Evinacumab and Cardiovascular Outcome in Patients With Homozygous Familial Hypercholesterolemia.

BACKGROUND: Patients with homozygous familial hypercholesterolemia (HoFH) remain at very high cardiovascular risk despite the best standard of care lipid-lowering treatment. The addition of evinacumab, an angiopoietin-like protein 3 monoclonal antibody, more than halves low-density lipoprotein cholesterol in short-term studies. This study evaluated whether the evinacumab response was durable in the long term and improved cardiovascular outcome. METHODS: The OLE ELIPSE HoFH (Open-Label Extension to Evinacumab Lipid Studies in Patients With HoFH) study included newly diagnosed patients and those completing the ELIPSE HoFH trial, on stable lipid-lowering therapy including lipoprotein apheresis but not lomitapide. All patients received evinacumab (15 mg/kg intravenously) every 4 weeks, with no change in concomitant lipid-lowering treatment during the first 6 months. The primary efficacy end points were the mean absolute and percentage changes in low-density lipoprotein cholesterol from baseline to 6 months. A key secondary end point was cardiovascular event-free survival, which was compared with a control HoFH cohort not treated with evinacumab or lomitapide and matched for age, sex, and lipoprotein apheresis, derived from French Registry of Familial hypercholesterolemia. RESULTS: Twelve patients, 5 women and 7 men (12-57 years), were enrolled in 3 centers in France. At 6 months, the mean low-density lipoprotein cholesterol reduction with evinacumab was 3.7 mmol/L or 56% (from 6.5 mmol/L at baseline to 2.8 mmol/L; P<0.0001) and was sustained over the median 3.5-year follow-up. No patients on evinacumab experienced cardiovascular events versus 13 events for 5/21 (24%) over 4 years in the control cohort (likelihood P=0.0267). CONCLUSIONS: Real-life, long-term evinacumab adjunctive to lipid-lowering therapy including lipoprotein apheresis led to sustained low-density lipoprotein cholesterol lowering and improved cardiovascular event-free survival of patients with HoFH.

Factors Predicting Statin Initiation During Childhood in Familial Hypercholesterolemia: Importance of Genetic Diagnosis.

OBJECTIVE: To identify childhood and parental factors associated with initiation of statin therapy in children with heterozygous familial hypercholesterolemia (HeFH), including underlying genetic diagnosis or parental premature atherosclerotic cardiovascular disease (ASCVD). STUDY DESIGN: This multicenter cohort study included 245 HeFH child-parent pairs from the REFERCHOL national register (2014-2020). Demographic and clinical characteristics at the last visit were collected. Vascular disease in parents was defined as a history of ASCVD, and/or a coronary artery calcium score >100, and/or stenosis of >50% in at least carotid artery. Statistical analyses included descriptive analysis, logistic regression for univariate and multivariate effects of statins, and a sensitivity analysis combining the characteristics of children and parents. RESULTS: Among the 245 children in the study cohort, 135 (58%), with a mean age of 14 ± 3 years, were treated with a statin. In multivariable analysis, the predictive childhood factors associated with statin treatment were genetic diagnosis (OR, 2.5; 95% CI, 1.3 to 4.9; P = .01), older age (OR, 4.4; 95% CI, 1.8-10.6; P = .01), more than 2 visits (OR, 2.36; 95% CI, 1.18-4.73; P = .015), and longer duration of follow-up (OR, 1.3; 95% CI, 1.1-1.6; P < .001). The predictive parental factor associated with childhood treatment was the presence of vascular disease (OR, 2.4; 95% CI, 1.0-5.7; P = .04). CONCLUSIONS: HeFH confirmed by DNA testing during childhood and a history of vascular disease in parents were independently associated with statin treatment in children with HeFH. Genetic diagnosis may be useful for cardiovascular prevention in children.

Impact of donor ferritin testing on iron deficiency prevention and blood availability in France: A cohort simulation study.

