Cancer Risk Prediction Research Articles

A breast cancer risk predication and classification model with ensemble learning and big data fusion

Breast cancer is a major health issue for women all over the world. Effective care and better patient outcomes depend on early identification and precise risk prediction. Ensemble learning algorithms and merging large data have recently emerged as potential ways to predict and classify breast cancer risk. This paper aims to investigate the feasibility of using ensemble learning and big data fusion for risk prediction and categorization of breast cancer. The key focus is building a reliable and precise model by combining the best features of various learning algorithms and a variety of vast and varied data sets. This research presents an Improved XGBoost Ensembling (I-XGBoost) technique with big data analytics to predict and diagnose breast cancer cells. The proposed work considers three critical identification phases to achieve greater accuracy: data pre-processing, feature extraction, and target role. The Wisconsin breast cancer diagnostic data was used for testing the proposed method. It is also compared in terms of performance with other classification methods like Decision Tree, Random Forest, Naive Bayes, K-Nearest Neighbors, Support Vector Machines, Adaboost, and XGBoost. This research aims to determine which characteristics are most beneficial in predicting cancer as malignant or benign. We show that I-XGBoost has an impressively high accuracy score of 99.84% using Spark’s Python Application Programming Interface (API). This research emphasizes the promise of ensemble learning and big data fusion for predicting and classifying breast cancer risks. By proposing a new method that draws on the best features of several models and uses the wealth of information available in big data, it adds to the current body of knowledge. Positive findings highlight the practical use of the suggested paradigm in clinical contexts.

Clinical Impact of Polygenic Risk Score for Breast Cancer Risk Prediction in 382 Individuals with Hereditary Breast and Ovarian Cancer Syndrome.

Single nucleotide polymorphisms are currently not considered in breast cancer (BC) risk predictions used in daily practice of genetic counselling and clinical management of familial BC in Germany. This study aimed to assess the clinical value of incorporating a 313-variant-based polygenic risk score (PRS) into BC risk calculations in a cohort of German women with suspected hereditary breast and ovarian cancer syndrome (HBOC). Data from 382 individuals seeking counselling for HBOC were analysed. Risk calculations were performed using the Breast and Ovarian Analysis of Disease Incidence and Carrier Estimation Algorithm with and without the inclusion of the PRS. Changes in risk predictions and their impact on clinical management were evaluated. The PRS led to changes in risk stratification based on 10-year risk calculations in 13.6% of individuals. Furthermore, the inclusion of the PRS in BC risk predictions resulted in clinically significant changes in 12.0% of cases, impacting the prevention recommendations established by the German Consortium for Hereditary Breast and Ovarian Cancer. These findings support the implementation of the PRS in genetic counselling for personalized BC risk assessment.

Critical Analysis of Risk Factors and Machine-Learning-Based Gastric Cancer Risk Prediction Models: A Systematic Review

The gastric cancer risk prediction model used for large-scale gastric cancer screening and individual risk stratification is an artificial intelligence tool that combines clinical diagnostic data with a classification algorithm. The ability to automatically make a quantitative assessment of complex clinical data contributes to increased accuracy for diagnosis with higher efficiency, significantly reducing the incidence of advanced gastric cancer. Previous studies have explored the predictive performance of gastric cancer risk prediction models, as well as the predictive factors and algorithms between each model, but have reached controversial conclusions. Thus, the performance of current machine-learning-based gastric cancer risk prediction models alongside the clinical relevance of different predictive factors needs to be evaluated to help build more efficient and feasible models in the future. In this systematic review, we summarize the current research progress related to the gastric cancer risk prediction model; discuss the predictive factors and methods used to construct the model; analyze the role of important predictive factors in gastric cancer, the preference of the selected classification algorithm, and the emphasis of evaluation criteria; and provide suggestions for the subsequent construction and improvement of the gastric cancer risk prediction model. Finally, we propose an improved approach based on the ethical issues of artificial intelligence in medicine to realize the clinical application of the gastric cancer risk prediction model in the future.

