Dataset Information

An assessment of existing models for individualized breast cancer risk estimation in a screening program in Spain.

ABSTRACT:

Background

The aim of this study was to evaluate the calibration and discriminatory power of three predictive models of breast cancer risk.

Methods

We included 13,760 women who were first-time participants in the Sabadell-Cerdanyola Breast Cancer Screening Program, in Catalonia, Spain. Projections of risk were obtained at three and five years for invasive cancer using the Gail, Chen and Barlow models. Incidence and mortality data were obtained from the Catalan registries. The calibration and discrimination of the models were assessed using the Hosmer-Lemeshow C statistic, the area under the receiver operating characteristic curve (AUC) and the Harrell's C statistic.

Results

The Gail and Chen models showed good calibration while the Barlow model overestimated the number of cases: the ratio between estimated and observed values at 5 years ranged from 0.86 to 1.55 for the first two models and from 1.82 to 3.44 for the Barlow model. The 5-year projection for the Chen and Barlow models had the highest discrimination, with an AUC around 0.58. The Harrell's C statistic showed very similar values in the 5-year projection for each of the models. Although they passed the calibration test, the Gail and Chen models overestimated the number of cases in some breast density categories.

Conclusions

These models cannot be used as a measure of individual risk in early detection programs to customize screening strategies. The inclusion of longitudinal measures of breast density or other risk factors in joint models of survival and longitudinal data may be a step towards personalized early detection of BC.

SUBMITTER: Arrospide A

PROVIDER: S-EPMC4029404 | biostudies-literature | 2013 Dec

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

An assessment of existing models for individualized breast cancer risk estimation in a screening program in Spain.

Arrospide Arantzazu A Forné Carles C Rué Montse M Torà Núria N Mar Javier J Baré Marisa M

BMC cancer 20131210

<h4>Background</h4>The aim of this study was to evaluate the calibration and discriminatory power of three predictive models of breast cancer risk.<h4>Methods</h4>We included 13,760 women who were first-time participants in the Sabadell-Cerdanyola Breast Cancer Screening Program, in Catalonia, Spain. Projections of risk were obtained at three and five years for invasive cancer using the Gail, Chen and Barlow models. Incidence and mortality data were obtained from the Catalan registries. The cali ...[more]

PMID: 24321553

Similar Datasets

Project description:Background:Pancreatic cancer (PC) is an aggressive disease with a dismal 5-year survival rate. Surveillance of high-risk individuals is hoped to improve survival outcomes by detection of precursor lesions or early-stage malignancy. Methods:Since 2011, a national high-risk cohort recruited through St Vincent's Hospital, Sydney, has undergone prospective PC screening incorporating annual endoscopic ultrasound, formal genetic counselling and mutation analysis as appropriate. PancPRO, a Bayesian PC risk assessment model, was used to estimate 5-year and lifetime PC risks for familial pancreatic cancer (FPC) participants and this was compared to their perceived chance of pancreatic and other cancers. Genetic counselling guidelines were developed to improve consistency. Follow-up questionnaires were used to assess the role of genetic counselling and testing. Results:We describe the Australian PC screening program design and recruitment strategy and the results of the first 102 individuals who have completed at least one-year of follow-up. Seventy-nine participants met the FPC criteria (? two first-degree relatives affected), 22 individuals had both a BRCA2 pathogenic variant and a close relative with PC and one had a clinical diagnosis of Peutz-Jeghers syndrome. Participants reported a high perceived chance of developing PC regardless of their genetic testing status. PancPRO reported FPC participants' mean 5-year and lifetime PC risks as 1.81% (range 0.2-3.2%) and 10.17% (range 2.4-14.4%), respectively. Participants' perceived PC chance did not correlate with their PancPRO 5-year (r?=?-?0.17, p?=?0.128) and lifetime PC risks (r?=?0.19, p?=?0.091). Two-thirds felt that current genetic testing would help them, and 91% of tested participants were glad to have undergone genetic testing. Overall, 79% of participants found genetic counselling to be helpful, and 88% reported they would recommend counselling to their relatives. Conclusions:Participants reported multiple benefits of genetic counselling and testing but continue to seek greater clarification about their individual PC risk. Extension of PancPRO is required to enable personalised PC risk assessment for all high-risk sub-groups. More detailed discussion of PC risk for BRCA2 pathogenic variant carriers, providing a written summary in all cases and a plan for genetics review were identified as areas for improvement.

Project description:ImportancePostoperative complications after head and neck surgery carry the potential for significant morbidity. Estimating the risk of complications in an individual patient is challenging.ObjectiveTo develop a statistical tool capable of predicting an individual patient's risk of developing a major complication after surgery for oral cavity squamous cell carcinoma.Design, setting, and participantsRetrospective case series derived from an institutional clinical oncologic database, augmented by medical record abstraction, at an academic tertiary care cancer center. Participants were 506 previously untreated adult patients with biopsy-proven oral cavity squamous cell carcinoma who underwent surgery between January 1, 2007, and December 31, 2012.Main outcomes and measuresThe primary end point was a major postoperative complication requiring invasive intervention (Clavien-Dindo classification grades III-V). Patients treated between January 1, 2007, and December 31, 2008 (354 of 506 [70.0%]) comprised the modeling cohort and were used to develop a nomogram to predict the risk of developing the primary end point. Univariable analysis and correlation analysis were used to prescreen 36 potential predictors for incorporation in the subsequent multivariable logistic regression analysis. The variables with the highest predictive value were identified with the step-down model reduction method and included in the nomogram. Patients treated between January 1, 2007, and December 31, 2008 (152 of 506 [30.0%]) were used to validate the nomogram.ResultsClinical characteristics were similar between the 2 cohorts for most comparisons. Thirty-six patients in the modeling cohort (10.2%) and 16 patients in the validation cohort (10.5%) developed a major postoperative complication. The 6 preoperative variables with the highest individual predictive value were incorporated within the nomogram, including body mass index, comorbidity status, preoperative white blood cell count, preoperative hematocrit, planned neck dissection, and planned tracheotomy. The nomogram predicted a major complication with a validated concordance index of 0.79. Inclusion of surgical operative variables in the nomogram maintained predictive accuracy (concordance index, 0.77).Conclusions and relevanceA statistical tool was developed that accurately estimates an individual patient's risk of developing a major complication after surgery for oral cavity squamous cell carcinoma.

Dataset Information

An assessment of existing models for individualized breast cancer risk estimation in a screening program in Spain.

Background

Methods

Results

Conclusions

Publications

An assessment of existing models for individualized breast cancer risk estimation in a screening program in Spain.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets