Epidemiology of heart failure in young adults: a French nationwide cohort study.

AIMS: Heart failure (HF) in young adults is uncommon, and changes in its incidence and prognosis in recent years are poorly described. METHODS AND RESULTS: The incidence and prognosis of HF in young adults (1850 years) were characterized using nationwide medico-administrative data from the French National Hospitalization Database (period 20132018). A total of 1,486 877 patients hospitalized for incident HF were identified, including 70 075 (4.7) patients aged 1850 years (estimated incidence of 0.44 for this age group). During the study period, the overall incidence of HF tended to decrease in the overall population but significantly increased by 0.041 in young adults (P 0.001). This increase was notably observed among young men (from 0.51 to 0.59, P 0.001), particularly those aged 3650 years. In these young men, ischaemic heart disease (IHD) was the most frequently reported cause of HF, whereas non-ischaemic HF was mainly observed in patients 35 years old. In contrast to non-ischaemic HF, the incidence of IHD increased over the study period, which suggests that IHD-related HF is progressively affecting younger patients. Concordantly, young HF patients presented with high rates of traditional IHD risk factors, including obesity, smoking, hypertension, dyslipidaemia, or diabetes. Lastly, the rates of re-hospitalization (for HF or for any cause) within two years after the first HF event and in-hospital mortality were high in all groups, indicating a poor-prognosis population. CONCLUSION: Strategies for the prevention of HF risk factors should be strongly considered for patients under 50 years old.

Safety of coronavirus disease 2019 vaccines in 213 adult patients with sickle cell disease.

Given the lack of information about safety of the COVID-19 vaccines for sickle cell disease (SCD) patients, we sought to determine whether COVID-19 vaccine was associated with subsequent hospital admission for vaso-occlusive events (VOEs). We included 402 patients with SCD, including 88 regularly transfused. As of July 31, 2021, 213 (53.0%) of them had received a least one dose of COVID vaccine (Pfizer 93.0%). We showed similar risk of hospital admission for a VOE among vaccinated patients (whether transfused or not) and among a control group of non-vaccinated patients matched for age, sex and genotype.

Acute chest syndrome in adult patients with sickle cell disease: The relationship with the time to onset after hospital admission.

Data on acute chest syndrome (ACS) in adult sickle cell disease patients are scarce. In this study, we describe 105 consecutive ACS episodes in 81 adult patients during a 32-month period and compare the characteristics as a function of the time to onset after hospital admission for a vaso-occlusive crisis (VOC), that is early-onset episodes (time to onset ≤24 h, 42%) versus secondary episodes (>24 h, 58%; median [interquartile range] time to onset: 2 [2-3] days). The median age was 27 [22-34] years, 89% of the patients had an S/S or S/ß0 -thalassaemia genotype; 81% of the patients had a history of ACS (median: 3 [2-5] per patient), only 61% were taking a disease-modifying treatment at the time of the ACS. Fever and chest pain were noted in respectively 54% and 73% of the episodes. Crackles (64%) and bronchial breathing (32%) were the main abnormal auscultatory findings. A positive microbiological test was found for 20% of episodes. Fifty percent of the episodes required a blood transfusion; ICU transfer and mortality rates were respectively 29% and 1%. Secondary and early-onset forms of ACS did not differ significantly. Disease-modifying treatments should be revaluated after each ACS episode because the recurrence rate is high.

What clindamycin dose should be administered by continuous infusion during combination therapy with rifampicin? A prospective population pharmacokinetics study.

