Development of Machine Learning Algorithms for the Prediction of Financial Toxicity in Localized Breast Cancer Following Surgical Treatment

Abstract

PURPOSE Financial burden caused by cancer treatment is associated with material loss, distress, and poorer outcomes. Financial resources exist to support patients but identification of need is difficult. We sought to develop and test a tool to accurately predict an individual's risk of financial toxicity based on clinical, demographic, and patient-reported data prior to initiation of breast cancer treatment. PATIENTS AND METHODS We surveyed 611 patients undergoing breast cancer therapy at MD Anderson Cancer Center. We collected data using the validated COmprehensive Score for financial Toxicity (COST) patient-reported outcome measure alongside other financial indicators (credit score, income, and insurance status). We also collected clinical and perioperative data. We trained and tested an ensemble of machine learning (ML) algorithms (neural network, regularized linear model, support vector machines, and a classification tree) to predict financial toxicity. Data were randomly partitioned into training and test samples (2:1 ratio). Predictive performance was assessed using area-under-the-receiver-operating-characteristics-curve (AUROC), accuracy, sensitivity, and specificity. RESULTS In our test sample (N = 203), 48 of 203 women (23.6%) reported significant financial burden. The algorithm ensemble performed well to predict financial burden with an AUROC of 0.85, accuracy of 0.82, sensitivity of 0.85, and specificity of 0.81. Key clinical predictors of financial burden from the linear model were neoadjuvant therapy (βregularized, .11) and autologous, rather than implant-based, reconstruction (βregularized, .06). Notably, radiation and clinical tumor stage had no effect on financial burden. CONCLUSION ML models accurately predicted financial toxicity related to breast cancer treatment. These predictions may inform decision making and care planning to avoid financial distress during cancer treatment or enable targeted financial support. Further research is warranted to validate this tool and assess applicability for other types of cancer.

Machine learning to predict individual patient-reported outcomes at 2-year follow-up for women undergoing cancer-related mastectomy and breast reconstruction (INSPiRED-001)

“…Choice of algorithms, algorithm development, and reporting on them was informed by recent guidelines on how to use machine learning in medicine [ 23 ], how to report findings of diagnostic tests [ 24 ] and multivariate prediction models [ 25 ], and previously published research by our group [ 9 , 11 , 12 , 17 ]. We developed and validated three algorithms to predict clinically meaningful changes in satisfaction with reconstructed breasts: Logistic Regression (LR) with Elastic Net Penalty: We chose this algorithm because of its known ability to attenuate the influence of certain predictors on the model, leading to greater generalizability to new datasets [ 26 , 27 ].…”

“…Machine learning algorithms identify complex patterns in data to make accurate outcome predictions of future events at an individual level [ [8] , [9] , [10] ]. Such algorithms have shown great performance in other areas of breast cancer treatment like identifying exceptional responders to neoadjuvant treatment or patients at risk of experiencing financial toxicity related to their cancer treatment [ 11 , 12 ]. As post-surgical satisfaction with breasts is a recommended key outcome for women undergoing cancer-related mastectomy and breast reconstruction [ 13 ], we hypothesized that machine learning algorithms may allow accurate, individualized predictions of long-term satisfaction with reconstructed breasts prior to the initiation of the breast reconstruction process to better inform the decision-making process for these women.…”

Vacuum-Assisted Breast Biopsy After Neoadjuvant Systemic Treatment for Reliable Exclusion of Residual Cancer in Breast Cancer Patients

“…Previous research on machine learning to improve diagnostic accuracy has shown promising results. [32][33][34] When omission of breast cancer surgery is considered, oncologic safety is of utmost importance. The fear of leaving residual disease behind is evident.…”

“…Previous research on machine learning to improve diagnostic accuracy has shown promising results. 32 – 34 …”

Solving the Malady of Financial Toxicity Using Augmented Intelligence

“…8 Most published data on financial toxicity are usually related to systemic anticancer therapy and rarely look at the impact of surgery. Development of machine learning algorithms for the Prediction of Financial Toxicity in Localized Breast Cancer following Surgical Treatment 9 published in this volume is an attempt to look at the impact of surgery on financial toxicity in patients with operable breast cancer and in subjects with high-risk predisposition to breast cancer (including prophylactic mastectomies) using machine learning and artificial intelligence (AI).…”