Abstract
Guidelines recommend morphine as the first-line pharmacological treatment for cancer dyspnoea. However, trials with other opioids have not been performed. Our aim was to demonstrate the non-inferiority of oxycodone to morphine for relieving dyspnoea in cancer patients.
We conducted a multicentre, open-label, parallel-group, randomized control trial. We randomly and equally assigned cancer patients on regular oxycodone who developed dyspnoea to get a single dose of oral immediate-release oxycodone or morphine. We evaluated the change in dyspnoea intensity (numeric rating scale: 0–10) and adverse events after the medication administration.
This study was preconfidence interval -maturely terminated. All 17 enroled patients (8 using oxycodone, 9 using morphine) completed evaluations. In the oxycodone group, dyspnoea intensity decreased 1.75 points [95% confidence interval, 0.72–2.78] at 60 min and 1.50 points (95% confidence interval, −0.11 to 3.11) at 120 min. In the morphine group, dyspnoea decreased 1.33 points (95% confidence interval, 0.41–2.25) at 60 min and 1.00 point (95% confidence interval, −0.08 to 2.08) at 120 min. The differences did not fulfil the non-inferiority requirement. Although two and four patients in morphine group developed drowsiness at 60 and 120 min, no patient in oxycodone group developed significant adverse events.
Although we did not show the non-inferiority of oxycodone, oxycodone may have some effectiveness for cancer dyspnoea without significant safety concern.
UMIN-CTR: UMIN 000005760
Introduction
Dyspnoea is defined as ‘a subjective experience of breathing discomfort’ (1). Dyspnoea results from physiological, psychological, social and environmental factor interactions and may induce secondary physiological and behavioural responses. Dyspnoea is common in patients with advanced cancer, with a prevalence of 54–76% (2–5). Moreover, dyspnoea is one of the most distressing symptoms for not only patients but also their families and caregivers. Therefore, relieving dyspnoea is important to improve the quality of life of patients with cancer.
Systemic morphine is the only pharmacological therapy proven to be efficacious for cancer dyspnoea in randomized controlled trials (6–9), and it is the first-line pharmacological treatment (10–13). However, the accumulation of active metabolites after morphine administration in cases of renal insufficiency increases the risk of adverse events, such as somnolence and myoclonus (14,15). Therefore, it is controversial to give systemic morphine for cancer dyspnoea to patients with renal insufficiency.
As an opioid analgesic for cancer pain, oxycodone has been used more popularly than morphine in Japan. Oxycodone seems relatively safe compared with morphine in the cases of mild to moderate renal insufficiency (14), because oxycodone is metabolized by liver enzymes (CYP2D6 and 3A4) to inactive metabolites (noroxycodone) and only low amounts of active metabolite (oxymorphone) (16). However, we found only few case reports or retrospective case series on the efficacy of oxycodone for cancer dyspnoea (17,18), and current guidelines do not give any conclusive recommendation on the use of oxycodone for cancer dyspnoea (19). In addition, physicians sometimes fear a potential risk of increasing adverse events when we use different types of opioid concurrently. Thus, whether we should use morphine or oxycodone for dyspnoea in cancer patients who have already used regular oxycodone is one of the controversies in daily practice among Japanese palliative care physicians.
The aim of the present study was to compare the efficacy of oxycodone with that of morphine in relieving dyspnoea in patients with cancer who were using oxycodone as regular analgesic. We hypothesized that oxycodone is not inferior to morphine in relieving dyspnoea in patients with cancer.
Material and methods
Study design
This multicentre, open-label, parallel-group, randomized control trial was undertaken from July 2012 to June 2014. We conducted the study in accordance with the Declaration of Helsinki, and the Institutional Review Board (IRB) of Osaka University as well as IRBs of all participating sites approved the study protocol. We registered the trial at UMIN-CTR (UMIN 000005760). We obtained written informed consents from all participants.