BACKGROUND: Despite its important drug-drug interaction, combined clindamycin/rifampicin therapy may achieve effective plasma clindamycin concentrations, provided clindamycin is administered by continuous infusion. However, the precise clindamycin dose remains unknown. OBJECTIVES: This study was undertaken to determine the daily clindamycin dose to be administered by continuous infusion in combination with rifampicin to achieve effective plasma clindamycin concentrations. PATIENTS AND METHODS: Two plasma clindamycin concentrations were determined prospectively for 124 patients with bone-and-joint infections treated with continuously infused clindamycin. Twenty patients received clindamycin monotherapy, 19 clindamycin combined with rifampicin and 85 received clindamycin successively without and with rifampicin. A population pharmacokinetic model was developed using NONMEM 7.5. Monte Carlo simulations were run to determine which regimens obtained clindamycin concentrations of at least 3 mg/L. RESULTS: A linear one-compartment model with first-order elimination accurately described the data. Clindamycin distribution volume was not estimated. Mean clindamycin clearances with rifampicin and without, respectively, were 33.6 and 10.9 L/h, with 12.8% interindividual variability. The lowest daily clindamycin dose achieving plasma concentrations of at least 3 mg/L in >90% of the patients, when combined with rifampicin, was 4200 mg/24 h. CONCLUSIONS: Our results support continuous infusion of 4200 mg of clindamycin/24 h, in combination with rifampicin. This high-dose regimen requires therapeutic drug monitoring-guided dose adaptation.

CT-based diagnostic algorithm to identify bowel and/or mesenteric injury in patients with blunt abdominal trauma.

OBJECTIVES: To develop a CT-based algorithm and evaluate its performance for the diagnosis of blunt bowel and/or mesenteric injury (BBMI) in patients with blunt abdominal trauma. METHODS: This retrospective study included a training cohort of 79 patients (29 with BBMI and 50 patients with blunt abdominal trauma without BBMI) and a validation cohort of 37 patients (13 patients with BBMI and 24 patients with blunt abdominal trauma without BBMI). CT examinations were blindly analyzed by two independent radiologists. For each CT sign, the kappa value, sensitivity, specificity, and accuracy were calculated. A diagnostic algorithm was built using a recursive partitioning model on the training cohort, and its performances were assessed on the validation cohort. RESULTS: CT signs with kappa value > 0.6 were extraluminal gas, hemoperitoneum, no or moderate bowel wall enhancement, and solid organ injury. CT signs yielding best accuracies in the training cohort were extraluminal gas (98%; 95% CI: 91-100), bowel wall defect (97%; 95% CI: 91-100), irregularity of mesenteric vessels (97%; 95% CI: 90-99), and mesenteric vessel extravasation (97%; 95% CI: 90-99). Using a recursive partitioning model, a decision tree algorithm including extraluminal gas and no/moderate bowel wall enhancement was built, achieving 86% sensitivity (95% CI: 74-99) and 96% specificity (95% CI: 91-100) in the training cohort and 92% sensitivity (95% CI: 78-97) and 88% specificity (95% CI: 74-100) in the validation cohort for the diagnosis of BBMI. CONCLUSIONS: An effective diagnostic algorithm was built to identify BBMI in patients with blunt abdominal trauma using only extraluminal gas and no/moderate bowel wall enhancement on CT examination. KEY POINTS: • A CT diagnostic algorithm that included extraluminal gas and no/moderate bowel wall enhancement was built for the diagnosis of surgical blunt bowel and/or mesenteric injury. • A decision tree combining only two reproducible CT signs has high diagnostic performance for the diagnosis of surgical blunt bowel and/or mesenteric injury.

Bioprosthetic Total Artificial Heart in Autoregulated Mode Is Biologically Hemocompatible: Insights for Multimers of von Willebrand Factor.

BACKGROUND: Carmat bioprosthetic total artificial heart (Aeson; A-TAH) is a pulsatile and autoregulated device. The aim of this study is to evaluate level of hemolysis potential acquired von Willebrand syndrome after A-TAH implantation. METHODS: We examined the presence of hemolysis and acquired von Willebrand syndrome in adult patients receiving A-TAH support (n=10) during their whole clinical follow-up in comparison with control subjects and adult patients receiving Heartmate II or Heartmate III support. We also performed a fluid structure interaction model coupled with computational fluid dynamics simulation to evaluate the A-TAH resulting shear stress and its distribution in the blood volume. RESULTS: The cumulative duration of A-TAH support was 2087 days. A-TAH implantation did not affect plasma free hemoglobin over time, and there was no association between plasma free hemoglobin and cardiac output or beat rate. For VWF (von Willebrand factor) evaluation, A-TAH implantation did not modify multimers profile of VWF in contrast to Heartmate II and Heartmate III. Furthermore, fluid structure interaction coupled with computational fluid dynamics showed a gradually increase of blood damage according to increase of cardiac output (P<0.01), however, the blood volume fraction that endured significant shear stresses was always inferior to 0.03% of the volume for both ventricles in all regimens tested. An inverse association between cardiac output, beat rate, and high-molecular weight multimers ratio was found. CONCLUSIONS: We demonstrated that A-TAH does not cause hemolysis or AWVS. However, relationship between HMWM and cardiac output depending flow confirms relevance of VWF as a biological sensor of blood flow, even in normal range.

