Risk prediction models of breast cancer: a systematic review of model performances

Following the High-Risk Patient: Breast Cancer Risk-Based Screening

Breast cancer (BC) is the most common cancer occurring in women in the USA today, and accounts for more than 40,000 deaths annually (Giaquinto in CA Cancer J Clin 72: 524-541, 2022). While breast cancer survival has improved over the past decades, incidence has increased, and diagnoses are being made at younger ages. This emphasizes the importance of risk evaluation, accurate prediction, and effective mitigation and risk reduction strategies. Enhanced screening can help detect cancers at an earlier stage, thus improving morbidity and mortality. This review addresses the recognition of women at high-risk for BC and monitoring strategies for those at high risk.

Understanding the contribution of lifestyle in breast cancer risk prediction: a systematic review of models applicable to Europe

BACKGROUND

Breast cancer (BC) is a significant health concern among European women, with the highest prevalence rates among all cancers. Existing BC prediction models account for major risks such as hereditary, hormonal and reproductive factors, but research suggests that adherence to a healthy lifestyle can reduce the risk of developing BC to some extent. Understanding the influence and predictive role of lifestyle variables in current risk prediction models could help identify actionable, modifiable, targets among high-risk population groups.

PURPOSE

To systematically review population-based BC risk prediction models applicable to European populations and identify lifestyle predictors and their corresponding parameter values for a better understanding of their relative contribution to the prediction of incident BC.

METHODS

A systematic review was conducted in PubMed, Embase and Web of Science from January 2000 to August 2021. Risk prediction models were included if (i) developed and/or validated in adult cancer-free women in Europe, (ii) based on easily ascertained information, and (iii) reported models' final predictors. To investigate further the comparability of lifestyle predictors across models, estimates were standardised into risk ratios and visualised using forest plots.

RESULTS

From a total of 49 studies, 33 models were developed and 22 different existing models, mostly from Gail (22 studies) and Tyrer-Cuzick and co-workers (12 studies) were validated or modified for European populations. Family history of BC was the most frequently included predictor (31 models), while body mass index (BMI) and alcohol consumption (26 and 21 models, respectively) were the lifestyle predictors most often included, followed by smoking and physical activity (7 and 6 models respectively). Overall, for lifestyle predictors, their modest predictive contribution was greater for riskier lifestyle levels, though highly variable model estimates across different models.

CONCLUSIONS

Given the increasing BC incidence rates in Europe, risk models utilising readily available risk factors could greatly aid in widening the population coverage of screening efforts, while the addition of lifestyle factors could help improving model performance and serve as intervention targets of prevention programmes.

Breast Cancer Risk Assessment Tools for Stratifying Women into Risk Groups: A Systematic Review

BACKGROUND

The benefits and harms of breast screening may be better balanced through a risk-stratified approach. We conducted a systematic review assessing the accuracy of questionnaire-based risk assessment tools for this purpose.

METHODS

Population: asymptomatic women aged ≥40 years; Intervention: questionnaire-based risk assessment tool (incorporating breast density and polygenic risk where available); Comparison: different tool applied to the same population; Primary outcome: breast cancer incidence; Scope: external validation studies identified from databases including Medline and Embase (period 1 January 2008-20 July 2021). We assessed calibration (goodness-of-fit) between expected and observed cancers and compared observed cancer rates by risk group. Risk of bias was assessed with PROBAST.

RESULTS

Of 5124 records, 13 were included examining 11 tools across 15 cohorts. The Gail tool was most represented (n = 11), followed by Tyrer-Cuzick (n = 5), BRCAPRO and iCARE-Lit (n = 3). No tool was consistently well-calibrated across multiple studies and breast density or polygenic risk scores did not improve calibration. Most tools identified a risk group with higher rates of observed cancers, but few tools identified lower-risk groups across different settings. All tools demonstrated a high risk of bias.

CONCLUSION

Some risk tools can identify groups of women at higher or lower breast cancer risk, but this is highly dependent on the setting and population.

Prediction of breast cancer risk among women of the Mariana Islands: the BRISK retrospective case-control study

OBJECTIVES

To develop a breast cancer risk prediction model for Chamorro and Filipino women of the Mariana Islands and compare its performance to that of the Breast Cancer Risk Assessment Tool (BCRAT).

DESIGN

Case-control study.

SETTING

Clinics/facilities and other community-based settings on Guam and Saipan (Northern Mariana Islands).

PARTICIPANTS

245 women (87 breast cancer cases and 158 controls) of Chamorro or Filipino ethnicity, age 25-80 years, with no prior history of cancer (other than skin cancer), residing on Guam or Saipan for at least 5 years.

PRIMARY AND SECONDARY OUTCOME MEASURES

Breast cancer risk models were constructed using combinations of exposures previously identified to affect breast cancer risk in this population, population breast cancer incidence rates and all-cause mortality rates for Guam.

RESULTS

Models using ethnic-specific relative risks performed better than those with relative risks estimated from all women. The model with the best performance among both ethnicities (the Breast Cancer Risk Model (BRISK) model; area under the receiver operating characteristic curve (AUC): 0.64 and 0.67 among Chamorros and Filipinos, respectively) included age at menarche, age at first live birth, number of relatives with breast cancer and waist circumference. The 10-year breast cancer risk predicted by the BRISK model was 1.28% for Chamorros and 0.89% for Filipinos. Performance of the BCRAT was modest among both Chamorros (AUC: 0.60) and Filipinos (AUC: 0.55), possibly due to incomplete information on BCRAT risk factors.

CONCLUSIONS

The ability to develop breast cancer risk models for Mariana Islands women is constrained by the small population size and limited availability of health services and data. Nonetheless, we have demonstrated that breast cancer risk prediction models with adequate discriminatory performance can be built for small populations such as in the Mariana Islands. Anthropometry, in particular waist circumference, was important for estimating breast cancer risk in this population.

Risk prediction models for head and neck cancer: A rapid review

Background

Cancer risk assessment models are used to support prevention and early detection. However, few models have been developed for head and neck cancer (HNC).

Methods

A rapid review of Embase and MEDLINE identified n = 3045 articles. Following dual screening, n = 14 studies were included. Quality appraisal using the PROBAST (risk of bias) instrument was conducted, and a narrative synthesis was performed to identify the best performing models in terms of risk factors and designs.

Results

Six of the 14 models were assessed as "high" quality. Of these, three had high predictive performance achieving area under curve values over 0.8 (0.87-0.89). The common features of these models were their inclusion of predictors carefully tailored to the target population/anatomical subsite and development with external validation.

Conclusions

Some existing models do possess the potential to identify and stratify those at risk of HNC but there is scope for improvement.

Exploring Clinical Risk Factors for Breast Cancer Among American Indian Women

Objective

Very little is known about the breast cancer risk profile among American Indian women. Previous research shows that the proportion of American Indian/Alaska Native women with baseline characteristics (commonly known breast cancer risk factors) differs from other ethnicities. This retrospective case control study was designed to the explore the association of these factors among American Indian women with and without breast cancer.

Methods

Cases and controls were retrospectively selected from the medical records of American Indian women who obtained their health care from Quentin N. Burdick Memorial Health Care Facility (Indian Health Service) in Belcourt, ND. For each woman with breast cancer (n = 141), two controls were selected when possible (n = 278). Risk factors examined included woman's age, age at first live birth, age of menarche, the number of previous benign breast biopsies, the total number of first-degree relatives with breast cancer, body mass index and parity. Odds ratios and 95% confidence intervals were calculated using logistic regression.

Results

Many of the associations found among American Indian women who obtained their health care from Quentin N. Burdick Memorial Health Care Facility (Indian Health Service) in Belcourt, ND, between risk factors commonly identified in other populations and breast cancer were weakly positive. Nulliparity was the only risk factor to consistently show a positive significant association (OR = 2.87, 95% CI 1.16–0.7.12).

Conclusion

Disparities in breast cancer incidence, mortality and screening among Northern Plains American Indian emphasize the need to better understand the risk factors associated with breast cancer in this population. Based on the results of this study, the value of current risk prediction models in American Indian communities is uncertain and clinicians should be cautious in using these models to inform American Indian patients of their risk for breast cancer.

