Breast cancer risk prediction combining a convolutional neural network-based mammographic evaluation with clinical factors.

PURPOSE: Deep learning techniques, including convolutional neural networks (CNN), have the potential to improve breast cancer risk prediction compared to traditional risk models. We assessed whether combining a CNN-based mammographic evaluation with clinical factors in the Breast Cancer Surveillance Consortium (BCSC) model improved risk prediction. METHODS: We conducted a retrospective cohort study among 23,467 women, age 35-74, undergoing screening mammography (2014-2018). We extracted electronic health record (EHR) data on risk factors. We identified 121 women who subsequently developed invasive breast cancer at least 1 year after the baseline mammogram. Mammograms were analyzed with a pixel-wise mammographic evaluation using CNN architecture. We used logistic regression models with breast cancer incidence as the outcome and predictors including clinical factors only (BCSC model) or combined with CNN risk score (hybrid model). We compared model prediction performance via area under the receiver operating characteristics curves (AUCs). RESULTS: Mean age was 55.9 years (SD, 9.5) with 9.3% non-Hispanic Black and 36% Hispanic. Our hybrid model did not significantly improve risk prediction compared to the BCSC model (AUC of 0.654 vs 0.624, respectively, p = 0.063). In subgroup analyses, the hybrid model outperformed the BCSC model among non-Hispanic Blacks (AUC 0.845 vs. 0.589; p = 0.026) and Hispanics (AUC 0.650 vs 0.595; p = 0.049). CONCLUSION: We aimed to develop an efficient breast cancer risk assessment method using CNN risk score and clinical factors from the EHR. With future validation in a larger cohort, our CNN model combined with clinical factors may help predict breast cancer risk in a cohort of racially/ethnically diverse women undergoing screening.

A systematic review of effective strategies for chronic disease management in humanitarian settings; opportunities and challenges.

Large number of people with non-communicable diseases (NCDs) face barriers to adequate healthcare in humanitarian settings. We conducted a systematic literature review in MEDLINE/PubMed, Web of Science, EMBASE/DARE, Cochrane, and grey literature from 1990 to 2021 to evaluate effective strategies in addressing NCDs (diabetes, cardiovascular diseases, COPD, cancer) in humanitarian settings. From 2793 articles, 2652 were eliminated through title/abstract screening; 141 articles were reviewed in full; 93 were eliminated for not meeting full criteria. Remaining 48 articles were reviewed qualitatively to assess populations, settings, interventions, outcome, and efficacy and effectiveness; 38 studies addressed treatments, 9 prevention, and 7 epidemiology. Prevention studies broadly addressed capacity-building. Treatment and epidemiology studies largely addressed hypertension and diabetes. Interventions included web-based/mobile health strategies, pharmacy-level interventions, portable imaging, and capacity building including physical clinics, staff training, forging collaborations, guideline development, point-of-care labs, health promotion activities, EMR, and monitoring interventions. Collaboration between academia and implementing agencies was limited. Models of care were largely not well-described and varied between studies due to contextual constraints. Barriers to interventions included financial, logistical, organizational, sociocultural, and security. Cancer care is significantly understudied. Simplified care models adapted to contexts and program evaluations of implemented strategies could address gaps in applied research. Inherent challenges in humanitarian settings pose unavoidable perils to evidence generation which requires a shift in research mindset to match aspirations with practicality, research collaborations at the inception of projects, reworking of desired conventional level of research evidence considering resource-intense constraints (HR, time, cost), and adapted research tools, methods, and procedures.

Patient, primary care provider, and stakeholder perspectives on mammography screening frequency: lessons learned from a qualitative study.

