Programme

Introduction: Donor-derived cell-free DNA (dd-cfDNA) has shown potential value in allograft surveillance by reducing the need for graft biopsies. However, limitations of dd-cfDNA-based allograft monitoring include multi-organ transplantation from the same donor and the early transplantation phase. Analysis of tissue-specific methylation signatures in cfDNA could serve as an alternative approach, potentially offering higher specificity to the relevant transplanted tissues. Yet, to detect tissues contributing at low levels, whole-genome approaches are not cost-effective and targeted methods are needed.

Methods: Utilizing a methylation atlas of 40 human cell types, we developed a panel for deep methylation sequencing of cfDNA. cfDNA from the urine and plasma of organ transplant recipients, along with genomic DNA from primary human cells was extracted and subjected to targeted deep methylation sequencing. Subsequently, the tissue-of-origin was determined through methylation-based deconvolution, employing a non-negative least squares algorithm.

Results: Analysis of genomic DNA from primary human kidney epithelial cells, hepatocytes and blood cells showed an accuracy of over 92% for these cell types. In-silico simulations revealed limited variation of rare tissue fractions at >750x coverage and the ability to accurately detect kidney epithelium- and hepatocyte-derived DNA fractions down to 0.5% and 0.1%, respectively. Analysis of plasma cfDNA collected within 24 hours after transplantation revealed a higher endothelial and a lower immune cell origin for kidney (n=9) and a higher hepatocyte and skeletal muscle origin for liver (n=9) transplant recipients compared to later (45-100 hours) after transplantation. In urine of kidney recipients, higher amounts of immune cell-derived cfDNA were observed in patients with graft injury (n=10) compared to patients without (n=10).

Conclusion: We developed a targeted deep methylation sequencing workflow that allows for accurate tissue-of-origin deconvolution of cfDNA from a single standard blood collection tube. Thus, enabling monitoring of cell turnover and damage to various tissues in different clinical contexts such as solid organ transplant and post-surgery recovery monitoring.

BACKGROUND: mild traumatic brain injury (mTBI) can affect anyone, with children being particularly vulnerable. Even mild, these traumas can lead to intracranial injuries (ICI) requiring prompt identification and monitoring of symptoms evolution at the emergency department (ED). The use of blood biomarkers for the management of these children is still to be explored.

OBJECTIVE: The objective of our study was to evaluate the performances of blood biomarkers, in discriminating between mTBI children who had ICI (CT+) and mTBI children without ICI (either CT- or without CT but with long in-hospital-observation stay). The aim was to rule-out the need of unnecessary CT scans and decrease the length of stay in observation in the ED.

METHODES: A multicenter prospective cohort study was conducted in five pediatric emergency departments in Switzerland. Newborns to teenagers (≤ 16 years old) with mTBI were included. Five blood biomarkers (S100b, GFAP, HFABP, IL6 and NfL) were measured by immunoassay. Their performances to identify patients with ICI were evaluated through ROC curves, where sensitivity was set at 100%. All CT scans were reviewed by one pediatric radiologist, following PECARN (Pediatric Emergency Care Applied Research Network) criteria to identify the presence of ICI.

RESULTS: Blood samples of 285 children with mTBI were analyzed. Twenty-two percent (n=62/285) of children had CT scan examination, while the other (n=223/285) were kept in observation at the ED for more than 6 hours. Twenty-three percent (n=14/62) of the children who underwent a CT scan had ICI, corresponding to five percent of all mTBI included patients. GFAP yielded 52% specificity, HFABP 41%, S100b 39% and NfL 33% to discriminate CT+ patients versus both in-hospital-observation patients and CT- patients with 100% negative predictive value (NPV), when sampled within six hours post-trauma. The cytokine IL6 shown 48% specificity and 100% sensitivity when sampled within twenty-four hours post-trauma.

CONCLUSION: This study demonstrated that biomarkers can significantly help in the management of mTBI patients in the ED, avoiding unnecessary CT scans, and reducing length of stay for children and their families. Further investigations remain needed, with increased sample size, to evaluate their performance, and considering their combinations within panels.

Background: Due to the existing data infrastructure, the medical laboratory is an ideal candidate to pioneer machine-learning algorithms (MLA) in medicine. In contrast to radiology, however, very few MLAs have been implemented yet. Recently, we developed an easy-to-use MLA for the screening of inherited mild bleeding disorders (MBDs) before surgery, for which there is no sensible diagnostic tool available.

Aims: This study aimed to overcome the implementation hurdle by performing a user-centric validation of a MLA for the screening for MBD.

Methods: Using detailed data from two prospective cohort studies of patients referred with suspected MBD (n=555; n=217), we trained, externally validated, and implemented a diagnostic MLA (https://toradi-hit.dbmr.unibe.ch/mbdcheck/). To assess user-friendliness, we developed a survey platform comprising a demographic questionnaire, four case vignettes, and the system usability scale (SUS), a validated questionnaire for software applications. The survey was sent out to surgeons, anesthesiologists, and hematologists, and various background data were collected.