Prediction models for gastric cancer risk in the general population: a systematic review

Risk prediction models for gastric cancer (GC) could identify high-risk individuals in the general population. The objective of this study was to systematically review the available evidence about the construction and verification of GC predictive models. We searched PubMed, Embase, and Cochrane Library databases for articles that developed or validated GC risk prediction models up to November 2021. Data extracted included study characteristics, predictor selection, missing data, and evaluation metrics. Risk of bias (ROB) was assessed using the Prediction model study Risk Of Bias Assessment Tool (PROBAST). We identified a total of 12 original risk prediction models that fulfilled the criteria for analysis. The area under the receiver operating characteristic curve ranged from 0.73 to 0.93 in derivation sets (n=6), 0.68 to 0.90 in internal validation sets (n=5), 0.71 to 0.92 in external validation sets (n=7). The higher-performing models usually include age, salt preference, Helicobacter pylori, smoking, BMI, family history, pepsinogen and sex. According to PROBAST, at least one domain with a high ROB was present in all studies mainly due to methodologic limitations in the analysis domain. In conclusion, although some risk prediction models including similar predictors have displayed sufficient discriminative abilities, many have a high ROB due to methodological limitations and are not externally validated efficiently. Future prediction models should adherence to well-established standards and guidelines to benefit GC screening.

Validation of Breast Cancer Risk Models by Race/Ethnicity, Family History and Molecular Subtypes

Simple Summary

Several statistical models exist to predict a person’s risk of breast cancer. Risk assessment models can guide cancer screening approaches by identifying individuals who would benefit from additional screening. In this study, we compared the performance of four models in predicting the 5-year risk of breast cancer in a cohort of women aged 40–84 years who underwent screening mammography at three large health systems. Models showed comparable discrimination (ability to distinguish between cases and non-cases) and calibration (ability to accurately predict risk) overall, with no difference by race. Model discrimination was poorer for some cancer subtypes, and better for women with high BMI. The combined BRCAPRO+BCRAT model had improved calibration and discrimination among women with a family history of breast cancer. Our results can inform risk-based screening approaches by identifying women at a high risk of breast cancer.

Abstract

(1) Background: The purpose of this study is to compare the performance of four breast cancer risk prediction models by race, molecular subtype, family history of breast cancer, age, and BMI. (2) Methods: Using a cohort of women aged 40–84 without prior history of breast cancer who underwent screening mammography from 2006 to 2015, we generated breast cancer risk estimates using the Breast Cancer Risk Assessment tool (BCRAT), BRCAPRO, Breast Cancer Surveillance Consortium (BCSC) and combined BRCAPRO+BCRAT models. Model calibration and discrimination were compared using observed-to-expected ratios (O/E) and the area under the receiver operator curve (AUC) among patients with at least five years of follow-up. (3) Results: We observed comparable discrimination and calibration across models. There was no significant difference in model performance between Black and White women. Model discrimination was poorer for HER2+ and triple-negative subtypes compared with ER/PR+HER2−. The BRCAPRO+BCRAT model displayed improved calibration and discrimination compared to BRCAPRO among women with a family history of breast cancer. Across models, discriminatory accuracy was greater among obese than non-obese women. When defining high risk as a 5-year risk of 1.67% or greater, models demonstrated discordance in 2.9% to 19.7% of patients. (4) Conclusions: Our results can inform the implementation of risk assessment and risk-based screening among women undergoing screening mammography.

Risk Prediction Model for Esophageal Cancer Among General Population: A Systematic Review

Objective: The risk prediction model is an effective tool for risk stratification and is expected to play an important role in the early detection and prevention of esophageal cancer. This study sought to summarize the available evidence of esophageal cancer risk predictions models and provide references for their development, validation, and application.

Methods: We searched PubMed, EMBASE, and Cochrane Library databases for original articles published in English up to October 22, 2021. Studies that developed or validated a risk prediction model of esophageal cancer and its precancerous lesions were included. Two reviewers independently extracted study characteristics including predictors, model performance and methodology, and assessed risk of bias and applicability with PROBAST (Prediction model Risk Of Bias Assessment Tool).

Results: A total of 20 studies including 30 original models were identified. The median area under the receiver operating characteristic curve of risk prediction models was 0.78, ranging from 0.68 to 0.94. Age, smoking, body mass index, sex, upper gastrointestinal symptoms, and family history were the most commonly included predictors. None of the models were assessed as low risk of bias based on PROBST. The major methodological deficiencies were inappropriate date sources, inconsistent definition of predictors and outcomes, and the insufficient number of participants with the outcome.

Conclusions: This study systematically reviewed available evidence on risk prediction models for esophageal cancer in general populations. The findings indicate a high risk of bias due to several methodological pitfalls in model development and validation, which limit their application in practice.

Breast Cancer Prevention: Time for Change

Agency breast cancer prevention guidelines for other than hereditary cancers have not materially changed in 20 years; endocrine-targeted agents (then, tamoxifen; now, adding raloxifene and aromatase inhibitors) reduce good prognosis estrogen receptor (ER)-positive, progesterone receptor (PR)-positive cancers without reducing deaths from breast cancer. Across three tamoxifen placebo-controlled prevention trials (N = 23,360) begun almost 30 years ago, although there were 226 fewer breast cancer cases, there were nine more deaths from breast cancer in the tamoxifen groups. Following clinical advances, currently more than half of breast cancer cases are solved problems with extremely low risk of death. As endocrine-targeted agents commonly prevent these cancers, widespread implementation of current prevention strategies may not reduce deaths from breast cancer. Compared with other breast cancers, ER-positive, PR-negative cancers and triple-negative cancers have inferior survival (90.6% v 83.8% v 78.1%, respectively; P < .001). Against this background, in the Women's Health Initiative Dietary Modification randomized trial (N = 48,835), ER-positive, PR-negative cancers were statistically significantly reduced in the intervention group (hazard ratio, 0.77; 95% CI, 0.64 to 0.94) and deaths from breast cancer were reduced 21% (P = .02). In the Women's Health Initiative randomized, placebo-controlled trial evaluating conjugated equine estrogen (N = 10,739), ER-positive, PR-negative cancers were statistically significantly reduced in the intervention group (hazard ratio, 0.44; 95% CI, 0.27 to 0.74) and deaths from breast cancer were reduced 40% (P = .04). These findings suggest that reexamination of breast cancer risk reduction strategies and clinical practice is needed.

Essentialism and Exclusion: Racism in Cancer Risk Prediction Models

Cancer risk prediction models have the potential to revolutionize the science and practice of cancer prevention and control by identifying the likelihood that a patient will develop cancer at some point in the future, likely experience more benefit than harm from a given intervention, and survive their cancer for a certain number of years. The ability of risk prediction models to produce estimates that are valid and reliable for people from diverse socio-demographic backgrounds-and consequently their utility for broadening the reach of precision medicine to marginalized populations-depends on ensuring that the risk factors included in the model are represented as thoroughly and as accurately as possible. However, cancer risk prediction models created in the United States have a critical limitation, the origins of which stem from the country's earliest days: they either erroneously treat the social construct of race as an immutable biological factor (ie, they "essentialize" race), or they exclude from the model those socio-contextual factors that are associated with both race and health outcomes. Models that essentialize race and/or exclude socio-contextual factors sometimes incorporate "race corrections" that adjust a patient's risk estimate up or down based on their race. This commentary discusses the origins of race corrections, potential flaws with such corrections, and strategies for developing cohorts for developing risk prediction models that do not essentialize race or exclude key socio-contextual factors. Such models will help move the science of cancer prevention and control towards its goal of eliminating cancer disparities and achieving health equity.

Development and Validation of a Breast Cancer Risk Prediction Model for Childhood Cancer Survivors Treated With Chest Radiation: A Report From the Childhood Cancer Survivor Study and the Dutch Hodgkin Late Effects and LATER Cohorts

PURPOSE

Women treated with chest radiation for childhood cancer have one of the highest risks of breast cancer. Models producing personalized breast cancer risk estimates applicable to this population do not exist. We sought to develop and validate a breast cancer risk prediction model for childhood cancer survivors treated with chest radiation incorporating treatment-related factors, family history, and reproductive factors.

METHODS

Analyses were based on multinational cohorts of female 5-year survivors of cancer diagnosed younger than age 21 years and treated with chest radiation. Model derivation was based on 1,120 participants in the Childhood Cancer Survivor Study diagnosed between 1970 and 1986, with median attained age 42 years (range 20-64) and 242 with breast cancer. Model validation included 1,027 participants from three cohorts, with median age 32 years (range 20-66) and 105 with breast cancer.

RESULTS

The model included current age, chest radiation field, whether chest radiation was delivered within 1 year of menarche, anthracycline exposure, age at menopause, and history of a first-degree relative with breast cancer. Ten-year risk estimates ranged from 2% to 23% for 30-year-old women (area under the curve, 0.63; 95% CI, 0.50 to 0.73) and from 5% to 34% for 40-year-old women (area under the curve, 0.67; 95% CI, 0.54 to 0.84). The highest risks were among premenopausal women older than age 40 years treated with mantle field radiation within a year of menarche who had a first-degree relative with breast cancer. It showed good calibration with an expected-to-observed ratio of the number of breast cancers of 0.92 (95% CI, 0.74 to 1.16).

CONCLUSION

Breast cancer risk varies among childhood cancer survivors treated with chest radiation. Accurate risk prediction may aid in refining surveillance, counseling, and preventive strategies in this population.