Participants
We recruited participants from 33 palliative care services (palliative care units or palliative care consultation teams) across Japan. The inclusion criteria were as follows: age ≥ 20 years; presence of locally advanced or metastatic malignant disease; regular use of control-release oxycodone for cancer pain; dyspnoea at rest with moderate/severe intensity [3 or greater on a 0–10 numeric rating scale (NRS; 0 = no symptom, 10 = worst possible symptom)] (20); O2 saturation (SpO2) > 90% with or without supplemental O2; Eastern Cooperative Oncology Group Performance Status 0–3 and no cognitive impairment with adequate communication capacity. Patients were excluded if they had anaemia (haemoglobin level <7 g/dl), acute respiratory failure, decompensated congestive heart failure, impaired liver function (AST ≥ 201 IU/L, ALT ≥ 201 IU/L, or total bilirubin ≥ 2 mg/dl), renal insufficiency (creatinine ≥ 2.1 mg/dl), uncontrolled pain (NRS ≥ 5), expected life expectancy of < 1 month, bacterial colitis, or hypersensitivity to opioids. To exclude the influence of rescue opioid, we excluded patients who had used rescue opioids for pain within 9 h before the administration of the study medication because 9-h interval is at least twice longer than T 1/2 of usual rescue medications (immediate-release morphine and oxycodone). We also exclude those who used regular opioids other than oxycodone.
Randomization
Local physicians enroled patients who met all eligible criteria at each site and assigned them randomly into the oxycodone or the morphine groups in a 1:1 ratio using a web-based central randomization system established by a data centre in blocks of four by site without stratification. Allocated medications were open to patients as well as the researchers who prescribed the medications and evaluated the patients.
Interventions
Patients received either oral oxycodone or oral morphine. Patients allocated to the oxycodone group received a single dose of immediate-release oxycodone powder (Oxinorm® Shionogi, Japan); patients allocated to the morphine group received a single dose of immediate-release morphine liquid (Opso® Sumitomo Dainippon Pharma, Japan). The dose of each study medication was equivalent to the rescue dose for breakthrough pain, which is 10–20% of the total daily dose of regular oral oxycodone. We used the conversion ratio of 2:3 between oxycodone and morphine, according to the Japanese clinical guideline for pharmacological management of cancer pain (21). Although we allowed continuation of any analgesic medications or oxygen supplementation already being used, we prohibited new medications or dose adjustments during the study period (until 120 min after the administration of the study medication).
Measurements
We evaluated ‘dyspnoea right now’ with NRS at baseline and at 60 and at 120 min after the administration of the study medication. The primary outcome in this study was the dyspnoea NRS change from the baseline. We also assessed the response rate, defining a response as ‘1 or more points reduction in the dyspnoea NRS from baseline’, because a 1-point reduction is considered to be a clinically meaningful change (22,23). We evaluated opioid-related adverse events (nausea, drowsiness and confusion) using a four-point Likert scale (0, absent; 1, mild; 2, moderate and 3, severe) which was used in previous opioid studies (24–27) at the same time points. We also obtained respiratory rate and SpO2 to evaluate the effect of study medications on respiration. Demographic information, current medication use (benzodiazepines, corticosteroids and antipsychotics) and supplemental oxygen use were obtained from medical charts before the treatment assignment. In addition, we evaluated the intensity of anxiety with NRS at baseline.
Statistical analysis
This study was designed to assess the non-inferiority of oxycodone to morphine in relieving dyspnoea in patients with cancer. The sample size was based on the requirement for student’s t-test to compare the mean of change from baseline in dyspnoea NRS at 60 min. We set a priori non-inferiority margin of 0.5 for the primary outcome. A sample size of 44 participants for each group provided 80% power in one-tailed type 1 error of 0.025. We estimated a 10% attrition rate. Therefore, our final estimated sample size was a total of 100 participants (50 in each group).
