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.

Smoking is Not Associated with Lung Metastasis in Pancreatic Cancer.

Smoking is a risk factor for pulmonary metastasis in various malignancies. We investigated this association for pancreatic ductal adenocarcinoma (PDAC). We conducted a retrospective 1:2 case-control study of consecutive patients who underwent PDAC resection (2011-2021). Cases ultimately developed lung metastases and controls did not. Of 744 patients we identified 53 cases and 106 matched controls. Twenty-five (47%) cases and 50 (47%) matched controls had a history of smoking (p = 1.0). This indicates that smoking is not associated with increased risk of pulmonary metastasis in resectable PDAC. Further research is needed to elucidate tumor and parenchymal factors influencing metastatic site.

Improving Quality of Life During Chemotherapy: Cannabinoids, Cryotherapy, and Scalp Cooling.

There have been significant advances in the treatment of cancer in the past decade. However, patients continue to suffer from significant side effects of antineoplastic agents that greatly affect their quality of life (QOL), including chemotherapy-induced nausea and vomiting (CINV), chemotherapy-induced peripheral neuropathy (CIPN), and chemotherapy-induced alopecia (CIA). This review aims to provide an updated overview of emerging strategies for the management and prevention of these immediate and long-lasting side effects. The use of integrative medicine including cannabis continues to evolve in the realm of CINV and cancer-related anorexia. Although no pharmaceutical agent has been approved for the prevention of CIPN, cryotherapy, compression therapy and, more recently, cryocompression therapy have shown benefit in small trials, but there are concerns with tolerability especially related to cryotherapy. More data are necessary to determine an effective and tolerable option to prevent CIPN in large, randomized studies. Scalp cooling (SC), which has a similar mechanism to cryotherapy and compression therapy for CIPN prevention, has proven to be an effective and tolerable approach in randomized studies and has significantly limited CIA, an entity that definitively affects the QOL of patients living with cancer. Taken together, cannabis, cryotherapy, compression and cryocompression therapy, and SC all strive to improve the QOL of patients living with cancer by minimizing the side effects of chemotherapeutic agents.

Time to Cancer Treatment and Chemotherapy Relative Dose Intensity for Patients With Breast Cancer Living With HIV.

Importance: Patients with breast cancer and comorbid HIV experience higher mortality than other patients with breast cancer. Objective: To compare time to cancer treatment initiation and relative dose intensity (RDI) of neoadjuvant and adjuvant chemotherapy among patients with breast cancer with vs without HIV. Design, Setting, and Participants: A retrospective, matched cohort study enrolled women who received a diagnosis of breast cancer from January 1, 2000, through December 31, 2018. The electronic medical records of 3 urban, academic cancer centers were searched for women with confirmed HIV infection prior to or simultaneous with diagnosis of stage I to III breast cancer. Tumor registry data were used to identify 2 control patients with breast cancer without HIV for each participant with HIV, matching for study site, stage, and year of cancer diagnosis. Statistical analysis was performed from December 2022 to October 2023. Exposure: HIV infection detected before or within 90 days of participants' breast cancer diagnosis. Main Outcomes and Measures: The primary outcome was time to breast cancer treatment initiation, defined as the number of days between cancer diagnosis and first treatment. The secondary outcome was overall RDI for patients who received chemotherapy. These outcomes were compared by HIV status using Cox proportional hazards regression and linear regression modeling, respectively, adjusting for confounding demographic and clinical factors. Exploratory outcomes included instances of anemia, neutropenia, thrombocytopenia, and liver function test result abnormalities during chemotherapy, which were compared using Fisher exact tests. Results: The study enrolled 66 women with comorbid breast cancer and HIV (median age, 51.1 years [IQR, 45.7-58.2 years]) and 132 with breast cancer alone (median age, 53.9 years [IQR, 47.0-62.5 years]). The median time to first cancer treatment was not significantly higher among patients with HIV than those without (48.5 days [IQR, 32.0-67.0 days] vs 42.5 days [IQR, 25.0-59.0 days]; adjusted hazard ratio, 0.78, 95% CI, 0.55-1.12). Among the 36 women with HIV and 62 women without HIV who received chemotherapy, the median overall RDI was lower for those with HIV vs without HIV (0.87 [IQR, 0.74-0.97] vs 0.96 [IQR, 0.88-1.00]; adjusted P = .01). Grade 3 or higher neutropenia during chemotherapy occurred among more women with HIV than those without HIV (13 of 36 [36.1%] vs 5 of 58 [8.6%]). Conclusions and Relevance: This matched cohort study suggests that patients with breast cancer and HIV may have experienced reduced adjuvant chemotherapy RDI, reflecting greater dose reductions, delays, or discontinuation. Strategies for supporting this vulnerable population during chemotherapy treatment are necessary.

Cancer genetic mutation prevalence in sub-Saharan Africa: A review of existing data.

BACKGROUND: Cancer represents a leading cause of death worldwide. Germline mutations in several genes increase the risk of developing several cancers, including cancers of the breast, ovary, pancreas, colorectum, and melanoma. An understanding of the population prevalence of pathogenic germline variants can be helpful in the design of public health interventions, such as genetic testing, which has downstream implications for cancer screening, prevention, and treatment. While population-based studies of pathogenic germline variants exist, most such studies have been conducted in White populations. Limited data exist regarding the prevalence of germline mutations within sub-Saharan African populations. MATERIALS AND METHODS: We identified countries defined as sub-Saharan Africa by the World Bank and conducted a scoping literature review using PubMed. For each country, we identified and summarized studies that focused on the prevalence of germline genetic mutations with sample sizes >10 and in a population directly from sub-Saharan Africa, either with or without diseases associated with the relevant genetic mutations. Studies that evaluated the prevalence of somatic or likely benign variants were excluded. RESULTS: Within the 48 countries in sub-Saharan Africa, we identified 34 studies which meet the inclusion criteria. Twenty studies were conducted in South Africa, Nigeria, or Burkina Faso; four countries had more than two published papers. We found that 33 of 48 countries in sub-Saharan Africa lacked any genetic studies. Notably, there has been an increase in relevant studies starting in 2020. Importantly, of the 34 studies identified, 29 included data on BRCA1 or BRCA2. Data on the prevalence of mutations contributing to familial cancer syndromes other than BRCA1 and BRCA2 was limited. CONCLUSIONS: While some progress has been made towards understanding the prevalence of germline mutations in cancer susceptibility genes, the characterization of genetic mutations among sub-Saharan African populations remains strikingly incomplete. Given the genetic diversity in the region, there remains a great need for large-scale, population-based studies to understand the prevalence of germline pathogenic variants and adequately capture all the subpopulations in this part of the world.