Breast cancer risk prediction in African women using Random Forest Classifier

INTRODUCTION

One of the most important steps in combating breast cancer is early and accurate diagnosis. Unfortunately, breast cancer is asymptomatic at the early stage, although some symptoms are presented at a later time, but at symptomatic stage treatment could be complicated or even become impossible thereby leading to death. Proper risk assessment is hence very important in reducing mortality. Some computational techniques have been developed for breast cancer risk assessment in the developed world, but such techniques do not work well in Africa because of the difference in risk profiles of African women e.g. later menarche, low drug abuse and low smoking rate.

AIM

In this work, we propose a bespoke risk prediction model for African women using Random Forest Classifier (RFC) machine learning technique.

METHODS

A total of 180 subjects were studied out of which 90 were confirmed cases of breast cancer and 90 were benign. Twenty-five risk factors were included, for example, smoking, alcohol intake, occupational hazards and age at menopause. Four approaches were empirically used in the feature selection, these are the use of Chi-Square, mutual information gain, Spearman correlation and the entire features. RFC algorithm was used to develop the prediction model.

RESULTS

We found that family history of breast cancer, dense breast, deliberate abortion, age at first child, fruit intake and regular exercise are predictors of breast cancer. The RFC model gave an accuracy of 91.67%, sensitivity of 87.10%, specificity of 96.55% and Area under curve (AUC) of 92% when all the risk factors were included in the model while an accuracy of 96.67%, sensitivity of 93.75%, specificity of 100% and AUC of 97% were obtained when correlation-selected features were included in the model. The Chi-Square selected features gave the best performance with 98.33% accuracy, 100% sensitivity, 96.55 specificity and 98% AUC. Mutual information gain selected feature gave the same results as Chi-Square selected features.

CONCLUSION

Random Forest Classifier has a good potential at predicting the risk of breast cancer in African women. The study helped to identify the risk factors of breast cancer in African women. This is a valuable information which can help African women to pay attention to those risk factors with the intention of reducing the incidence of breast cancer in Africa.

Prediction of Incident Cancers in the Lifelines Population-Based Cohort

Simple Summary

The accurate prediction of incident cancers could be relevant to understanding and reducing cancer incidence. The aim of this study was to develop machine learning (ML) models that could predict an incident diagnosis of cancer. Data were available for 116,188 cancer-free participants and 4232 incident cancer cases. The main outcome was an incident cancer (excluding skin cancer) during follow-up assessment in a population-based cohort. The performance of three ML algorithms was evaluated using supervised binary classification to identify incident cancers among participants. An overall area under the receiver operator curve (AUC) < 0.75 was obtained; the highest AUC was for prostate cancer AUC > 0.80. Linear and non-linear ML algorithms including socioeconomic, lifestyle, and clinical variables produced a moderate predictive performance of incident cancers in the Lifelines cohort.

Abstract

Cancer incidence is rising, and accurate prediction of incident cancers could be relevant to understanding and reducing cancer incidence. The aim of this study was to develop machine learning (ML) models that could predict an incident diagnosis of cancer. Participants without any history of cancer within the Lifelines population-based cohort were followed for a median of 7 years. Data were available for 116,188 cancer-free participants and 4232 incident cancer cases. At baseline, socioeconomic, lifestyle, and clinical variables were assessed. The main outcome was an incident cancer during follow-up (excluding skin cancer), based on linkage with the national pathology registry. The performance of three ML algorithms was evaluated using supervised binary classification to identify incident cancers among participants. Elastic net regularization and Gini index were used for variables selection. An overall area under the receiver operator curve (AUC) <0.75 was obtained, the highest AUC value was for prostate cancer (random forest AUC = 0.82 (95% CI 0.77–0.87), logistic regression AUC = 0.81 (95% CI 0.76–0.86), and support vector machines AUC = 0.83 (95% CI 0.78–0.88), respectively); age was the most important predictor in these models. Linear and non-linear ML algorithms including socioeconomic, lifestyle, and clinical variables produced a moderate predictive performance of incident cancers in the Lifelines cohort.

Preventing Breast Cancer Through Identification and Pharmacologic Management of High-Risk Patients

Breast cancer remains the most common cancer in women in the United States. For certain women at high risk for breast cancer, endocrine therapy (ET) can greatly decrease the risk. Tools such as the Breast Cancer Risk Assessment Tool (or Gail Model) and the International Breast Cancer Intervention Study risk calculator are available to help identify women at increased risk for breast cancer. Physician awareness of family history, reproductive and lifestyle factors, dense breast tissue, and history of benign proliferative breast disease are important when identifying high-risk women. The updated US Preventive Services Task Force and American Society of Clinical Oncology guidelines encourage primary care providers to identify at-risk women and offer risk-reducing medications. Among the various ETs, which include tamoxifen, raloxifene, anastrozole, and exemestane, tamoxifen is the only one available for premenopausal women aged 35 years and older. A shared decision-making process should be used to increase the usage of ET and must be individualized. This individualized approach must account for each woman's medical history and weigh the benefits and risks of ET in combination with the personal values of the patient.

Discriminatory Accuracy of the Gail Model for Breast Cancer Risk Assessment among Iranian Women

Background:

The Gail model is the most well-known tool for breast cancer risk assessment worldwide. Although it was validated in various Western populations, inconsistent results were reported from Asian populations. We used data from a large case-control study and evaluated the discriminatory accuracy of the Gail model for breast cancer risk assessment among the Iranian female population.

Methods:

We used data from 942 breast cancer patients and 975 healthy controls at the Cancer Institute of Iran, Tehran, Iran, in 2016. We refitted the Gail model to our case-control data (the IR-Gail model). We compared the discriminatory power of the IR-Gail with the original Gail model, using ROC curve analyses and estimation of the area under the ROC curve (AUC).

Results:

Except for the history of biopsies that showed an extremely high relative risk (OR=9.1), the observed ORs were similar to the estimates observed in Gail’s study. Incidence rates of breast cancer were extremely lower in Iran than in the USA, leading to a lower average absolute risk among the Iranian population (2.78, ±SD 2.45). The AUC was significantly improved after refitting the model, but it remained modest (0.636 vs. 0.627, ΔAUC = 0.009, bootstrapped P=0.008). We reported that the cut-point of 1.67 suggested in the Gail study did not discriminate between breast cancer patients and controls among the Iranian female population.

Conclusion:

Although the coefficients from the local study improved the discriminatory accuracy of the model, it remained modest. Cohort studies are warranted to evaluate the validity of the model for Iranian women.

Performance of Breast Cancer Risk-Assessment Models in a Large Mammography Cohort

BACKGROUND

Several breast cancer risk-assessment models exist. Few studies have evaluated predictive accuracy of multiple models in large screening populations.

METHODS

We evaluated the performance of the BRCAPRO, Gail, Claus, Breast Cancer Surveillance Consortium (BCSC), and Tyrer-Cuzick models in predicting risk of breast cancer over 6 years among 35 921 women aged 40-84 years who underwent mammography screening at Newton-Wellesley Hospital from 2007 to 2009. We assessed model discrimination using the area under the receiver operating characteristic curve (AUC) and assessed calibration by comparing the ratio of observed-to-expected (O/E) cases. We calculated the square root of the Brier score and positive and negative predictive values of each model.

RESULTS

Our results confirmed the good calibration and comparable moderate discrimination of the BRCAPRO, Gail, Tyrer-Cuzick, and BCSC models. The Gail model had slightly better O/E ratio and AUC (O/E = 0.98, 95% confidence interval [CI] = 0.91 to 1.06, AUC = 0.64, 95% CI = 0.61 to 0.65) compared with BRCAPRO (O/E = 0.94, 95% CI = 0.88 to 1.02, AUC = 0.61, 95% CI = 0.59 to 0.63) and Tyrer-Cuzick (version 8, O/E = 0.84, 95% CI = 0.79 to 0.91, AUC = 0.62, 95% 0.60 to 0.64) in the full study population, and the BCSC model had the highest AUC among women with available breast density information (O/E = 0.97, 95% CI = 0.89 to 1.05, AUC = 0.64, 95% CI = 0.62 to 0.66). All models had poorer predictive accuracy for human epidermal growth factor receptor 2 positive and triple-negative breast cancers than hormone receptor positive human epidermal growth factor receptor 2 negative breast cancers.

CONCLUSIONS

In a large cohort of patients undergoing mammography screening, existing risk prediction models had similar, moderate predictive accuracy and good calibration overall. Models that incorporate additional genetic and nongenetic risk factors and estimate risk of tumor subtypes may further improve breast cancer risk prediction.

Comparative Validation of Breast Cancer Risk Prediction Models and Projections for Future Risk Stratification

BACKGROUND

External validation of risk models is critical for risk-stratified breast cancer prevention. We used the Individualized Coherent Absolute Risk Estimation (iCARE) as a flexible tool for risk model development and comparative model validation and to make projections for population risk stratification.

