Modeling Longitudinal Quality of Life Changes in Patients Diagnosed with Lymphomas: Heatmap Analysis of Electronic Patient-Reported Outcomes (ePRO) Data from the Bison-PRO Quality of Life Study

Background

Patient-reported outcomes (PRO) measures are used in lymphoma clinical trials to assess health-related quality of life (QoL) and tolerability of novel treatment regimens, including CAR T-cell and bispecific antibody therapies. However, outcomes in the clinical trial setting often differ from those in real-world patient cohorts. The Blood DIsorder Symptomatology Outcomes Network—Patient Reported Outcomes (BISON-PRO) Quality of Life Study prospectively administers longitudinal electronic PRO (ePRO) instruments and is currently enrolling patients (pts) with lymphomas and CLL to measure QoL outcomes in the real-world setting.

Methods

All pts who present for a new patient visit in lymphoma/CLL clinics are approached to participate in BISON-PRO. Pts are enrolled in 1 of 3 study arms: Arm 1: active surveillance, Arm 2: cellular therapies, and Arm 3: starting active systemic or radiation therapy. Enrolled pts receive the following ePRO instruments at baseline, 6, and 12 months: PROMIS Global Health 10, PROMIS 29+2 Profile, FACT-LYM/ (FACT-LEU if CLL, and Acceptability of Intervention Measure (AIM). PROs were also collected at 3 and 9 mos in Arms 2 and 3. BISON-PRO enrollment began in May 2023; this analysis was performed over a 6-month time horizon. PROMIS T-scores were converted to percentile scores to construct a normal distribution for heatmap analysis. Statistical tests included one-way ANOVA for comparisons between group means.

Results

There were 65 pts at baseline, 20 at 3 mos, and 19 at 6 mos who completed ePROs. PRO retention rates were 71% at 3 mos and 59% at 6 mos. Median age was 60 years (range 25-85), 49% identified as women, 8% were Black, and 4% were Hispanic. Mean AIM scores at baseline were 4.0 for Arm 1, 3.5 for Arm 2, and 3.9 for Arm 3.

At baseline, pts with lymphoma (n = 45) on active surveillance (Arm 1) had significantly better QoL compared to Arm 2 and 3 on FACT-LYM total scores (142 vs 126 and 128, respectively; p = .047 and p = .032). By 6 months, pts with lymphoma on active therapy (Arm 3) had significant improvement in QoL compared to baseline (FACT-LYM-TOT score 146, p = .048). For pts with CLL (n = 20), FACT-LEU total scores at baseline were similar for Arms 1 and 3 (143 and 147, respectively) with no significant changes at 6 mos (142 and 140, respectively).

For the entire cohort of pts, global physical health and social role functioning mean PROMIS scores were significantly higher in Arm 1 compared with Arms 2 and 3 over time (ANOVA p = .036 and p = .035, respectively). In pts on active therapy (Arm 3), global physical health improved from 3 to 6 months by 17 percentile points (p = .096).

In the heatmap analysis of PROMIS domains for the entire cohort, there were significant improvements in global physical health between baseline and 6 months for Arms 2 and 3, while Arm 1 remained nearly unchanged (26 and 12 percentile points vs -1, respectively). Social role functioning improved 23 and 9 percentile points by 6 mos in Arms 2 and 3, respectively. There were no changes in cognitive function for Arms 1 or 2, while pts in Arm 3 experienced a 16 percentile point decline in cognitive function between 3 and 6 mos. Anxiety improved by 6 mos in all arms (14, 27, and 15 percentile improvements for Arms 1-3, respectively). Fatigue also improved by 6 mos in Arms 2 and 3 (19 and 13 percentile points, respectively).

Conclusions

In the BISON-PRO Quality of Life Study, meaningful differences in PROs were detected between study cohorts, with higher QoL (measured by FACT and PROMIS) in pts undergoing surveillance as compared to active therapies. Importantly, pts with lymphoma who underwent systemic or radiation therapy had a significant improvement in lymphoma-related QoL on FACT-LYM by 6 mos, which was not seen in pts with CLL receiving therapy on FACT-LEU. Heatmap modeling enabled granular analysis of longitudinal changes in these QoL domains, revealing improvements in physical health, social role functioning, anxiety, and fatigue in pts receiving cellular and other active therapies, but worsening cognitive function in the active therapy cohort. High acceptability of BISON-PRO (AIM scores approaching 4) demonstrates the utility of ePRO platforms to collect real-world QoL and treatment tolerability data. Use of BISON-PRO for further study of subgroups at risk for inferior QoL trajectories, as well as comparison of psychometric properties of PROMIS versus FACT instruments in pts receiving novel therapies, is warranted.