Search

Background: Stratified breast cancer (BC) prevention is a major option for the future but requires clinically meaningful internationally validated risk models. Non parametric models may be alternate methods for modeling in very large cohorts. We have previously shown that a non-parametric similarity-based k-nearest neighbors' (kNN) model performs better than the BCRAT/Gail model to on 65 000 women of the E3N French national cohort (Dartois et al 2015). We used this method to develop and validate a mammographic density-based model in larger general screening populations (pops). Methods: A modified version of a data-mining based algorithm, the kNN method, was implemented and adapted as previously described [ref Dartois]. Core concept of kNN algorithm is to gather similar profiles using a distance computation. We developed a BC risk prediction model on 629 229 women (wn) from the US Breast Cancer Research Consortium (BCSC), with 5 times random selection of learning and validation sets (75/25 %) within the cohort. 5 parameters were included: age, family history, previous breast biopsy, mammographic density and race. The model's performances (discrimination using c-stat (AUC) and calibration using E/0 ratio) were evaluated and compared to the parametric model developed on the same pop (param/BCSC)(Tice et al 2008). This kNN/BCSC model was then tested on two French screening pops after adjustment on French BC incidence: an urban area (Paris suburbs, N=316 775) and a rural area (Côte d'Or, N=32 930). Its performances were compared to those of a model directly developed (same methods) on the Paris cohort (kNN/Paris). Levels of individual risks assessed by the models were assigned into 4 risk categories. The sensitivity of the models was defined as the number of wn who had BC whose 5 yrs-risk category was intermediate (median risk at 50-yrs - 1.66%), high (> 1.66%) or very high (> 20% lifetime) divided by the total number of wn who had BC. Results: The performances of the different models are shown in Table 1. The kNN model developed on BCSC performed well (c-stat 0.653 and E/0 1.001). It had equivalent performances as the parametric model developed previously on the same pop. This kNN/BCSC had a good discrimination on French pops, although slightly lower than that on US pops. This is expected since French screening starts at 50 (vs 40 in BCSC) and French parameters do not include race. The calibration of such model was excellent on Paris, while it overestimated the risk on Côte d'Or, in which BC incidence is lower. It performed as well as the kNN/Paris model directly developed on Paris' pop. The sensitivity of the kNN/BCSC on US and French pops was good. Conclusions: A new non parametric kNN breast cancer risk model developed on an American screening cohort (BCSC) was successfully validated on two French screening cohorts. This new international model could allow stratified prevention. Performances of the models Param/BCSC on BCSCParam/BCSC on PariskNN/BCSC on BCSCkNN/BCSC on PariskNN/BCSC on Cote d'OrkNN/Paris on ParisN629 229313 817629 229313 81732 930313 817c-statistic0.6580.6020.6530.6020.5930.605E/0 global1.031.071.0011.061.521.00Sensitivity--76.18%72.87%72.58%- Citation Format: Ragusa S, Gauthier E, Dartois L, Tice J, Dancourt V, Arveux P, Brixi Z, Bernoux A, Soyer P, Delattre H, Brechenade S, Catajar N, Kaufmanis A, Hélin VM, Clavel F, Kerlikowske K, Miglioretti D, Delaloge S. Development and validation of a new non-parametric breast cancer risk assessment model on US and European screening populations [abstract]. In: Proceedings of the 2016 San Antonio Breast Cancer Symposium; 2016 Dec 6-10; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2017;77(4 Suppl):Abstract nr P2-06-05.

BackgroundNo study has investigated factors associated with non-participation or partial participation in the different combination patterns of screening programmes for all three cancers, that is, breast, colorectal and cervical cancer.MethodsIn a retrospective cohort study, we sought to describe combinations of cancer screening participation rates among women in the Val-de-Marne area of France and to identify individual and contextual factors associated with non-participation or partial participation.ResultsWomen aged between 50 and 65 and who were eligible for all three screening programmes (n = 102 219) were analysed in multilevel logistic models, with the individual as the Level 1 variable and the place of residence as the Level 2 variable. The women who did not participate in any of the screening programmes were 34.4%, whereas 30.1%, 24% and 11.5% participated in one, two or all three screening programmes, respectively. Age below 55, a previous false-positive mammography, prior opportunistic mammography only, no previous mammography, membership of certain health insurance schemes (all P < 0.05) and residence in a deprived area (P < 0.001) were independently associated with non-participation or partial participation. We observed a stronger effect of deprivation on non-participation in all three cancers than in combinations of screening programmes.ConclusionOur findings suggest that the health authorities should focus on improving cancer screenings in general rather than screenings for specific types of cancer, especially among younger women and those living in the most socially deprived areas.

