Search

International audience; Background and Objectives: Some medications require specific medical procedures in the weeks before their start. Such procedures may meet the definition of instrumental variables (IVs). We examined how they may influence treatment effect estimation in propensity score (PS)-adjusted comparative studies, and how to remedy. Study Design and Setting: Different covariate assessment periods (CAPs) did and did not include the month preceding treatment start were used to compute PS in the French claims database (Syt eme National des Donn ees de Sant e-SNDS), and 1:1 match patients with metastatic castration resistant prostate cancer initiating abiraterone acetate or docetaxel. The 36-month survival was assessed. Results: Among 1, 213 docetaxel and 2, 442 abiraterone initiators, the PS distribution resulting from the CAP [-12; 0 months] distinctly separated populations (c 5 0.93; 273 matched pairs). The CAPs [-12;-1 months] identified 765 pairs (c 5 0.81). Strong docetaxel treatment predictors during the month before treatment start were implantable delivery systems (1% vs. 59%), which fulfilled IV conditions. The 36-month survival was not meaningfully different under the [-12; 0 months] CAP but differed by 10% points (38% vs. 28%) after excluding month À1. Conclusion: In the setting of highly predictive pretreatment procedures, excluding the immediate pre-exposure time from the CAP will reduce the risk of including potential IVs in PS models and may reduce bias.

To conduct the direct comparison of abiraterone acetate and docetaxel for first-line treatment of metastatic castration-resistant prostate cancer (mCRPC) in real-life settings. Data were extracted from the French nationwide claims database (SNDS) on all men aged ≥40 years starting first-line treatment with abiraterone acetate or docetaxel for mCRPC in 2014. A high-dimensional propensity score including 100 baseline characteristics was used to match patients of both groups and form two comparative cohorts. Three-year overall survival and treatment discontinuation-free survival were determined using Kaplan–Meier analysis. In 2014, 2,444 patients started abiraterone for treatment of mCRPC and 1,214 started docetaxel. After trimming and matching, 716 patients were available in each group. Median overall survival tended to be longer in the abiraterone acetate cohort (23.8 months, 95% confidence interval = [21.5; 26.0]) than in the docetaxel cohort (20.3 [18.4; 21.6] months). Survival at 36 months was 34.6% for abiraterone acetate and 27.9% for docetaxel (p = 0.0027). Treatment discontinuation-free median was longer in the abiraterone acetate cohort compared to the docetaxel cohort (10.8 [10.1; 11.7] versus 7.4 [7.0; 8.0] months). The findings underline the interest of oral abiraterone acetate over intravenous docetaxel as the first-line treatment option in mCRPC.

Studies have not yet found conclusive results on the risk of cancer in patients with multiple sclerosis (MS). This study aimed to compare the incidence of all cancers and of specific types of cancer between MS patients and the general population by age and by sex.All prevalent MS patients identified between 2008 and 2014 in the nationwide French health care database (Système National des Données de Santé) and without history of malignancy were included in a cohort study and followed up until cancer occurrence, date of death, or 31 December 2015, whichever came first. MS patients were matched based on sex and year of birth to non-MS controls from the general population without cancer before index date. Incidence rate was reported per 100,000 person-years (PY), and risk of cancer was estimated by type of cancer, age, and sex using a Cox model (hazard ratio [HR] and its 95% confidence interval [CI]).Overall, 576 cancers per 100,000 PY were observed in MS patients versus 424 per 100,000 PY in the control population. The risk of cancer was higher among MS patients than among population controls whether considered overall (HR = 1.36, 95% CI = 1.29-1.43) or for prostate (HR = 2.08, 95% CI = 1.68-2.58), colorectal and anal (HR = 1.35, 95% CI = 1.16-1.58), trachea, bronchus, and lung (HR = 2.36, 95% CI = 1.96-2.84), and to a lesser extent, breast cancer (HR = 1.12, 95% CI = 1.03-1.23).MS patients were associated with increased risk of cancer compared to population controls.

AIMS Poor efficacy has been reported for patients with BRAF mutations for metastatic colorectal cancer (mCRC). METHODS EREBUS is a French cohort study of wild-type (wt) KRAS unresectable mCRC patients initiating a first-line treatment with cetuximab from 2009 to 2010, followed for two years (five years for vital status). Molecular genetics platforms have provided additional RAS and BRAF mutation testing results. Progression-free survival (PFS) and overall survival (OS) were assessed according to tumour mutation (mt) status: RASmt/BRAFany, RASwt/BRAFmt and RASwt/BRAFwt. Multivariate Cox analyses were used to evaluate association between mutation status and death or progression. RESULTS A total of 389 patients were included in 65 centres and with a known tumour mutation status: 64 RASmt/BRAFany (21%), 33 RASwt/BRAFmt (13%) and 213 RASwt/BRAFwt (87%). Respective baseline characteristics were: median age 65, 64 and 63 years, male gender 63%, 64% and 69%, Eastern Cooperative Oncology Group performance status ≤ 1 75%, 76% and 79%, and liver-only metastases 39%, 33% and 40%. Median progression-free survival was 8.0 months [5.9-9.3] for patients with RASmt/BRAFany, 6.0 months [2.3-7.2] for patients with RASwt/BRAFmt, and 10.4 months [9.5-11.0] for patients with RASwt/BRAFwt. Respectively, median overall survival was 18.4 months [10.9-23.3], 9.7 months [6.9-16.6] and 29.3 months [26.3-36.1]. In multivariate analyses, progression (HR = 2.71 [1.79-4.10]) and death (HR = 2.79 [1.81-4.30]) were more likely for RASwt/BRAFmt vs RASwt/BRAFwt patients. CONCLUSIONS BRAF mutations were associated with markedly poorer outcomes in initially unresectable RASwt mCRC patients treated by cetuximab in first-line treatment.

International audience; Background: There is a lack of information about the burden of metastatic castration-resistant prostate cancer (mCRPC). The present work aims to estimate the incidence and prevalence of mCRPC in 2014 using the French nationwide healthcare database (SNDS).Methods: Prevalence and incidence were estimated based on an SNDS extraction of men covered by the general healthcare insurance (86 % of the French population), and aged ≥40. Patients with mCRPC were identified amongst prostate cancer cases using an algorithm estimating a date of first metastasis management and a date of castration resistance. This algorithm was validated by clinical experts through a blind review of 200 anonymized medical charts from SNDS data. Prevalence and incidence were standardized on the European Standard Population (2013 edition).Results: Prevalence and incidence of mCRPC were estimated as, respectively, 62 and 21 cases per 100 000 men in 2014. Less than one mCRPC case per 100 000 was observed in men aged 40-49. Maximum mCRPC incidence was in men aged 80-89 (175 per 100 000). The algorithm used for mCRPC identification had 97 % positive and 99 % negative predictive values.Conclusion: The good performances of the algorithm for mCRPC identification and the consistency of the generated results with the existing data highlight the robustness of these first estimates of mCRPC prevalence and incidence. Future updates will call for algorithm adjustment as practices evolve over time. These first real-life data will serve for future follow-up of the impact of changes in the management of prostate cancer.

Aims The present study aims to describe real-life outcomes in stable patients after-myocardial infarction (MI) similar to those in the PEGASUS-TIMI 54 trial (PEGASUS), which found long-term benefits of ticagrelor in patients with a history of MI. Methods One-year event-free post-MI patients were identified in the French claims database representative 1/97 sample (2005–2010) and followed for up to 3 years. A PEGASUS-like (PL) population included patients with age ≥ 65 years, or age ≥ 50 and diabetes, renal dysfunction or prior MI, without stroke, end-stage renal failure or oral anticoagulation. Outcomes were: a composite of all-cause death or hospital admission for MI or stroke; individual events; major bleeding. Results There were 1585 post-MI patients totalling 3926 person–years including 865 PL patients (2114 PY); 68% were male; mean age was 66 (standard deviation 15) in post-MI, 74 (10) in PL. Outcomes per 100 person–years [95% confidence interval] were, respectively, in post-MI and PL 6.3 [5.6–7.1] and 7.8 [6.7–8.9] for the composite outcome; 5.1 [4.4–5.8] and 6.5 [5.5–7.6] for death; 1.0 [0.7–1.3] and 1.0 [0.6–1.4] for MI; 0.6 [0.4–0.9] and 0.9 [0.5–1.2] for stroke; 1.3 [0.9–1.6] and 1.4 [0.9–1.9] for major bleeding. Event rates were stable over the 3 study years. Placebo patients in the PEGASUS-TIMI54 Study were younger, more often male and had lower event rates, especially for all-cause death and major bleeding. Conclusions Patients selected using the criteria described in PEGASUS were older with more comorbidities, resulting in higher all-cause death and bleeding rates, but similar MI recurrence rates.

