Comparing the effectiveness, cost-effectiveness and implementation of low-dose oral modified release morphine in...

Chronic breathlessness is common and disabling in chronic medical conditions. There is good rationale to use opioids from neuroscience and exercise laboratory studies.

We assessed the 56-day effectiveness, cost-consequences and safety of long-acting oral morphine for breathlessness in people with cardio-respiratory diseases or cancer.

We conducted a parallel-group, randomised, placebo-controlled trial in 11 UK outpatient services, comparing 10-20 mg daily oral morphine with placebo (and blinded laxative) in people with moderate to severe breathlessness, with embedded health economic evaluation. The primary outcome was worst breathlessness/24 hours (0-10 numerical rating scale) at day 28. Secondary outcomes included: physical activity levels; worst cough numerical rating scale/24 hours; Short Form questionnaire-12 items; EuroQol-5 Dimensions, five-level version; morphine-related toxicities; caregiver burden. Analyses were by modified intention-to-treat (modified intention-to-treat; ≥ 1 dose of study drug). Primary analysis used repeated measures of covariance adjusted for baseline worst breathlessness and stratification variables. The planned sample size (90% power, 20% withdrawal) was 158. We undertook two exploratory substudies, a process theory-informed evaluation and an open-label follow-on study.

Between 18 March 2021 and 26 October 2023, 143 were randomised (75 morphine, 68 placebo); 140 formed the modified intention-to-treat population [90% power; males 66%; mean age 70.5 (standard deviation 9.4)]. By day 28, 60/73 (82%) morphine versus 56/67 (84%) placebo participants had ≥ 90% adherence. At day 28, we found no evidence of a primary outcome difference [adjusted mean difference; 0.09 (95% confidence interval -0.57 to 0.75), p = 0.78], or any secondary measure except improved cough seen at day 56 [adjusted mean difference; -1.41 (-2.18 to -0.64)]. Increased moderate/vigorous physical activity was seen at day 28 [adjusted mean difference; 9.51 minutes/day (0.54 to 18.48)], but this was not significant after multiple measures correction. There were 413 adverse events: 251 versus 162 events in 63/73 (86%) morphine versus 51/67 (76%) placebo participants; 18 serious adverse events in 15 participants (morphine, 3/15 related; placebo, 0/3 related); no treatment-related deaths. Morphine participants used more inpatient and fewer outpatient services. In the 56-day base-case analysis, morphine was associated with similar mean per-patient costs and quality-adjusted life-years: an increase of £24 (95% confidence interval -£395 to £552) and 0.002 (95% confidence interval -0.004 to 0.008) quality-adjusted life-years.

Interpretation is cautious, but findings are plausible and consistent with other evidence. We infer improved exercise endurance. Most participants had chronic lung disease and minoritised ethnic communities are under-represented.

We found no evidence that morphine improved worst breathlessness intensity. The pattern of improved physical activity and cough, without increased sleepiness, sedation or cognitive impairment, with good adherence provides rationale for future research. A therapeutic trial of morphine should be evaluated using an exertional task.

Further research is needed to understand if there is any role of morphine in chronic breathlessness, but our findings do not support its use in this setting. Future studies should consider physical activity levels (actigraphy), exercise endurance (isotime/isoload breathlessness measures) or cough as the primary outcome. We need better representation from minoritised ethnic communities and people with heart failure.

This synopsis presents independent research funded by the National Institute for Health and Care Research (NIHR) Health Technology Assessment programme as award number 17/34/01.