Reminders have been used in various settings, but failed to produce convincing evidence of benefits on patient adherence to colorectal cancer (CRC) screening. The aim of this study was to assess the effectiveness of sending general practitioners (GPs) printed reminders about CRC screening. We conducted a cluster-randomized controlled trial involving 144 GPs in the Val-de-Marne district (France), who provided care for any reason to 20 778 patients eligible for CRC screening between June 2010 and November 2011. Data were collected from the main statutory health-insurance programme and local cancer screening agency. GPs were randomly assigned in a 1 : 1 proportion to the intervention or the control group. Every 4 months, intervention-group GPs received a computer-generated printed list of patients who had not performed scheduled faecal occult blood test (FOBT) screening. The primary outcome was patient adherence to FOBT screening or exclusion from CRC screening for medical reasons. The screening adherence rate was 31.2% [95% confidence interval (CI) 30.3-32.1] in the control group and 32.9% (95% CI 32.0-33.8) in the intervention group [crude relative risk, 1.05 (95% CI 1.01-1.09), P

Objective Our aim was to determine whether physician-related factors influenced patient participation in colorectal cancer (CRC) screening programs and to identify patient characteristics associated with lower participation in order to facilitate the development of targeted actions to improve participation. Method A retrospective cohort study was conducted in a French department during its first CRC screening campaign from June 2007 to May 2010. Data for 157,766 patients followed by 903 general practitioners (GPs) were analyzed. Patient participation was assessed using multilevel logistic modeling. Results The overall participation rate was 30% (95% confidence interval [95% CI], 29.8–30.2) and varied across the 903 GPs from 0% to 75.5% (median, 30; interquartile range, 24–35). Inter-GP variance explained only 5.5% of the participation rate variance. Participation was significantly lower in males (odds ratio [OR], 0.79; 95% CI, 0.78–0.91), the youngest age group (55–59 years, OR, 0.61; 95% CI, 0.58–0.63), and patients living in socioeconomically deprived areas (OR, 0.82; 95% CI, 0.77–0.87). Conclusion Targeted actions to improve CRC screening participation should focus on patients younger than 60 years, males, and individuals living in deprived areas. Actions to enhance the influence of GPs on patient participation should be directed to the overall population of GPs.

Introduction En France, il existe un depistage organise pour le cancer du sein et le cancer colo-rectal. Le depistage du cancer du col n’est pas encore generalise. Cependant les taux de participation observes sont inferieurs aux taux esperes pour diminuer la mortalite par cancer. Des determinants a cette participation ont ete identifies dans la litterature pour chacun des trois depistages, mais pas pour la participation conjointe, ou partielle, des femmes aux trois depistages. Materiel et methode Etude de cohorte retrospective menee dans le departement du Val-de-Marne (94) en population generale, concernant les femmes âgees de 50 a 65 ans eligibles aux trois depistages. Au total, 102 219 femmes ont ete incluses, issues de la base de donnees de l’Association pour le depistage organise des cancers (Adoc). Notre variable d’interet avait quatre modalites : participation conjointe aux trois depistages (PC3), participation partielle a deux depistages parmi trois (PP2), participation partielle a un depistage parmi trois (PP1) et non-participation conjointe (NPC0). La categorie de reference etait la participation conjointe aux trois depistages (PC3). Des modeles de regression logistique multivariee multiniveau (niveau 1 individuel ; niveau 2 contextuel [le quartier d’habitation (IRISs) ou le medecin generaliste (MG)]) a intercept aleatoire ont ete utilises. Des analyses en sous-groupes selon la localisation ont ete realisees. Resultats Le taux de participation conjointe aux trois depistages etait de 11,5 %. Les taux de participation partielle a deux ou un depistage etaient respectivement 24 % et 30,1 %. Le taux de non-participation conjointe etait de 34,4 %. Les facteurs independamment associes a la non-participation conjointe et a la participation partielle etaient l’âge Conclusions Notre etude propose plusieurs leviers d’actions en replacant le MG au cœur du dispositif de ces trois depistages, en generalisant le depistage du cancer du col, en revalorisant la demarche de prevention dans la consultation du MG, dans le but de lever les freins individuels et de permettre aux femmes de devenir actrices de leur sante. Enfin, des actions locales sous la coordination des structures de gestion devraient etre envisagees de maniere ciblee aupres des populations defavorisees selon le principe de l’universalisme proportionne.