BackgroundAlthough the efficacy and safety of existing therapies of heart failure (HF) have been demonstrated in clinical trials, little is known about the treatment patterns in clinical practice, especially in France.ObjectivesTo describe the treatment initiation patterns and the subsequent treatment changes among HF patients, in the first year following an incident hospitalization for HF, in a French real-world setting.MethodsA cohort of patients aged ≥ 40 years, with an incident hospitalization for HF between 01/01/2008 and 31/12/2013, was identified in the 1/97th permanent random sample of the French nationwide claims database and followed 1 year. HF drug exposure—beta blockers (BB), angiotensin-converting enzyme inhibitors (ACEI), angiotensin receptor blockers (ARBs), aldosterone antagonists (AA), diuretics, digoxin, or ivabradine—was assessed quarterly using a Proportion of Days Covered ≥ 66% (≥ 60 days out of the 90 days of the quarter), by considering HF drugs individually or in combination. Drug changes were assessed between each quarter.ResultsBetween 2008 and 2013, 7387 patients were included. Their mean age was 77.7 years (± 12.0 years) and 51.6% were women. During the follow-up, 24.4% died, 20% were not exposed to any HF treatment, 48.3 to 43.2% had diuretics, one third had BB or ACEI, 9% had ARB or AA, 6% had digoxin, and 2% had ivabradine. The main change occurred between the first and the second quarter for 53.1% of the initially untreated patients.ConclusionThis study provides valuable information on treatment patterns after an initial hospitalization for HF.

Graphical abstract, Users of newly marketed drugs often differ from the patients included in randomized clinical trials, and from patients prescribed similar drugs. Cohorts of such users may be compared using propensity score adjustment, or similar user cohorts may be built using high-dimensional propensity score matching in large population databases. One such database is SNDS, the French nationwide claims and hospitalization database, which covers 99 % of the French population. It has yet been rarely used. To study the comparative effectiveness and safety in secondary coronary prevention of ticagrelor, compared to clopidogrel or prasugrel, we identified in SNDS patients who were dispensed any of the three antiplatelet agents of interest (± aspirin) within a month after discharge from hospital for acute coronary syndrome (ACS) and followed them one year for recurrence of ACS, stroke, acute bleeding, or death. High-dimensional propensity scores were developed to identify matched cohorts. Drug performances were also compared in the whole population using adjustment on the same parameters. Here we describe the database that was used, and the methods developed for the high-dimensional propensity score matching, resulting in standardized mean differences between the matched populations of less than 2 % for all of the 500+ variables included in the model. • This study was done in a newly available large-scale claims database, which may differ from other population databases, by it size and exhaustiveness • The methods elaborate on standard high-dimensional propensity scores as adapted to this claims database

Background Although the efficacy and safety of existing therapies of heart failure (HF) have been demonstrated in clinical trials in the last 35 years, little is known about the treatment patterns of HF in clinical practice, especially in France. Objectives To describe the treatment initiation patterns and the subsequent treatment changes among HF patients, in the first year following an incident hospitalisation for HF, in a French real-world setting. Methods A cohort of patients aged 40 years old and older, with an incident hospitalisation for HF between January 1, 2008 and December 31, 2013, was identified in the EGB, a 1/97 permanent random sample of the French nationwide claims database. All patients who died during the index hospitalization or with a period of at least 3 consecutive months with no healthcare dispensing recorded were excluded. All included patients were followed one year. HF drugs of interest were: beta blockers (BB), angiotensin-converting enzyme inhibitors (ACEI), angiotensin receptor blockers (ARBs), aldosterone antagonists (AA), diuretics, digoxin or ivabradine. Drug exposure was assessed quarterly using a Proportion of Days Covered >66% (>60 days out of the 90 days of the quarter covered by the treatment of interest), by considering HF drugs individually or in combination. Drug changes were assessed between each quarter over the first year of follow-up. Results Between 2008 and 2013, 7,387 from the EGB were included in the cohort study. The mean age at baseline was 77.7 years (±12.0 years) and 51.6% were women. During the follow-up, 24.4% of patients died and 20% did not receive any HF treatment. During the first quarter following initial hospitalisation, 42.7% of patients had diuretics, 26.0% had BB, 25.7% had ACEI, 7.4% had ARB, 7.6% had AA, 4.7% had digoxin and 1.3% had ivabradine. the most frequent combination was BB/ACE/ARB (23.4%). These proportions remained globally constant in each quarter of the follow-up. The main change occurred between thee first and the second quarter and concerned 53.1% of the initially untreated patients; by the second quarter, 22.2% of them initiated a BB/ACI/ARB combination, 13% a diuretic alone, 7.4% a BB and 4.9% a BB/ACI/ARB/AA combination. Conclusion This study provides precious information on treatment patterns after an initial hospital admission for HF at a time when new treatments for HF are emerging. Acknowledgement/Funding None

Background Cabazitaxel is a treatment of metastatic castration-resistant prostate cancer (mCRPC) after docetaxel failure. The FUJI cohort aimed to confirm the real-life overall and progression-free survival (OS, PFS) and safety of cabazitaxel. Methods Multicentre, non-interventional cohort of French mCRPC patients initiating cabazitaxel between 2013 and 2015, followed 18 months. Results Four hundred one patients were recruited in 42 centres. At inclusion, median age was 70, main metastatic sites were bones (87%), lymph nodes (42%) and visceral (20%). 18% had cabazitaxel in 2nd-line treatment, 39% in 3rd-line and 43% in 4th-line or beyond. All had prior docetaxel, and 82% prior abiraterone, enzalutamide or both. Median duration of cabazitaxel treatment was 3.4 months. Median OS from cabazitaxel initiation was 11.9 months [95% CI: 10.1–12.9]. In multivariate analyses, grade ≥ 3 adverse events, visceral metastases, polymedication, and >5 bone metastases were associated with a shorter OS. Main grade ≥ 3 adverse events were haematological with 8% febrile neutropenia. Conclusion Real-life survival with cabazitaxel in FUJI was shorter than in TROPIC (pivotal trial, median OS 15.1 months) or PROSELICA (clinical trial 20 vs 25 mg/m2, median OS, respectively, 13.4 and 14.5 months). There was no effect of treatment-line on survival. No unexpected adverse concerns were identified. Study registration It was registered with the European Medicines Agency EUPASS registry, available at www.encepp.eu, as EUPAS10391. It has been approved as an ENCEPP SEAL study.

Contexte L’acetate d’abiraterone en association avec la prednisone/prednisolone et le docetaxel peuvent etre utilises en premiere ou en deuxieme intention dans le traitement du cancer de la prostate resistant a la castration et metastatique (mCRPC). Objectifs Evaluer si le fait de commencer par une premiere ligne de traitement d’acetate d’abiraterone suivie d’une deuxieme ligne de docetaxel (sequence ABI–DOCE) ou d’utiliser la sequence de traitement inverse DOCE–ABI a un impact sur la survie des patients mCRPC en situation reelle de soins. Methodes Il s’agit d’une etude de cohorte a partir du Systeme National des Donnees de Sante (SNDS) couvrant environ 86 % de la population au moment de l’etude : les patients mCRPC ont ete identifies en 2014 comme initiant la sequence de traitement ABI–DOCE ou DOCE–ABI avec un suivi de 3 ans. Les patients avec la sequence ABI–DOCE et ceux avec DOCE-ABI ont ete apparies 1 : 1 sur le stade de la maladie avant le statut mCRPC, l’anciennete du cancer de la prostate, et un score de propension a haute dimension. La survie globale ainsi que la survie jusqu’au changement de traitement ou au deces (proxy de la survie sans progression) a 36 mois ont ete comparees a l’aide de modeles de Cox multivaries. Resultats Sur les 3949 patients mCRPC ayant initie un traitement de premiere intention en 2014, 1162 sont decedes au cours de cette premiere ligne et 2283 ont beneficie d’un traitement de deuxieme ligne. Parmi ces derniers, 693 ont recu la sequence ABI–DOCE et 354 DOCE–ABI. Au total, 159 patients par groupe etaient apparies 1 : 1. La duree mediane de la premiere ligne de traitement etait de 8,4 mois dans la sequence ABI–DOCE et de 6,6 mois dans la sequence DOCE–ABI. La duree mediane de la deuxieme ligne etait respectivement de 6,3 mois et 6,5 mois dans les sequences ABI–DOCE et DOCE–ABI. Environ 60 % des patients ont recu une troisieme ligne de traitement dans les deux groupes. Les probabilites de survie globale a 36 mois etaient de 33,8 % [27,4 - 41,7] pour ABI–DOCE et 34,4 % [7,8 - 42,5] pour DOCE–ABI avec des medianes de survie de respectivement 26,2 mois [22,2 - 29,8] et 26,6 mois [22,6 - 30,5] (valeur de p : 0,91). Les probabilites de survie jusqu’au changement de traitement ou au deces a 36 mois etaient de 13,7 % [9,6 - 19,7] pour ABI-DOCE et 9,9 % [6,4 - 15,1] pour DOCE–ABI avec des medianes de survie de respectivement 17,5 mois [15,4 - 18,9] et 16,1 mois [14,5 - 17,6] (valeur de p : 0,1983). Conclusion En situation reelle de soins, les sequences de traitement (ABI–DOCE versus DOCE–ABI) semblent n’avoir aucun impact differentiel sur les resultats de survie chez les patients mCRPC partageant les memes caracteristiques.