We used student’s t-test to assess the difference between the two treatment groups in change in dyspnoea NRS from baseline and 60 and 120 min after the administration of the study medication. We also determined the mean differences in change in dyspnoea NRS from baseline at each evaluation time point with 95% confidence intervals (CIs) between two groups. For the analysis of response rate at each evaluation time point, we used χ2 test to assess the difference between two groups.
We performed all statistical analyses using SPSS for Windows software, version 19.0 (IBM Japan Institute, Tokyo, Japan).
Results
This study was prematurely terminated at the end of the planned study period (end of June 2014) because of slow accrual. Although we had planned for a total of 100 patients to be enroled, we were able to enrol only 17 patients; 8 were assigned to the oxycodone group and 9 to the morphine group. Of those 17 patients, all received the allocated medication and completed the study assessment (Fig. 1). Table 1 shows the baseline characteristics of the participants. Mean administration dose of study medications were 5.31 ± 4.91 (median 2.5: range 2.5–17.5) mg of immediate-release oxycodone in oxycodone group and 18.89 ± 15.23 (median 15: range 5–50) mg of immediate-release morphine in morphine group.
CONSORT participant flow diagram.
CONSORT participant flow diagram.
Patient characteristics
| Oxycodone (n = 8) | Morphine (n = 9) | |
|---|---|---|
| Age (years) | ||
| Mean ± SD | 71.38 ± 6.37 | 61.56 ± 13.11 |
| Median (range) | 72 (58–81) | 69 (41–74) |
| Sex [male, n (%)] | 5 (62.5) | 5 (56.6) |
| Primary site [n (%)] | ||
| Lung | 5 (62.5) | 4 (44.4) |
| Colo-rectal | 1 (12.5) | 1 (11.1) |
| Others | 2 (25) | 4 (44.4) |
| Lung metastasis [n (%)] | 3 (37.5) | 8 (88.9) |
| Pleural effusion [n (%)] | 3 (37.5) | 7 (77.8) |
| Lymphangitis carcinomatosa [n (%)] | 0 (0) | 1 (11.1) |
| Comorbidities [n (%)] | ||
| COPD | 2 (25) | 0 (0) |
| Interstitial lung disease | 1 (12.5) | 0 (0) |
| Baseline regular oxycodone dose (mg/day) | ||
| Mean ± SD | 33.75 ± 27.74 | 77.22 ± 63.30 |
| Median (range) | 20 (20–100) | 60 (20–200) |
| Concurrent therapies [n (%)] | ||
| Benzodiazepine | 1 (12.5) | 3 (33.3) |
| Corticosteroids | 3 (37.5) | 5 (55.6) |
| Antipsychotics | 1 (12.5) | 1 (11.1) |
| Supplemental oxygen | 2 (25) | 5 (55.6) |
| Baseline symptom intensity (mean ± SD, NRS: 0–10) | ||
| Dyspnoea | 4.25 ± 1.45 | 3.89 ± 0.93 |
| Anxiety | 2.63 ± 3.20 | 4.11 ± 2.85 |
| Baseline respiratory rate (mean ± SD, /min) | 14.13 ± 3.18 | 15.56 ± 3.95 |