METHODS

Performance of two recently developed models, one based on the Breast and Prostate Cancer Cohort Consortium analysis (iCARE-BPC3) and another based on a literature review (iCARE-Lit), were compared with two established models (Breast Cancer Risk Assessment Tool and International Breast Cancer Intervention Study Model) based on classical risk factors in a UK-based cohort of 64 874 white non-Hispanic women (863 patients) age 35-74 years. Risk projections in a target population of US white non-Hispanic women age 50-70 years assessed potential improvements in risk stratification by adding mammographic breast density (MD) and polygenic risk score (PRS).

RESULTS

The best calibrated models were iCARE-Lit (expected to observed number of cases [E/O] = 0.98, 95% confidence interval [CI] = 0.87 to 1.11) for women younger than 50 years, and iCARE-BPC3 (E/O = 1.00, 95% CI = 0.93 to 1.09) for women 50 years or older. Risk projections using iCARE-BPC3 indicated classical risk factors can identify approximately 500 000 women at moderate to high risk (>3% 5-year risk) in the target population. Addition of MD and a 313-variant PRS is expected to increase this number to approximately 3.5 million women, and among them, approximately 153 000 are expected to develop invasive breast cancer within 5 years.

CONCLUSIONS

iCARE models based on classical risk factors perform similarly to or better than BCRAT or IBIS in white non-Hispanic women. Addition of MD and PRS can lead to substantial improvements in risk stratification. However, these integrated models require independent prospective validation before broad clinical applications.

Development of a comprehensive health-risk prediction tool for postmenopausal women

OBJECTIVE

The aim of the study was to develop a web-based calculator that predicts the likelihood of experiencing multiple, competing outcomes prospectively over 5, 10, and 15 years.

METHODS

Baseline demographic and medical data from a healthy and racially and ethnically diverse cohort of 161,808 postmenopausal women, aged 50 to 79 at study baseline, who participated in the Women's Health Initiative (WHI), were used to develop and evaluate a risk-prediction calculator designed to predict individual risk for morbidity and mortality outcomes. Women were enrolled from 40 sites arranged in four regions of the United States. The calculator predicts all-cause mortality, adjudicated outcomes of health events (ie, myocardial infarction [MI], stroke, and hip fracture), and disease (lung, breast, and colorectal cancer). A proportional subdistribution hazards regression model was used to develop the calculator in a training dataset using data from three regions. The calculator was evaluated using the C-statistic in a test dataset with data from the fourth region.

RESULTS

The predictive validity of our calculator measured by the C-statistic in the test dataset for a first event at 5 and 15 years was as follows: MI 0.77, 0.61, stroke 0.77, 0.72, lung cancer 0.82, 0.79, breast cancer 0.60, 0.59, colorectal cancer 0.67, 0.60, hip fracture 0.79, 0.76, and death 0.74, 0.72.

CONCLUSION

This study represents the first large-scale study to develop a risk prediction calculator that yields health risk prediction for several outcomes simultaneously. Development of this tool is a first step toward enabling women to prioritize interventions that may decrease these risks. : Video Summary:http://links.lww.com/MENO/A463.

Identifying Women at High Risk for Breast Cancer Using Data From the Electronic Health Record Compared With Self-Report

PURPOSE

A barrier to chemoprevention uptake among high-risk women is the lack of routine breast cancer risk assessment in the primary care setting. We calculated breast cancer risk using the Breast Cancer Surveillance Consortium (BCSC) model, accounting for age, race/ethnicity, first-degree family history of breast cancer, benign breast disease, and mammographic density, using data collected from the electronic health records (EHRs) and self-reports.

PATIENTS AND METHODS

Among women undergoing screening mammography, we enrolled those age 35 to 74 years without a prior history of breast cancer. Data on demographics, first-degree family history, breast radiology, and pathology reports were extracted from the EHR. We assessed agreement between the EHR and self-report on information about breast cancer risk.

RESULTS

Among 9,514 women with known race/ethnicity, 1,443 women (15.2%) met high-risk criteria based upon a 5-year invasive breast cancer risk of 1.67% or greater according to the BCSC model. Among 1,495 women with both self-report and EHR data, more women with a first-degree family history of breast cancer (14.6% v 4.4%) and previous breast biopsies (21.3% v 11.3%) were identified by self-report versus EHR, respectively. However, more women with atypia and lobular carcinoma in situ were identified from the EHR. There was moderate agreement in identification of high-risk women between EHR and self-report data (κ, 0.48; 95% CI, 0.42-0.54).

CONCLUSION

By using EHR data, we determined that 15% of women undergoing screening mammography had a high risk for breast cancer according to the BCSC model. There was moderate agreement between information on breast cancer risk derived from the EHR and self-report. Examining EHR data may serve as an initial screen for identifying women eligible for breast cancer chemoprevention.

Individual and joint performance of DNA methylation profiles, genetic risk score and environmental risk scores for predicting breast cancer risk

DNA methylation patterns in the blood, genetic risk scores (GRSs), and environmental risk factors can potentially improve breast cancer (BC) risk prediction. We assessed the individual and joint predictive performance of methylation, GRS, and environmental risk factors for BC incidence in a prospective cohort study. In a cohort of 5462 women aged 50–75 from Germany, 101 BC cases were identified during 14 years of follow‐up and were compared to 263 BC‐free controls in a nested case–control design. Three previously suggested methylation risk scores (MRSs) based on methylation of 423, 248, and 131 cytosine‐phosphate‐guanine (CpG) loci, and a GRS based on the risk alleles from 269 recently identified single nucleotide polymorphisms were constructed. Additionally, multiple previously proposed environmental risk scores (ERSs) were built based on environmental variables. Areas under the receiver operating characteristic curves (AUCs) were estimated for evaluating BC risk prediction performance. MRS and ERS showed limited accuracy in predicting BC incidence, with AUCs ranging from 0.52 to 0.56 and from 0.52 to 0.59, respectively. The GRS predicted BC incidence with a higher accuracy (AUC = 0.61). Adjusted odds ratios per standard deviation increase (95% confidence interval) were 1.07 (0.84–1.36) and 1.40 (1.09–1.80) for the best performing MRS and ERS, respectively, and 1.48 (1.16–1.90) for the GRS. A full risk model combining the MRS, GRS, and ERS predicted BC incidence with the highest accuracy (AUC = 0.64) and might be useful for identifying high‐risk populations for BC screening.

The use of rigorous methods was strongly warranted among prognostic prediction models for obstetric care

OBJECTIVE

The objective of the study was to examine methodological characteristics about the design and conduct in prognostic prediction models used for obstetric care.

STUDY DESIGN AND SETTING

We searched PubMed for studies on prognostic prediction models for obstetric care, published in top general medicine or major specialty journals between January 2011 and February 2018. Teams of method-trained investigators independently screened titles and abstracts and collected data using a prespecified, pilot-tested, structured questionnaire.

RESULTS

In total, 91 studies were eligible, of which two were published in top general medicine journals, 20 (22.0%) involved an epidemiologist or statistician, 18 (19.4%) published study protocols, 53 (58.2%) did not include any model validation, 20 (22.0%) did not clearly state the intended timing of use, 23 (25.3%) had no eligibility criteria, 15 (16.5%) did not use clear criteria for ascertaining outcome, and 69 (75.82%) did not apply blinding to outcome assessment. Among those models, 11 (12.1%) included participants fewer than 200 events, 41 (48.8%) had fewer than 100 events, and 19 (24.7%) had fewer than 10 events per variable.

CONCLUSION

The prognostic prediction models have important limitations in design and conduct. Substantial efforts are needed to strengthen the production of reliable prognostic prediction models for obstetric care.

A systematic review and quality assessment of individualised breast cancer risk prediction models

Background

Individualised breast cancer risk prediction models may be key for planning risk-based screening approaches. Our aim was to conduct a systematic review and quality assessment of these models addressed to women in the general population.

Methods

We followed the Cochrane Collaboration methods searching in Medline, EMBASE and The Cochrane Library databases up to February 2018. We included studies reporting a model to estimate the individualised risk of breast cancer in women in the general population. Study quality was assessed by two independent reviewers. Results are narratively summarised.

Results

We included 24 studies out of the 2976 citations initially retrieved. Twenty studies were based on four models, the Breast Cancer Risk Assessment Tool (BCRAT), the Breast Cancer Surveillance Consortium (BCSC), the Rosner & Colditz model, and the International Breast Cancer Intervention Study (IBIS), whereas four studies addressed other original models. Four of the studies included genetic information. The quality of the studies was moderate with some limitations in the discriminative power and data inputs. A maximum AUROC value of 0.71 was reported in the study conducted in a screening context.