Background: Non-persistence to oral hormonal therapy (HT) in breast cancer (BC) is an emerging health issue, and estimations vary according to the population selected and/or the statistical method applied. This study aimed to estimate non-persistence over 5 years to HT in an unselected sample of women with BC using a French national population-based database and accounting for competing risks. Methods: A retrospective cohort of 600 women initiating a HT between 2006 and 2007 was constituted using a representative sample of the French national healthcare insurance system database. The Cumulative Incidence Function method was used to estimate the probability of first treatment discontinuation of at least 90 days accounting for competing risk of death from any cause over the theoretical 5-year period of treatment. Results: Thirty one percent of patients who initiated a HT were identified as non-persistent at the fifth year of follow-up. Patients who switched to another HT (HR 3.10, 95% CI (2.20; 4.36)) or had metastatic BC (HR 3.07, 95% CI (1.73; 5.46)) were more likely to be non-persistent. Women who initiated aromatase inhibitors as compared with tamoxifen (HR 0.62, 95% CI (0.46; 0.83)), had administrative registration for BC (HR 0.21, 95% CI (0.13; 0.32)), or had received an adjuvant chemotherapy (HR 0.65, 95% CI (0.48; 0.89)) were less likely to discontinue. Conclusions: The estimate of long-term non-persistence in an unselected sample of women treated in France by oral hormonal therapy is substantial, even accounting for competing risks.

91 Background: Abiraterone acetate in association with prednisone/prednisolone and docetaxel can both be used as first or second line treatments for metastatic castration-resistant prostate cancer (mCRPC). As a result, one may wonder if it is better to start with an abiraterone acetate first line followed by a docetaxel second line (ABI–DOCE sequence) or to use the inverse sequence DOCE-ABI. Methods: A new user cohort study design with a 3-year follow-up: patients initiating a first-line treatment for mCRPC in 2014 followed by a second treatment line were identified from the French nationwide claims database (SNDS), which covers about 86% of the population at the time of the study. Patients with sequence ABI–DOCE and those with DOCE-ABI were 1:1 matched on prostate cancer stage before mCRPC status, duration from prostate cancer diagnosis, and a high-dimensional propensity score. The 36-month overall survival and the 36-month survival time until treatment switch or death (proxy of progression-free survival) were compared using Cox proportional hazards risk model. Results: Out of the 3 949 mCRPC patients that initiated a first-line treatment in 2014, 1 162 died during this first line, and 2 283 had a second-line treatment. Among them, 693 received the sequence ABI–DOCE and 354 DOCE–ABI. A total of 159 patients per group were 1:1 matched. The median duration of the first treatment line was 8.4 months in the ABI–DOCE sequence and 6.6 months in the DOCE–ABI sequence. The median duration of the second line was 6.3 months and 6.5 months in the ABI–DOCE and DOCE–ABI sequences, respectively. Results are presented in the table below. Median survivals were similar in both groups with no significant differences observed in the 36-month survival probabilities. Around 60% of the patients received a third-line treatment in both groups. Conclusions: In real life settings, treatment sequences (ABI–DOCE versus DOCE–ABI) seem to have no differential impact on survival outcome in mCRPC patients sharing same characteristics. [Table: see text]

International audience; Background and aims: We aimed to compare the effectiveness of ticagrelor vs. clopidogrel or prasugrel on recurrence of acute coronary syndromes (ACS) in real-life conditions, as requested by regulatory authorities at the time of marketing.Methods: We performed a cohort study in SNDS, the French national healthcare database. All patients with a hospital admission for ACS in 2013 were followed one year. Patients on ticagrelor, clopidogrel or prasugrel were matched 1: 1 using age, gender, index ACS type, and high-dimensional propensity scores (hdPS). Outcomes were ACS, stroke, all-cause death, and major bleeding, compared within matched groups using Cox proportional hazards models analysis during treatment.Results: 54,048 ACS were hospitalized in 2013. At discharge, 19,796 were dispensed clopidogrel, 8242 prasugrel, and 13,916 ticagrelor. Per group, 9224 ticagrelor vs. clopidogrel, 6752 ticagrelor vs. prasugrel, and 4676 prasugrel vs. clopidogrel patients were matched. Compared to clopidogrel, ticagrelor was associated with a lower hazard ratio of death 0.73 [0.59-0.90] and composite criterion (0.88, 95% CI [0.79-0.99] but not ACS 0.92 [0.80-1.06], stroke (0.96 [017-5.53]) or major bleeding (1.02 [0.82-1.26]). Prasugrel was not different from ticagrelor or clopidogrel for any outcome, in matched patients.Conclusions: Ticagrelor in real-life conditions in matched populations was associated with a lower risk of all-cause death than clopidogrel, and a lower risk of composite outcome, as in the main pivotal clinical trial. Ticagrelor and prasugrel were not different, nor were prasugrel and clopidogrel.

5570 Background: Therapeutic strategy in metastatic castration-resistant prostate cancer (mCRPC) has evolved significantly with the introduction of abiraterone acetate in association with prednisone/prednisolone in first-line treatment in December 2012. This work aimed to compare the effectiveness of abiraterone acetate and docetaxel as first-line treatments for mCRPC, in real-life setting. Methods: Patients with mCRPC were identified in the main scheme of the National Healthcare System database (SNDS), which covers about 86% of the French population, and capturing all reimbursed healthcare expenditures and hospital discharge summaries. Those initiating docetaxel or abiraterone acetate in 1st line in 2014 were included and 1:1 matched on the previous prostate cancer stage before mCRPC status, the delay from the date of initial diagnosis and a high-dimensional propensity score. The 36-month overall survival and the 36-month discontinuation-free survival (i.e. survival time until treatment switch or death) were compared using Cox proportional hazards risk model. Results: In 2014, out of the 12,951 patients with prevalent mCRPC, 1,214 initiated docetaxel in 1st line and 2 444 initiated abiraterone. A total of 716 patients per group were matched with good comparability (C-statistic = 0.6). The median duration of docetaxel–defined as the time between the first and the last infusion–was 7.3 months with a median of 6 infusions. The median duration of abiraterone acetate–corresponding to the period covered by the dispensed drug–was 9.1 months. Near 70% of the docetaxel and 62% of the abiraterone acetate patients received a 2nd line of treatment. Results related to the main survival outcomes are presented in the table below. Conclusions: First-line treatment with abiraterone acetate in mCRPC patients results in a better 36-month overall survival and discontinuation-free survival compared to docetaxel in real-life setting. [Table: see text]

Objectives Metastatic colorectal cancer (mCRC) is increasingly treated with targeted therapies, but little is known about real-life mCRC treatment in older adults. The aims were to describe the real-life first-line treatment modalities in older adult mCRC patients, to identify factors associated with treatment modalities, and to evaluate survival with regard to treatment modalities. Patients and methods A cohort of mCRC patients aged 65 years and older at diagnosis was identified between 2009 and 2013 using French national healthcare insurance system claims data. Treatment modalities were: treatment with one or more anticancer medication vs best supportive care and, among treated patients, treatment with targeted therapy vs conventional chemotherapy alone. Multivariate logistic regression was used to identify factors associated with treatment by anticancer medication and by targeted therapy. Cox proportional hazards models were used to assess the independent effect of treatment modalities on overall survival while adjusting for baseline covariates identified with logistic regression. Results A total of 503 patients were included with a median age of 78 years (54% were men). Of these, 299 (59%) were treated with anticancer medications. Among treated patients, 131 (44%) received targeted therapy. In multivariate analysis, age 75 years or older, renal failure, malnutrition, and five or more concomitant medications were associated with a lower likelihood of treatment with anticancer medications. Among treated patients, age 75 years or older, history of cancer, lymph node metastases, and a single metastatic site were associated with a lower likelihood of treatment with targeted therapy. Multivariate Cox proportional hazards models found that treatment with any anticancer medication tended to be associated with a lower risk of death; treatment with targeted therapy was not significantly associated. Conclusion A more appropriate prescription of anticancer medications in the older adult will require the definition of more explicit criteria to avoid undertreatment. The real benefit of targeted therapies vs conventional chemotherapy alone needs to be confirmed in this population. J Am Geriatr Soc 67:913-919, 2019.