| Baseline SpO2 (mean ± SD, %) | 96.25 ± 2.25 | 96.00 ± 1.50 |
| Baseline ECOG PS [n (%)] | ||
| 1 | 1 (12.5) | 2 (22.2) |
| 2 | 5 (62.5) | 4 (44.4) |
| 3 | 2 (25) | 3 (33.3) |
| Oxycodone (n = 8) | Morphine (n = 9) | |
|---|---|---|
| Age (years) | ||
| Mean ± SD | 71.38 ± 6.37 | 61.56 ± 13.11 |
| Median (range) | 72 (58–81) | 69 (41–74) |
| Sex [male, n (%)] | 5 (62.5) | 5 (56.6) |
| Primary site [n (%)] | ||
| Lung | 5 (62.5) | 4 (44.4) |
| Colo-rectal | 1 (12.5) | 1 (11.1) |
| Others | 2 (25) | 4 (44.4) |
| Lung metastasis [n (%)] | 3 (37.5) | 8 (88.9) |
| Pleural effusion [n (%)] | 3 (37.5) | 7 (77.8) |
| Lymphangitis carcinomatosa [n (%)] | 0 (0) | 1 (11.1) |
| Comorbidities [n (%)] | ||
| COPD | 2 (25) | 0 (0) |
| Interstitial lung disease | 1 (12.5) | 0 (0) |
| Baseline regular oxycodone dose (mg/day) | ||
| Mean ± SD | 33.75 ± 27.74 | 77.22 ± 63.30 |
| Median (range) | 20 (20–100) | 60 (20–200) |
| Concurrent therapies [n (%)] | ||
| Benzodiazepine | 1 (12.5) | 3 (33.3) |
| Corticosteroids | 3 (37.5) | 5 (55.6) |
| Antipsychotics | 1 (12.5) | 1 (11.1) |
| Supplemental oxygen | 2 (25) | 5 (55.6) |
| Baseline symptom intensity (mean ± SD, NRS: 0–10) | ||
| Dyspnoea | 4.25 ± 1.45 | 3.89 ± 0.93 |
| Anxiety | 2.63 ± 3.20 | 4.11 ± 2.85 |
| Baseline respiratory rate (mean ± SD, /min) | 14.13 ± 3.18 | 15.56 ± 3.95 |
| Baseline SpO2 (mean ± SD, %) | 96.25 ± 2.25 | 96.00 ± 1.50 |
| Baseline ECOG PS [n (%)] | ||
| 1 | 1 (12.5) | 2 (22.2) |
| 2 | 5 (62.5) | 4 (44.4) |
| 3 | 2 (25) | 3 (33.3) |
SD, standard deviation; COPD, chronic obstructive pulmonary disease; NRS, numeric rating scale, ECOG PS, Eastern Cooperative Oncology Group Performance Status.
Patient characteristics
| Oxycodone (n = 8) | Morphine (n = 9) | |
|---|---|---|
| Age (years) | ||
| Mean ± SD | 71.38 ± 6.37 | 61.56 ± 13.11 |
| Median (range) | 72 (58–81) | 69 (41–74) |
| Sex [male, n (%)] | 5 (62.5) | 5 (56.6) |
| Primary site [n (%)] | ||
| Lung | 5 (62.5) | 4 (44.4) |
| Colo-rectal | 1 (12.5) | 1 (11.1) |
| Others | 2 (25) | 4 (44.4) |
| Lung metastasis [n (%)] | 3 (37.5) | 8 (88.9) |
| Pleural effusion [n (%)] | 3 (37.5) | 7 (77.8) |
| Lymphangitis carcinomatosa [n (%)] | 0 (0) | 1 (11.1) |
| Comorbidities [n (%)] | ||
| COPD | 2 (25) | 0 (0) |
| Interstitial lung disease | 1 (12.5) | 0 (0) |
| Baseline regular oxycodone dose (mg/day) | ||
| Mean ± SD | 33.75 ± 27.74 | 77.22 ± 63.30 |
| Median (range) | 20 (20–100) | 60 (20–200) |
| Concurrent therapies [n (%)] | ||
| Benzodiazepine | 1 (12.5) | 3 (33.3) |
| Corticosteroids | 3 (37.5) | 5 (55.6) |
| Antipsychotics | 1 (12.5) | 1 (11.1) |
