Patient- vs Physician-Initiated Response to Symptom Monitoring and Health-Related Quality of Life

Abstract

Online symptom monitoring through patient-reported outcomes can enhance health-related quality of life and survival. However, widespread adoption in clinical care remains limited due to various barriers including the need to reduce health care practitioners' workload. To report the effects of patient-reported outcome (PRO) symptom monitoring on HRQOL and survival up to 1 year after initiation of any treatment in patients with lung cancer. SYMPRO-Lung is a multicenter stepped-wedge cluster randomized trial including patients with stage I to IV lung cancer. The inclusion period was from October 24, 2019, until September 16, 2021, and data collection ended October 8, 2022. Data analysis was conducted from November 9, 2023, until March 18, 2024. Patients in the intervention group reported PRO symptoms weekly using the Patient Reported Outcomes version of the Common Toxicity Criteria for Adverse Events lung cancer subset. If symptoms exceeded a validated threshold, an alert was sent to the health care practitioner (active intervention subgroup) or to the patient (reactive intervention subgroup). Patients in the control group received standard care. Health-related quality of life was measured by the European Organisation for Research and Treatment of Cancer Quality of Life Questionnaire at baseline, 15 weeks (T1), 6 months (T2), and 1 year (T3), with the summary score (SS) and physical functioning (PF) as primary end points. Linear mixed-effects modeling was used to assess mean differences over time. Effect size (ES) of 0.40 or greater was considered clinically relevant. Cox proportional hazards regression survival analyses were performed to estimate the effect of the intervention on progression-free survival and overall survival (OS). Data were analyzed on an intention-to-treat basis. A total of 515 patients (266 [51.7%] men; mean [SD] age, 65.4 [9.4] years) were included in the study (266 in the control group; 249 in the pooled intervention group). Most baseline characteristics were balanced between groups; however, the most notable exception was the distribution in cancer staging: the intervention group had a higher proportion of patients with stage IV cancer compared with the control group (139 [56%] vs 118 [44%]). The pooled intervention group had a significantly better SS (mean difference T1, 5.22; 95% CI, 2.72-7.73; P < .001; ES = 0.33; mean difference T2, 6.28; 95% CI, 3.65-8.92; P < .001; ES = 0.40; mean difference T3, 3.97; 95% CI, 1.15-6.80; P = .006; ES = 0.25) compared with the control group. Group differences improved more in PF but did not meet the ES greater than or equal to 0.40 threshold (mean difference T1, 7.00; 95% CI, 3.65-10.35; P < .001; ES = 0.27; mean difference T2, 6.79; 95% CI, 3.26-10.31; P < .001; ES = 0.26; mean difference T3, 5.01; 95% CI, 1.23-8.79; P = .009; ES = 0.19). No significant differences in HRQOL were observed between the reactive (n = 89) and active (n = 160) intervention groups. The HR for progression-free survival for the active intervention group compared with the control group was 0.78 (95% CI, 0.58-1.04); the finding was not statistically significant. The HR for overall survival for both interventions groups compared with the control group were not statistically significant.(active: HR, 0.80; 95% CI, 0.55-1.15; reactive: HR, 0.69; 95% CI, 0.42-1.15). In this 1-year follow-up of a stepped-wedge cluster randomized trial, PRO symptom monitoring yielded improvements in long-term HRQOL in patients with lung cancer. The reactive approach proved equally effective as the active approach. A nonsignificant potential survival benefit was observed for the intervention group. These positive results provide further evidence for the usefulness of routine PRO symptom monitoring in lung cancer care. The Netherlands trial register Identifier: NL7897.