Application of machine learning techniques in the prediction of excess lifetime cancer risks of agricultural byproducts used as building and construction materials

Application of machine learning techniques in the prediction of excess lifetime cancer risks of agricultural byproducts used as building and construction materials

Spatiotemporal Mammography-based Deep Learning Model for Improved Breast Cancer Risk Prediction.

Breast cancer is one of the most prevalent cancers among women. It is the second leading cause of death in cancer-related deaths. Early detection and personalized risk assessment can reduce the mortality rate and improve survival rates. Classical risk prediction models which rely on traditional risk factors produce inconsistent results among the different populations. Thus, they are not routinely used in screening programs. Deep learning was proven to improve the results of breast cancer risk prediction. CNNs can detect risk cues from screening mammograms. However, the deep learning models utilize the spatial information of each screening mammogram independently. This study aims to further improve the risk prediction models by exploiting the spatiotemporal information in multiple screening time points. We implemented a Siamese neural network for spatiotemporal risk prediction and compared the results against CNN trained using two different time points (T1 and T2) independently. We tested our results on 191 cases, 61 of which were diagnosed with cancer. The Siamese model showed a superior AUC of 0.81 against 0.75 and 0.77 at T1 and T2 respectively. The Siamese network also exhibited higher accuracy and F1-score with values of 0.78 and 0.61 while CNNs have the same accuracy of 0.76 with an F1-score of 0.54 at T1, and 0.59 at T2. The results suggest that spatiotemporal risk prediction can be a more reliable risk assessment tool.

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

BackgroundBreast 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.PurposeTo 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.MethodsA 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.ResultsFrom 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.ConclusionsGiven 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.

Development of lung cancer risk prediction models based on F-18 FDG PET images.

We aimed to evaluate whether the degree of F-18 fluorodeoxyglucose (FDG) uptake in the lungs is associated with an increased risk of lung cancer and to develop lung cancer risk prediction models using metabolic parameters on F-18 FDG positron emission tomography (PET). We retrospectively included 795 healthy individuals who underwent F-18 FDG PET/CT scans for a health check-up. Individuals who developed lung cancer within 5years of the PET/CT scan were classified into the lung cancer group (n = 136); those who did not were classified into the control group (n = 659). The healthy individuals were then randomly assigned to either the training (n = 585) or validation sets (n = 210). Clinical factors including age, sex, body mass index (BMI), and smoking history were collected. The standardized uptake value ratio (SUVR) and metabolic heterogeneity (MH) index were obtained for the bilateral lungs. Logistic regression models including clinical factors, SUVR, and MH index were generated to quantify the probability of lung cancer development using a training set. The prediction models were validated using a validation set. The lung SUVR and lung MH index in the lung cancer group were significantly higher than in the control group (p < 0.001 and p < 0.001, respectively). In the combined prediction model 1, age, sex, BMI, smoking history, and lung SUVR were significantly associated with lung cancer development (age: OR 1.07, p < 0.001; male: OR 2.08, p = 0.015; BMI: OR 0.93, p = 0.057; current or past smoker: OR 5.60, p < 0.001; lung SUVR: OR 1.13, p < 0.001). In the combined prediction model 2, age, sex, BMI, smoking history, and lung MH index showed a significant association with lung cancer development (age: OR 1.06, p < 0.001; male: OR 1.87, p = 0.045; BMI: OR 0.93, p = 0.010; current or past smoker: OR 4.78, p < 0.001; lung MH index: OR 1.33, p < 0.001). In the validation data, combined prediction models 1 and 2 exhibited very good discrimination [area under the receiver operator curve (AUC): 0.867 and 0.901, respectively]. The metabolic parameters on F-18 FDG PET are related to an increased risk of lung cancer. Metabolic parameters can be used as biomarkers to provide information independent of the clinical parameters, related to lung cancer risk.