BACKGROUND: U.S. professional organizations have provided conflicting recommendations on annual vs. biennial mammography screening. Potential harms of more frequent screening include increased anxiety and costs of false positive results, including unnecessary breast biopsies and overdiagnosis. OBJECTIVE: To characterize current practices and beliefs surrounding mammography screening frequency and perspectives on using risk-based screening to inform screening intervals. DESIGN: Semi-structured interviews informed by the Consolidated Framework for Implementation Research (CFIR). PARTICIPANTS: Patients, primary care providers (PCPs), third-party stakeholders (breast radiologists, radiology administrators, patient advocates). MAIN MEASURES: Qualitative data, with a codebook developed based upon prespecified implementation science constructs. KEY RESULTS: We interviewed 25 patients, 11 PCPs, and eight key stakeholders, including three radiologists, two radiology administrators, and three patient advocates. Most patients reported having annual mammograms, however, half believed having mammograms every two years was acceptable. Some women were worried early breast cancer would be missed if undergoing biennial screening. PCPs were equally split between recommending annual and biennial mammograms. Although PCPs were interested in using breast cancer risk models to inform screening decisions, concerns raised include time burden and lack of familiarity with breast cancer risk assessment tools. All breast radiologists believed patients should receive annual mammograms, while patient advocates and radiology administrators were split between annual vs. biennial. Radiologists were worried about missing breast cancer diagnoses when mammograms are not performed yearly. Patient advocates and radiology administrators were more open to biennial mammograms and utilizing risk-based screening. CONCLUSIONS: Uncertainty remains across stakeholder groups regarding appropriate mammogram screening intervals. Radiologists recommend annual mammography, whereas patients and PCPs were evenly split between annual vs. biennial screening, although both favored annual screening among higher-risk women. Breast cancer risk assessment tools may help facilitate decisions about screening intervals, but face barriers to widespread implementation in the primary care setting. These results will inform future implementation strategies to adopt risk-stratified breast cancer screening.

Use of a convolutional neural network-based mammographic evaluation to predict breast cancer recurrence among women with hormone receptor-positive operable breast cancer.

PURPOSE: We evaluated whether a novel, fully automated convolutional neural network (CNN)-based mammographic evaluation can predict breast cancer relapse among women with operable hormone receptor (HR)-positive breast cancer. METHODS: We conducted a retrospective cohort study among women with stage I-III, HR-positive unilateral breast cancer diagnosed at Columbia University Medical Center from 2007 to 2017, who received adjuvant endocrine therapy and had at least two mammograms (baseline, annual follow-up) of the contralateral unaffected breast for CNN analysis. We extracted demographics, clinicopathologic characteristics, breast cancer treatments, and relapse status from the electronic health record. Our primary endpoint was change in CNN risk score (range, 0-1). We used two-sample t-tests to assess for difference in mean CNN scores between patients who relapsed vs. remained in remission, and conducted Cox regression analyses to assess for association between change in CNN score and breast cancer-free interval (BCFI), adjusting for known prognostic factors. RESULTS: Among 848 women followed for a median of 59 months, there were 67 (7.9%) breast cancer relapses (36 distant, 25 local, 6 new primaries). There was a significant difference in mean absolute change in CNN risk score from baseline to 1-year follow-up between those who relapsed vs. remained in remission (0.001 vs. - 0.022, p = 0.030). After adjustment for prognostic factors, a 0.01 absolute increase in CNN score at 1-year was significantly associated with BCFI, hazard ratio = 1.05 (95% Confidence Interval 1.01-1.09, p = 0.011). CONCLUSION: Short-term change in the CNN-based breast cancer risk model on adjuvant endocrine therapy predicts breast cancer relapse, and warrants further evaluation in prospective studies.

Association Between Genetic Testing for Hereditary Breast Cancer and Contralateral Prophylactic Mastectomy Among Multiethnic Women Diagnosed With Early-Stage Breast Cancer.

PURPOSE: Increasing usage of multigene panel testing has identified more patients with pathogenic or likely pathogenic (P or LP) variants in low-moderate penetrance genes or variants of uncertain significance (VUS). Our study evaluates the association between genetic test results and contralateral prophylactic mastectomy (CPM) among patients with breast cancer. METHODS: We conducted a retrospective cohort study among women diagnosed with unilateral stage 0-III breast cancer between 2013 and 2020 who underwent genetic testing. We examined whether genetic test results were associated with CPM using multivariable logistic regression models. RESULTS: Among 707 racially or ethnically diverse women, most had benign or likely benign (B or LB) variants, whereas 12.5% had P or LP and 17.9% had VUS. Racial or ethnic minorities were twice as likely to receive VUS. Patients with P or LP variants had higher CPM rates than VUS or B or LB (64.8% v 25.8% v 25.9%), and highest among women with P or LP variants in high-penetrance genes (74.6%). On multivariable analysis, P or LP compared with B or LB variants were significantly associated with CPM (odds ratio = 4.24; 95% CI, 2.48 to 7.26). CONCLUSION: Women with P or LP variants on genetic testing were over four times more likely to undergo CPM than B or LB. Those with VUS had similar CPM rates as B or LB. Our findings suggest appropriate genetic counseling and communication of cancer risk to multiethnic breast cancer survivors.

