1. For fault-tolerant operation, quantum computers must correct errors occurring owing to unavoidable decoherence and limited control accuracy2. Here we demonstrate quantum error correction using the surface code, which is known for its exceptionally high tolerance to errors3-6. Using 17 physical qubits in a superconducting circuit, we encode quantum information in a distance-three logical qubit, building on recent distance-two error-detection experiments7-9. In an error-correction cycle taking only 1.1 µs, we demonstrate the preservation of four cardinal states of the logical qubit. Repeatedly executing the cycle, we measure and decode both bit-flip and phase-flip error syndromes using a minimum-weight perfect-matching algorithm in an error-model-free approach and apply corrections in post-processing. We find a low logical error probability of 3% per cycle when rejecting experimental runs in which leakage is detected. The measured characteristics of our device agree well with a numerical model. Our demonstration of repeated, fast and high-performance quantum error-correction cycles, together with recent advances in ion traps10, support our understanding that fault-tolerant quantum computation will be practically realizable.

Experimental deterministic correction of qubit loss.

The successful operation of quantum computers relies on protecting qubits from decoherence and noise, which-if uncorrected-will lead to erroneous results. Because these errors accumulate during an algorithm, correcting them is a key requirement for large-scale and fault-tolerant quantum information processors. Besides computational errors, which can be addressed by quantum error correction1-9, the carrier of the information can also be completely lost or the information can leak out of the computational space10-14. It is expected that such loss errors will occur at rates that are comparable to those of computational errors. Here we experimentally implement a full cycle of qubit loss detection and correction on a minimal instance of a topological surface code15,16 in a trapped-ion quantum processor. The key technique used for this correction is a quantum non-demolition measurement performed via an ancillary qubit, which acts as a minimally invasive probe that detects absent qubits while imparting the smallest quantum mechanically possible disturbance to the remaining qubits. Upon detecting qubit loss, a recovery procedure is triggered in real time that maps the logical information onto a new encoding on the remaining qubits. Although the current demonstration is performed in a trapped-ion quantum processor17, the protocol is applicable to other quantum computing architectures and error correcting codes, including leading two- and three-dimensional topological codes. These deterministic methods provide a complete toolbox for the correction of qubit loss that, together with techniques that mitigate computational errors, constitute the building blocks of complete and scalable quantum error correction.

A genome-wide genetic association study reveals SNPs significantly associated with environmental variables and specific leaf area in European beech.

European beech is negatively affected by climate change and a further growth decline is predicted for large parts of its distribution range. Despite the importance of this species, little is known about its genetic adaptation and especially the genetic basis of its physiological traits. Here, we used genotyping by sequencing to identify SNPs in 43 German European beech populations growing under different environmental conditions. In total, 28 of these populations were located along a precipitation and temperature gradient in northern Germany, and single tree-based hydraulic and morphological traits were available. We obtained a set of 13,493 high-quality SNPs that were used for environmental and SNP-trait association analysis. In total, 22 SNPs were identified that were significantly associated with environmental variables or specific leaf area (SLA). Several SNPs were located in genes related to stress response. The majority of the significant SNPs were located in non-coding (intergenic and intronic) regions. These may be in linkage disequilibrium with the causative coding or regulatory regions. Our study gives insights into the genetic basis of abiotic adaptation in European beech, and provides genetic resources that can be used in future studies on this species. Besides clear patterns of local adaptation to environmental conditions of the investigated populations, the analyzed morphological and hydraulic traits explained most of the explainable genetic variation. Thus, they could successfully be altered in tree breeding programs, which may help to increase the adaptation of European beech to changing environmental conditions in the future.

Capillaroscopic differences between primary Raynaud phenomenon and healthy controls indicate potential microangiopathic involvement in benign vasospasms.

BACKGROUND: For primary Raynaud phenomenon (PRP), an otherwise unexplained vasospastic disposition is assumed. To test the hypothesis of an additional involvement of distinct ultrastructural microvascular alterations, we compared the nailfold capillary pattern of patients with PRP and healthy controls. METHODS: A total of 120 patients with PRP (with a median duration of vasospastic symptoms of 60 [IQR: 3-120] months) were compared against 125 controls. In both groups, nailfold capillaroscopy was performed to record the presence of dilatations, capillary edema, tortuous capillaries, ramifications, hemorrhages, and reduced capillary density and to determine a semiquantitative rating score. Further, the capacity of finger skin rewarming was investigated by performing infrared thermography in combination with cold provocation. RESULTS: Unspecific morphologic alterations were found in both, PRP, such as controls, whereby the risk for PRP was four times as high in the presence of capillary dilations (CI: 2.3-7.6) and five times as high if capillary density was reduced (CI: 1.9-13.5). Capillary density correlated with thermoregulatory capacity in both hands in the PRP group, but not in controls. In addition, a negative correlation between the microangiopathy score and the percentage degree of rewarming in both hands was found for patients with PRP only. CONCLUSION: We found specific differences within the microvascular architecture between patients with PRP and controls. As a conclusion, PRP may not be an entirely benign vasospastic phenomenon, but might be associated with subtle microcirculatory vasculopathy. In addition, we suggest that the implementation of a scoring system might serve as guidance in the diagnostic process at least of patients with long-standing PRP.