| Supplemental oxygen | 2 (25) | 5 (55.6) |
| Baseline symptom intensity (mean ± SD, NRS: 0–10) | ||
| Dyspnoea | 4.25 ± 1.45 | 3.89 ± 0.93 |
| Anxiety | 2.63 ± 3.20 | 4.11 ± 2.85 |
| Baseline respiratory rate (mean ± SD, /min) | 14.13 ± 3.18 | 15.56 ± 3.95 |
| Baseline SpO2 (mean ± SD, %) | 96.25 ± 2.25 | 96.00 ± 1.50 |
| Baseline ECOG PS [n (%)] | ||
| 1 | 1 (12.5) | 2 (22.2) |
| 2 | 5 (62.5) | 4 (44.4) |
| 3 | 2 (25) | 3 (33.3) |
| Oxycodone (n = 8) | Morphine (n = 9) | |
|---|---|---|
| Age (years) | ||
| Mean ± SD | 71.38 ± 6.37 | 61.56 ± 13.11 |
| Median (range) | 72 (58–81) | 69 (41–74) |
| Sex [male, n (%)] | 5 (62.5) | 5 (56.6) |
| Primary site [n (%)] | ||
| Lung | 5 (62.5) | 4 (44.4) |
| Colo-rectal | 1 (12.5) | 1 (11.1) |
| Others | 2 (25) | 4 (44.4) |
| Lung metastasis [n (%)] | 3 (37.5) | 8 (88.9) |
| Pleural effusion [n (%)] | 3 (37.5) | 7 (77.8) |
| Lymphangitis carcinomatosa [n (%)] | 0 (0) | 1 (11.1) |
| Comorbidities [n (%)] | ||
| COPD | 2 (25) | 0 (0) |
| Interstitial lung disease | 1 (12.5) | 0 (0) |
| Baseline regular oxycodone dose (mg/day) | ||
| Mean ± SD | 33.75 ± 27.74 | 77.22 ± 63.30 |
| Median (range) | 20 (20–100) | 60 (20–200) |
| Concurrent therapies [n (%)] | ||
| Benzodiazepine | 1 (12.5) | 3 (33.3) |
| Corticosteroids | 3 (37.5) | 5 (55.6) |
| Antipsychotics | 1 (12.5) | 1 (11.1) |
| Supplemental oxygen | 2 (25) | 5 (55.6) |
| Baseline symptom intensity (mean ± SD, NRS: 0–10) | ||
| Dyspnoea | 4.25 ± 1.45 | 3.89 ± 0.93 |
| Anxiety | 2.63 ± 3.20 | 4.11 ± 2.85 |
| Baseline respiratory rate (mean ± SD, /min) | 14.13 ± 3.18 | 15.56 ± 3.95 |
| Baseline SpO2 (mean ± SD, %) | 96.25 ± 2.25 | 96.00 ± 1.50 |
| Baseline ECOG PS [n (%)] | ||
| 1 | 1 (12.5) | 2 (22.2) |
| 2 | 5 (62.5) | 4 (44.4) |
| 3 | 2 (25) | 3 (33.3) |
SD, standard deviation; COPD, chronic obstructive pulmonary disease; NRS, numeric rating scale, ECOG PS, Eastern Cooperative Oncology Group Performance Status.
Change in dyspnoea
Table 2 shows the changes in dyspnoea intensity at 60 and 120 min. At 60 min, the mean reductions of dyspnoea NRS were 1.75 (95% CI, 0.72–2.78) and 1.00 (95% CI, −0.08 to 2.08) in the oxycodone and morphine groups, respectively. The mean intergroup difference was 0.75 (95% CI, −0.89–2.39), and the lower limit of the 95% CI was beyond the non-inferiority margin of −0.5. At 120 min, the mean reductions were 1.50 (95% CI, −0.11–3.11) and 1.33 (95% CI, 0.41–2.25) in the oxycodone and morphine groups, respectively. Mean intergroup difference was 0.17 (95% CI, −1.80–2.13), and the lower limit of the 95% CI was beyond the non-inferiority margin of −0.5.