International audience; BACKGROUND: Metastatic colorectal cancer (mCRC) is increasingly treated using targeted therapies. Their real-life evaluation is insufficient, especially in elderly and frail patients. The aim was to describe use, safety, and effectiveness of targeted therapies in first-line mCRC treatment according to age. PATIENTS AND METHODS: Two field cohorts of patients initiating bevacizumab or cetuximab for first-line mCRC were pooled. Patients characteristics, use, and safety were compared between younger and elderly patients (/=75 years). Two-year overall survival (OS) and progression-free survival (PFS) were estimated in both age groups using the Kaplan-Meier method adjusted on factors associated with death or progression identified with Cox multivariate modeling. RESULTS: Eight hundred patients (n = 411, 51.4% bevacizumab) were included: 498 (62.3%) male, median age 64 years, 118 (14.8%) Eastern Cooperative Oncology Group performance status (ECOG-PS) >/=2. Elderly patients (n = 126, 15.8%) were more often treated with 5-fluorouracil alone than younger. Severe adverse events were equivalent across age groups. ECOG-PS >/=1, abnormal hemoglobin, and abnormal alkaline phosphatases were associated with a higher risk of death; OS adjusted on these factors was similar between elderly and younger patients. ECOG-PS >/=1, lung metastases, abnormal hemoglobin, and abnormal creatinine clearance were associated with a higher risk of progression or death; PFS adjusted on these factors was similar across groups. CONCLUSION: Despite treatment adaptations, elderly patients could benefit from targeted therapies as younger without safety warning.

Resection of metastases after chemotherapy improves survival outcomes of patients with initially inoperable metastatic colorectal cancer (mCRC), yet little data is available for those treated in the first-line setting with bevacizumab plus irinotecan. To provide data on this, the present study described the subgroup of the ETNA cohort who underwent metastases surgery. The population of operated patients was described according to metastatic site (exclusively hepatic, non-exclusively hepatic, and non-hepatic). Factors associated with overall survival (OS) and progression-free survival (PFS) were evaluated using multivariable Cox analysis. A total of 76 patients (21.1 % of the ETNA cohort) underwent metastases resection: 50 % male, median age 61.9 years, 85.5 % ECOG ≤ 1, and median duration of bevacizumab use 7.2 months. No surgery-related deaths were observed and 30.6 % of patients had at least one post-operative complication, mainly infections (11.8 % of resections), bleeding complications (3.5 %), or delayed wound healing (2.4 %). Complete remission was higher for those with exclusively hepatic metastases (22/32, 68.8 %) than those with non-exclusively hepatic metastases (12/24, 50.0 %), or non-hepatic metastases (12/20, 60.0 %). Among operated patients, 52.6 % had died after 5 years of follow-up. In multivariable analysis at 2 years of follow-up, death (HR 0.09 [95 % CI 0.02-0.35]) and progression (HR 0.35 [95 % CI 0.23-0.56]) were less likely for patients with complete remission (CR) after surgery R0-R1 or radiofrequency ablation (RFA) [CR RFA] compared with those who were not resected or with R2 resection. In real-life practice, bevacizumab with irinotecan in first-line therapy for mCRC allows secondary resection of metastases and survival is more favourable in those with complete remission (R0-R1/CR RFA).

Background and purpose The incidence and prevalence of Parkinson's disease are important for public health planning yet there is a lack of representative, up-to-date estimations for France. Methods For this cross-sectional study, subjects with suspected Parkinson's were identified in the EGB database, a 1/97 random sample of the national healthcare insurance database, linked to the national hospital-discharge summary database. Incidence and prevalence were estimated using a specific definition that included those with a diagnosis (hospitalization or listed as a long-term chronic disease for full reimbursement) and a sensitive definition that also included those with an indicative drug reimbursement profile. Estimations were extrapolated to the national population, standardizing on age and gender. Results According to either the specific or the sensitive definitions, the annual incidence of Parkinson's disease during the study period was respectively 36 and 49 per 100 000 person-years and prevalence in 2010 was 308–410 per 100 000 persons in the population as a whole. According to the age groups 55–64, 65–74, 75–84 and ≥85 years incidence was respectively 33–46, 139–172, 301–363 and 442–560 per 100 000 person-years amongst men and 32–55, 81–117, 203–270 and 251–313 per 100 000 person-years amongst women. The 2010 prevalence stratified by the same age groups was 293–376, 898–1161, 2524–3011 and 3760–4578 per 100 000 persons amongst men and 199–351, 618–889, 1910–2433 and 2504–3263 per 100 000 persons amongst women. Conclusions The specific and sensitive definitions of disease bracket the true values; the relatively small range indicates that the current study provides good estimations of incidence and prevalence of Parkinson's disease for recent years in France.

Objectif La prise en charge du cancer de la prostate metastatique resistant a la castration (mCRPC) a fortement evolue depuis 2011. L’etude CAMERRA vise a decrire cette evolution via l’identification des cas prevalents de mCRPC dans le SNDS. En l’absence de marqueur direct du mCRPC au sein du SNDS, l’identification des cas necessite la construction d’un algorithme complexe. Methodes Un algorithme d’identification des patients mCRPC a ete construit a partir d’une extraction d’hommes âges de plus de 40 ans, couverts par le Regime general de l’Assurance maladie, avec un indicateur de cancer de la prostate (ALD, hospitalisations et actes CCAM, medicaments) et disposant d’un historique de soins de cinq ans sans absence de remboursement > 1 an. L’identification des mCRPC repose sur la construction de deux indicateurs : la date de premiere prise en charge des metastases se basant sur des actes specifiques (radiotherapie, hospitalisations, traitements ciblant les metastases [denosumab, acide zoledronique, radio-isotopes, radiofrequences, etc.], et des traitements specifiques du mCRPC (acetate d’abiraterone, enzalutamide, docetaxel et cabazitaxel) associes a des actes d’imageries ; la date de resistance a la castration s’appuyant sur les changements entre les sequences de traitements par anti-androgenes et analogues de la LH-RH, les actes chirurgicaux (orchidectomie et pulpectomie) ou la premiere delivrance d’estramustine ou d’un traitement specifique du mCRPC. Un patient est considere comme mCRPC a partir du moment ou il dispose d’une date de premiere prise en charge des metastases et d’une date de resistance a la castration. Resultats En 2014, 3192 patients avec un cancer de la prostate ont ete identifies dans l’EGB, parmi lesquels 499 metastatiques et 254 resistants a la castration. Chez ces 3192 patients, 141 ont ete identifies comme ayant un mCRPC. Par extrapolation, 468 142 cancers de la prostate sont attendus dans le SNDS en 2014. Cette estimation concorde avec celle de l’Institut national du cancer (INCa) : 508 699 en 2008. Parmi ces cas prevalents de cancer de la prostate, 20 137 devraient etre des mCRPC. Conclusion Cette etude preliminaire a permis de construire un algorithme fonctionnel d’identification des mCRPC en tenant compte d’elements complexes et de leur sequence, permettant une premiere estimation de la prevalence des mCRPC dans la population francaise d’apres le SNDS. La prise en charge des patients selectionnes sera par la suite decrite afin de rendre compte de l’evolution du parcours de soin.

Objectif Estimer l’incidence de survenue d’IHAT chez l’adulte apres exposition medicamenteuse. Methode Etude cas-population multicentrique sur une periode de deux ans (2015–2016). Les donnees des patients inscrits sur liste de transplantation hepatique pour hepatite aigue etaient collectees a partir des dossiers medicaux dans les centres de transplantation hepatique. Les cas d’IHAT etaient classes en deux groupes : IHAT « avec cause clinique identifiee » (origine non-medicamenteuse : hepatite virale, auto-immune, etc.) et IHAT « sans cause clinique identifiee » (origine medicamenteuse clairement identifiee, origine indeterminee avec exposition medicamenteuse et origine indeterminee sans exposition medicamenteuse). L’exposition medicamenteuse etait evaluee pour tous les cas quelle que soit la cause d’IHAT, sur une fenetre de 30 jours avant la Date Index (DI = date des premiers symptomes de la maladie hepatique). Le taux d’incidence, exprime par million de traitements-annees, est calcule a partir des donnees de l’Echantillon generaliste des beneficiaires de l’Assurance maladie (EGB). Resultats Les resultats presentes sont preliminaires, l’analyse statistique etant en cours. Les resultats definitifs seront presentes. Sur la periode 2015-2016, 140 IHAT ont ete identifies par 17 centres participants : 60 (42,9 %) « avec cause clinique identifiee » et 80 (57,1 %) « sans cause clinique identifiee ». Parmi les 60 IHAT « avec cause clinique identifiee », 43 (30,7 % des IHAT) ont ete exposees a des medicaments dans les 30 jours precedant la DI, et 17 (12,1 % des IHAT) ont ete classees sans exposition medicamenteuse. Les 80 IHAT « sans cause clinique identifiee » ont ete classees selon les categories suivantes : 60 cas (42,9 % des IHAT) avec surdosage medicamenteux aigu (SMA), 16 cas (11,4 %) avec exposition medicamenteuse (EM) dans les 30 jours precedant la DI, et 4 cas (2,9 %) sans exposition medicamenteuse (SEM). Concernant les IHAT« sans cause clinique identifiee », la proportion de femmes etait plus importante chez les EM (75,0 %) comparativement aux SMA (58,3 %) et aux SEM (50,0 %). L’âge moyen etait de 45,4 ans (± 13,3) pour les EM, 39,4 ans (± 10,7) pour les SMA et 40,0 ans (± 13,6) pour les SEM. La transplantation hepatique etait realisee pour 51,3 % des cas : 68,8 % pour les EM, 45,0 % pour les SMA et 75,0 % pour les SEM. Parmi les 16 EM, 31,3 % etaient exposes au paracetamol seul ou combine. Parmi les 60 SMA, 96,7 % (n = 59) etaient exposes au paracetamol ou combine : 27 surdosages (45,0 %) etaient consideres non-intentionnels, 31 (51,7 %) intentionnels et 1 (1,7 %) non evaluable. Conclusion En France sur la periode 2015–2016, comme sur la periode 2005-2013 etudiee dans le cadre des etudes SALT-I et SALT-II, les IHAT restent majoritairement exposes au paracetamol a dose therapeutique ou en surdosage. L’exposition aux autres medicaments les plus frequemment retrouves et le calcul du risque d’IHAT chez les patients exposes seront presentes. Les resultats seront compares au groupe d’IHAT « avec cause clinique identifiee » afin d’identifier les medicaments pouvant accroitre le risque d’IHAT.