Weight changes of younger and older early breast cancer patients-a meta regression.

BACKGROUND: Weight gain during chemotherapy for breast cancer is quite common and has a major impact on the quality of life. Post-treatment weight gain can also impact on primary endpoints such as tumor recurrence and overall survival. Parameters thought to impact weight gain include menopausal status, age and chemotherapy regimen. Using meta-regression, we studied the effect of age on weight change by menopausal status and chemotherapy regimen. METHODS: Twenty-four studies were identified, and extracted for weight change, mean/median age, menopausal status and chemotherapy regimen. A meta-regression was performed using a random-effects model for high heterogeneity and fixed-effects inverse-variance model for low heterogeneity. Subgroup analyses by menopausal status and chemotherapy regimen were conducted. P values <0.05 were considered statistically significant. RESULTS: There exists no relationship between weight change and age (ß=0.00; P=0.987). Stratifying by menopausal status (ß=0.05 and P=0.150 for premenopausal patients; ß=0.09 and P=0.588 for postmenopausal patients) and chemotherapy regimens (ß=-0.07 and P=0.562 for patients receiving CMF alone; ß=0.08 and P=0.707 for patients receiving CMF in addition to others; ß=0.02 and P=0.807 for patients not receiving CMF), there likewise was no relationship between weight change and age. CONCLUSIONS: Management of weight gain due to chemotherapy has been focused on relatively young women who are generally at higher risk of mortality and tumor recurrence. However, our results suggest that age should not be used for differential care.

Risks and Preventive Strategies for Clostridioides difficile Transmission to Household or Community Contacts during Transition in Healthcare Settings.

The burden of Clostridioides difficile infection (CDI) has greatly increased. We evaluated the risks for CDI transmission to community members after hospitalized patients are discharged. We conducted a systematic literature review in MEDLINE/PubMed, EMBASE, CINAHL plus EBSCO, Web of Science, Cochrane Library, and gray literature during January 2000‒February 2019 and identified 4,798 citations were identified. We eliminated 4,554 citations through title and abstract screening; 217 additional citations did not meet full criteria. We reviewed texts for the 27 remaining articles qualitatively for internal/external validity. A few identified studies describing risks to community members lacked accurate risk measurement or preventative strategies. Primary data are needed to assess efficacy of and inform current expertise-driven CDI prevention practices. Raising awareness among providers and researchers, conducting clinical and health services research, linking up integrated monitoring and evaluation processes at hospitals and outpatient settings, and developing and integrating CDI surveillance systems are warranted.

Systematic Review of Electronic Cigarette Use (Vaping) and Mental Health Comorbidity Among Adolescents and Young Adults.

INTRODUCTION: The prevalence of electronic cigarette (EC) use has risen dramatically among adolescents and young adults (AYA, ages 12-26) over the past decade. Despite extensive established relationships between combustible cigarette use and mental health problems, the mental health comorbidities of EC use remain unclear. AIMS AND METHODS: To provide a systematic review of existing literature on mental health comorbidities of EC use among AYA. Database searches using search terms related to EC, AYA, and mental health identified 1168 unique articles, 87 of which prompted full-text screening. Multiple authors extracted data, applied the Effective Public Health Practice Project Quality Assessment Tool to evaluate the evidence, and synthesized findings. RESULTS: Forty articles met eligibility criteria (n = 24 predominantly adolescent and 16 predominantly young adult). Analyses yielded three main categories of focus: internalizing disorders (including depression, anxiety, suicidality, eating disorders, post-traumatic stress disorder), externalizing disorders (attention-deficit/hyperactivity disorder and conduct disorder), and transdiagnostic concepts (impulsivity and perceived stress). Significant methodological limitations were noted. CONCLUSIONS: Youth EC use is associated with greater mental health problems (compared with nonuse) across several domains, particularly among adolescents. Because many existing studies are cross-sectional, directionality remains uncertain. Well-designed longitudinal studies to investigate long-term mental health sequelae of EC use remain needed. IMPLICATIONS: Forty recent studies demonstrate a variety of mental health comorbidities with AYA EC use, particularly among adolescents. Mental health comorbidities of EC use generally parallel those of combustible cigarette use, with a few exceptions. Future EC prevention and treatment strategies may be enhanced by addressing mental health.