Background: Contextual socio-economic factors, health-care access, and general practitioner (GP) involvement may influence colonoscopy uptake and its timing after positive faecal occult blood testing (FOBT). Our objectives were to identify predictors of delayed or no colonoscopy and to assess the role for GPs in colonoscopy uptake. Methods: We included all residents of a French district with positive FOBTs (n=2369) during one of the two screening rounds (2007–2010). Multilevel logistic regression analysis was performed to identify individual and area-level predictors of delayed colonoscopy, no colonoscopy, and no information on colonoscopy. Results: A total of 998 (45.2%) individuals underwent early, 989 (44.8%) delayed, and 102 (4.6%) no colonoscopy; no information was available for 119 (5.4%) individuals. Delayed colonoscopy was independently associated with first FOBT (odds ratio, (OR)), 1.61; 95% confidence interval ((95% CI), 1.16–2.25); and no colonoscopy and no information with first FOBT (OR, 2.01; 95% CI, 1.02–3.97), FOBT kit not received from the GP (OR, 2.29; 95% CI, 1.67–3.14), and socio-economically deprived area (OR, 3.17; 95% CI, 1.98–5.08). Colonoscopy uptake varied significantly across GPs (P=0.01). Conclusion: Socio-economic factors, GP-related factors, and history of previous FOBT influenced colonoscopy uptake after a positive FOBT. Interventions should target GPs and individuals performing their first screening FOBT and/or living in socio-economically deprived areas.

Background The participation rate in organized breast cancer screening in France is lower than recommended. Non-participants either use opportunistic screening or do not use either screening modality. Objective To assess any differences in perceptions, attitudes and knowledge related to breast cancer screening between users of opportunistic screening and non-users of any screening mammograms and to identify potential barriers to participation in organized screening. Methods Six focus groups were conducted in May 2010 with 34 French non-participants in organized screening, 15 who used opportunistic screening (OpS group) and 19 who used no screening (NoS group). The guide used for both groups explored perceptions and attitudes related to health, cancer and screening; perceptions of femininity; and knowledge about breast cancer screening. Thematic content analysis was performed. Results Perceptions, attitudes and knowledge differed between the two groups. Women in the OpS group perceived a high susceptibility to breast cancer, visited their gynaecologist regularly, were unfamiliar with organized screening modalities and had doubts about its quality. NoS women had very high- or low-perceived susceptibility to breast cancer, knew about screening modalities, had doubts about its usefulness and expressed negative opinions of mammograms. Conclusions Differences in perceptions and attitudes related to breast cancer screening partially explain why some women choose opportunistic screening or no screening. General practitioners and gynaecologists are in a unique position to provide individually tailored preventative messages to improve participation in organized screening.

Le programme national de depistage du cancer du sein en France, initie en 1989, puis generalise en 2004, est precisement organise car soumis a des regles strictes de mise en œuvre.

Alzheimer disease (AD) is characterized by excessive deposition of amyloid beta-peptides (Abeta peptides) in the form of senile plaques as well as neurofibrillary tangles (NFTs) in the brain. In the amyloidogenic pathway, the amyloid-beta precursor protein (APP) is cleaved by beta-secretase first, followed by gamma-secretase cleavage producing therefore Abeta. This review summarizes the recent findings in the AD field and focuses on the different gamma-secretase inhibitors that have been developed as a therapeutic approach toward AD.