Conclusion

Individualised risk prediction models are promising tools for implementing risk-based screening policies. However, it is a challenge to recommend any of them since they need further improvement in their quality and discriminatory capacity.

Comparison of Questionnaire-Based Breast Cancer Prediction Models in the Nurses' Health Study

BACKGROUND

The Gail model and the model developed by Tyrer and Cuzick are two questionnaire-based approaches with demonstrated ability to predict development of breast cancer in a general population.

METHODS

We compared calibration, discrimination, and net reclassification of these models, using data from questionnaires sent every 2 years to 76,922 participants in the Nurses' Health Study between 1980 and 2006, with 4,384 incident invasive breast cancers identified by 2008 (median follow-up, 24 years; range, 1-28 years). In a random one third sample of women, we also compared the performance of these models with predictions from the Rosner-Colditz model estimated from the remaining participants.

RESULTS

Both the Gail and Tyrer-Cuzick models showed evidence of miscalibration (Hosmer-Lemeshow P < 0.001 for each) with notable (P < 0.01) overprediction in higher-risk women (2-year risk above about 1%) and underprediction in lower-risk women (risk below about 0.25%). The Tyrer-Cuzick model had slightly higher C-statistics both overall (P < 0.001) and in age-specific comparisons than the Gail model (overall C, 0.63 for Tyrer-Cuzick vs. 0.61 for the Gail model). Evaluation of net reclassification did not favor either model. In the one third sample, the Rosner-Colditz model had better calibration and discrimination than the other two models. All models had C-statistics <0.60 among women ages ≥70 years.

CONCLUSIONS

Both the Gail and Tyrer-Cuzick models had some ability to discriminate breast cancer cases and noncases, but have limitations in their model fit.

IMPACT

Refinements may be needed to questionnaire-based approaches to predict breast cancer in older and higher-risk women.

Breast cancer risk prediction in women aged 35-50 years: impact of including sex hormone concentrations in the Gail model

Background

Models that accurately predict risk of breast cancer are needed to help younger women make decisions about when to begin screening. Premenopausal concentrations of circulating anti-Müllerian hormone (AMH), a biomarker of ovarian reserve, and testosterone have been positively associated with breast cancer risk in prospective studies. We assessed whether adding AMH and/or testosterone to the Gail model improves its prediction performance for women aged 35–50.

Methods

In a nested case-control study including ten prospective cohorts (1762 invasive cases/1890 matched controls) with pre-diagnostic serum/plasma samples, we estimated relative risks (RR) for the biomarkers and Gail risk factors using conditional logistic regression and random-effects meta-analysis. Absolute risk models were developed using these RR estimates, attributable risk fractions calculated using the distributions of the risk factors in the cases from the consortium, and population-based incidence and mortality rates. The area under the receiver operating characteristic curve (AUC) was used to compare the discriminatory accuracy of the models with and without biomarkers.

Results

The AUC for invasive breast cancer including only the Gail risk factor variables was 55.3 (95% CI 53.4, 57.1). The AUC increased moderately with the addition of AMH (AUC 57.6, 95% CI 55.7, 59.5), testosterone (AUC 56.2, 95% CI 54.4, 58.1), or both (AUC 58.1, 95% CI 56.2, 59.9). The largest AUC improvement (4.0) was among women without a family history of breast cancer.

Conclusions

AMH and testosterone moderately increase the discriminatory accuracy of the Gail model among women aged 35–50. We observed the largest AUC increase for women without a family history of breast cancer, the group that would benefit most from improved risk prediction because early screening is already recommended for women with a family history.

Electronic supplementary material

The online version of this article (10.1186/s13058-019-1126-z) contains supplementary material, which is available to authorized users.

Modeling risk assessment for breast cancer in symptomatic women: a Saudi Arabian study

BACKGROUND

Despite the continuing increase in the breast cancer incidence rate among Saudi Arabian women, no breast cancer risk-prediction model is available in this population. The aim of this research was to develop a risk-assessment tool to distinguish between high risk and low risk of breast cancer in a sample of Saudi women who were screened for breast cancer.

METHODS

A retrospective chart review was conducted on symptomatic women who underwent breast mass biopsies between September 8, 2015 and November 8, 2017 at King Abdulaziz Medical City, Riyadh, Saudi Arabia.

RESULTS

A total of 404 (63.8%) malignant breast biopsies and 229 (36.2%) benign breast biopsies were analyzed. Women ≥40 years old (aOR: 6.202, CI 3.497-11.001, P=0.001), hormone-replacement therapy (aOR 24.365, 95% CI 8.606-68.987, P=0.001), postmenopausal (aOR 3.058, 95% CI 1.861-5.024, P=0.001), and with a family history of breast cancer (aOR 2.307, 95% CI 1.142-4.658, P=0.020) were independently associated with an increased risk of breast cancer. This model showed an acceptable fit and had area under the receiver-operating characteristic curve of 0.877 (95% CI 0.851-0.903), with optimism-corrected area under the curve of 0.865.

CONCLUSION

The prediction model developed in this study has a high ability in predicting increased breast cancer risk in our facility. Combining information on age, use of hormone therapy, postmenopausal status, and family history of breast cancer improved the degree of discriminatory accuracy of breast cancer prediction. Our risk model may assist in initiating population-screening programs and prompt clinical decision making to manage cases and prevent unfavorable outcomes.

Risk prediction for estrogen receptor-specific breast cancers in two large prospective cohorts

BACKGROUND

Few published breast cancer (BC) risk prediction models consider the heterogeneity of predictor variables between estrogen-receptor positive (ER+) and negative (ER-) tumors. Using data from two large cohorts, we examined whether modeling this heterogeneity could improve prediction.

METHODS

We built two models, for ER+ (ModelER+) and ER- tumors (ModelER-), respectively, in 281,330 women (51% postmenopausal at recruitment) from the European Prospective Investigation into Cancer and Nutrition cohort. Discrimination (C-statistic) and calibration (the agreement between predicted and observed tumor risks) were assessed both internally and externally in 82,319 postmenopausal women from the Women's Health Initiative study. We performed decision curve analysis to compare ModelER+ and the Gail model (ModelGail) regarding their applicability in risk assessment for chemoprevention.

RESULTS

Parity, number of full-term pregnancies, age at first full-term pregnancy and body height were only associated with ER+ tumors. Menopausal status, age at menarche and at menopause, hormone replacement therapy, postmenopausal body mass index, and alcohol intake were homogeneously associated with ER+ and ER- tumors. Internal validation yielded a C-statistic of 0.64 for ModelER+ and 0.59 for ModelER-. External validation reduced the C-statistic of ModelER+ (0.59) and ModelGail (0.57). In external evaluation of calibration, ModelER+ outperformed the ModelGail: the former led to a 9% overestimation of the risk of ER+ tumors, while the latter yielded a 22% underestimation of the overall BC risk. Compared with the treat-all strategy, ModelER+ produced equal or higher net benefits irrespective of the benefit-to-harm ratio of chemoprevention, while ModelGail did not produce higher net benefits unless the benefit-to-harm ratio was below 50. The clinical applicability, i.e. the area defined by the net benefit curve and the treat-all and treat-none strategies, was 12.7 × 10- 6 for ModelER+ and 3.0 × 10- 6 for ModelGail.

CONCLUSIONS

Modeling heterogeneous epidemiological risk factors might yield little improvement in BC risk prediction. Nevertheless, a model specifically predictive of ER+ tumor risk could be more applicable than an omnibus model in risk assessment for chemoprevention.

Estimating Risk of Breast Cancer Occurrences at Different Ages: Application of Survival Techniques

Background: Awareness is the primary means to control breast cancer occurrence. The purpose of the present work is to study the risk of breast cancer occurrence in different age group, for the study area, Assam, India, by means of survival analysis techniques. Methods: Survival and hazard functions are key concepts in survival analysis for describing the distribution of event times. In the present research a new individialized model has been proposed for cumulative hazard function, taking gamma probability distribution as probability distribution of breast cancer occurrences. Kaplan Meier Survival method has been applied to find out the probability of diseases occurrence in the early menarche and late menarche group. The data used for implementation were collected from the Record Department of a prime local cancer institute, for the period 2010-2012. The information for the risk factor age at menarche were collected from the patients registered during August 2011 to February 2012. Results: The study reveals that in the study area, cumulative hazard of the women belonging to 35 to 50 years is higher than the early and late aged women. The cumulative hazard plot with shape parameter 0.5, 1 and 10 shows that cumulative risk for early aged women are greater than the late age women but when this values is increased from 10, the opposite trend is observed. Further, the median age of disease occurrence among early menarche group is 52 years and for late menarche it is 54 years. Conclusion: The model developed could successfully point out the age group for women lying at higher risk of breast cancer occurrence. Additionally the important risk factor, age at menarche, was effectively applied to supplement to this calculation. It is hoped that practical use of this method would enhance not only awareness but also early detection of the said disease.