Impact of long-term lipid-lowering therapy on clinical outcomes in breast cancer.

INTRODUCTION: The use of statins has been associated with improved survival in patients with breast cancer in several studies but results have been mixed. This study evaluates the impact of duration of statin use on breast cancer patient outcomes. METHODS: This is a single-institution, retrospective cohort, examining the impact of statin use on the outcomes of 1523 women diagnosed with operable breast cancer between1995 and 2015. Clinical variables were compared using Student's t test, Fisher's exact and Chi square tests. Overall (OS) and disease-free (DFS) survival were performed using Kaplan-Meier and Cox-Proportional Hazard (Cox-PH) analysis in the statistical software R. RESULTS: Patients were grouped by duration of statin use: never-statin user [N] (n = 1092), short (< 3 years) [S] (n = 115), moderate [M] (3-5 years) (n = 109) and long [L] (> 5 years) (n = 207) term. Over a median follow-up of 70.2 months, 138 women died (84 died of breast cancer) and 125 had disease recurrence. On multivariable Cox-PH analysis adjusting for clinical variables including metabolic comorbidities using the Charlson comorbidity index, OS in the [S] and [M] subgroups did not differ [N], while OS was improved in [L] (adjusted hazard ratio (AHR) 0.38, confidence interval (CI) 0.17-0.85, p < 0.018). DFS was also significantly improved in the [L] subgroup (adjusted HR 0.15, CI 0.05-0.48, p < 0.001). Subanalysis stratified by receptor status showed a trend towards improved DFS in all tumor subtypes including triple-negative breast cancer. CONCLUSIONS: Our retrospective analyses suggest that long-term statin use (> 5 years) was associated with improved OS and DFS in women with breast cancer regardless of receptor subtype, even after adjusting for metabolic comorbidities.

The impact of aspirin use on breast cancer subtype and clinical course.

BACKGROUND: The use of aspirin has been associated with improved survival in patients with breast cancer, but the results have been mixed. We aim to analyze the impact of aspirin use before or after breast cancer diagnosis on breast cancer clinical characteristics and outcomes. MATERIALS AND METHODS: We performed a single-institution, retrospective analysis of 1113 women diagnosed with operable breast cancer between 1995 and 2015. Patients were grouped according to their aspirin use: never (944), before diagnosis (79), and after diagnosis (90). Clinical variables, overall survival (OS), and disease-free survival (DFS) were compared between groups. RESULTS: Women using aspirin before diagnosis were older, more likely to be black, and to have associated medical comorbidities than patients in other groups (all P <0.001). These patients were also more likely to present with hormone receptor-negative cancers, including triple-negative breast cancer (P = 0.002). Aspirin use before diagnosis was associated with a worse OS in univariate and multivariate analyses (both P <0.001), but there were no other differences in OS or DFS related to aspirin use. CONCLUSIONS: Despite a potential impact on tumor subtype in patients using aspirin before their breast cancer diagnosis, aspirin use does not appear to alter breast cancer-related survival.

Obesity, Metabolic Syndrome, and Breast Cancer: From Prevention to Intervention.

Obesity contributes to an estimated forty-percent, or 630,000 cases, of malignant neoplasms diagnosed in the United States[1] and higher body mass index (BMI) has been associated with at least seventeen types of solid tumors, including 9% of all breast cancer cases. In this review, we discuss the impact of obesity and consequences of obesity, including the metabolic syndrome and type 2 diabetes mellitus, on breast cancer risk and recurrence. Recent work has identified multiple molecular mechanisms that may underlie the association between obesity and breast cancer. In particular, insulin resistance, increased inflammatory cytokines, leptin signaling, and adipokine signaling have been shown to affect breast cancer risk and outcomes. While obesity is associated with higher breast cancer incidences and worse breast cancer outcomes, several risk reduction methods have been shown to attenuate these risks. Both metformin and statins have been shown to improve disease free survival and overall survival compared to non-users. Metformin also has been associated with lower risk of breast cancer incidence. Furthermore, increased physical activity and weight loss have been shown to decrease risk of breast cancer, especially in post-menopausal women. These studies have emphasized the potential impact that lifestyle changes can have on breast cancer risk and outcomes, and demonstrate the need for randomized control trials to evaluate the roles of metformin and statins for the treatment and chemoprevention of breast cancer.