Tracking clusters of patients over time enables extracting information from medico-administrative databases.

CONTEXT: Identifying clusters (i.e., subgroups) of patients from the analysis of medico-administrative databases is particularly important to better understand disease heterogeneity. However, these databases contain different types of longitudinal variables which are measured over different follow-up periods, generating truncated data. It is therefore fundamental to develop clustering approaches that can handle this type of data. OBJECTIVE: We propose here cluster-tracking approaches to identify clusters of patients from truncated longitudinal data contained in medico-administrative databases. MATERIAL AND METHODS: We first cluster patients at each age. We then track the identified clusters over ages to construct cluster-trajectories. We compared our novel approaches with three classical longitudinal clustering approaches by calculating the silhouette score. As a use-case, we analyzed antithrombotic drugs used from 2008 to 2018 contained in the Échantillon Généraliste des Bénéficiaires (EGB), a French national cohort. RESULTS: Our cluster-tracking approaches allow us to identify several cluster-trajectories with clinical significance without any imputation of data. The comparison of the silhouette scores obtained with the different approaches highlights the better performances of the cluster-tracking approaches. CONCLUSION: The cluster-tracking approaches are a novel and efficient alternative to identify patient clusters from medico-administrative databases by taking into account their specificities.

New insights into epidemiological data and impact of the COVID-19 pandemic on IgA vasculitis in children and adults: a French nationwide cohort.

IgA vasculitis (IgAV) is a small size vasculitis for which epidemiologic data are strikingly lacking, especially about the adult form. Additionally, the COVID-19 pandemic seems to have profoundly modified the incidence of this disease. Here, we aimed to establish some relevant epidemiological data in both pediatric and adult IgAV. We performed an observational study using a national database called "BNDMR" on IgAV, which gathers patients managed in the French network of experts on rare diseases. We primarily performed descriptive statistics over the 2010-2022 period. Then, we compared the North-South geographical areas, the seasonality, and the impact of COVID-19 with that of other patients reported in the same centers. We collected data from 1988 IgAV patients. The sex ratio was 1.57 for adults and 1.05 for children. The annual incidence in 2021 was 0.06 for 100,000 adults and 0.50 for 100,000 children. Compared with other diseases reported into the BNDMR, IgAV was more common in the South than in the North of France (OR 4.88 [4.17-5.74] in adults and OR 1.51 [1.35-1.68] in children). IgAV was also observed more frequently in winter and autumn. Strikingly, we observed a decrease in incidence during the COVID-19 pandemic period in children (OR 0.62 [0.47-0.81]). Our study provides both new insights and confirmations of IgAV epidemiological data: winter and autumn seasonality, more pronounced male predominance in adults, decreasing incidence of pediatric IgAV during the COVID-19 pandemic and increasing incidence in the South of France.

Life expectancy and likelihood of surgery in multiple endocrine neoplasia type 1: AFCE and GTE cohort study.