Review of non-clinical risk models to aid prevention of breast cancer

A disease risk model is a statistical method which assesses the probability that an individual will develop one or more diseases within a stated period of time. Such models take into account the presence or absence of specific epidemiological risk factors associated with the disease and thereby potentially identify individuals at higher risk. Such models are currently used clinically to identify people at higher risk, including identifying women who are at increased risk of developing breast cancer. Many genetic and non-genetic breast cancer risk models have been developed previously. We have evaluated existing non-genetic/non-clinical models for breast cancer that incorporate modifiable risk factors. This review focuses on risk models that can be used by women themselves in the community in the absence of clinical risk factors characterization. The inclusion of modifiable factors in these models means that they can be used to improve primary prevention and health education pertinent for breast cancer. Literature searches were conducted using PubMed, ScienceDirect and the Cochrane Database of Systematic Reviews. Fourteen studies were eligible for review with sample sizes ranging from 654 to 248,407 participants. All models reviewed had acceptable calibration measures, with expected/observed (E/O) ratios ranging from 0.79 to 1.17. However, discrimination measures were variable across studies with concordance statistics (C-statistics) ranging from 0.56 to 0.89. We conclude that breast cancer risk models that include modifiable risk factors have been well calibrated but have less ability to discriminate. The latter may be a consequence of the omission of some significant risk factors in the models or from applying models to studies with limited sample sizes. More importantly, external validation is missing for most of the models. Generalization across models is also problematic as some variables may not be considered applicable to some populations and each model performance is conditioned by particular population characteristics. In conclusion, it is clear that there is still a need to develop a more reliable model for estimating breast cancer risk which has a good calibration, ability to accurately discriminate high risk and with better generalizability across populations.

Comparison of Logistic Regression and Bayesian Networks for Risk Prediction of Breast Cancer Recurrence

Purpose. For individualized follow-up, accurate prediction of locoregional recurrence (LRR) and second primary (SP) breast cancer risk is required. Current prediction models employ regression, but with large data sets, machine-learning techniques such as Bayesian Networks (BNs) may be better alternatives. In this study, logistic regression was compared with different BNs, built with network classifiers and constraint- and score-based algorithms. Methods. Women diagnosed with early breast cancer between 2003 and 2006 were selected from the Netherlands Cancer Registry (NCR) ( N = 37,320). BN structures were developed using 1) Bayesian network classifiers, 2) correlation coefficients with different cutoffs, 3) constraint-based learning algorithms, and 4) score-based learning algorithms. The different models were compared with logistic regression using the area under the receiver operating characteristic curve, an external validation set obtained from the NCR from 2007 and 2008 ( N = 12,308), and subgroup analyses for a high- and low-risk group. Results. The BNs with the most links showed the best performance in both LRR and SP prediction (c-statistic of 0.76 for LRR and 0.69 for SP). In the external validation, logistic regression generally outperformed the BNs in both SP and LRR (c-statistic of 0.71 for LRR and 0.64 for SP). The differences were nonetheless small. Although logistic regression performed best on most parts of the subgroup analysis, BNs outperformed regression with respect to average risk for SP prediction in low- and high-risk groups. Conclusions. Although estimates of regression coefficients depend on other independent variables, there is no assumed dependence relationship between coefficient estimators and the change in value of other variables as in the case of BNs. Nonetheless, this analysis suggests that regression is still more accurate or at least as accurate as BNs for risk estimation for both LRRs and SP tumors.

Antidepressant drug-specific prediction of depression treatment outcomes from genetic and clinical variables

Individuals with depression differ substantially in their response to treatment with antidepressants. Specific predictors explain only a small proportion of these differences. To meaningfully predict who will respond to which antidepressant, it may be necessary to combine multiple biomarkers and clinical variables. Using statistical learning on common genetic variants and clinical information in a training sample of 280 individuals randomly allocated to 12-week treatment with antidepressants escitalopram or nortriptyline, we derived models to predict remission with each antidepressant drug. We tested the reproducibility of each prediction in a validation set of 150 participants not used in model derivation. An elastic net logistic model based on eleven genetic and six clinical variables predicted remission with escitalopram in the validation dataset with area under the curve 0.77 (95%CI; 0.66-0.88; p = 0.004), explaining approximately 30% of variance in who achieves remission. A model derived from 20 genetic variables predicted remission with nortriptyline in the validation dataset with an area under the curve 0.77 (95%CI; 0.65-0.90; p < 0.001), explaining approximately 36% of variance in who achieves remission. The predictive models were antidepressant drug-specific. Validated drug-specific predictions suggest that a relatively small number of genetic and clinical variables can help select treatment between escitalopram and nortriptyline.

Assessment of performance of the Gail model for predicting breast cancer risk: a systematic review and meta-analysis with trial sequential analysis

BACKGROUND

The Gail model has been widely used and validated with conflicting results. The current study aims to evaluate the performance of different versions of the Gail model by means of systematic review and meta-analysis with trial sequential analysis (TSA).

METHODS

Three systematic review and meta-analyses were conducted. Pooled expected-to-observed (E/O) ratio and pooled area under the curve (AUC) were calculated using the DerSimonian and Laird random-effects model. Pooled sensitivity, specificity and diagnostic odds ratio were evaluated by bivariate mixed-effects model. TSA was also conducted to determine whether the evidence was sufficient and conclusive.

RESULTS

Gail model 1 accurately predicted breast cancer risk in American women (pooled E/O = 1.03; 95% CI 0.76-1.40). The pooled E/O ratios of Caucasian-American Gail model 2 in American, European and Asian women were 0.98 (95% CI 0.91-1.06), 1.07 (95% CI 0.66-1.74) and 2.29 (95% CI 1.95-2.68), respectively. Additionally, Asian-American Gail model 2 overestimated the risk for Asian women about two times (pooled E/O = 1.82; 95% CI 1.31-2.51). TSA showed that evidence in Asian women was sufficient; nonetheless, the results in American and European women need further verification. The pooled AUCs for Gail model 1 in American and European women and Asian females were 0.55 (95% CI 0.53-0.56) and 0.75 (95% CI 0.63-0.88), respectively, and the pooled AUCs of Caucasian-American Gail model 2 for American, Asian and European females were 0.61 (95% CI 0.59-0.63), 0.55 (95% CI 0.52-0.58) and 0.58 (95% CI 0.55-0.62), respectively. The pooled sensitivity, specificity and diagnostic odds ratio of Gail model 1 were 0.63 (95% CI 0.27-0.89), 0.91 (95% CI 0.87-0.94) and 17.38 (95% CI 2.66-113.70), respectively, and the corresponding indexes of Gail model 2 were 0.35 (95% CI 0.17-0.59), 0.86 (95% CI 0.76-0.92) and 3.38 (95% CI 1.40-8.17), respectively.

CONCLUSIONS

The Gail model was more accurate in predicting the incidence of breast cancer in American and European females, while far less useful for individual-level risk prediction. Moreover, the Gail model may overestimate the risk in Asian women and the results were further validated by TSA, which is an addition to the three previous systematic review and meta-analyses.

TRIAL REGISTRATION

PROSPERO CRD42016047215 .

Using New and Emerging Technologies to Identify and Respond to Suicidality Among Help-Seeking Young People: A Cross-Sectional Study

BACKGROUND

Suicidal thoughts are common among young people presenting to face-to-face and online mental health services. The early detection and rapid response to these suicidal thoughts and other suicidal behaviors is a priority for suicide prevention and early intervention efforts internationally. Establishing how best to use new and emerging technologies to facilitate person-centered systematic assessment and early intervention for suicidality is crucial to these efforts.

OBJECTIVE

The aim of this study was to examine the use of a suicidality escalation protocol to respond to suicidality among help-seeking young people.

METHODS

A total of 232 young people in the age range of 16-25 years were recruited from either a primary mental health care service or online in the community. Each young person used the Synergy Online System and completed an initial clinical assessment online before their face-to-face or online clinical appointment. A suicidality escalation protocol was used to identify and respond to current and previous suicidal thoughts and behaviors.

RESULTS

A total of 153 young people (66%, 153/232) reported some degree of suicidality and were provided with a real-time alert online. Further levels of escalation (email or phone contact and clinical review) were initiated for the 35 young people (15%, 35/232) reporting high suicidality. Higher levels of psychological distress (P<.001) and a current alcohol or substance use problem (P=.02) predicted any level of suicidality compared with no suicidality. Furthermore, predictors of high suicidality compared with low suicidality were higher levels of psychological distress (P=.01), psychosis-like symptoms in the last 12 months (P=.01), a previous mental health problem (P=.01), and a history of suicide planning or attempts (P=.001).