A risk prediction model of cervical cancer developed based on nested case-control design

Objective: To construct a cervical cancer risk prediction model based on nested case-control study design and Yinzhou Health Information Platform in Ningbo, and provide reliable reference for self-risk assessment of cervical cancer in local women. Methods: In local women aged 25-75 years old who had no history of cervical cancer registered in Yinzhou before October 31, 2018, a follow up was conducted for at least three years, the patients who developed cervical cancer during the follow up period were selected as the case group and matched with a control group at a ratio of 1∶10. The prediction indicators before the onset was used in model construction. Variables were selected by Lasso-logistic regression, the variables with non-zero β were selected to fit the logistic regression model and Bootstrap was used for internal validation. The discrimination of the model was evaluated by area under the receiver operating characteristic curve(AUROC), and the calibration was evaluated by calibration curve and Hosmer-Lemeshow test. Results: The prediction indicators included in the final model were age, smoking status, history of cervicitis, history of adenomyosis, HPV testing, and thinprep cytologic test. The AUROC calculated in the internal validation was 0.740 (95%CI:0.739-0.740), and the calibration curve was almost identical with the ideal curve, P=0.991 in Hosmer-Lemeshow test, indicating that the model discrimination and calibration were good. Conclusions: In this study, a simple and practical cervical cancer risk prediction model was developed. The model can be used in general population with strong interpretability, good discrimination and calibration in internal validation, which can provide a reference for women to assess their risk of cervical cancer.

Lung cancer risk score for ever and never smokers in China.

Most lung cancer risk prediction models were developed in European and North-American cohorts of smokers aged ≥ 55 years, while less is known about risk profiles in Asia, especially for never smokers or individuals aged < 50 years. Hence, we aimed to develop and validate a lung cancer risk estimate tool for ever and never smokers across a wide age range. Based on the China Kadoorie Biobank cohort, we first systematically selected the predictors and explored the nonlinear association of predictors with lung cancer risk using restricted cubic splines. Then, we separately developed risk prediction models to construct a lung cancer risk score (LCRS) in 159,715 ever smokers and 336,526 never smokers. The LCRS was further validated in an independent cohort over a median follow-up of 13.6 years, consisting of 14,153 never smokers and 5,890 ever smokers. A total of 13 and 9 routinely available predictors were identified for ever and never smokers, respectively. Of these predictors, cigarettes per day and quit years showed nonlinear associations with lung cancer risk (Pnon-linear < 0.001). The curve of lung cancer incidence increased rapidly above 20 cigarettes per day and then was relatively flat until approximately 30 cigarettes per day. We also observed that lung cancer risk declined sharply within the first 5 years of quitting, and then continued to decrease but at a slower rate in the subsequent years. The 6-year area under the receiver operating curve for the ever and never smokers' models were respectively 0.778 and 0.733 in the derivation cohort, and 0.774 and 0.759 in the validation cohort. In the validation cohort, the 10-year cumulative incidence of lung cancer was 0.39% and 2.57% for ever smokers with low (< 166.2) and intermediate-high LCRS (≥ 166.2), respectively. Never smokers with a high LCRS (≥ 21.2) had a higher 10-year cumulative incidence rate than those with a low LCRS (< 21.2; 1.05% vs. 0.22%). An online risk evaluation tool (LCKEY; http://ccra.njmu.edu.cn/lckey/web) was developed to facilitate the use of LCRS. The LCRS can be an effective risk assessment tool designed for ever and never smokers aged 30 to 80 years.

Deep Learning Analysis of Mammography for Breast Cancer Risk Prediction in Asian Women.

The purpose of this study was to develop a mammography-based deep learning (DL) model for predicting the risk of breast cancer in Asian women. This retrospective study included 287 examinations in 153 women in the cancer group and 736 examinations in 447 women in the negative group, obtained from the databases of two tertiary hospitals between November 2012 and March 2022. All examinations were labeled as either dense breast or nondense breast, and then randomly assigned to either training, validation, or test sets. DL models, referred to as image-level and examination-level models, were developed. Both models were trained to predict whether or not the breast would develop breast cancer with two datasets: the whole dataset and the dense-only dataset. The performance of DL models was evaluated using the accuracy, precision, sensitivity, specificity, F1 score, and area under the receiver operating characteristic curve (AUC). On a test set, performance metrics for the four scenarios were obtained: image-level model with whole dataset, image-level model with dense-only dataset, examination-level model with whole dataset, and examination-level model with dense-only dataset with AUCs of 0.71, 0.75, 0.66, and 0.67, respectively. Our DL models using mammograms have the potential to predict breast cancer risk in Asian women.