Objectifs Le cancer de la prostate est le 1er cancer chez l’homme. Il peut evoluer vers un stade metastatique et resistant a la castration (mCRPC). Tres peu de donnees epidemiologiques sont disponibles sur le mCRPC. La base du Systeme national des donnees de sante (SNDS) couvrant l’ensemble de la population francaise offre les informations medicales individuelles necessaires pour estimer la prevalence et l’incidence du mCRPC. Methodes Un algorithme visant a identifier les patients ayant un mCRPC en 2014 parmi les hommes de plus de 40 ans affilies au regime general, a ete developpe, valide et applique dans le SNDS sur la periode 2009–2016. La date de resistance a la castration et la date de premiere prise en charge des metastases ont ete estimees en se basant sur les actes medicaux (prostatectomie, radiotherapie, etc.), les diagnostics hospitaliers, la realisation d’examen medicaux, et les dispensations et administration de medicaments captures dans le SNDS (hormonotherapie classique et de nouvelle generation, chimiotherapie, etc.). Les resultats ont ete extrapoles a l’ensemble de la population francaise. Resultats En 2014, le nombre estime de cas prevalents de cancer de la prostate etait de 488 618 dont 36 590 metastatiques (7,5 %) et 24 096 resistants a la castration (4,9 %), pour un total de 16 423 mCRPC (3,4 %). Dans 52 % des cas, la resistance a la castration et la prise en charge des metastases ont ete detectees dans un intervalle de 4 mois. Les metastases ont precede de plus de 4 mois la resistance a la castration dans 30 % des cas. Les mCRPC incidents etaient 5561 en 2014. Au 31/12/2014, la prevalence partielle a 5 ans du cancer de la prostate etait de 185 322 et de 3596 chez les mCRPC. Conclusion La richesse des donnees du SNDS a permis la mise en œuvre d’un algorithme complexe capable d’identifier les patients mCRPC et d’actualiser les donnees epidemiologiques associees.

Objectif Un algorithme d’identification des cas de cancer de la prostate resistant a la castration et metastatique (mCRPC) a ete developpe a partir des donnees du Systeme national des donnees de sante (SNDS). La validation d’un tel algorithme est usuellement basee sur la revue de dossiers medicaux des patients correspondants par des experts cliniciens. Une alternative pour effectuer cette revue consiste a utiliser la richesse des donnees de consommations de soins pseudonymisees disponibles dans le SNDS. Methodes Un tirage au sort a ete effectue pour selectionner 100 cas parmi les 14 050 patients mCRPC identifies par l’algorithme au sein du SNDS et 100 cas parmi 372 273 patients non-mCRPC. L’historique de tous les soins medicaux sur six annees a ete edite pour chacun de ces 200 cas (ALD, hospitalisations, actes CCAM, notamment chirurgie et radiotherapie, remboursements de medicaments, tests biologiques, anatomopathologie, etc.) avec suppression de toutes les variables dites « sensibles » pour eviter tout risque de re-identification (exemple : date de soins remplacee par son delai depuis le diagnostic initial de cancer de la prostate). Un Comite d’experts constitue de deux urologues et de deux oncologues repartis par binome (urologue-oncologue) a evalue le statut mCRPC a partir de ces informations et en aveugle des resultats de l’algorithme. Chaque binome s’est vu attribuer la relecture de 100 cas. Lors de discordance au sein d’un binome sur un statut, le cas etait revu de facon collegiale par les 4 experts. Les valeurs predictives positives (VPP) et negatives (VPN) de l’algorithme ont ete calculees par rapport aux conclusions du Comite. Resultats Apres confrontation des resultats de l’algorithme aux conclusions du Comite, 92 des 100 cas mCRPC et 93 des 100 cas non-mCRPC ont ete confirmes. Il en resulte une VPP de 0,92 et une VPN de 0,93. Conclusion La richesse des donnees a disposition dans le SNDS rend possible la mise en œuvre d’algorithmes de detection de pathologies complexes, mais aussi d’etudes de validation via la reconstitution de dossiers medicaux 100 % anonymises. Les resultats montrent que l’algorithme d’identification presente permet une bonne identification des cas de mCRPC dans le SNDS. En outre, l’etude de validation a permis de detecter quelques parametres a ajuster afin d’en optimiser les performances.

Introduction Le systeme national des donnees de sante (SNDS) est un outil offrant de multiples possibilites d’evaluation des produits de sante en vie reelle. Cependant, son objectif primaire etant la tarification, son architecture est centree sur la prestation rendant son usage a des fins de recherche centrees sur le patient complexe. L’objectif de ces travaux est de presenter un modele de donnees simplifie qui permettrait de faciliter et d’optimiser son utilisation. Methodes Les donnees d’une extraction SNDS de 32 000 sujets portant sur la periode 2006–2017 ont ete verifiees et traitees selon les recommandations de la Cnam. A partir d’une cartographie des variables existantes dans les cinq univers du SNDS (DCIR, HAD, MCO, RIP et SSR), un modele de donnees relationnel centre sur le patient a ete defini puis deploye. Douze tables ont ete creees : – Patient ; – Medicament ; – Affiliation ; – Deces ; – Prestation ; – Visite ; – Hospitalisation ; – Diagnostic ; – Procedure ; – Biologie ; – Materiel ; – Transport. La table Medicament a ete appariee au referentiel de medicaments du SNDS (IR_PHA_R). Les programmes ont ete valides par relecture et execution croisees. La completude des donnees relatives aux variables retenues a ete verifiee. Resultats L’extraction brute initiale comportait 783 tables fractionnees par annee provenant des cinq univers. Les liaisons indirectes inter-tables initiales necessitaient la jointure de 1 a 9 variables. Apres application du modele simplifie, les donnees retenues ont pu etre centralisees au sein des 12 tables finales reliees par une cle unique. Les effectifs des tables finales correspondaient a ceux des tables sources. Le gain en terme d’espace disque a ete majeur : environ 40 %. Les donnees ainsi formatees sont facilement exploitables : les programmes developpes dans le cadre d’autres travaux peuvent y etre directement appliques. Discussion/Conclusion Le passage d’un modele centre sur la prestation a un modele simplifie avec cle de jointure unique basee sur le patient selon une procedure standardisee permettra d’optimiser le travail des statisticiens et de diminuer la variabilite inter-programmeur. Il en resulte une meilleure reproductibilite des resultats, un gain notable en termes de temps, et de qualite scientifique.

In response to a regulatory request for real-life data on patterns of use and survival outcomes, 793 patients initiating bortezomib for multiple myeloma in France (May 2004-April 2006) were included in this observational study. Data were collected from medical files and patients were followed for 2 years, with vital status collected after 3 years. In total 779 patients were analyzed: 83.1% had immunoglobulin G (IgG) or IgA M-component, mean age was 65.7 years and 46.5% were female. Bortezomib was initiated as third-or-later line in 82.0%. For 75.9%, the starting dose was 1.3 mg/m(2); 42.6% had bortezomib alone, 54.0% with dexamethasone. The mean number of bortezomib cycles was 5.0. Three-year overall survival from bortezomib initiation was 31.4% (95% confidence interval, CI [28.1; 34.7]) and median overall survival was 19.6 months. Two-year progression-free survival was 12.0% (95% CI [9.8; 14.4]), and median progression-free survival was 7.2 months. Overall best response was 44.0%. Survival outcomes during real-life use of bortezomib were within the range of those reported in clinical trials.