CONCLUSIONS

This study demonstrates the use of new and emerging technologies to facilitate the systematic assessment and detection of help-seeking young people presenting with suicidality. This protocol empowered the young person by suggesting pathways to care that were based on their current needs. The protocol also enabled an appropriate and timely response from service providers for young people reporting high suicidality that was associated with additional comorbid issues, including psychosis-like symptoms, and a history of suicide plans and attempts.

Association Between Imaging Characteristics and Different Molecular Subtypes of Breast Cancer

RATIONALE AND OBJECTIVE

Breast cancer can be divided into four major molecular subtypes based on the expression of hormone receptor (estrogen receptor and progesterone receptor), human epidermal growth factor receptor 2, HER2 status, and molecular proliferation rate (Ki67). In this study, we sought to investigate the association between breast cancer subtype and radiological findings in the Chinese population.

MATERIALS AND METHODS

Medical records of 300 consecutive invasive breast cancer patients were reviewed from the database: the Breast Imaging Reporting and Data System. The imaging characteristics of the lesions were evaluated. The molecular subtypes of breast cancer were classified into four types: luminal A, luminal B, HER2 overexpressed (HER2), and basal-like breast cancer (BLBC). Univariate and multivariate logistic regression analyses were performed to assess the association between the subtype (dependent variable) and mammography or 15 magnetic resonance imaging (MRI) indicators (independent variables).

RESULTS

Luminal A and B subtypes were commonly associated with "clustered calcification distribution," "nipple invasion," or "skin invasion" (P <0.05). The BLBC subtype was more commonly associated with "rim enhancement" and persistent inflow type enhancement in delayed phase (P <0.05). HER2 overexpressed cancers showed association with persistent enhancement in the delayed phase on MRI and "clustered calcification distribution" on mammography (P <0.05).

CONCLUSION

Certain radiological features are strongly associated with the molecular subtype and hormone receptor status of breast tumor, which are potentially useful tools in the diagnosis and subtyping of breast cancer.

Risk Prediction Models for Incident Heart Failure: A Systematic Review of Methodology and Model Performance

BACKGROUND

Numerous models predicting the risk of incident heart failure (HF) have been developed; however, evidence of their methodological rigor and reporting remains unclear. This study critically appraises the methods underpinning incident HF risk prediction models.

METHODS AND RESULTS

EMBASE and PubMed were searched for articles published between 1990 and June 2016 that reported at least 1 multivariable model for prediction of HF. Model development information, including study design, variable coding, missing data, and predictor selection, was extracted. Nineteen studies reporting 40 risk prediction models were included. Existing models have acceptable discriminative ability (C-statistics > 0.70), although only 6 models were externally validated. Candidate variable selection was based on statistical significance from a univariate screening in 11 models, whereas it was unclear in 12 models. Continuous predictors were retained in 16 models, whereas it was unclear how continuous variables were handled in 16 models. Missing values were excluded in 19 of 23 models that reported missing data, and the number of events per variable was < 10 in 13 models. Only 2 models presented recommended regression equations. There was significant heterogeneity in discriminative ability of models with respect to age (P < .001) and sample size (P = .007).

CONCLUSIONS

There is an abundance of HF risk prediction models that had sufficient discriminative ability, although few are externally validated. Methods not recommended for the conduct and reporting of risk prediction modeling were frequently used, and resulting algorithms should be applied with caution.

Using patient self-reports to study heterogeneity of treatment effects in major depressive disorder

BACKGROUNDS

Clinicians need guidance to address the heterogeneity of treatment responses of patients with major depressive disorder (MDD). While prediction schemes based on symptom clustering and biomarkers have so far not yielded results of sufficient strength to inform clinical decision-making, prediction schemes based on big data predictive analytic models might be more practically useful.

METHOD

We review evidence suggesting that prediction equations based on symptoms and other easily-assessed clinical features found in previous research to predict MDD treatment outcomes might provide a foundation for developing predictive analytic clinical decision support models that could help clinicians select optimal (personalised) MDD treatments. These methods could also be useful in targeting patient subsamples for more expensive biomarker assessments.

RESULTS

Approximately two dozen baseline variables obtained from medical records or patient reports have been found repeatedly in MDD treatment trials to predict overall treatment outcomes (i.e., intervention v. control) or differential treatment outcomes (i.e., intervention A v. intervention B). Similar evidence has been found in observational studies of MDD persistence-severity. However, no treatment studies have yet attempted to develop treatment outcome equations using the full set of these predictors. Promising preliminary empirical results coupled with recent developments in statistical methodology suggest that models could be developed to provide useful clinical decision support in personalised treatment selection. These tools could also provide a strong foundation to increase statistical power in focused studies of biomarkers and MDD heterogeneity of treatment response in subsequent controlled trials.

CONCLUSIONS

Coordinated efforts are needed to develop a protocol for systematically collecting information about established predictors of heterogeneity of MDD treatment response in large observational treatment studies, applying and refining these models in subsequent pragmatic trials, carrying out pooled secondary analyses to extract the maximum amount of information from these coordinated studies, and using this information to focus future discovery efforts in the segment of the patient population in which continued uncertainty about treatment response exists.

Methodological issues in current practice may lead to bias in the development of biomarker combinations for predicting acute kidney injury

Individual biomarkers of renal injury are only modestly predictive of acute kidney injury (AKI). Using multiple biomarkers has the potential to improve predictive capacity. In this systematic review, statistical methods of articles developing biomarker combinations to predict acute kidney injury were assessed. We identified and described three potential sources of bias (resubstitution bias, model selection bias and bias due to center differences) that may compromise the development of biomarker combinations. Fifteen studies reported developing kidney injury biomarker combinations for the prediction of AKI after cardiac surgery (8 articles), in the intensive care unit (4 articles) or other settings (3 articles). All studies were susceptible to at least one source of bias and did not account for or acknowledge the bias. Inadequate reporting often hindered our assessment of the articles. We then evaluated, when possible (7 articles), the performance of published biomarker combinations in the TRIBE-AKI cardiac surgery cohort. Predictive performance was markedly attenuated in six out of seven cases. Thus, deficiencies in analysis and reporting are avoidable and care should be taken to provide accurate estimates of risk prediction model performance. Hence, rigorous design, analysis and reporting of biomarker combination studies are essential to realizing the promise of biomarkers in clinical practice.

Variation in Mammographic Breast Density Assessments Among Radiologists in Clinical Practice: A Multicenter Observational Study

BACKGROUND

About half of the United States has legislation requiring radiology facilities to disclose mammographic breast density information to women, often with language recommending discussion of supplemental screening options for women with dense breasts.

OBJECTIVE

To examine variation in breast density assessment across radiologists in clinical practice.

DESIGN

Cross-sectional and longitudinal analyses of prospectively collected observational data.

SETTING

30 radiology facilities within the 3 breast cancer screening research centers of the Population-based Research Optimizing Screening through Personalized Regimens (PROSPR) consortium.

PARTICIPANTS

Radiologists who interpreted at least 500 screening mammograms during 2011 to 2013 (n = 83). Data on 216 783 screening mammograms from 145 123 women aged 40 to 89 years were included.

MEASUREMENTS

Mammographic breast density, as clinically recorded using the 4 Breast Imaging Reporting and Data System categories (heterogeneously dense and extremely dense categories were considered "dense" for analyses), and patient age, race, and body mass index (BMI).

RESULTS

Overall, 36.9% of mammograms were rated as showing dense breasts. Across radiologists, this percentage ranged from 6.3% to 84.5% (median, 38.7% [interquartile range, 28.9% to 50.9%]), with multivariable adjustment for patient characteristics having little effect (interquartile range, 29.9% to 50.8%). Examination of patient subgroups revealed that variation in density assessment across radiologists was pervasive in all but the most extreme patient age and BMI combinations. Among women with consecutive mammograms interpreted by different radiologists, 17.2% (5909 of 34 271) had discordant assessments of dense versus nondense status.

LIMITATION

Quantitative measures of mammographic breast density were not available for comparison.

CONCLUSION

There is wide variation in density assessment across radiologists that should be carefully considered by providers and policymakers when considering supplemental screening strategies. The likelihood of a woman being told she has dense breasts varies substantially according to which radiologist interprets her mammogram.

PRIMARY FUNDING SOURCE

National Institutes of Health.