How do women at increased risk of breast cancer make sense of their risk? An interpretative phenomenological analysis.

Offering breast cancer risk prediction for all women of screening age is being considered globally. For women who have received a clinically derived estimate, risk appraisals are often inaccurate. This study aimed to gain an in-depth understanding of women's lived experiences of receiving an increased breast cancer risk. One-to-one semi-structured telephone interviews. Eight women informed that they were at a 10-year above-average (moderate) or high risk in a breast cancer risk study (BC-Predict) were interviewed about their views on breast cancer, personal breast cancer risk and risk prevention. Interviews lasted between 40 and 70 min. Data were analysed using Interpretative Phenomenological Analysis. Four themes were generated: (i) encounters with breast cancer and perceived personal significance, where the nature of women's lived experiences of others with breast cancer impacted their views on the significance of the disease, (ii) 'It's random really': difficulty in seeking causal attributions, where women encountered contradictions and confusion in attributing causes to breast cancer, (iii) believing versus identifying with a clinically-derived breast cancer risk, where personal risk appraisals and expectations influenced women's ability to internalize their clinically derived risk and pursue preventative action and (iv) perceived utility of breast cancer risk notification, where women reflected on the usefulness of knowing their risk. Providing (numerical) risk estimates appear to have little impact on stable yet internally contradictory beliefs about breast cancer risk. Given this, discussions with healthcare professionals are needed to help women form more accurate appraisals and make informed decisions.

Analisa Gambar X-Ray Mammography dengan Convolution Neural Network pada Deep Learning dengan Arsitektur Resnet

Cancer is a disease that occurs when cells in the body undergo changes and grow uncontrollably. Breast cancer is one of the common types of cancer that affects women worldwide. Early detection of breast cancer is crucial to improve the survival rate. Mammography is a medical imaging method used for the early detection of breast cancer. In this context, deep learning technology and computerized classifiers, such as Convolutional Neural Network (CNN) with the Resnet model, have been used for the analysis and prediction of mammography images with promising results. Previous studies have shown high accuracy in classifying breast masses as benign or malignant using CNN and Resnet. Furthermore, CNN has also been employed for the classification of malignant and benign breast cancer, prediction of breast cancer risk, as well as detection and classification of breast masses with satisfactory accuracy rates. The use of deep learning in medical image analysis, including mammograms and X-ray images, has proven to be an effective tool in improving cancer diagnosis and treatment. The dataset used consisted of 322 images divided into 7 classes. After testing, an accuracy of 72% was achieved with a 90:10 ratio of test data to training data, along with the corresponding confusion matrix values. Therefore, it can be concluded that the Resnet method is capable of identifying breast cancer.

Construction and evaluation of the functional polygenic risk score for gastric cancer in a prospective cohort of the European population.