Cancer patients receiving chemotherapy are at increased risk of venous thromboembolism (VTE). The presence of cancer and anticoagulant use are risk factors for bleeding, yet data on bleeding risk are limited in these patients. This analysis evaluated the risk of major bleeding in cancer patients receiving chemotherapy using a US claims database. This retrospective cohort study used the MarketScan® databases, a nationwide database containing data from about 100 payers and covering > 30 million patients annually. Adult cancer patients receiving chemotherapy within 6 months of cancer diagnosis between January 2004 and December 2010 were included. Cancers of interest were: lung, colon/rectum, pancreas, bladder, stomach, and ovary. The index date was the first date of chemotherapy. Patients were followed until the earliest of: 1) first diagnosis of major bleeding; 2) termination of enrolment in the health plan; 3) end of study. The primary outcome was the first occurrence of major bleeding, based on selected ICD-9-CM/CPT codes, following chemotherapy initiation. Of 74,575 patients identified, exclusion of those with prior history of bleeding at baseline (∼5%) resulted in 70,822 patients included in the analysis. Mean age was 62 years, 37% were ≥ 65 years, and 52% were male. Average time of follow up and chemotherapy were 14.3 and 8.6 months, respectively; 6% had a history of VTE within 6 months prior to the index date. Major bleeding occurred in 5.8% of patients and the incidence rate for all cancers combined was 4.9 per 100 person-year (PY) and 10.5, 9.3, 6.2, 4.3, 3.6, and 3.3/100 PY for pancreatic, stomach, lung, bladder, colon/rectum, and ovarian cancer, respectively. Approximately 14% of patients (N = 10,456) developed VTE after chemotherapy initiation (> half in the first 3 months of chemotherapy treatment). Of these, 7.8% experienced major bleeding with incidence rates ranging from 5.9-17.7/100 PY after VTE. Major bleeding incidence in cancer patients receiving chemotherapy varies by cancer type with the highest rates in patients with upper gastrointestinal cancer. Compared to the overall cohort, major bleeding risk was higher in cancer patients who developed VTE. Disclosure J.H. Zong: Employee of Sanofi. L. Eckert: Employee of Sanofi. L. Zhang: Employee of Sanofi. W.S. Dai: Employee of Sanofi. A.T. Cohen: Consult: Astellas, AZ, Bayer, BI, BMS, Daiichi, GSK, JJ ResFund: AZ, Bayer, BI, BMS, Daiichi, GSK, JJ BoardSpeakerAdvis comm: Bayer, BI, BMS, Daiichi, GSK, J&J, Mitsubishi, Pfizer, Sanofi. All other authors have declared no conflicts of interest.

Objectifs Determiner le profil semiologique des cancers detectes en deuxieme lecture par rapport aux cancers detectes en premiere lecture. Materiels et methodes Etude retrospective ayant collige sur 6 annees les 100 derniers cancers detectes en deuxieme lecture avec determination du type d’erreur du premier lecteur (erreur de detection ou erreur d’interpretation) du classement ACR du deuxieme lecteur, du type semiologique de l’anomalie et enfin du diagnostic definitif avec la taille de la lesion. Resultats Les 100 cancers detectes en deuxieme lecture representaient 10,5 % des cancers detectes pendant cette periode trente cinq etaient des foyers de microcalcifications purs, 14 des lesions mixtes associant des microcalcifications a une autre anomalie « fibreuse » (masse, distorsion ou asymetrie de densite) et 51 etaient des lesions purement fibreuses sans microcalcifications (41 masses, 4 asymetries et 6 distorsions) soixante huit etaient des carcinomes invasifs et 32 des carcinomes in situ. La taille moyenne des cancers detectes etait de 13 mm (taille en premiere lecture 29 mm), (13,5 mm pour les invasifs) (extremes 2-60 mm). Les cancers invasifs de moins de 10 mm representaient 27,4 % du total (30,8 % en premiere lecture) et les cancers in situ 32,3 % (12 %). Conclusion Les cancers detectes en deuxieme lecture sont de plus petite taille mais ont une semiologie peu differente de ceux detectes en premiere lecture.