BACKGROUND AND OBJECTIVES: Implementing a ferritin testing policy for whole blood (WB) donors may prevent iron deficiency (ID, ferritin <26 ng/mL) and anaemia, but may induce donation losses. As part of a national prevention plan in France, we aimed to estimate its impact on ID, anaemias and WB donations among donors at high risk of ID. MATERIALS AND METHODS: A micro-simulation model was developed to evaluate different scenarios compared to the current situation without ferritin testing as a reference scenario. The following scenarios were simulated: a minimum scenario with a 6-month deferral for donors with absent iron store (AIS, ferritinemia <15 ng/ml), a main scenario with additional delayed invitations for donors with ferritinemia 15-25 ng/ml and a supplementation scenario with additional iron supplementation for 50% of the donors with AIS. RESULTS: In the main scenario, 52,699 WB donations per year were estimated to be lost after 1 year (-8%), falling to 27,687 (-4.7%) after 5 years. IDs and anaemias were reduced by 13.6% and 29.3%, respectively, after 1 year. The supplementation scenario increased the number of prevented IDs and anaemias to 24.1% and 35.4%, respectively, after 1 year, and halved the number of anaemias at 5 years. The latter scenario also had the least impact on the number of donations (-3.2% after 5 years). CONCLUSION: A ferritin testing policy resulting in delayed donations for ID donors is effective in reducing IDs and anaemias, but significantly impacts the number of donations, thereby posing a self-sufficiency challenge.

Study of the diffusion properties of zeolite mixtures by combined gravimetric analysis, IR spectroscopy and inversion methods (IRIS).

We report the development of a new method of investigation of the mass transport properties of acidic zeolite-based materials aiming to overcome the limitations of classical approaches. It consists in hyphenating gravimetric analysis and infrared spectroscopy. The former allows assessing the diffusion from the gas phase to all the porosity, while IR allows for selective assessment of diffusion to the zeolite active sites located in the micropores. Furthermore, the data are processed by an original methodology allowing the recovery of the distribution of diffusion domains by inversion of the integral equations describing the uptake curves or the evolution of the infrared spectra. The combination of gravimetric analysis and IR spectroscopy makes it possible to monitor and distinguish diffusion within the various components of the material. The methodology has been applied to the isooctane uptake in the mechanical mixture of FAU and MFI zeolites. Analysis of both gravimetric uptake curves and evolving infrared spectra allows distinguishing and assigning diffusion domains to the H-FAU and H-MFI components of the mixture, with high and low effective diffusion rate constants, respectively. The advantages and limits of the methodology are discussed.

Impact of the Si/Al ratio on the ethanol/water coadsorption on MFI zeolites revealed using original quantitative IR approaches.

Advanced IR vibrational spectroscopic techniques, e.g., using a coupled gravimetric-IR surface analyzer (AGIR) and a high-throughput in situ IR cell (Carroucell), have been used for the quantitative studies of the adsorption and coadsorption of ethanol and water on MFI zeolites with different Si/Al ratios. The AGIR coupling is a powerful tool for the accurate determination of the molar adsorption coefficients during coadsorption experiments since their evaluation is based on the measurement of the exact amount of adsorbed species. The use of the Carroucell set up allows characterizing all the samples simultaneously, strictly in the same gaseous and temperature environment. The molar absorption coefficients of pure adsorbed ethanol and water are determined: their values are constant whatever the Si/Al ratio of the MFI zeolites. Moreover, these coefficients are found to be identical in the case of the water-ethanol coadsorption experiments. Their use allows obtaining the exact quantity of each adsorbate specie in the binary system. At low partial pressures, the unary water adsorption experiments suggest that the amount of adsorbed water results mainly from the preferential adsorption on Brønsted acid sites in tetrameric clusters. In contrast, the adsorption of EtOH occurs on both silanol groups and Brønsted acid sites (BASs). The effect of the Si/Al ratio is only observed at relatively low partial pressures. The effect of the Si/Al ratio on the ethanol adsorption capacity is also investigated. This study directs the choice of an appropriate zeolite once it is used in membranes for drying ethanol.

The Ambulatory Medical Assistance (AMA) programme during active-phase treatment in patients with haematological malignancies: A cost-effectiveness analysis.

CONTEXT: The need for patient navigator is growing, and there is a lack of cost evaluation, especially during survivorship. OBJECTIVE: The objective of this study is to evaluate the cost-effectiveness of an Ambulatory Medical Assistance (AMA) programme in patients with haematological malignancies (HM). DESIGN: A cost-effectiveness analysis of the AMA programme was performed compared to a simulated control arm. SETTING: An interventional, single-arm and prospective study was conducted in a French reference haematology-oncology centre between 2016 and 2020. PARTICIPANTS: Adult patients were enrolled with histologically documented malignant haematology, during their active therapy phase, and treated either by intravenous chemotherapy or oral therapy. METHODS: An extrapolation of the effectiveness was derived from a similar nurse monitoring programme (CAPRI study). Cost effectiveness of the programme was evaluated through adverse events of Grade 3 or 4 avoided in different populations. RESULTS: Included patient (n = 797) from the AMA programme were followed during 125 days (IQR: 0-181), and adverse events (Grade 3/4) were observed in 10.1% of patients versus 13.4% in the simulated control arm. The overall cost of AE avoided was estimated to €81,113, leading to an ICER of €864. CONCLUSION: The AMA programme was shown to be cost-effective compared to a simulated control arm with no intervention.