BACKGROUND: Several methods have been proposed to assess causality in drug-induced liver injury but none have been tested in the specific context of acute liver failure leading to transplantation (ALFT). OBJECTIVE: We took advantage of the Study of Acute Liver Transplant (SALT), a European case-population study of ALFT, to test different causality scales. METHODS: Causality was assessed by experts in SALT, a 7-country case-population study from 2005 to 2007 of adult otherwise unexplained ALFT, for all drugs found within 30 days prior to the date of initial symptoms of liver disease (index date), using information content, causality scales, and data circuit determined from a pilot study, Salome. RESULTS: The consensus points from Salome were to provide full data on drugs including international non-proprietary name (INN) and doses except for non-steroidal anti-inflammatory drugs (NSAIDs) and to use the World Health Organization (WHO) causality scale. In SALT, among the 9,479 identified patients, 600 (6.3 %) were cases of ALFT, of which 187 had been exposed to drugs within 30 days, without overdose. In 130 (69.5 %) of these the causality score was possible, probable, or highly probable. CONCLUSION: In ALFT cases, once other clinical causes have been excluded and drug exposure established within 30 days, the main discriminant characteristic for causality will be previous knowledge of possible hepatotoxicity.

Introduction Few real-life data are available on cetuximab benefit. The EREBUS cohort was performed to assess metastases resection rate, use, safety, and survival outcomes in wild-type KRAS (Kirsten rat sarcoma viral oncogene) patients with initially unresectable metastatic colorectal cancer (mCRC) treated by cetuximab in real practice. Patients and Methods The study cohort comprised patients initiating cetuximab between January 2009 and December 2010 in 65 French centers, with initially unresectable mCRC and wild-type KRAS . Kaplan-Meier analysis estimated 24-month probability of metastases resection and progression-free survival, and 36-month overall survival (OS). Cox proportional hazards models investigated factors associated with survival outcomes. Results Among the 389 patients included, median age was 64 years, 67.4% were male, 77.9% had Eastern Cooperative Oncology Group performance status ≤ 1, and hepatic metastases were most frequent at baseline (n = 146 exclusively, n = 149 not exclusively, n = 94 nonliver only). Median duration of cetuximab use was 4.8 months. Metastases resection was performed in 106 patients (27.2%) (n = 60 liver exclusively, n = 33 not exclusively, n = 13 nonliver only). The 24-month probability (95% confidence interval) of metastases resection occurrence was 33.6% (28.5-39.3). Median progression-free survival was 9.2 (8.5-9.8) months for the total cohort and 13.0 (11.6-15.1) for those resected; median OS was 23.0 (20.6-26.3) months for the total cohort and was not reached after 36 months for those who were resected. The strongest factor associated with higher OS was metastases resection with complete remission (hazard ratio, 0.41; 95% confidence interval, 0.19-0.88). Conclusion This cohort study highlights in French real-life practice the benefit of cetuximab in first-line mCRC therapy, notably in case of metastases resection with complete remission.

Background Most NSAIDs are thought to be able to cause hepatic injury and acute liver failure (ALF), but the event rates of those leading to transplantation (ALFT) remain uncertain. Objectives The aim of the study was to estimate population event rates for NSAID-associated ALFT Methods This was a case-population study of ALFT in 57 eligible liver transplant centres in seven countries (France, Greece, Ireland, Italy, The Netherlands, Portugal and the UK). Cases were all adults registered from 2005 to 2007 for a liver transplant following ALFT without identified clinical aetiology, exposed to an NSAID or paracetamol (acetaminophen) within 30 days before the onset of clinical symptoms. NSAID and paracetamol population exposures were assessed using national sales data from Intercontinental Marketing Services (IMS). Risk was estimated as the rate of ALFT per million treatment-years (MTY). Results In the 52 participating centres, 9479 patients were registered for transplantation, with 600 for ALFT, 301 of whom, without clinical aetiology, had been exposed to a drug within 30 days. Of these 301 patients, 40 had been exposed to an NSAID and 192 to paracetamol (81 of whom were without overdose). Event rates per MTY were 1.59 (95 % CI 1.1–2.2) for all NSAIDs pooled, 2.3 (95 % CI 1.2–3.9) for ibuprofen, 1.9 (95 % CI 0.8–3.7) for nimesulide, 1.6 (95 % CI 0.6–3.4) for diclofenac and 1.6 (95 % CI 0.3–4.5) for ketoprofen. For paracetamol, the event rate was 3.3 per MTY (95 % CI 2.6–4.1) without overdoses and 7.8 (95 % CI 6.8–9.0) including overdoses. Conclusions ALF leading to registration for transplantation after exposure to an NSAID was rare, with no major difference between NSAID. Non-overdose paracetamol-exposed liver failure was twice more common than NSAID-exposed liver failure. Electronic supplementary material The online version of this article (doi:10.1007/s40264-012-0013-7) contains supplementary material, which is available to authorized users.

Purpose The effect of denominator options on event rates was tested on the French part of the Study of Acute Liver Transplant (SALT). Methods SALT is a case population study of acute liver failure registered for transplantation (ALFT), exposed to non-steroidal anti-inflammatory drugs (NSAIDs) or non-overdose paracetamol, from 2005 to 2007. Population exposure was computed from the Intercontinental Medical Services' (IMS) and the French national healthcare insurance system's data as the number of defined daily doses (DDDs) sold or dispensed and the number of exposed patients. Results Nine ALFT cases were exposed to 10 NSAIDs and 49 to non-overdose paracetamol. NSAID sales ranged from 0.04 billion (niflumic acid) to 0.5 billion (ibuprofen) DDDs, amounting to 2.5 billion DDDs for all NSAIDs. The mean per-person exposure ranged from 13.1 (niflumic acid) to 43.2 (ketoprofen) DDDs, reaching 60.5 DDDs for any NSAID. The number of users ranged from 2 million (niflumic acid) to 13 million (ibuprofen), which was 26.6 million for all NSAIDs. The ALFT rates per billion DDDs ranged from 0 to 26 for individual NSAIDs and amounted to 4.6 for all NSAIDs. The ALFT rates per billion DDDs were inversely correlated with the average per-patient exposure (R2 = 0.935, p = 0.0016). The ALFT rates per million users ranged from 0.31 to 0.49, which was 0.41 for all NSAIDs, with no difference between drugs and no effect of mean per-patient exposure (R2 = 0.01, p = 0.9). Whether measured per DDD or per user, the event rate with paracetamol was three to five times higher than with NSAIDs. Conclusion ALFT risk with NSAID seems to be user dependent rather than person-time (exposure) dependent. Choosing the wrong denominator in case population studies might give erroneous results. Copyright © 2012 John Wiley & Sons, Ltd.

Purpose Fear of discontinuing concomitant anti-epileptic drugs (AEDs) may lead to potentially unnecessary and perhaps unsafe polypharmacy. The effect of withdrawing concomitant AEDs on epilepsy control was therefore studied in long-term users of levetiracetam. Methods The EULEV cohort followed patients initiating levetiracetam in France in 2005 or 2006 for one year. In those maintaining levetiracetam throughout the study period, the association of a reduction in the number of concomitant AEDs during the first six months with seizure-freedom during the last six months of follow-up was investigated using logistic regression. Results Of the 356 patients continuing levetiracetam for at least 1 year, 140 (39.3%) were seizure-free during the last six months of follow-up. Partial symptomatic or generalised idiopathic epilepsy were associated with greater seizure-freedom than partial cryptogenic disease. Factors associated with seizures were: longer disease duration, initial incapacity, increased number of seizures in the six months preceding levetiracetam initiation, and number of consultations for epilepsy in the six months preceding levetiracetam initiation. There was a trend for the association between the early reduction in the number of concomitant AEDs and seizure-free status later during follow-up, which however did not reach statistical significance in the final propensity score-adjusted multivariate model (OR = 1.8, 95%CI [0.8;4.0]). Conclusions Taking into account the various risk factors for seizures, the early reduction of concomitant AEDs was not associated with worse seizure rates during follow-up in real-life users of levetiracetam. Copyright © 2012 John Wiley & Sons, Ltd.

