Patient-reported symptom monitoring in patients with lung cancer (SYMPRO-Lung trial): long-term, post-hoc survival results of a multicentre, stepped-wedged, cluster-randomised clinical trial.

Abstract

Patient-reported outcome (PRO) symptom monitoring has been shown to improve overall survival. Yet, available evidence largely derives from resource-intensive studies limited to patients with advanced cancer. We aimed to assess the effect of PRO symptom monitoring on overall and progression-free survival in patients with stage I-IV lung cancer in a real-world clinical setting, with follow-up extending to 5 years. The SYMPRO-Lung trial was a multicentre, stepped-wedged, cluster-randomised clinical trial across 14 medical centres in the Netherlands. Patients were eligible for inclusion if they were aged 18 years or older with stage I-IV lung cancer (new-onset or recurrent; cytologically or histologically confirmed or radiological suspected) and were starting new treatment. All hospitals started with a control period (standard of care) and, according to randomisation by a computer-generated allocation sequence, crossed over at predefined timepoints to the intervention period (completing weekly online symptom assessments for 1 year). The primary outcome for SYMPRO-Lung trial was health-related quality of life, which has been previously reported. In this post-hoc analysis, we assessed long-term survival outcomes of overall survival and progression-free survival, with overall survival as the primary outcome. Survival analyses in the intention-to-treat population was performed using Cox mixed-effects modelling, adjusted for age, sex, stage, histology, Eastern Cooperative Oncology Group performance status, systemic treatment line, and a random effect for site clustering. The study is registered with Netherlands Trial Register, NL7897, and is complete. Of the 515 patients initially enrolled to the SYMPRO-Lung trial between Oct 24, 2019, and Sept 16, 2021, 446 (87%) patients were included in this post-hoc analysis (230 [52%] male; 216 [48%] female). 235 (53%) were assigned to the pooled intervention group and 211 (47%) patients to the control group. 69 (13%) of 515 patients were excluded due to the absence of informed consent for the use of their data in follow-up research or because linkage to the Netherlands Cancer Registry could not be established. At a median follow-up 50 months (IQR 46-55), median overall survival was 26 months in the intervention group (IQR 9-not reached, 95% CI 21-33) compared with 21 months (IQR 10-54, 95% CI 17-26) in the control group. Analyses adjusting for confounders showed that PRO symptom monitoring might be associated with improved overall survival (hazard ratio [HR] 0·80, 95% CI 0·63-1·00, p=0·049). Analyses adjusting for confounders showed that PRO symptom monitoring showed no significant difference in progression-free survival between groups (0·86, 0·70-1·07; p=0·18). Weekly PRO symptom monitoring was associated with meaningful improvement in long-term overall survival among patients with stage I-IV lung cancer. Median overall survival was 5 months longer in the intervention group than in the control group. These findings underscore the value of implementing PRO symptom monitoring within routine care for patients with lung cancer. Roche Netherlands, Stichting Kwaliteitsgelden Medisch Specialisten, and Zorginnovatiefonds. For the Dutch translation of the abstract see Supplementary Materials section.