Ana Tergas

Explainable artificial intelligence analysis of brachytherapy boost receipt in cervical cancer during the COVID-19 era

Brachytherapy is a critical component of the standard-of-care curative radiotherapy regimen for women with locally advanced cervical cancer (LACC). However, existing literature suggests that many patients will not receive the brachytherapy boost. We used machine learning (ML) and explainable artificial intelligence to characterize this disparity. Patients with LACC diagnosed from 2004 to 2020 who received definitive radiation were identified in the National Cancer Database. Five ML models were trained to predict if a patient received a brachytherapy boost. The best-performing model was explained using SHapley Additive exPlanation (SHAP) values. To identify trends that may be attributable to the coronavirus disease 2019 (COVID-19) pandemic, the previous analysis was repeated and limited to 2019 to 2020. A total of 37,564 patients with LACC were identified; 5799 were diagnosed from 2019 to 2020 (COVID cohort). Of these patients, 59.3% received a brachytherapy boost, with 76.4% of patients diagnosed in 2019 to 2020 receiving a boost. The random forest model achieved the best performance for both the overall and COVID cohorts. In the overall cohort, the most important predictive features were the year of diagnosis, stage, age, and insurance status. In the COVID cohort, the most important predictive features were FIGO stage, age, insurance status, and hospital type. Of the 26 patients who tested positive for COVID-19 during their course of radiotherapy, 19 (73.1%) received a brachytherapy boost. A gradual increase in brachytherapy boost utilization has been noted, which did not seem to be significantly impacted by the onset of the COVID-19 pandemic. ML could be considered to identify patient populations where brachytherapy is underutilized, which can provide actionable feedback for improving access.

Region of origin and cervical cancer stage in multiethnic Hispanic/Latinx patients living in the United States

AbstractBackgroundHispanic/Latinx people have the second highest cervical cancer incidence rates in the U.S. However, there is a lack of disaggregated data on clinical outcomes for this diverse and populous group, which is critical to direct resources and funding where they are most needed. This study assessed differences in stage at diagnosis of cervical cancer among Hispanic/Latinx subpopulations and associated factors.MethodsWe analyzed patients with primary cervical cancer from 2004 to 2019 in the National Cancer Database. Hispanic/Latinx patients were further categorized into Mexican, Puerto Rican (PR), Cuban, Dominican, and Central/South American, as per standard NCDB categories, and evaluated based on stage at diagnosis and sociodemographic characteristics. Multinomial logistic regression quantified the odds of advanced stage at presentation. Regression models were adjusted for age, education, neighborhood income, insurance status, and additional factors.ResultsHispanic/Latinx cervical cancer patients were more likely to be uninsured (18.9% vs. 6.0%, p < 0.001) and more likely to live in low‐income neighborhoods (28.6% vs. 16.9%, p < 0.001) when compared to non‐Hispanic White populations. Uninsured Hispanic/Latinx patients had 37.0% higher odds of presenting with regional versus localized disease (OR 1.37; 95% CI, 1.19–1.58) and 47.0% higher odds of presenting with distant versus. Localized disease than insured patients (OR 1.47; 95% CI, 1.33–1.62). When adjusting for age, education, neighborhood income, and insurance status, PR patients were 48% more likely than Mexican patients to present with stage IV versus stage I disease (OR 1.48; 95% CI, 1.34–1.64).ConclusionDisaggregating health data revealed differences in stage at cervical cancer presentation among Hispanic/Latinx subpopulations, with insurance status as a major predictor. Further work targeting structural factors, such as insurance status, within specific Hispanic/Latinx subpopulations is needed.