Ultrasmall Copper Nanoclusters in Zirconium Metal-Organic Frameworks for the Photoreduction of CO2.

Encapsulating ultrasmall Cu nanoparticles inside Zr-MOFs to form core-shell architecture is very challenging but of interest for CO2 reduction. We report for the first time the incorporation of ultrasmall Cu NCs into a series of benchmark Zr-MOFs, without Cu NCs aggregation, via a scalable room temperature fabrication approach. The Cu NCs@MOFs core-shell composites show much enhanced reactivity in comparison to the Cu NCs confined in the pore of MOFs, regardless of their very similar intrinsic properties at the atomic level. Moreover, introducing polar groups on the MOF structure can further improve both the catalytic reactivity and selectivity. Mechanistic investigation reveals that the CuI sites located at the interface between Cu NCs and support serve as the active sites and efficiently catalyze CO2 photoreduction. This synergetic effect may pave the way for the design of low-cost and efficient catalysts for CO2 photoreduction into high-value chemical feedstock.

Probing the Brønsted Acidity of the External Surface of Faujasite-Type Zeolites.

We outline two methodologies to selectively characterize the Brønsted acidity of the external surface of FAU-type zeolites by IR and NMR spectroscopy of adsorbed basic probe molecules. The challenge and goal are to develop reliable and quantitative IR and NMR methodologies to investigate the accessibility of acidic sites in the large pore FAU-type zeolite Y and its mesoporous derivatives often referred to as ultra-stable Y (USY). The accessibility of their Brønsted acid sites to probe molecules (n-alkylamines, n-alkylpyridines, n-alkylphosphine- and phenylphosphine-oxides) of different molecular sizes is quantitatively monitored either by IR or 31 P NMR spectroscopy. It is now possible, for the first time to quantitatively discriminate between the Brønsted acidity located in the microporosity and on the external surface of large pore zeolites. For instance, the number of external acid sites on a Y (LZY-64) zeolite represents 2 % of its total acid sites while that of a USY (CBV760) represents 4 % while the latter has a much lower framework Si/Al ratio.

Tuning Cellular Biological Functions Through the Controlled Release of NO from a Porous Ti-MOF.

Materials for the controlled release of nitric oxide (NO) are of interest for therapeutic applications. However, to date, many suffer from toxicity and stability issues, as well as poor performance. Herein, we propose a new NO adsorption/release mechanism through the formation of nitrites on the skeleton of a titanium-based metal-organic framework (MOF) that we named MIP-177, featuring a suitable set of properties for such an application: (i) high NO storage capacity (3 µmol mg-1solid ), (ii) excellent biocompatibility at therapeutic relevant concentrations (no cytotoxicity at 90 µg mL-1 for wound healing) due to its high stability in biological media (<9 % degradation in 72 hours) and (iii) slow NO release in biological media (≈2 hours for 90 % release). The prospective application of MIP-177 is demonstrated through NO-driven control of mitochondrial respiration in cells and stimulation of cell migration, paving the way for the design of new NO delivery systems for wound healing therapy.

Selective nitrogen capture by porous hybrid materials containing accessible transition metal ion sites.

Selective dinitrogen binding to transition metal ions mainly covers two strategic domains: biological nitrogen fixation catalysed by metalloenzyme nitrogenases, and adsorptive purification of natural gas and air. Many transition metal-dinitrogen complexes have been envisaged for biomimetic nitrogen fixation to produce ammonia. Inspired by this concept, here we report mesoporous metal-organic framework materials containing accessible Cr(III) sites, able to thermodynamically capture N2 over CH4 and O2. This fundamental study integrating advanced experimental and computational tools confirmed that the separation mechanism for both N2/CH4 and N2/O2 gas mixtures is driven by the presence of these unsaturated Cr(III) sites that allows a much stronger binding of N2 over the two other gases. Besides the potential breakthrough in adsorption-based technologies, this proof of concept could open new horizons to address several challenges in chemistry, including the design of heterogeneous biomimetic catalysts through nitrogen fixation.