Driveline infections in left ventricular assist devices-Incidence, epidemiology, and staging proposal.

BACKGROUND: Driveline infections (DLI) are a serious complication in patients with left ventricular assist devices (LVAD). Apart from the differentiation between superficial and deep DLI, there is no consensus on the classification of the severity of DLI. Little is known about risk factors and typical bacteria causing DLI in centrifugal-flow LVADs. METHODS: In this single-center study with 245 patients, DLI were classified by their local appearance using a modification of a score suggested by the Sharp Memorial group. The driveline exit site was inspected routinely every 6 months. RESULTS: Severe DLI were detected in 34 patients (15%) after 6 months and in 24 patients (22%) after 24 months. The proportion of patients with DLI increased significantly during the follow-up (p = 0.0096). The most common bacteria in local smears were Corynebacterium, coagulase-negative Staphylococcus, and Staphylococcus aureus. Fifty-nine patients were hospitalized more than once for DLI. In these patients, S. aureus was the most common bacterium. It was also the most common bacterium in blood cultures. Higher BMI, no partnership, and a HeartMate 3 device were identified as risk factors for DLI in a multivariable cause-specific Cox regression. CONCLUSION: This study is a standardized analysis of DLI in a large cohort with centrifugal-flow LVADs.

Fourier phase index for extracting signatures of determinism and nonlinear features in time series.

Detecting determinism and nonlinear properties from empirical time series is highly nontrivial. Traditionally, nonlinear time series analysis is based on an error-prone phase space reconstruction that is only applicable for stationary, largely noise-free data from a low-dimensional system and requires the nontrivial adjustment of various parameters. We present a data-driven index based on Fourier phases that detects determinism at a well-defined significance level, without using Fourier transform surrogate data. It extracts nonlinear features, is robust to noise, provides time-frequency resolution by a double running window approach, and potentially distinguishes regular and chaotic dynamics. We test this method on data derived from dynamical models as well as on real-world data, namely, intracranial recordings of an epileptic patient and a series of density related variations of sediments of a paleolake in Tlaxcala, Mexico.

[The Role of the Anaesthesiologist in Liver Transplantation - Preoperative Evaluation]. / Die Rolle des Anästhesisten bei Lebertransplantationen ­ präoperative Evaluation.

Preoperative evaluation prior to listing for orthotopic liver transplantation (LT) requires a careful multidisciplinary approach with specialized teams including surgeons, hepatologists and anesthesiologists in order to improve short- and long-term clinical outcomes. Due to inadequate supply of donor organs and changing demographics, patients listed for LT have become older, sicker and share more comorbidities. As cardiovascular events are the leading cause for early mortality precise evaluation of risk factors is mandatory. This review focuses on the detection and management of coronary artery disease, cirrhotic cardiomyopathy, portopulmonary hypertension and hepatopulmonary syndrome in patients awaiting LT. Further insights are being given into scoring systems, patients with Acute-on-chronic-liver-failure (ACLF), frailty, NASH cirrhosis and into psychologic evaluation of patients with substance abuse.

Efficient Tandem Copper-Catalyzed Click Synthesis of Multisugar-Modified Oligonucleotides.

Nucleic acids in the form of siRNA, antisense oligonucleotides or mRNA are currently explored as new promising modalities in the pharmaceutical industry. Particularly, the success of mRNA-vaccines against SARS-CoV-2, along with the successful development of the first sugar-modified siRNA therapeutics has inspired the field. The development of nucleic acid therapeutics requires efficient chemistry to link oligonucleotides to chemical structures that can improve stability, boost cellular uptake, or enable specific targeting. For the siRNA therapeutics currently in use, modification of the 3'-end of the oligonucleotides with triple-N-acetylgalactosamine (GalNAc)3 was shown to be of significance. This modification is currently achieved through cumbersome multistep synthesis and subsequent loading onto the solid support material. Herein, we report the development of a bifunctional click-reactive linker that allows the modification of oligonucleotides in a tandem click reaction with multiple sugars, regardless of the position within the oligonucleotide, with remarkable efficiency and in a one-pot reaction.