A polygenic risk score (PRS) derived from 112 single-nucleotide polymorphisms (SNPs) for gastric cancer has been reported in Chinese populations (PRS-112). However, its performance in other populations is unknown. A functional PRS (fPRS) using functional SNPs (fSNPs) may improve the generalizability of the PRS across populations with distinct ethnicities. We performed functional annotations on SNPs in strong linkage disequilibrium (LD) with the 112 previously reported SNPs to identify fSNPs that affect protein-coding or transcriptional regulation. Subsequently, we constructed an fPRS based on the fSNPs by using the LDpred2-infinitesimal model and then analyzed the performance of the PRS-112 and fPRS in the risk prediction of gastric cancer in 457,521 European participants of the UK Biobank cohort. Finally, the performance of the fPRS in combination with lifestyle factors were evaluated in predicting the risk of gastric cancer. During 4,582,045 person-years of follow-up with a total of 623 incident gastric cancer cases, we found no significant association between the PRS-112 and gastric cancer risk in the European population (hazard ratio [HR] = 1.00 [95% confidence interval (CI) 0.93-1.09], P = 0.846). We identified 125 fSNPs, including seven deleterious protein-coding SNPs and 118 regulatory non-coding SNPs, and used them to constructed the fPRS-125. Our result showed that the fPRS-125 was significantly associated with gastric cancer risk (HR = 1.11 [95% CI, 1.03-1.20], P = 0.009). Compared to participants with a low fPRS-125 (bottom quintile), those with a high fPRS-125 (top quintile) had a higher risk of incident gastric cancer (HR = 1.43 [95% CI, 1.12-1.84], P = 0.005). Moreover, we observed that participants with both an unfavorable lifestyle and a high genetic risk had the highest risk of incident gastric cancer (HR = 4.99 [95% CI, 1.55-16.10], P = 0.007) compared to those with both a favorable lifestyle and a low genetic risk. These results indicate that the fPRS-125 derived from fSNPs may act as an indicator to measure the genetic risk of gastric cancer in the European population.

Development and Validation of Risk Prediction Models for Colorectal Cancer in Patients with Symptoms.

We aimed to develop and validate prediction models incorporating demographics, clinical features, and a weighted genetic risk score (wGRS) for individual prediction of colorectal cancer (CRC) risk in patients with gastroenterological symptoms. Prediction models were developed with internal validation [CRC Cases: n = 1686/Controls: n = 963]. Candidate predictors included age, sex, BMI, wGRS, family history, and symptoms (changes in bowel habits, rectal bleeding, weight loss, anaemia, abdominal pain). The baseline model included all the non-genetic predictors. Models A (baseline model + wGRS) and B (baseline model) were developed based on LASSO regression to select predictors. Models C (baseline model + wGRS) and D (baseline model) were built using all variables. Models' calibration and discrimination were evaluated through the Hosmer-Lemeshow test (calibration curves were plotted) and C-statistics (corrected based on 1000 bootstrapping). The models' prediction performance was: model A (corrected C-statistic = 0.765); model B (corrected C-statistic = 0.753); model C (corrected C-statistic = 0.764); and model D (corrected C-statistic = 0.752). Models A and C, that integrated wGRS with demographic and clinical predictors, had a statistically significant improved prediction performance. Our findings suggest that future application of genetic predictors holds significant promise, which could enhance CRC risk prediction. Therefore, further investigation through model external validation and clinical impact is merited.

Biomarkers in Breast Cancer: An Old Story with a New End.

Breast cancer is the second most frequent cancer in the world. It is a heterogeneous disease and the leading cause of cancer mortality in women. Advances in molecular technologies allowed for the identification of new and more specifics biomarkers for breast cancer diagnosis, prognosis, and risk prediction, enabling personalized treatments, improving therapy, and preventing overtreatment, undertreatment, and incorrect treatment. Several breast cancer biomarkers have been identified and, along with traditional biomarkers, they can assist physicians throughout treatment plan and increase therapy success. Despite the need of more data to improve specificity and determine the real clinical utility of some biomarkers, others are already established and can be used as a guide to make treatment decisions. In this review, we summarize the available traditional, novel, and potential biomarkers while also including gene expression profiles, breast cancer single-cell and polyploid giant cancer cells. We hope to help physicians understand tumor specific characteristics and support decision-making in patient-personalized clinical management, consequently improving treatment outcome.

Regulating cancer risk prediction: legal considerations and stakeholder perspectives on the Canadian context