Treatment of pediatric heterozygous familial hypercholesterolemia 7 years after the EAS recommendations: Real-world results from a large French cohort.

BACKGROUND: Heterozygous familial hypercholesterolemia (HeFH) predisposes to premature cardiovascular diseases. Since 2015, the European Atherosclerosis Society has advocated initiation of statins at 8-10 years of age and a low-density lipoprotein cholesterol (LDL-C) target of <135 mg/dL. Longitudinal data from large databases on pharmacological management of pediatric HeFH are lacking. OBJECTIVE: Here, we describe treatment patterns and LDL-C goal attainment in pediatric HeFH using longitudinal real-world data. METHODS: This was a retrospective and prospective multicenter cohort study (2015-2021) of children with HeFH, diagnosed genetically or clinically, aged <18 years, and followed up in the National French Registry of FH (REFERCHOL). Data on the study population as well as treatment patterns and outcomes are summarized as mean±SD. RESULTS: We analyzed the data of 674 HeFH children (age at last visit: 13.1 ± 3.6 years; 82.0 % ≥10 years; 52.5 % females) who were followed up for a mean of 2.8 ± 3.5 years. Initiation of lipid-lowering therapy was on average at 11.8 ± 3.0 years of age for a duration of 2.5 ± 2.8 years. At the last visit, among patients eligible for treatment (573), 36 % were not treated, 57.1 % received statins alone, 6.4 % statins with ezetimibe, and 0.2 % ezetimibe alone. LDL-C was 266±51 mg/dL before treatment and 147±54 mg/dL at the last visit (-44.7 %) in treated patients. Regarding statins, 3.3 %, 65.1 %, and 31.6 % of patients received high-, moderate-, and low-intensity statins, respectively. Overall, 59 % of children on statin therapy alone and 35.1 % on bitherapy did not achieve the LDL-C goal; fewer patients in the older age group did not reach the treatment goal. CONCLUSION: Pediatric patients with FH followed up in specialist lipid clinics in France receive late treatment, undertreatment, or suboptimal treatment and half of them do not reach the therapeutic LDL-C goal. Finding a more efficient framework for linking scientific evidence to clinical practice is needed.

Room Temperature Reduction of Nitrogen Oxide on Iron Metal-Organic Frameworks.

Nitrogen oxides represent one of the main threats for the environment. Despite decades of intensive research efforts, a sustainable solution for NOx removal under environmental conditions is still undefined. Using theoretical modelling, material design, state-of-the-art investigation methods and mimicking enzymes, it is found that selected porous hybrid iron(II/III) based MOF material are able to decompose NOx, at room temperature, in the presence of water and oxygen, into N2 and O2 and without reducing agents. This paves the way to the development of new highly sustainable heterogeneous catalysts to improve air quality.

A Scalable Robust Microporous Al-MOF for Post-Combustion Carbon Capture.

Herein, a robust microporous aluminum tetracarboxylate framework, MIL-120(Al)-AP, (MIL, AP: Institute Lavoisier and Ambient Pressure synthesis, respectively) is reported, which exhibits high CO2 uptake (1.9 mmol g-1 at 0.1 bar, 298 K). In situ Synchrotron X-ray diffraction measurements together with Monte Carlo simulations reveal that this structure offers a favorable CO2 capture configuration with the pores being decorated with a high density of µ2-OH groups and accessible aromatic rings. Meanwhile, based on calculations and experimental evidence, moderate host-guest interactions Qst (CO2) value of MIL-120(Al)-AP (-40 kJ mol-1) is deduced, suggesting a relatively low energy penalty for full regeneration. Moreover, an environmentally friendly ambient pressure green route, relying on inexpensive raw materials, is developed to prepare MIL-120(Al)-AP at the kilogram scale with a high yield while the Metal- Organic Framework (MOF) is further shaped with inorganic binders as millimeter-sized mechanically stable beads. First evidences of its efficient CO2/N2 separation ability are validated by breakthrough experiments while operando IR experiments indicate a kinetically favorable CO2 adsorption over water. Finally, a techno-economic analysis gives an estimated production cost of ≈ 13 $ kg-1, significantly lower than for other benchmark MOFs. These advancements make MIL-120(Al)-AP an excellent candidate as an adsorbent for industrial-scale CO2 capture processes.