Results: The final model included the following predictors and performance in the external validation was better than any other diagnostic tool (AUROC: 0.86; 95% CI: 0.81, 0.90): (1) sex, (2) activated partial thromboplastin time, (3) PFA-200 closure time with an epinephrine collagen cartridge, and (4) a simplified bleeding history based on the ISTH bleeding assessment tool (ISTH BAT). Thirty-three surgeons, 29 anesthesiologists, and 24 hematologists participated in the survey; most physicians had 10-14 years of experience in their respective fields (29.0 %). The median time needed to fill out the tool was 72 seconds (interquartile range [IQR]: 49.0, 79.5). The median SUS score was 82.5 (IQR: 72.5, 90, > 80 = Software with excellent usability).

Conclusion: Using the screening of MBD as a case study, we developed and implemented a user-friendly MLA that performed well in external validation. This experience can serve as a starting point for ML applications in a wide range of clinical problems.

Background and aim: HIV infection increases autoantibodies against apolipoprotein A1 (AAA1). We investigated their possible relationship with high-density lipoproteins (HDL) particle proteome and subfractions.

Methods: This case-control study consisted in three groups: healthy volunteers (n=50), people living with HIV (PLWH) on antiretroviral therapy (ART) (n=50), PLWH ART-naïve (n=44), in whom AAA1, HDL proteome and subfractions were characterized. AAA1 serum levels were assessed with an in-house immunoassay. Proteome profiling was performed using LC-MS/MS (DIA). HDL particle subfractions were analysed using the Lipoprint® system, and inflammatory biomarkers were measured using the Meso Scale Discovery® platform. Framingham risk score (FRS), carotid intima thickness (cIMT) and flow-mediated dilation (FMD) were determined individually.

Results: PLWH on ART revealed the highest significant modulations in HDL particles:  37 HDL features were modified compared to 18 in PLWH ART-naïve and 8 in healthy volunteers. In PLWH with ART, AAA1 was associated with proteins related of HDL metabolism and dysfunction, such as decreased apolipoprotein AII, apolipoprotein CIII, lecithin cholesterol acyltransferase or paraoxonase-1. This latter was negatively correlated with levels of circulating VCAM-1 and ICAM-1. In PLWH ART-naïve, AAA1 was associated with higher HDL content of phospholipid transfer protein and with larger HDL particles. In healthy volunteers, AAA1 was associated with higher serum amyloid A protein content in HDL, which was positively correlated with circulating c-reactive protein levels. In AAA1 positive subjects, large HDL subfraction was found to be negatively associated with FRS in all groups.

Conclusion: These results indicate that AAA1 response is associated with HDL proteome signatures known to be associated with HDL dysfunction or metabolism. In PLWH with ART these HDL-associated proteins tend to be associated with an inflammatory profile, while in PLWH ART-naïve AAA1 were associated with larger HDL subfractions, controversially associated with cardiovascular risk. These results highlighted that group-specific HDL profiles are differentially associated with AAA1. Knowing whether these would translate in changes of HDL functions and atherosclerosis burden in PLWH warrants further investigations.

Background and Aims: Atherosclerosis is characterized by the accumulation of low-density lipoprotein (LDL) in the arterial wall, a process requiring LDL to pass the endothelial barrier. Several studies suggest that trans-endothelium transit of lipoproteins is an active process regulated by proteins that limit the binding to, the internalization by and the transport through endothelial cells (EC). Endothelial lipase (EL), a phospholipase expressed by EC, binds to LDL in the circulation and regulates lipoprotein plasma level. While EL is known to regulate trans-endothelial transport of HDL, its role in LDL remains unknown. Angiopoietin-like protein 3 (ANGPTL3) is a natural inhibitor of EL and its pharmacological inhibition is under clinical investigation with the aim to reduce plasma LDL levels in patients suffering of homozygous familial hypercholesterolemia. However if EL promotes transport of LDL through EC, disinhibition of ANGPTL3 and the subsequent increase in EL activity might determine the risk of atherosclerosis beyond their plasma levels.

Methods: Here we transfected human primary aortic EC (hAEC) with either siRNA against EL or adenoviruses encoding for catalytically active or inactive EL. We measured binding (4oC), cell association and transport (37oC) of 125I-LDL. 
Results: After pharmacolocial inhibition using orlistat or RNA interference against EL, binding and association of LDL were significantly reduced. Conversely, binding and association of LDL were increased after overexpression of catalytically active and inactive EL. Orlistat treatment reduced transport of LDL while only overexpression of catalytically active EL increased LDL transport.  Addition of ANGPTL3 reduced cell association and transport of LDL through hAEC. Finally, ANGPTL3 and orlistat also prevent the entry of LDL into bovine aortic wall as measured ex vivo.   
Conclusions: Our results confirm the role of EL and ANGPTL3 in LDL transport.