BACKGROUND: The overall natural history, risk of death and surgical burden of patients with multiple endocrine neoplasia type 1 (MEN1) is not well known. METHODS: Patients with MEN1 from a nationwide cohort were included. The survival of patients with MEN1 was compared with that of the general population using simulated controls. The cumulative probabilities of MEN1-specific operations and postoperative mortality were assessed, and surgical sequences were analysed using sunburst charts and Venn diagrams. RESULTS: A total of 1386 patients with MEN1 were included. Life expectancy was significantly reduced in patients with MEN1 compared with simulated controls from the general population, with a lifetime difference of 15 years. Mutations affecting the JunD interaction domain had a significant negative impact on survival. Survival for patients with MEN1 compared with the general population improved over time. The probability of experiencing at least one specific MEN1 operation was above 95 per cent after 75 years, and most patients had surgery at least twice during their lifetime. Time to a 50 per cent risk of MEN1 surgery was 30.5 years for patients born after 1960, compared with 47.9 years for those born before 1960. Sex and mutations affecting the JunD interacting domain had no impact on time to first surgery. There was considerable heterogeneity in surgical sequences, with no specific clinical pathway. CONCLUSION: Life expectancy was significantly lower among patients with MEN1 compared with the general population, and further decreased in patients with mutations affecting the JunD interacting domain. Almost all patients underwent at least one MEN1-specific operation during their lifetime, but there was no standardized sequence of surgery.

Association between antidepressant use and reduced risk of intubation or death in hospitalized patients with COVID-19: results from an observational study.

A prior meta-analysis showed that antidepressant use in major depressive disorder was associated with reduced plasma levels of several pro-inflammatory mediators, which have been associated with severe COVID-19. Recent studies also suggest that several antidepressants may inhibit acid sphingomyelinase activity, which may prevent the infection of epithelial cells with SARS-CoV-2, and that the SSRI fluoxetine may exert in-vitro antiviral effects on SARS-CoV-2. We examined the potential usefulness of antidepressant use in patients hospitalized for COVID-19 in an observational multicenter retrospective cohort study conducted at AP-HP Greater Paris University hospitals. Of 7230 adults hospitalized for COVID-19, 345 patients (4.8%) received an antidepressant within 48 h of hospital admission. The primary endpoint was a composite of intubation or death. We compared this endpoint between patients who received antidepressants and those who did not in time-to-event analyses adjusted for patient characteristics, clinical and biological markers of disease severity, and other psychotropic medications. The primary analysis was a multivariable Cox model with inverse probability weighting. This analysis showed a significant association between antidepressant use and reduced risk of intubation or death (HR, 0.56; 95% CI, 0.43-0.73, p < 0.001). This association remained significant in multiple sensitivity analyses. Exploratory analyses suggest that this association was also significant for SSRI and non-SSRI antidepressants, and for fluoxetine, paroxetine, escitalopram, venlafaxine, and mirtazapine (all p < 0.05). These results suggest that antidepressant use could be associated with lower risk of death or intubation in patients hospitalized for COVID-19. Double-blind controlled randomized clinical trials of antidepressant medications for COVID-19 are needed.

Binacox: automatic cut-point detection in high-dimensional Cox model with applications in genetics.

We introduce binacox, a prognostic method to deal with the problem of detecting multiple cut-points per feature in a multivariate setting where a large number of continuous features are available. The method is based on the Cox model and combines one-hot encoding with the binarsity penalty, which uses total-variation regularization together with an extra linear constraint, and enables feature selection. Original nonasymptotic oracle inequalities for prediction (in terms of Kullback-Leibler divergence) and estimation with a fast rate of convergence are established. The statistical performance of the method is examined in an extensive Monte Carlo simulation study, and then illustrated on three publicly available genetic cancer data sets. On these high-dimensional data sets, our proposed method outperforms state-of-the-art survival models regarding risk prediction in terms of the C-index, with a computing time orders of magnitude faster. In addition, it provides powerful interpretability from a clinical perspective by automatically pinpointing significant cut-points in relevant variables.

The French Covid-19 vaccination policy did not solve vaccination inequities: a nationwide study on 64.5 million people.