Merged Molecular Switches Excel as Optoacoustic Dyes: Azobenzene-Cyanines Are Loud and Photostable NIR Imaging Agents.

Optoacoustic (or photoacoustic) imaging promises micron-resolution noninvasive bioimaging with much deeper penetration (>cm) than fluorescence. However, optoacoustic imaging of enzyme activity would require loud, photostable, NIR-absorbing molecular contrast agents: which remain unknown. Most organic molecular contrast agents are repurposed fluorophores, with severe shortcomings of photoinstability or phototoxicity under optoacoustic imaging, as consequences of their slow S1→S0 electronic relaxation. We now report that known fluorophores can be rationally modified to reach ultrafast S1→S0 rates, without much extra molecular complexity, simply by merging them with molecular switches. Here, we merge azobenzene switches to cyanine dyes to give ultrafast relaxation (<10 ps, >100-fold faster). Without even adapting instrument settings, these azohemicyanines display outstanding improvements in signal longevity (>1000-fold increase of photostability) and signal loudness (here: >3-fold even at time zero). We show why this simple but unexplored design strategy can still offer stronger performance in the future, and can also increase the spatial resolution and the quantitative linearity of photoacoustic response over extended longitudinal imaging. By bringing the world of molecular switches and rotors to bear on problems facing optoacoustic agents, this practical strategy will help to unleash the full potential of optoacoustic imaging in fundamental studies and translational uses.

Synthesis and Structure Elucidation of Glutamyl-Queuosine.

Queuosine is one of the most complex hypermodified RNA nucleosides found in the Wobble position of tRNAs. In addition to Queuosine itself, several further modified derivatives are known, where the cyclopentene ring structure is additionally modified by a galactosyl-, a mannosyl-, or a glutamyl-residue. While sugar-modified Queuosine derivatives are found in the tRNAs of vertebrates, glutamylated Queuosine (gluQ) is only known in bacteria. The exact structure of gluQ, particularly with respect to how and where the glutamyl side chain is connected to the Queuosine cyclopentene side chain, is unknown. Here we report the first synthesis of gluQ and, using UHPLC-MS-coinjection and NMR studies, we show that the isolated natural gluQ is the α-allyl-connected gluQ compound.

Ultrahigh Sensitivity PTR-MS Instrument with a Well-Defined Ion Chemistry.

Proton-transfer-reaction mass spectrometry (PTR-MS) is widely used for measuring organic trace gases in air. In traditional PTR-MS, both nonpolar and polar analytes are ionized with unit efficiency, as predicted from ion-molecule collision theories. This well-defined ion chemistry allows for direct quantification of analytes without prior calibration and therefore is an important characteristic of PTR-MS. In an effort to further increase the sensitivity, recently developed ultrahigh sensitivity chemical ionization mass spectrometry (CIMS) analyzers have, however, been reported to have sacrificed unit ionization efficiency for selected analytes or classes of analytes. We herein report on the development of a novel ultrasensitive PTR-MS instrument, the FUSION PTR-TOF 10k, which exhibits the same universal unit response as conventional PTR-MS analyzers. The core component of this analyzer is the newly designed FUSION ion-molecule reactor, which is a stack of concentric ring electrodes generating a static longitudinal electric field superimposed by a focusing transversal radiofrequency (RF) field. The FUSION PTR-TOF 10k instrument is equipped with an improved ion source, capable of switching between different reagent ions (H3O+, O2+, NO+, NH4+) in less than one second. The improved time-of-flight mass spectrometer analyzes m/z signals with a mass resolution in the 10000-15000 range. FUSION PTR-TOF 10k achieves sensitivities up to 80000 cps ppbV-1 and detection limits down to 0.5 pptV for a 1 s measurement time. We show time-series of naphthalene and 13C-napthalene as measured in ambient air in Innsbruck for demonstrating the sub-pptV detection capability of this novel FUSION PTR-TOF 10k.

Small volume bone marrow aspirates with high progenitor cell concentrations maximize cell therapy dose manufacture and substantially reduce donor hemoglobin loss.