N/S-heterocyclic contaminant removal from fuels by the mesoporous metal-organic framework MIL-100: the role of the metal ion.

The influence of the metal ion in the mesoporous metal trimesate MIL-100(Al(3+), Cr(3+), Fe(3+), V(3+)) on the adsorptive removal of N/S-heterocyclic molecules from fuels has been investigated by combining isotherms for adsorption from a model fuel solution with microcalorimetric and IR spectroscopic characterizations. The results show a clear influence of the different metals (Al, Fe, Cr, V) on the affinity for the heterocyclic compounds, on the integral adsorption enthalpies, and on the uptake capacities. Among several factors, the availability of coordinatively unsaturated sites and the presence of basic sites next to the coordinative vacancies are important factors contributing to the observed affinity differences for N-heterocyclic compounds. These trends were deduced from IR spectroscopic observation of adsorbed indole molecules, which can be chemisorbed coordinatively or by formation of hydrogen bonded species. On the basis of our results we are able to propose an optimized adsorbent for the deep and selective removal of nitrogen contaminants out of fuel feeds, namely MIL-100(V).

Synthesis modulation as a tool to increase the catalytic activity of metal-organic frameworks: the unique case of UiO-66(Zr).

The catalytic activity of the zirconium terephthalate UiO-66(Zr) can be drastically increased by using a modulation approach. The combined use of trifluoroacetic acid and HCl during the synthesis results in a highly crystalline material, with partial substitution of terephthalates by trifluoroacetate. Thermal activation of the material leads not only to dehydroxylation of the hexanuclear Zr cluster but also to post-synthetic removal of the trifluoroacetate groups, resulting in a more open framework with a large number of open sites. Consequently, the material is a highly active catalyst for several Lewis acid catalyzed reactions.

Isomorphous substitution in a flexible metal-organic framework: mixed-metal, mixed-valent MIL-53 type materials.

Mixed-metal iron-vanadium analogues of the 1,4-benzenedicarboxylate (BDC) metal-organic framework MIL-53 have been synthesized solvothermally in N,N'-dimethylformamide (DMF) from metal chlorides using initial Fe:V ratios of 2:1 and 1:1. At 200 °C and short reaction time (1 h), materials (Fe,V)(II/III)BDC(DMF(1-x)F(x)) crystallize directly, whereas the use of longer reaction times (3 days) at 170 °C yields phases of composition [(Fe,V)(III)0.5(Fe,V)0.5(II)(BDC)(OH,F)](0.5-)·0.5DMA(+) (DMA = dimethylammonium). The identity of the materials is confirmed using high-resolution powder X-ray diffraction, with refined unit cell parameters compared to known pure iron analogues of the same phases. The oxidation states of iron and vanadium in all samples are verified using X-ray absorption near edge structure (XANES) spectroscopy at the metal K-edges. This shows that in the two sets of materials each of the vanadium and the iron centers are present in both +2 and +3 oxidation states. The local environment and oxidation state of iron is confirmed by (57)Fe Mössbauer spectrometry. Infrared and Raman spectroscopies as a function of temperature allowed the conditions for removal of extra-framework species to be identified, and the evolution of µ2-hydroxyls to be monitored. Thus calcination of the mixed-valent, mixed-metal phases [(Fe,V)(III)0.5(Fe,V)0.5(II)(BDC)(OH,F)](0.5-)·0.5DMA(+) yields single-phase MIL-53-type materials, (Fe,V)(III)(BDC)(OH,F). The iron-rich, mixed-metal MIL-53 shows structural flexibility that is distinct from either the pure Fe material or the pure V material, with a thermally induced pore opening upon heating that is reversible upon cooling. In contrast, the material with a Fe:V content of 1:1 shows an irreversible expansion upon heating, akin to the pure vanadium analogue, suggesting the presence of some domains of vanadium-rich regions that can be permanently oxidized to V(IV).

Prognostic models for short-term annual risk of severe complications and mortality in patients living with type 2 diabetes using a national medical claim database.