BACKGROUND: To encourage Covid-19 vaccination, France introduced during the Summer 2021 a 'Sanitary Pass', which morphed into a 'Vaccine Pass' in early 2022. While the sanitary pass led to an increase in Covid-19 vaccination rates, spatial heterogeneities in vaccination rates remained. To identify potential determinants of these heterogeneities and evaluate the French sanitary and vaccine passes' efficacies in reducing them, we used a data-driven approach on exhaustive nationwide data, gathering 141 socio-economic, political and geographic indicators. METHODS: We considered the association between vaccination rates and each indicator at different time points: before the sanitary pass announcement (week 2021-W27), before the sanitary pass came into force (week 2021-W31) and 1 month after (week 2021-W35) and the equivalent dates for the vaccine pass (weeks 2021-W49, 2022-W03 and 2022-W07). RESULTS: The indicators most associated with vaccination rates were the share of local income coming from unemployment benefits, overcrowded households rate, immigrants rate and vote for an 'anti-establishment' candidate at the 2017 Presidential election. These associations increase over time. Consequently, living in a district below the median of such indicator decreases the probability to be vaccinated by about 30% at the end of the studied period, and this probability gradually decreases by deciles of these indicators. CONCLUSIONS: Our analysis reveals that factors related to poverty, immigration and trust in the government are strong determinants of vaccination rate, and that vaccination inequities tended to increase after the introduction of the French sanitary and vaccination passes.

Diagnostic Algorithm to Differentiate Benign Atypical Leiomyomas from Malignant Uterine Sarcomas with Diffusion-weighted MRI.

Background Improving the differentiation of uterine sarcomas from atypical leiomyomas remains a clinical challenge and is needed to avoid inappropriate surgery. Purpose To develop a diagnostic algorithm including diffusion-weighted MRI criteria to differentiate malignant uterine sarcomas from benign atypical leiomyomas. Materials and Methods This case-control retrospective study identified women with an atypical uterine mass at MRI between January 2000 and April 2017, with surgery or MRI follow-up after 1 year or longer. A diagnostic algorithm including T2-weighted MRI and diffusion-weighted imaging (DWI) signal and apparent diffusion coefficient (ADC) values was developed to predict for sarcoma. The training set consisted of 51 sarcomas and 105 leiomyomas. Two external validation sets were used to evaluate interreader reproducibility (16 sarcomas; 26 leiomyomas) and impact of reader experience (29 sarcomas; 30 leiomyomas). Wilson confidence intervals (CIs) were calculated for sensitivity and specificity. Results Evaluated were 156 women (median age, 50 years; interquartile range, 44-63 years). Predictive MRI criteria for malignancy were enlarged lymph nodes or peritoneal implants, high DWI signal greater than that in endometrium, and ADC less than or equal to 0.905 × 10-3 mm2/sec. Conversely, a global or focal area of low T2 signal intensity and a low or an intermediate DWI signal less than that in endometrium or lymph nodes allowed readers to confidently diagnose as benign a uterine mass demonstrating one or more of these signs (P < .001) in 100% cases in all three data sets. The sensitivities and specificities of the algorithm for diagnosis of malignancy were 98% (50 of 51 masses; 95% CI: 90%, 100%) and 94% (99 of 105 masses; 95% CI: 88%, 98%) in the training set; 88% (14 of 16 masses; 95% CI: 64%, 97%) and 100% (26 of 26 masses; 95% CI: 87%, 100%) in the validation set; and 83% (24 of 29 masses; 95% CI: 65%, 92%) and 97% (29 of 30 masses; 95% CI: 83%, 99%) for the less experienced reader, respectively. Conclusion A diagnostic algorithm with predictive features including lymphadenopathy, high diffusion-weighted imaging signal with reference to endometrium, and low apparent diffusion coefficient enabled differentiation of malignant sarcomas from atypical leiomyomas, and it may assist inexperienced readers. © RSNA, 2020 Online supplemental material is available for this article. See also the editorial by Méndez in this issue.

Conditional prescriptions of oral antihypertensive drugs for the management of hypertension urgencies in the inpatient setting: An observational study.