BACKGROUND: Bone marrow (BM) transplantation is a life-saving therapy for hematological diseases, and the BM harbors also highly useful (progenitor) cell types for novel cell therapies manufacture. Yet, the BM collection technique is not standardized. METHODS: Benchmarking our collection efficiency to BM collections worldwide (N = 1248), we noted a great variability of total nucleated cell (TNC) yields in BM products (HPC-M) with superior performance of our center, where we have implemented a small volume aspirate policy. Thus, we next prospectively aimed to assess the impact of BM collection technique on HPC-M quality. For each BM collection (N = 20 donors), small volume (3 mL) and large volume (10 mL) BM aspirates were sampled at 3 time points and analyzed for cell composition. RESULTS: Compared to large volume aspirates, small volume aspirates concentrated more TNCs, immune cells, platelets, hematopoietic stem/progenitor cells, mesenchymal stromal cells (MSCs), and endothelial progenitors. Inversely, the hemoglobin concentration was higher in large volume aspirates indicating more hemoglobin loss. Manufacturing and dosing scenarios showed that small volume aspirates save up to 42% BM volume and 44% hemoglobin for HPC-M donors. Moreover, MSC production efficiency can be increased by more than 150%. CONCLUSIONS: We propose to consider small volume BM aspiration as standard technique for BM collection.

Multiple mobile excitons manifested as sidebands in quasi-one-dimensional metallic TaSe3.

Charge neutrality and their expected itinerant nature makes excitons potential transmitters of information. However, exciton mobility remains inaccessible to traditional optical experiments that only create and detect excitons with negligible momentum. Here, using angle-resolved photoemission spectroscopy, we detect dispersing excitons in the quasi-one-dimensional metallic trichalcogenide, TaSe3. The low density of conduction electrons and the low dimensionality in TaSe3 combined with a polaronic renormalization of the conduction band and the poorly screened interaction between these polarons and photo-induced valence holes leads to various excitonic bound states that we interpret as intrachain and interchain excitons, and possibly trions. The thresholds for the formation of a photo-hole together with an exciton appear as side valence bands with dispersions nearly parallel to the main valence band, but shifted to lower excitation energies. The energy separation between side and main valence bands can be controlled by surface doping, enabling the tuning of certain exciton properties.

Ganglioglioma with adverse clinical outcome and atypical histopathological features were defined by alterations in PTPN11/KRAS/NF1 and other RAS-/MAP-Kinase pathway genes.

Exome-wide sequencing studies recently described PTPN11 as a novel brain somatic epilepsy gene. In contrast, germline mutations of PTPN11 are known to cause Noonan syndrome, a multisystem disorder characterized by abnormal facial features, developmental delay, and sporadically, also brain tumors. Herein, we performed a deep phenotype-genotype analysis of a comprehensive series of ganglioglioma (GG) with brain somatic alterations of the PTPN11/KRAS/NF1 genes compared to GG with common MAP-Kinase signaling pathway alterations, i.e., BRAFV600E. Seventy-two GG were submitted to whole exome sequencing and genotyping and 84 low grade epilepsy associated tumors (LEAT) to DNA-methylation analysis. In 28 tumours, both analyses were available from the same sample. Clinical data were retrieved from hospital files including disease onset, age at surgery, brain localization, and seizure outcome. A comprehensive histopathology staining panel was available in all cases. We identified eight GG with PTPN11 alterations, copy number variant (CNV) gains of chromosome 12, and the commonality of additional CNV gains in NF1, KRAS, FGFR4 and RHEB, as well as BRAFV600E alterations. Histopathology revealed an atypical glio-neuronal phenotype with subarachnoidal tumor spread and large, pleomorphic, and multinuclear cellular features. Only three out of eight patients with GG and PTPN11/KRAS/NF1 alterations were free of disabling-seizures 2 years after surgery (38% had Engel I). This was remarkably different from our series of GG with only BRAFV600E mutations (85% had Engel I). Unsupervised cluster analysis of DNA methylation arrays separated these tumours from well-established LEAT categories. Our data point to a subgroup of GG with cellular atypia in glial and neuronal cell components, adverse postsurgical outcome, and genetically characterized by complex alterations in PTPN11 and other RAS-/MAP-Kinase and/or mTOR signaling pathways. These findings need prospective validation in clinical practice as they argue for an adaptation of the WHO grading system in developmental, glio-neuronal tumors associated with early onset focal epilepsy.

Preprocedural imaging modalities in patients undergoing iliocaval venous recanalization and stent placement.

