Implementation of a Remote Tracking System for Acute Toxicities Using Patient-Reported Outcomes in Patients Treated with Radiation for Breast Cancer

Our aim was to develop and study the implementation of an electronic survey to collect and respond to patient-reported outcomes (PROs) regarding acute breast radiation side effects. A response-adapted survey using PRO-CTCAE items was deployed in 5 clinics from June to December 2019. Surveys were sent weekly during radiation and for 8 weeks after via an established patient portal to all patients treated for primary breast cancer. The care team received alerts and contacted patients when severe symptoms (> grade 2) were reported. Treatment, clinic, and socio-demographic characteristics were collected from patient records. Univariable logistic regression was used to assess factors associated with survey response and high-frequency response (>3 surveys completed). A Chi-square test was used to assess toxicity across treatment groups. A subsample of patients was qualitatively interviewed at follow-up. Overall, 3,319 surveys were sent to 333 patients, of which 247 (74%) responded. On average, responders completed 54% (n = 1,290) of assigned surveys. Survey response was associated with younger age (OR 0.97, 95% CI 0.94-0.99, p<0.01), higher level of education (OR 2.10, 95% CI 1.35-3.11, p<0.01), and full-time employment (OR 2.64, 95% CI 1.31-5.34, p<0.01). 211 (85%) responders reported at least one instance of grade 2 or higher toxicity (G2+). High-frequency response was associated with G2+ toxicity (OR 2.91, 95% CI 1.38-6.13, p<0.01). G2+ toxicities peaked post-radiation (Table 1). 56%, 61%, and 75% of surveys reported G2+ toxicities for the partial-breast, whole-breast, and regional nodal/postmastectomy group, respectively (p<0.01). 146 alerts were generated for 73 unique patients, 62% reported severe radiation dermatitis. 81% of alerts occurred post-treatment. Interviews (n = 15) indicated surveys to be acceptable to patients, who expressed interest in viewing symptom trajectories. Interviews also identified redundant items to be modified in the next iteration of the PRO-CTCAE-based tool. For patients undergoing breast radiation, acute toxicity PROs are well received. Rate of any response is associated with socio-demographic factors, while clinical toxicity is associated with high-frequency response. Toxicity severity and the frequency of alerts peaked post-treatment, indicating that a remote symptom monitoring tool is useful in the post-treatment period, particularly as treatment regimens become shorter.Abstract 2073; Table 1Patient-reported acute toxicityWeeks after treatment start0123456789101112Group% Respondents with G2+ Skin ToxicityPartial-breast irradiation031385042675050140Whole-breast irradiation (WBI)31343547672625437202550Postmastectomy radiation/WBI with regional nodal irradiation382355848394846745443035* Shaded cells denote treatment Open table in a new tab