Objectif Comparer l’incidence des syndromes coronaires aigus (SCA), des accidents vasculaires cerebraux, des deces, et des hemorragies majeures a un an chez les patients traites par antiagregant plaquettaire (AAP) apres une hospitalisation pour SCA : ticagrelor (T) versus clopidogrel (C) et T versus prasugrel (P) (EUPAS5987). Methode Cohorte des patients hospitalises en 2013 avec un diagnostic principal d’infarctus du myocarde avec ou sans susdecalage ST (IDM ST+, IDM ST−) ou d’angor instable avec passage en unite de soins intensifs (USI) durant l’hospitalisation, identifies et suivis pendant un an dans le Sniiram. Le groupe de traitement (T, C, ou P ± aspirine) etait defini par le premier AAP delivre dans le mois suivant la sortie de la premiere hospitalisation pour SCA (hospitalisation index). Pour chaque comparaison, les patients etaient apparies 1:1 sur le sexe, l’âge, le diagnostic de l’hospitalisation index et un score de propension haute dimension incluant de nombreuses variables de l’hospitalisation index et de l’annee precedente. L’incidence cumulee a un an du premier evenement pendant l’exposition au traitement a ete estimee par la methode de Kaplan-Meier et les risques compares avec un modele de Cox ou de Poisson. Resultats En 2013, 54 097 patients (13 916 T, 19 796 C, 8242 P, 12 143 autres) ont ete identifies dans le Sniiram. Les patients des groupes T, C ou P etaient differents : âge moyen respectivement de 63, 72, 58 ans ; 76 %, 68 %, 86 % d’hommes ; 19 %, 24 %, 20 % de diabetiques ; 54,9 %, 41,6 %, 72,4 % d’IDM ST+. Les differences standardisees devenaient normalisees ou nulles apres appariement. Conclusion Cette etude de plus de 50 000 patients montre une reduction significative de 27 % du risque de deces chez les patients traites par T comparativement au C, une reduction de 12 % du critere composite (hospitalisation pour ACS avec passage en USI, AVC ou deces), sans augmentation du risque d’hemorragie majeure. En revanche, cette etude montre une frequence d’evenements proche pour T et P, sans difference statistique. Le nombre de patients apparies pour chaque comparaison, les resultats d’incidences cumulees et les risques relatifs sont presentes dans le Tableau S1 .

WHAT IS ALREADY KNOWN ABOUT THIS SUBJECT • Levetiracetam has shown good safety/tolerability and efficacy in regulatory trials. This was confirmed in observational investigations performed soon after marketing by using continuation or retention rates as a composite measure. • When an anti-epileptic drug first becomes available; however, there is evidence of channelling to more severe patients than thereafter. WHAT THIS STUDY ADDS • This study was performed several years after marketing of levetiracetam and found high rates of continuation. • It also further explores this measure by determining the continuation in the absence of initiation of additional anti-epileptic drugs. AIMS To investigate real-life effectiveness of levetiracetam in patients initiating treatment in a stable market situation. METHODS Epileptic adults who had initiated levetiracetam between 1 January and 31 August in 2005 or 2006 were included and followed for 1 year by hospital or nonhospital neurologists practising in France. One-year continuation rates were estimated using Kaplan–Meier analysis. Among those still treated at end of study, treatment goals were investigated. Factors associated with discontinuation were investigated using Cox proportional hazards regression. RESULTS A total of 794 subjects were included in the cohort, and 753 subjects were followed up and included in the analysis. Among these, mean (SD) age was 42.6 (±17.0) years, 51.1% were female, 76.6% had partial epilepsy, 93.5% had seizures in the 6 months preceding levetiracetam initiation and 82.9% had at least one concomitant anti-epileptic drug when starting levetiracetam. One-year levetiracetam continuation rate was 83.5% (95% confidence interval, 80.5–86.0%). Of the 579 patients still using levetiracetam at end of study, 46.8% were seizure free during the last 6 months, and 24% were on levetiracetam monotherapy. Reasons for discontinuation (n= 122) were adverse events (45%), lack of efficacy (38%) or both (9%). Levetiracetam discontinuation was most strongly associated with previous exposure to more than four anti-epileptic drugs, whereas continuation was most strongly associated with presence of seizure-related falls in the 6 months preceding levetiracetam initiation. CONCLUSIONS This population-based cohort study in a stable market situation found a high 1 year levetiracetam continuation rate compared with previous studies done sooner after market introduction.

Introduction Le cabazitaxel, commercialise en France depuis mars 2012, a demontre une amelioration de la survie globale (SG) en 2e ligne apres docetaxel, des patients atteints d’un mCRPC. Objectifs Evaluer la SG, la tolerance, la qualite de vie (QdV) et la douleur des patients traites par cabazitaxel en situation reelle de soins. Methode FUJI est une etude de cohortes observationnelles multicentriques : (i) de patients debutant le cabazitaxel entre septembre 2013 et aout 2015 suivis 18 mois ; (ii) de patients initiant le cabazitaxel entre mars 2016 et mars 2017 suivis 6 mois (cohorte QdV). La SG a ete estimee par la methode de Kaplan–Meier. Un modele de Cox multivarie a ete utilise pour etudier les facteurs associes au deces. La QdV et la douleur ont ete evaluees avec des auto-questionnaires specifiques (FACT-P et BPI-SF). Resultats Au total, 401 patients (âge median 70 ans) ont ete inclus dans la cohorte principale, 18 % d’entre eux ont recu le cabazitaxel en 2e ligne, 39 % en 3e ligne, 23 % en 4e ligne et 21 % en 5e ligne ou plus. Les principaux traitements anterieurs etaient le docetaxel (100 %), l’abiraterone (77 %) et l’enzalutamide (33 %). La duree mediane de traitement par cabazitaxel etait de 3,4 mois. La SG mediane etait de 11,9 mois [10,1–12,9]. En analyse multivariee, les facteurs associes a une plus faible SG etaient : un evenement indesirable (EI) de grade ≥ 3 sous cabazitaxel (HR = 2,05 [1,53–2,73]), des metastases viscerales (HR = 1,98 [1,40–2,80]), une polymedication > 5 traitements non anti-cancereux (HR = 1,74 [1,23–2,45]), > 5 lesions metastatiques osseuses (HR = 1,74 [1,20–2,53]), une progression sous docetaxel (HR = 1,69 [1,13–2,53]) ou dans les 3 mois suivant la derniere perfusion de docetaxel (HR = 1,51 [1,07–2,14]), ≥ 3 traitements anti-cancereux avant cabazitaxel (HR = 1,39 [1,00–1,92]), PSA ≥ 135 ng/mL (HR = 1,36 [1,01–1,82]). Les facteurs associes a une meilleure SG etaient : une anciennete du cancer ≥ 10 ans avant cabazitaxel (HR = 0,66 [0,46–0,96]), un delai ≥ 6 mois entre la derniere perfusion de docetaxel et l’initiation du cabazitaxel (HR = 0,71 [0,52–0,97]). Des EI de grade ≥ 3 etaient observes chez 55 % des patients, principalement des anemies (27 %), neutropenies (15 %), neutropenies febriles (8 %), insuffisances renales (7 %), septicemies/chocs septiques (5 %). Les 61 patients inclus dans la cohorte QdV (âge median 72 ans) etaient precedemment traites par docetaxel (98 %), abiraterone (61 %) et/ou enzalutamide (61 %) ; 49 patients etaient evaluables pour la QdV et 44 patients pour la douleur. La QdV a ete amelioree chez 41 % des patients et maintenue chez 29 %. Une diminution du palier de douleur a ete observee chez 25 % des patients et un maintien de palier chez 50 % des patients. Conclusion La mediane de survie globale en vie reelle dans FUJI est inferieure a celle observee dans l’essai pivot TROPIC (11,9 versus 15,1 mois), mais tres peu de patients FUJI correspondaient aux criteres d’inclusion de TROPIC. Aucun nouveau signal de pharmacovigilance n’a ete detecte. Une amelioration/stabilite de la QdV et de la douleur etait reportee respectivement par 70 % et 75 % des patients traites par cabazitaxel.

5025Background: Cabazitaxel (CAB) was marketed in March 2012 in France, based on overall survival (OS) benefit in mCRPC in 2nd-line (2L) post-docetaxel (DOC). FUJI is a post-authorisation study of ...

Background: Pain is a frequent symptom during the course of multiple sclerosis (MS) but its frequency and impact at the early clinical stages remain unknown. Objectives: The aim of this study was to establish prevalence and severity of pain in a cohort of patients recently diagnosed with MS and to determine the evolution of pain prevalence over 2 years. Other objectives were to investigate the presence of baseline clinical predictors of pain after 2 years and to establish its impact on quality of life (QOL). Methods: In a population-based sample of 69 patients recently diagnosed with MS (< 6 mo), pain was measured using questions from the SEP-59 QOL questionnaire. A standardized bedside neurologic examination was performed to establish sensory function, sensory Kurtzke functional system score, and disability scales. Patients were reassessed after 1 and 2 years. Results: Pain was reported by 73.5% of MS patients at baseline and by 70.6% and 61.8% at 1 and 2-year follow-ups, respectively. Clinically significant pain (grades between 3 and 6 using a 6-graded verbal scale) was reported by 63.2% of patients at baseline and by 51.5% and 45.6%, at 1 and 2-year follow-ups, respectively. Pain significantly altered daily activities in 44% of patients. Low overall QOL scores were significantly associated with pain. At 2 years time point, occurrence of pain was associated with baseline depressive symptoms after controlling for the presence of pain at baseline. Conclusions: Pain is frequent in the early stages of MS and affects the daily QOL.