WHAT IS KNOWN AND OBJECTIVES: The management of hypertension urgencies during hospitalization may generally not necessitate urgent care. However, physicians frequently prescribe 'as needed' antihypertensive drugs for which administration is triggered by blood pressure thresholds. The lack of rationale for this hospital practice led us to study oral conditional antihypertensive (OCA) prescriptions. We aimed to estimate the prevalence of OCA prescriptions and to establish their characteristics. METHODS: In our institution, prescriptions are computerized. The study was retrospectively performed using a hospital clinical data warehouse over a 5-year period. RESULTS AND DISCUSSION: The prevalence of OCA prescriptions was 6.9% among subjects treated with an antihypertensive drug. The median duration of these prescriptions was 4 days, until the day of the patient discharge in 78.8% stays. The calcium channel inhibitors were the main (79.9%) pharmacological class prescribed, with mostly prescriptions of nicardipine. OCA prescriptions were associated with another antihypertensive medication in 58.8% of the prescriptions; for 19.3%, it was a medication belonging to the same pharmacological class than the OCA drug prescribed. Regarding the computerized drafting, 39.6% of the conditional prescriptions were considered uninterpretable. At least one administration by nurses concerned 65.1% of the OCA prescriptions. The mean SBP and DBP before the initiation of an OCA drug was 142.9 ± 28.2 and 75.8 ± 24.5 mm Hg, respectively, relative to 143.0 ± 24.9 and 77.6 ± 19.9 mm Hg after the initiation (P = .8 for SBP and P = .06 for DBP). WHAT IS NEW AND CONCLUSION: The originality of this study lies in the use of a clinical data warehouse to evaluate OCA prescriptions in hospital. These prescriptions are current, often uninterpretable and mostly ordered until patient discharge. Such drug orders could be associated with an increased risk of iatrogenic events and/or administration errors. This underlies the need for developing decision support tools and computerized protocols to manage hypertension urgencies.

NKp46 is a diagnostic biomarker and may be a therapeutic target in gastrointestinal T-cell lymphoproliferative diseases: a CELAC study.

OBJECTIVES: Primary GI T-cell lymphoproliferative diseases (T-LPD) are heterogeneous entities, which raise difficult diagnosis and therapeutic challenges. We have recently provided evidences that lymphomas complicating coeliac disease (CD) arise from innate-like lymphocytes, which may carry NK receptors (NKRs). DESIGN: NKRs expression was compared by flow cytometry in intraepithelial lymphocytes (IEL) from CD, type I or type II refractory CD (RCD). NKp46 was next assessed by immunohistochemistry in paraffin-embedded biopsies from 204 patients with CD, RCDI, RCDII or GI T-cell lymphomas and from a validation cohort of 61 patients. The cytotoxic properties of an anti-NKp46 monoclonal antibody conjugated to pyrrolobenzodiazepine (PBD) was tested ex vivo in human primary tumour cells isolated from fresh duodenal biopsies. RESULTS: NKp46 (but not CD94, NKG2A, NKG2C, NKG2D) was significantly more expressed by malignant RCDII IEL than by normal IEL in CD and RCDI. In paraffin biopsies, detection of >25 NKp46+ IEL per 100 epithelial cells discriminated RCDII from CD and RCDI. NKp46 was also detected in enteropathy-associated T-cell lymphomas (EATL, 24/29) and in monomorphic epitheliotropic intestinal T-cell lymphomas (MEITL, 4/4) but not in indolent T-LPD (0/15). Treatment with anti-NKp46-PBD could efficiently and selectively kill human NKp46+ primary IEL ex vivo. CONCLUSION: NKp46 is a novel biomarker useful for diagnosis and therapeutic stratification of GI T-LPD. Strong preclinical rationale identifies anti-NKp46-PBD as a promising therapy for RCDII, EATL and MEITL.

Bayesian variable selection based on clinical relevance weights in small sample studies-Application to colon cancer.

Using clinical data to model the medical decisions behind sequential treatment actions raises methodological challenges. Physicians often have access to many covariates that may be used when making sequential treatment decisions for individual patients. Statistical variable selection methods may help finding which of these variables are used for this decision in everyday practice. When the sample size is not large, Bayesian variable selection methods can address this setting and allow for expert information to be incorporated into prior distributions. Motivated by clinical practice data involving repeated dose adaptation for Irinotecan in colorectal metastatic cancer, we propose a modification of the stochastic search variable selection (SSVS) method, which we call weight-based SSVS (WBS). We use clinical relevance weights elicited from physician experts to construct prior distributions, with the goal to identify the most influential toxicities and other covariates used for dose adjustment. We evaluate and compare the WBS model performance to the Lasso and SSVS through an extensive simulation study. The simulations show that WBS has better performance and lower rates of false positives and false negatives than the other methods but depends strongly on the covariate weights.

