Real-world ePRO use and clinical outcomes using electronic patient-reported symptom monitoring for patients with advanced non-small-cell lung cancer receiving first-line pembrolizumab.

Aim: This ambispective observational study assessed the impact of Noona, an electronic patient-reported outcomes (ePRO) platform, for patients with non-small cell lung cancer (NSCLC) treated in a community oncology setting. Methods: Adults with advanced NSCLC, ECOG performance status of 0-2, who received first-line (1L) pembrolizumab (monotherapy or with chemotherapy) were eligible. Those initiating pembrolizumab from 1 July 2017 to 30 June 2019, identified retrospectively (historical cohort), were compared with those initiating pembrolizumab from 1 October 2019 to 30 September 2021 who were prospectively offered Noona (standard of care [SoC] cohort). The Kaplan-Meier method and Cox proportional hazards models were used to compare pembrolizumab real-world time on treatment (rwToT; primary outcome measure) and rw time to next treatment or death (rwTTNTD) between historical and SoC cohorts. Healthcare resource use (HCRU) was compared using generalized linear models with Poisson distribution. Analyses were repeated to compare outcomes in the SoC cohort between Noona users (created a profile and used any function ≥one-time during 1L therapy) and nonusers with >42 days on 1L pembrolizumab. Data cutoff was 30 June 2020 and 30 September 2022 for historical and SoC cohorts, respectively. Results: Median pembrolizumab rwToT was 4.4 months (95% CI: 3.9-5.1) in the historical cohort (n = 448) versus 4.1 months (95% CI: 3.3-4.8) in the SoC cohort (n = 462; adjusted hazard ratio [aHR], 0.9; 95% CI: 0.8-1.0; p = 0.14 vs historical cohort). In the SoC cohort, 147 of 341 eligible patients (43%) established a Noona profile; 122/341 (36%) were Noona users. Median rwToT was 6.4 months (95% CI: 5.1-7.4) and 6.9 months (95% CI: 5.6-7.6) among Noona users and Noona nonusers (n = 219), respectively (aHR, 1.1; 95% CI: 0.8-1.4; p = 0.95 vs Noona users). The rwTTNTD and HCRU were comparable in historical versus SoC cohorts and for Noona users versus nonusers. During the first year after establishing a Noona profile, 92 of 147 patients (63%) used the platform; monthly use was 32-42%, and checking laboratory results was the most used function overall (by 52% of the 147). Conclusion: Notwithstanding the null findings of this study, positive results of ePRO use in clinical trials and observational studies support the treatment-related symptom monitoring and survival benefits of ePRO utilization.