Risk prediction models hold great promise to reduce the impact of cancer in society through advanced warning of risk and improved preventative modalities. These models are evolving and becoming more complex, increasingly integrating genetic screening data and polygenic risk scores as well as calculating risk for multiple types of a disease. However, unclear regulatory compliance requirements applicable to these models raise significant legal uncertainty and new questions about the regulation of medical devices. This paper aims to address these novel regulatory questions by presenting an initial assessment of the legal status likely applicable to risk prediction models in Canada, using the CanRisk tool for breast and ovarian cancer as an exemplar. Legal analysis is supplemented with qualitative perspectives from expert stakeholders regarding the accessibility and compliance challenges of the Canadian regulatory framework. While the paper focuses on the Canadian context, it also refers to European and U.S. regulations in this domain to contrast them. Legal analysis and stakeholder perspectives highlight the need to clarify and update the Canadian regulatory framework for Software as a Medical Device as it applies to risk prediction models. Findings demonstrate how normative guidance perceived as convoluted, contradictory or overly burdensome can discourage innovation, compliance, and ultimately, implementation. This contribution aims to initiate discussion about a more optimal legal framework for risk prediction models as they continue to evolve and are increasingly integrated into landscape for public health.

Analyzing Big EHR Data—Optimal Cox Regression Subsampling Procedure with Rare Events

Massive sized survival datasets become increasingly prevalent with the development of the healthcare industry, and pose computational challenges unprecedented in traditional survival analysis use cases. In this work we analyze the UK-biobank colorectal cancer data with genetic and environmental risk factors, including a time-dependent coefficient, which transforms the dataset into “pseudo-observation” form, thus, critically inflating its size. A popular way for coping with massive datasets is downsampling them, such that the computational resources can be afforded by the researcher. Cox regression has remained one of the most popular statistical models for the analysis of survival data to-date. This work addresses the settings of right censored and possibly left-truncated data with rare events, such that the observed failure times constitute only a small portion of the overall sample. We propose Cox regression subsampling-based estimators that approximate their full-data partial-likelihood-based counterparts, by assigning optimal sampling probabilities to censored observations, and including all observed failures in the analysis. The suggested methodology is applied on the UK-biobank for building a colorectal cancer risk-prediction model, while reducing the computation time and memory requirements. Asymptotic properties of the proposed estimators are established under suitable regularity conditions, and simulation studies are carried out to evaluate their finite sample performance. Supplementary materials for this article are available online.

3 QUESTIONS ON . . . Using Artificial Intelligence in Breast Cancer Risk Prediction

Celine Vachon, PhD: Celine Vachon, PhDA rtificial intelligence (AI) technology is making headlines in many areas of research and sectors of business. Now, epidemiologists and others who study cancer screening methodology say they've found new evidence it could improve breast cancer detection efforts, too. In a study published in the Journal of Clinical Oncology (JCO), researchers found that using an AI algorithm designed to improve breast cancer detection on mammograms could improve breast cancer risk prediction, including interval and advanced cancers—particularly when combined with consideration of breast density (2023; doi: 10.1200/JCO.22.01153). “This adds further evidence to support computer-aided imaging assessments and will help motivate prospective studies to disentangle improved earlier diagnosis from an improved risk model for subsequent cancer,” Paul H. Frankel, PhD, JCO Associate Editor, wrote in a relevance statement published with the article. “Such prospective studies will also provide estimates that can more easily guide clinical practice.” In an interview with Oncology Times, the study's lead author, Celine Vachon, PhD, Professor and Consultant in the Division of Epidemiology within the Department of Quantitative Health Sciences at Mayo Clinic, shared more about the work. 1 What were the key findings from this research? “We evaluated whether the AI measure assessed on mammograms taken years prior to the cancer (2-5.5 years prior to the cancer diagnosis) contributed additional information to breast density for risk of breast cancer—in particular interval and advanced cancers. The AI algorithm was developed to be used on all women, not just those with dense breasts. It was run on the mammograms from women in our clinical practices who did and did not develop breast cancer. “AI score assessed on mammograms 2-5.5 years before cancer diagnosis was associated with invasive breast cancer, and similarly predictive of interval and advanced cancers in dense and non-dense breasts. These findings imply that the AI score is robust for long-term risk prediction by severity of cancer and independent of breast density for predicting cancer. “Furthermore, the AI score improved prediction of all cancer types in models with breast density measures. AI imaging algorithms coupled with breast density independently contribute to long-term risk prediction of invasive breast cancers, including advanced cancers.” 2 What are the implications of these findings and how might they improve real-world screening efforts? “Data from our study and others suggests that AI algorithms will help improve both breast cancer detection and breast cancer risk prediction, including interval cancers and advanced cancers. These will likely help better identify women for supplemental screening and, potentially, for identification of women who would benefit from early interventions. “The AI algorithm we used is commercially available. And there are several similar AI algorithms from multiple vendors available that likely would provide similar findings. However, it will be important to understand how these algorithms work among women from various races, BMI (those with obesity, for example), and how they contribute to factors already known for risk prediction (such as genetic factors). Also, these algorithms need to be able to identify prognostically poor cancers, not just DCIS or early cancers that may not progress. This research is underway by many groups to inform how best to use these algorithms in the clinical setting.” 3 What is the next step of this work? “We are pursuing additional research to understand the imaging factors that determine the high-risk score, including calcifications. We are also examining other AI models, in particular those designed for longer-term risk, and how the combination of these algorithms with genetic factors and breast density can better inform future risk. The bottom line that practicing oncologists and cancer care providers need to know now is that this research confirms the importance of both breast density and AI models to predict invasive breast cancer, including both interval and advanced cancers. Our data suggest that this AI algorithm is not only relevant at the time of cancer [for detection], but also 2-5.5 years prior to the breast cancer.”