PURPOSE: To determine the diagnostic accuracy of preinterventional imaging modalities in patients being evaluated for iliocaval venous recanalization and stent placement. METHODS: Consecutive patients with iliocaval postthrombotic obstructions or nonthrombotic iliac vein lesions (NIVL), who were scheduled for recanalization, underwent duplex ultrasound (DUS), magnetic resonance venography (MRV), multiplanar venography (MPV), and intravascular ultrasound (IVUS). The diagnostic accuracies of DUS, MRV, and MPV were analyzed using IVUS as reference. RESULTS: A total of 216 limbs in 108 patients (80 patients with postthrombotic obstructions, 28 patients with NIVL) were examined. In patients with postthrombotic obstructions, the diagnostic sensitivities for the detection of lesions of the common femoral vein were 81% (95% CI 71-89%) for DUS, 76% (95% CI 65-85%) for MRV, and 86% (95% CI 76-93%) for MPV. The sensitivities for detecting lesions of the iliac veins were 96% (95% CI 89-99%) for DUS, 99% (95% CI 92-100%) for MRV, and 100% (95% CI 94-100%) for MPV. Regarding the inferior vena cava, the sensitivities were 44% (95% CI 24-65%) for DUS, 52% (95% CI 31-73%) for MRV, and 70% (95% CI 47-86%) for MPV. The sensitivities for detecting NIVL were 58% (95% CI 34-79%) for DUS, 90% (95% CI 68-97%) for MRV, and 95% (95% CI 73-99%) for MPV. CONCLUSION: In patients scheduled for recanalization of iliocaval postthrombotic obstructions, the sensitivities of DUS, MRV, and MPV were similar. In patients with suspected inferior vena cava involvement and in patients with NIVL, additional imaging with MR or conventional venography is required.

A decrease in the age of respired carbon from the terrestrial biosphere and increase in the asymmetry of its distribution.

We provide here a model-based estimate of the transit time of carbon through the terrestrial biosphere, since the time of carbon uptake through photosynthesis until its release through respiration. We explored the consequences of increasing productivity versus increasing respiration rates on the transit time distribution and found that while higher respiration rates induced by higher temperature increase the transit time because older carbon is respired, increases in productivity cause a decline in transit times because more young carbon is available to supply increased metabolism. The combined effect of increases in temperature and productivity results in a decrease in transit times, with the productivity effect dominating over the respiration effect. By using an ensemble of simulation trajectories from the Carbon Data Model Framework (CARDAMOM), we obtained time-dependent transit time distributions incorporating the twentieth century global change. In these simulations, transit time declined over the twentieth century, suggesting an increased productivity effect that augmented the amount of respired young carbon, but also increasing the release of old carbon from high latitudes. The transit time distribution of carbon becomes more asymmetric over time, with more carbon transiting faster through tropical and temperate regions, and older carbon being respired from high latitude regions. This article is part of the Theo Murphy meeting issue 'Radiocarbon in the Anthropocene'.

German Patient Blood Management Network: effectiveness and safety analysis in 1.2 million patients.

BACKGROUND: Patient Blood Management (PBM) is a patient-centred, systematic, evidence-based approach to improve patient outcomes by managing and preserving a patient's own blood whilst promoting patient safety and empowerment. The effectiveness and safety of PBM over a longer period have not yet been investigated. METHODS: We performed a prospectively designed, multicentre follow-up study with non-inferiority design. Data were retrospectively extracted case-based from electronic hospital information systems. All in-hospital patients (≥18 yr) undergoing surgery and discharged between January 1, 2010 and December 31, 2019 were included in the analysis. The PBM programme focused on three domains: preoperative optimisation of haemoglobin concentrations, blood-sparing techniques, and guideline adherence/standardisation of allogeneic blood product transfusions. The outcomes were utilisation of blood products, composite endpoint of in-hospital mortality and postoperative complications (myocardial infarction/ischaemic stroke/acute renal failure with renal replacement therapy/sepsis/pneumonia), anaemia rate at admission and discharge, and hospital length of stay. RESULTS: A total of 1 201 817 (pre-PBM: n=441 082 vs PBM: n=760 735) patients from 14 (five university/nine non-university) hospitals were analysed. Implementation of PBM resulted in a substantial reduction of red blood cell utilisation. The mean number of red blood cell units transfused per 1000 patients was 547 in the PBM cohort vs 635 in the pre-PBM cohort (relative reduction of 13.9%). The red blood cell transfusion rate was significantly lower (P<0.001) with odds ratio 0.86 (0.85-0.87). The composite endpoint was 5.8% in the PBM vs 5.6% in the pre-PBM cohort. The non-inferiority aim (safety of PBM) was achieved (P<0.001). CONCLUSIONS: Analysis of >1 million surgical patients showed that the non-inferiority condition (safety of Patient Blood Management) was fulfilled, and PBM was superior with respect to red blood cell transfusion. CLINICAL TRIAL REGISTRATION: NCT02147795.