Comparison of methods for early-readmission prediction in a high-dimensional heterogeneous covariates and time-to-event outcome framework.

BACKGROUND: Choosing the most performing method in terms of outcome prediction or variables selection is a recurring problem in prognosis studies, leading to many publications on methods comparison. But some aspects have received little attention. First, most comparison studies treat prediction performance and variable selection aspects separately. Second, methods are either compared within a binary outcome setting (where we want to predict whether the readmission will occur within an arbitrarily chosen delay or not) or within a survival analysis setting (where the outcomes are directly the censored times), but not both. In this paper, we propose a comparison methodology to weight up those different settings both in terms of prediction and variables selection, while incorporating advanced machine learning strategies. METHODS: Using a high-dimensional case study on a sickle-cell disease (SCD) cohort, we compare 8 statistical methods. In the binary outcome setting, we consider logistic regression (LR), support vector machine (SVM), random forest (RF), gradient boosting (GB) and neural network (NN); while on the survival analysis setting, we consider the Cox Proportional Hazards (PH), the CURE and the C-mix models. We also propose a method using Gaussian Processes to extract meaningfull structured covariates from longitudinal data. RESULTS: Among all assessed statistical methods, the survival analysis ones obtain the best results. In particular the C-mix model yields the better performances in both the two considered settings (AUC =0.94 in the binary outcome setting), as well as interesting interpretation aspects. There is some consistency in selected covariates across methods within a setting, but not much across the two settings. CONCLUSIONS: It appears that learning withing the survival analysis setting first (so using all the temporal information), and then going back to a binary prediction using the survival estimates gives significantly better prediction performances than the ones obtained by models trained "directly" within the binary outcome setting.

Features from Computerized Texture Analysis of Breast Cancers at Pretreatment MR Imaging Are Associated with Response to Neoadjuvant Chemotherapy.

Purpose To evaluate whether features from texture analysis of breast cancers were associated with pathologic complete response (pCR) after neoadjuvant chemotherapy and to explore the association between texture features and tumor subtypes at pretreatment magnetic resonance (MR) imaging. Materials and Methods Institutional review board approval was obtained. This retrospective study included 85 patients with 85 breast cancers who underwent breast MR imaging before neoadjuvant chemotherapy between April 10, 2008, and March 12, 2015. Two-dimensional texture analysis was performed by using software at T2-weighted MR imaging and contrast material-enhanced T1-weighted MR imaging. Quantitative parameters were compared between patients with pCR and those with non-pCR and between patients with triple-negative breast cancer and those with non-triple-negative cancer. Multiple logistic regression analysis was used to determine independent parameters. Results Eighteen tumors (22%) were triple-negative breast cancers. pCR was achieved in 30 of the 85 tumors (35%). At univariate analysis, mean pixel intensity with spatial scaling factor (SSF) of 2 and 4 on T2-weighted images and kurtosis on contrast-enhanced T1-weighted images showed a significant difference between triple-negative breast cancer and non-triple-negative breast cancer (P = .009, .003, and .001, respectively). Kurtosis (SSF, 2) on T2-weighted images showed a significant difference between pCR and non-pCR (P = .015). At multiple logistic regression, kurtosis on T2-weighted images was independently associated with pCR in non-triple-negative breast cancer (P = .033). A multivariate model incorporating T2-weighted and contrast-enhanced T1-weighted kurtosis showed good performance for the identification of triple-negative breast cancer (area under the receiver operating characteristic curve, 0.834). Conclusion At pretreatment MR imaging, kurtosis appears to be associated with pCR to neoadjuvant chemotherapy in non-triple-negative breast cancer and may be a promising biomarker for the identification of triple-negative breast cancer. © RSNA, 2017.