Novel predictions of invasive breast cancer risk in mammography screening cohorts.

Mammography screening programs are aimed at reducing mortality due to breast cancer by detecting tumors at an early stage. There is currently interest in moving away from the age-based screening programs, and toward personalized screening based on individual risk factors. To accomplish this, risk prediction models for breast cancer are needed to determine who should be screened, and when. We develop a novel approach using a (random effects) continuous growth model, which we apply to a large population-based, Swedish screening cohort. Unlike existing breast cancer prediction models, this approach explicitly incorporates each woman's individual screening visits in the prediction. It jointly models invasive breast cancer tumor onset, tumor growth rate, symptomatic detection rate, and screening sensitivity. In addition to predicting the overall risk of invasive breast cancer, this model can make separate predictions regarding specific tumor sizes, and the mode of detection (eg, detected at screening, or through symptoms between screenings). It can also predict how these risks change depending on whether or not a woman will attend her next screening. In our study, we predict, given a future diagnosis, that the probability of having a tumor less than (as opposed to greater than) 10-mm diameter, at detection, will be, on average, 2.6 times higher if a woman in the cohort attends their next screening. This indicates that the model can be used to evaluate the short-term benefit of screening attendance, at an individual level.

Development of a Breast Cancer Risk Prediction Model Incorporating Polygenic Risk Scores and Nongenetic Risk Factors for Korean Women.

To develop a breast cancer prediction model for Korean women using published polygenic risk scores (PRS) combined with nongenetic risk factors (NGRF). Thirteen PRS models generated from single or multiple combinations of the Asian and European PRSs were evaluated among 20,434 Korean women. The AUC and increase in OR per SD were compared for each PRS. The PRSs with the highest predictive power were combined with NGRFs; then, an integrated prediction model was established using the Individualized Coherent Absolute Risk Estimation (iCARE) tool. The absolute breast cancer risk was stratified for 18,142 women with available follow-up data. PRS38_ASN+PRS190_EB, a combination of Asian and European PRSs, had the highest AUC (0.621) among PRSs, with an OR per SD increase of 1.45 (95% confidence interval: 1.31-1.61). Compared with the average risk group (35%-65%), women in the top 5% had a 2.5-fold higher risk of breast cancer. Incorporating NGRFs yielded a modest increase in the AUC of women ages >50 years. For PRS38_ASN+PRS190_EB+NGRF, the average absolute risk was 5.06%. The lifetime absolute risk at age 80 years for women in the top 5% was 9.93%, whereas that of women in the lowest 5% was 2.22%. Women at higher risks were more sensitive to NGRF incorporation. Combined Asian and European PRSs were predictive of breast cancer in Korean women. Our findings support the use of these models for personalized screening and prevention of breast cancer. Our study provides insights into genetic susceptibility and NGRFs for predicting breast cancer in Korean women.