Financial Toxicity (FT) in Patients (pts) with Breast Cancer (BC) Receiving Radiation Therapy (RT): A Prospective Observational Study

Abstract

Cancer pts often face FT due to direct medical expenses and indirect expenses like lost income due to disability or missed work time. The incidence and severity of FT in pts receiving RT for BC is unknown. We aimed to describe the incidence of FT and the distribution of the COST score, a measure of FT validated in cancer pts. We also aimed to associate COST with the length of RT (long course, 25-35 fractions vs. short/intermediate (S/I) course, 1-24 fractions), disease burden, treatment and demographic factors, and the FACT-G7 quality of life score. We hypothesized that pts receiving long course RT vs. S/I would have greater FT. After signing informed consent, pts with stage 0-III BC completed a FT questionnaire, COST and FACT-G7 within one month of completion of RT. Group comparisons were performed with two-sample t-test for continuous variables and Chi-square for categorical. Pearson correlation was used to associate COST with FACT-G7. Linear and multiple regression were used to evaluate explanatory variables as predictors of COST. Between 12/2018-1/2019, 88 pts were eligible and 65 (35 long, 30 S/I) consented. Participants were well matched with non-participants with respect to stage, demographic and treatment factors; however, participants were less likely to be underserved (p = 0.04) or to have Medicaid (p = 0.02). Mean COST and FACT-G7 scores were 28.3 (95% CI 25.8-30.7) and 18.8 (95% CI 17.5-20.2), respectively. Due to RT, 26% (n = 17) reported they were unable to work during RT, 14% (n = 9) missed ≥10 work days, 17% (n = 11) lost wages, 20% (n = 13) had high transportation costs, 9% (n = 6) had high medication costs, and 19% (n = 12) had a caregiver that lost time or wages from work. Pts in long and S/I RT groups were well matched for income, education level, insurance, underserved status, dependent children, marital and employment status, but dissimilar with respect to stage, age, surgery and RT type, chemotherapy use, nodal RT and time since BC diagnosis (all p<0.001). There was significant correlation between COST and FACT-G7 score for all pts (correlation coefficient 0.62, p<0.001), long course RT (0.60, p<0.001) and S/I (0.65, p<0.001). In a multivariable regression model for COST, marital status (married vs. unmarried, p = 0.010), high medication costs (no vs. yes, p = 0.025), surgery type (lumpectomy vs. mastectomy, p = 0.008) and time since BC diagnosis (p = 0.015) retained significance, but RT duration did not (p = 0.340). After adjusting for FACT-G7 in multivariable regression, FACT-G7 (p<0.001) and surgery type (p = 0.003) retained significance, with marital status (p = 0.058) having borderline significance. Among study participants, indirect expenses due to RT were common. RT duration does not appear to predict FT, but pts who are unmarried, have mastectomy, and a longer course of BC treatment are at highest risk of FT during RT. Clinicians should recognize FT as an important determinant of quality of life in women receiving RT for BC.