Objectifs Evaluer l’apport d’un systeme d’aide a la detection par ordinateur (CAD) en complement de la seconde lecture (L2) des mammographies BIRADS 1 a 3, en premiere lecture (Ll) en depistage organise. Materiels et methodes Entre janvier et decembre 2005, 8357 mammographies negatives en Ll ont ete numerisees par un systeme CAD analogique (Cadx, Icad®) puis interpretees sans et avec les resultats du CAD par un second lecteur, issu d’un pool de 35. Resultats En L2 sans et avec CAD (VN), 1393 dossiers etaient negatifs ; 6528 dossiers positifs avec CAD etaient negatives en L2 (FP) ; 9 cancers etaient detectes sur 303 dossiers positives par le CAD et le L2. Le CAD a positive seul 40 cas : 21 etaient negatives en L3, et 19 explores secondairement. Parmi ceux-ci, 17 etaient negatifs, et 1 cancer et un cas d’hyperplasie atypique ont ete diagnostiques. Enfin, sur 93 images retenues en L2 non detectees par le CAD, 4 cancers etaient detectes. Conclusion Cette etude ne retrouve pas l’amelioration de 20 % presentee dans la litterature. Il faut cependant ponderer ces resultats par le nombre important de lecteurs, qui rend plus heterogene le recueil des donnees. La sensibilite de detection est encore difficile a evaluer.

Objectifs Connaitre le principe du CAD en mammographie. Connaitre les performances du CAD dans le depistage du cancer du sein. Apprehender les roles comparatifs de la double lecture et du CAD en depistage organise. Resume Apres une dizaine d’annees d’experience, le CAD est une technique adulte. L’essentiel des questions initiales ont ete resolues : le CAD est-il sensible pour la detection des cancers ? La reponse est oui, puisque la sensibilite est de 95 % pour la detection des microcalcifications et de 85 % pour la detection des masses. Le CAD peut-il se substituer a une premiere lecture ? Non, a cause de la specificite faible inherente a tous les CAD. Le CAD beneficie-t-il du couplage avec la mammographie numerique ? Oui, meme s’il est possible de l’utiliser en analogique, l’utilisation est plus pratique en mammographie numerique. Quelques questions subsistent : le CAD peut-il aider en routine un radiologue expert ? A-t-il des vertus de « pedagogie ou de rattrapage » pour un lecteur moins experimente ? Il apparait aujourd’hui que le CAD en mammographie est a la fois mieux accepte en meme temps qu’il est banalise. Il ne se substitue pas a la lecture humaine mais il en facilite le deroulement. Un nouvel horizon sera de lui demander non plus d’analyser un seul cliche, mais deux cliches du meme patient a deux temps differents ; il pourrait faciliter la detection des anomalies subtiles et se verrait ainsi debarrasse d’un taux trop eleve de faux positifs. Conclusion Le CAD en mammographie a un interet certain d’aide complementaire pour le radiologue sans pour autant le remplacer.

Background Current screening guidelines recommend a complete colon evaluation with colonoscopy after a positive fecal occult blood test (FOBT). However, no timing guidelines are provided. A recent study showed that delay from FOBT to colonoscopy was associated with an increased risk of neoplasia. Our aim was to identify individual and contextual predictors of not undergoing an early colonoscopy after a positive FOBT. Methods All average-risk residents of a French county who have had a positive FOBT from June 2007 to December 2010 (n = 2369) during organized colorectal cancer (CRC) screening campaigns were included. Individual data were abstracted from the Organized Cancer Screening Association database and aggregated socioeconomic data (physician density, Townsend deprivation index) from the National Institute of Statistics and Economic Studies database. Multilevel and multinomial logistic regression analyses were performed to assess predictors of delayed colonoscopy (> median delay, 58 days), no colonoscopy and no response to cancer screening program after one year. Early colonoscopy was the reference category. Results The rate of colonoscopy was 86.9%. 1 037 patients (45.3%) had an early, and 1 021 (44.6%) a delayed colonoscopy, 106 (4.7%) did not perform colonoscopy and 123 (5.4%) were nonrespondents. The multilevel analysis displayed a significant (p Conclusion Actions to improve follow-up after a positive FOBT should focus on first CRC screening and population living in the most deprived areas. This study emphasizes the role of GPs in CRC screening. Disclosure All authors have declared no conflicts of interest.