Testing a machine-learning algorithm to predict the persistence and severity of major depressive disorder from baseline self-reports

Heterogeneity of major depressive disorder (MDD) illness course complicates clinical decision-making. Although efforts to use symptom profiles or biomarkers to develop clinically useful prognostic subtypes have had limited success, a recent report showed that machine-learning (ML) models developed from self-reports about incident episode characteristics and comorbidities among respondents with lifetime MDD in the World Health Organization World Mental Health (WMH) Surveys predicted MDD persistence, chronicity and severity with good accuracy. We report results of model validation in an independent prospective national household sample of 1056 respondents with lifetime MDD at baseline. The WMH ML models were applied to these baseline data to generate predicted outcome scores that were compared with observed scores assessed 10-12 years after baseline. ML model prediction accuracy was also compared with that of conventional logistic regression models. Area under the receiver operating characteristic curve based on ML (0.63 for high chronicity and 0.71-0.76 for the other prospective outcomes) was consistently higher than for the logistic models (0.62-0.70) despite the latter models including more predictors. A total of 34.6-38.1% of respondents with subsequent high persistence chronicity and 40.8-55.8% with the severity indicators were in the top 20% of the baseline ML-predicted risk distribution, while only 0.9% of respondents with subsequent hospitalizations and 1.5% with suicide attempts were in the lowest 20% of the ML-predicted risk distribution. These results confirm that clinically useful MDD risk-stratification models can be generated from baseline patient self-reports and that ML methods improve on conventional methods in developing such models.

Relationship of Predicted Risk of Developing Invasive Breast Cancer, as Assessed with Three Models, and Breast Cancer Mortality among Breast Cancer Patients

PURPOSE

Breast cancer risk prediction models are used to plan clinical trials and counsel women; however, relationships of predicted risks of breast cancer incidence and prognosis after breast cancer diagnosis are unknown.

METHODS

Using largely pre-diagnostic information from the Breast Cancer Surveillance Consortium (BCSC) for 37,939 invasive breast cancers (1996-2007), we estimated 5-year breast cancer risk (<1%; 1-1.66%; ≥1.67%) with three models: BCSC 1-year risk model (BCSC-1; adapted to 5-year predictions); Breast Cancer Risk Assessment Tool (BCRAT); and BCSC 5-year risk model (BCSC-5). Breast cancer-specific mortality post-diagnosis (range: 1-13 years; median: 5.4-5.6 years) was related to predicted risk of developing breast cancer using unadjusted Cox proportional hazards models, and in age-stratified (35-44; 45-54; 55-69; 70-89 years) models adjusted for continuous age, BCSC registry, calendar period, income, mode of presentation, stage and treatment. Mean age at diagnosis was 60 years.

RESULTS

Of 6,021 deaths, 2,993 (49.7%) were ascribed to breast cancer. In unadjusted case-only analyses, predicted breast cancer risk ≥1.67% versus <1.0% was associated with lower risk of breast cancer death; BCSC-1: hazard ratio (HR) = 0.82 (95% CI = 0.75-0.90); BCRAT: HR = 0.72 (95% CI = 0.65-0.81) and BCSC-5: HR = 0.84 (95% CI = 0.75-0.94). Age-stratified, adjusted models showed similar, although mostly non-significant HRs. Among women ages 55-69 years, HRs approximated 1.0. Generally, higher predicted risk was inversely related to percentages of cancers with unfavorable prognostic characteristics, especially among women 35-44 years.

CONCLUSIONS

Among cases assessed with three models, higher predicted risk of developing breast cancer was not associated with greater risk of breast cancer death; thus, these models would have limited utility in planning studies to evaluate breast cancer mortality reduction strategies. Further, when offering women counseling, it may be useful to note that high predicted risk of developing breast cancer does not imply that if cancer develops it will behave aggressively.

Bayesian joint ordinal and survival modeling for breast cancer risk assessment

We propose a joint model to analyze the structure and intensity of the association between longitudinal measurements of an ordinal marker and time to a relevant event. The longitudinal process is defined in terms of a proportional‐odds cumulative logit model. Time‐to‐event is modeled through a left‐truncated proportional‐hazards model, which incorporates information of the longitudinal marker as well as baseline covariates. Both longitudinal and survival processes are connected by means of a common vector of random effects.

General inferences are discussed under the Bayesian approach and include the posterior distribution of the probabilities associated to each longitudinal category and the assessment of the impact of the baseline covariates and the longitudinal marker on the hazard function. The flexibility provided by the joint model makes possible to dynamically estimate individual event‐free probabilities and predict future longitudinal marker values.

The model is applied to the assessment of breast cancer risk in women attending a population‐based screening program. The longitudinal ordinal marker is mammographic breast density measured with the Breast Imaging Reporting and Data System (BI‐RADS) scale in biennial screening exams. © 2016 The Authors. Statistics in Medicine Published by John Wiley & Sons Ltd.

[Breast cancer screening: On our way to the future]

Breast cancer remains a potentially lethal disease, which requires aggressive treatments and is associated with long-term consequences. Its prognosis is linked to both tumor biology and burden at diagnosis. Although treatments have allowed important improvements in prognosis over the past 20 years, breast cancer screening remains necessary. Mammographic screening allows earlier stage diagnoses and a decrease of breast cancer specific mortality. However, breast cancer screening modalities should be revised with the objective to address demonstrated limitations of mammographic screening (limited benefit, imperfect sensitivity and specificity, overdiagnoses, radiation-induced morbidity). Furthermore, both objective and perceived performances of screening procedures should be improved. Numerous large international efforts are ongoing, leading to scientific progresses that should have rapid clinical implications in this area. Among them is improvement of imaging techniques performance, development of real time diagnosis, and development of new non radiological screening techniques such as the search for circulating tumor DNA, development of biomarkers able to allow precise risk evaluation and stratified screening. As well, overtreatment is currently addressed by biomarker-based de-escalation clinical trials. These advances need to be associated with strong societal support, as well as major paradigm changes regarding the way health and cancer prevention is perceived by individuals.

Identifying a clinical signature of suicidality among patients with mood disorders: A pilot study using a machine learning approach

OBJECTIVE

A growing body of evidence has put forward clinical risk factors associated with patients with mood disorders that attempt suicide. However, what is not known is how to integrate clinical variables into a clinically useful tool in order to estimate the probability of an individual patient attempting suicide.

METHOD

A total of 144 patients with mood disorders were included. Clinical variables associated with suicide attempts among patients with mood disorders and demographic variables were used to 'train' a machine learning algorithm. The resulting algorithm was utilized in identifying novel or 'unseen' individual subjects as either suicide attempters or non-attempters. Three machine learning algorithms were implemented and evaluated.

RESULTS

All algorithms distinguished individual suicide attempters from non-attempters with prediction accuracy ranging between 65% and 72% (p<0.05). In particular, the relevance vector machine (RVM) algorithm correctly predicted 103 out of 144 subjects translating into 72% accuracy (72.1% sensitivity and 71.3% specificity) and an area under the curve of 0.77 (p<0.0001). The most relevant predictor variables in distinguishing attempters from non-attempters included previous hospitalizations for depression, a history of psychosis, cocaine dependence and post-traumatic stress disorder (PTSD) comorbidity.

CONCLUSION

Risk for suicide attempt among patients with mood disorders can be estimated at an individual subject level by incorporating both demographic and clinical variables. Future studies should examine the performance of this model in other populations and its subsequent utility in facilitating selection of interventions to prevent suicide.

The Proliferative Activity of Mammary Epithelial Cells in Normal Tissue Predicts Breast Cancer Risk in Premenopausal Women

The frequency and proliferative activity of tissue-specific stem and progenitor cells are suggested to correlate with cancer risk. In this study, we investigated the association between breast cancer risk and the frequency of mammary epithelial cells expressing p27, estrogen receptor (ER), and Ki67 in normal breast tissue. We performed a nested case-control study of 302 women (69 breast cancer cases, 233 controls) who had been initially diagnosed with benign breast disease according to the Nurses' Health Studies. Immunofluorescence for p27, ER, and Ki67 was performed on tissue microarrays constructed from benign biopsies containing normal mammary epithelium and scored by computational image analysis. We found that the frequency of Ki67(+) cells was positively associated with breast cancer risk among premenopausal women [OR = 10.1, 95% confidence interval (CI) = 2.12-48.0]. Conversely, the frequency of ER(+) or p27(+) cells was inversely, but not significantly, associated with subsequent breast cancer risk (ER(+): OR = 0.70, 95% CI, 0.33-1.50; p27(+): OR = 0.89, 95% CI, 0.45-1.75). Notably, high Ki67(+)/low p27(+) and high Ki67(+)/low ER(+) cell frequencies were significantly associated with a 5-fold higher risk of breast cancer compared with low Ki67(+)/low p27(+) and low Ki67(+)/low ER(+) cell frequencies, respectively, among premenopausal women (Ki67(hi)/p27(lo): OR = 5.08, 95% CI, 1.43-18.1; Ki67(hi)/ER(lo): OR = 4.68, 95% CI, 1.63-13.5). Taken together, our data suggest that the fraction of actively cycling cells in normal breast tissue may represent a marker for breast cancer risk assessment, which may therefore impact the frequency of screening procedures in at-risk women. Cancer Res; 76(7); 1926-34. ©2016 AACR.