Effectiveness of bevacizumab for metastatic colorectal cancer in elderly patients has been investigated in observational studies, mainly associated with oxaliplatin-based regimens. Here, using the ETNA cohort in which the majority of patients received bevacizumab+FOLFIRI, the effectiveness of this combination in elderly patients is explored.Patients initiating first-line therapy with bevacizumab between January 2006 and December 2007 were identified in 28 French centres and followed for 24months. Vital status was collected over 36months. In the present analysis those who received FOLFIRI were retained (85% of those included), and patients were stratified by age (70/≥70years). The Kaplan-Meier method estimated progression-free survival (PFS) and overall survival (OS), and Cox models were used to assess the independent effect of age on survival outcomes.Among the 351 patients who received bevacizumab+FOLFIRI, 33.9% were aged ≥70years, 66.1%70years. Respectively 15.1% and 9.5% of patients had ECOG-PS ≥2; 49.6% and 40.1% used 'stop-and-go' treatment scheduling; and 56.3% and 44.4% experienced grade 3/4 adverse events. Overall response rate was 58.8% and 62.5%. Median [95% confidence interval, CI] OS was respectively 24.1 [20.4; 26.2] and 28.5 [25.0; 31.0] months; age≥70years and ECOG-PS≥2 were significantly associated with death. Median PFS [95% CI] was respectively 10.9 [9.4; 12.6] and 9.8 [9.2; 11.2] months; hepatic metastases was associated with progression, and age ≥70years was associated with progression after 14months of follow-up but not before.The present study adds to the literature on the safe and beneficial effect of bevacizumab in the elderly receiving FOLFIRI regimen.

Therapeutical options for older multiple myeloma patients have been improved with the advent of new drugs, yet there is a lack of observational data for such patients. To address this issue, an age-stratified analysis of the VESUVE cohort of bortezomib users was performed. Among the 779 patients included in the analysis, 358 (46%) were aged ≤ 65 years, 282 (36%) were between 65-75 years and 139 (18%) were more than 75 years old. There were few significant differences in treatment parameters across age groups; notably, older patients received a lower dose of bortezomib and more frequently experienced general or administration site conditions, metabolism or nutrition disorders and cardiac disorders. Overall best response rate and progression-free survival were similar across age groups. Taken together, these results indicate that older patients do benefit from bortezomib and that tailored treatment in real-life clinical practice does not compromise effectiveness.

Objectifs Compte tenu du vieillissement de la population, l’oncogeriatrie est devenue une priorite de sante publique. Les patients âges etant le plus souvent exclus des essais therapeutiques, il n’existe pas de recommandations specifiques validees pour le traitement du cancer chez les patients âges en France. De plus, il n’y a pas de donnees observationnelles disponibles sur l’utilisation du cetuximab pour cette population. Dans le cadre de la cohorte EREBUS, la population âgee de plus de 70 ans a ete comparee a la population d’âge inferieur ou egal a 70 ans, en termes de conditions d’utilisation du cetuximab, de securite d’emploi et d’efficacite avec la survie sans progression et le taux de reponse tumorale en situation reelle de soins. Methode EREBUS est une etude de cohorte francaise multicentrique (n = 92), qui a inclus 389 patients inoperables initiant, en situation reelle de soins, du cetuximab en premiere ligne de traitement pour un cancer colorectal metastatique entre 2009 et 2010. Ces patients ont ete identifies a partir des registres de delivrance des pharmacies hospitalieres et suivis 24 mois. Resultats Un total de 116 patients (29,8 %) etaient âges de plus de 70 ans (« âges ») et 273 patients (70,2 %) avaient un âge ≤70 ans (« non âges »). Les patients « âges » etaient relativement comparables aux patients « non âges » selon le sexe et le score ECOG (62,1 % hommes versus 69,6 % [p = 0,15] ; ECOG 0–1 : 72,4 % versus 80,2 % [p = 0,12]). La duree mediane du traitement par cetuximab etait significativement plus courte chez les « âges » (3,7 versus 5,3 mois, p = 0,03). La frequence d’association a une chimiotherapie a base d’irinotecan (54,3 % versus 56,8 %, p = 0,64) ou a base d’oxaliplatine (36,2 % versus 38,5 %, p = 0,67) et l’incidence des evenements indesirables tout grade confondu n’etaient pas statistiquement differentes (hematologiques [91,4 % versus 90,8 %, p = 0,87], digestifs [79,3 % versus 85,7 %, p = 0,12], dermatologiques [77,6 % versus 85,0 %, p = 0,08], et neurosensitifs [37,9 % versus 47,6 %, p = 0,08]). La survie sans progression et la reponse tumorale au traitement n’etaient pas differentes entre les « âges » et « non âges » : mediane de survie sans progression : 9,5 mois (IC 95 % [7,1–10,5]) versus 9,2 mois [8,1–9,8] et taux de reponse : 46,5 % [37,4–55,6] versus 56,7 % [50,8–62,6]. Conclusion Dans la cohorte EREBUS, l’efficacite et la securite d’emploi du cetuximab associe a une chimiotherapie chez les patients « âges » n’etait pas differente de celles observees chez les patients « non âges ». Des resultats similaires ont ete rapportes pour une autre therapie ciblee dans le traitement du cancer colorectal metastatique ; le bevacizumab dans la cohorte ETNA en France et dans la cohorte BRITE aux Etats-Unis.

Background The French HTA agency requested a real-life benefit-risk study of ticagrelor (T) compared to clopidogrel (C) and prasugrel (P) in the secondary prevention of acute coronary syndrome (ACS) (EUPAS5987). Purpose This study aimed to identify the risk factors associated with the primary effectiveness endpoint for patients with antiplatelet agents after a hospitalisation for ACS. Methods Patients hospitalised in 2013 for STEMI, NSTEMI or unstable angina, with intensive care unit (ICU) stay, and followed for 1 year in the nationwide French claims and hospitalisation database, SNIIRAM. Treatment group (T, C, or P ± aspirin) was defined by the first drug dispensed in the month after discharge. The primary effectiveness endpoint was a composite of the first event of ACS with ICU stay, stroke, or all-cause death. Hazard ratios were estimated with Cox proportional hazard risk model (multivariate analysis). Results A total of 41,954 patients were included (19,796 C, 13,916 T, 8242 P) with different characteristics: mean age of 72, 63, 58 years, respectively; 68%, 76%, 86% male; 42%, 55%, 72% STEMI. A Charlson comorbidity index > 5 was associated with a risk increased by at least 50% for the primary effectiveness endpoint; age, a poverty index, ICU stay duration > 5 days and no PCI during index hospitalisation, and congestive heart failure, peripheral arterial disease, ischemic stroke, abnormal renal function, C and P before index hospitalisation were associated with a risk increased by 20–50%; diabetes, hypertension, aspirin before index hospitalisation, and C in the month after discharge (versus T) were associated with a risk increased by 10–20%. Conclusions In this nationwide real-life study, the risk factor most associated with the primary effectiveness endpoint was a Charlson comorbidity index > 5.

Aims Acute drug overdose, especially with paracetamol, may cause acute liver failure leading to registration for transplantation (ALFT). Population statistics and between-country differences for ALFT related to overdose have been poorly described. The aim of the present study was to evaluate overdose ALFT in the multi-country Study of Acute Liver Transplantation (SALT). Methods All adult overdose-related ALFT, with or without suicidal intent, in France, Greece, Ireland, Italy, the Netherlands, Portugal and the UK between 2005 and 2007 were identified from liver transplant registries and hospital records. These were compared with whole-country and per capita use of paracetamol. Results Six hundred cases of ALFT were identified in 52 of 57 eligible transplant centres, of which 114 involved overdose (72 intentional, 10 non-intentional, 32 uncertain). Overdose represented 20% of all-cause ALFT: Ireland 52%, UK 28%, France 18%, the Netherlands 8%, and Italy 1%. Overdose ALFT were mostly females (61%), mean age 33.6 ± 10.9 years. A total of 111 (97%) of the overdoses involved paracetamol. Event rates ranged from one ALFT for 20.7 tons of paracetamol in Ireland, to one for 1074 tons in Italy and one case in 60 million inhabitants over 3 years in Italy to one case in 286 000 inhabitants per year in Ireland. Per-country event rates for non-overdose ALFT exposed to paracetamol were between 2.5 and 4.0 per million treatment-years sold. Conclusions Paracetamol overdose was found to represent one-sixth of all-cause ALFT. There was a 50-fold difference in Europe in the rates of paracetamol overdose ALFT, and a 200-fold difference per million inhabitants.