OBJECTIVE: Prognostic models in patients living with diabetes allow physicians to estimate individual risk based on medical records and biological results. Clinical risk factors are not always all available to evaluate these models so that they may be complemented with models from claims databases. The objective of this study was to develop, validate and compare models predicting the annual risk of severe complications and mortality in patients living with type 2 diabetes (T2D) from a national claims data. RESEARCH DESIGN AND METHODS: Adult patients with T2D were identified in a national medical claims database through their history of treatments or hospitalizations. Prognostic models were developed using logistic regression (LR), random forest (RF) and neural network (NN) to predict annual risk of outcome: severe cardiovascular (CV) complications, other severe T2D-related complications, and all-cause mortality. Risk factors included demographics, comorbidities, the adjusted Diabetes Severity and Comorbidity Index (aDSCI) and diabetes medications. Model performance was assessed using discrimination (C-statistics), balanced accuracy, sensibility and specificity. RESULTS: A total of 22,708 patients with T2D were identified, with mean age of 68 years and average duration of T2D of 9.7 years. Age, aDSCI, disease duration, diabetes medications and chronic cardiovascular disease were the most important predictors for all outcomes. Discrimination with C-statistic ranged from 0.715 to 0.786 for severe CV complications, from 0.670 to 0.847 for other severe complications and from 0.814 to 0.860 for all-cause mortality, with RF having consistently the highest discrimination. CONCLUSION: The proposed models reliably predict severe complications and mortality in patients with T2D, without requiring medical records or biological measures. These predictions could be used by payers to alert primary care providers and high-risk patients living with T2D.

Discovering the active sites for C3 separation in MIL-100(Fe) by using operando IR spectroscopy.

A reducible MIL-100(Fe) metal-organic framework (MOF) was investigated for the separation of a propane/propene mixture. An operando methodology was applied (for the first time in the case of a MOF) in order to shed light on the separation mechanism. Breakthrough curves were obtained as in traditional separation column experiments, but monitoring the material surface online, thus providing evidences on the adsorption sites. The qualitative and quantitative analyses of Fe(II) and, to some extent, Fe(III) sites were possible, upon different activation protocols. Moreover, it was possible to identify the nature and the role of the active sites in the separation process by selective poisoning of one family of sites: it was clearly evidenced that the unsaturated Fe(II) sites are mainly responsible for the separation effect of the propane/propene mixture, thanks to their affinity for the unsaturated bonds, such as the C=C entities in propene. The activity of the highly concentrated Fe(III) sites was also highlighted.

Why hybrid porous solids capture greenhouse gases?

Hybrid porous solids, with their tunable structures, their multifunctional properties and their numerous applications, are currently topical, particularly in the domain of adsorption and storage of greenhouse gases. Most of the data reported so far concern the performances of these solids in this domain, particularly in terms of adsorbed amounts of gas but do not explain at the atomic level why and how adsorption and storage occur. From a combination of structural, spectroscopic, thermodynamic experiments and of molecular simulations, this tutorial review proposes answers to these open questions with a special emphasis on CO(2) and CH(4) storage by some rigid and flexible hybrid porous materials.

Machine learning versus regression modelling in predicting individual healthcare costs from a representative sample of the nationwide claims database in France.

BACKGROUND: Innovative provider payment methods that avoid adverse selection and reward performance require accurate prediction of healthcare costs based on individual risk adjustment. Our objective was to compare the performances of a simple neural network (NN) and random forest (RF) to a generalized linear model (GLM) for the prediction of medical cost at the individual level. METHODS: A 1/97 representative sample of the French National Health Data Information System was used. Predictors selected were: demographic information; pre-existing conditions, Charlson comorbidity index; healthcare service use and costs. Predictive performances of each model were compared through individual-level (adjusted R-squared (adj-R2), mean absolute error (MAE) and hit ratio (HiR)), and distribution-level metrics on different sets of covariates in the general population and by pre-existing morbid condition, using a quasi-Monte Carlo design. RESULTS: We included 510,182 subjects alive on 31st December, 2015. Mean annual costs were 1894€ (standard deviation 9326€) (median 393€, IQ range 95€; 1480€), including zero-claim subjects. All models performed similarly after adjustment on demographics. RF model had better performances on other sets of covariates (pre-existing conditions, resource counts and past year costs). On full model, RF reached an adj-R2 of 47.5%, a MAE of 1338€ and a HiR of 67%, while GLM and NN had an adj-R2 of 34.7% and 31.6%, a MAE of 1635€ and 1660€, and a HiR of 58% and 55 M, respectively. RF model outperformed GLM and NN for most conditions and for high-cost subjects. CONCLUSIONS: RF should be preferred when the objective is to best predict medical costs. When the objective is to understand the contribution of predictors, GLM was well suited with demographics, conditions and base year cost.