Change in dyspnoea intensity in oxycodone and morphine groups
| Oxycodone (n = 8) | Morphine (n = 9) | Mean difference between arms (95% CI) | |||
|---|---|---|---|---|---|
| Dyspnoea intensity, mean (SD) | Change from baseline (95% CI) | Dyspnoea intensity, mean (SD) | Change from baseline (95% CI) | ||
| Baseline | 4.25 (1.45) | – | 3.89 (0.93) | – | – |
| 60 min | 2.50 (1.41) | 1.75 (0.72–2.78) | 2.89 (2.09) | 1.00 (−0.08–2.08) | 0.75 (−0.89–2.39) |
| 120 min | 2.75 (1.67) | 1.50 (−0.11–3.11) | 2.56 (1.67) | 1.33 (0.41–2.25) | 0.17 (−1.80–2.13) |
| Oxycodone (n = 8) | Morphine (n = 9) | Mean difference between arms (95% CI) | |||
|---|---|---|---|---|---|
| Dyspnoea intensity, mean (SD) | Change from baseline (95% CI) | Dyspnoea intensity, mean (SD) | Change from baseline (95% CI) | ||
| Baseline | 4.25 (1.45) | – | 3.89 (0.93) | – | – |
| 60 min | 2.50 (1.41) | 1.75 (0.72–2.78) | 2.89 (2.09) | 1.00 (−0.08–2.08) | 0.75 (−0.89–2.39) |
| 120 min | 2.75 (1.67) | 1.50 (−0.11–3.11) | 2.56 (1.67) | 1.33 (0.41–2.25) | 0.17 (−1.80–2.13) |
CI, confidence interval.
Change in dyspnoea intensity in oxycodone and morphine groups
| Oxycodone (n = 8) | Morphine (n = 9) | Mean difference between arms (95% CI) | |||
|---|---|---|---|---|---|
| Dyspnoea intensity, mean (SD) | Change from baseline (95% CI) | Dyspnoea intensity, mean (SD) | Change from baseline (95% CI) | ||
| Baseline | 4.25 (1.45) | – | 3.89 (0.93) | – | – |
| 60 min | 2.50 (1.41) | 1.75 (0.72–2.78) | 2.89 (2.09) | 1.00 (−0.08–2.08) | 0.75 (−0.89–2.39) |
| 120 min | 2.75 (1.67) | 1.50 (−0.11–3.11) | 2.56 (1.67) | 1.33 (0.41–2.25) | 0.17 (−1.80–2.13) |
| Oxycodone (n = 8) | Morphine (n = 9) | Mean difference between arms (95% CI) | |||
|---|---|---|---|---|---|
| Dyspnoea intensity, mean (SD) | Change from baseline (95% CI) | Dyspnoea intensity, mean (SD) | Change from baseline (95% CI) | ||
| Baseline | 4.25 (1.45) | – | 3.89 (0.93) | – | – |
| 60 min | 2.50 (1.41) | 1.75 (0.72–2.78) | 2.89 (2.09) | 1.00 (−0.08–2.08) | 0.75 (−0.89–2.39) |
| 120 min | 2.75 (1.67) | 1.50 (−0.11–3.11) | 2.56 (1.67) | 1.33 (0.41–2.25) | 0.17 (−1.80–2.13) |
CI, confidence interval.
In terms of the response rate in both treatment groups, we found no significant difference between the treatment groups at 60 min (oxycodone, 75% (6/8:95%CI, 40.6–92.9); morphine, 44.4% (4/9:95%CI, 18.7–73.6); P = 0.201) and 120 min (oxycodone, 75% (6/8:95%CI, 40.6–92.9); morphine, 66.7% (6/9:95%CI, 35.2–88.1); P = 0.707) (Fig. 2).
Response rates of oxycodone and morphine groups.
Response rates of oxycodone and morphine groups.
Adverse events
No patients in the oxycodone group developed any moderate/severe opioid-related adverse events during the study period. On the other hand, two and four patients in the morphine group developed moderate/severe drowsiness at 60 and 120 min, respectively (Table 3). The respiratory rate and SpO2 showed no significant changes from baseline to 60 and 120 min after administration of the study medications in either treatment groups (Table 4).
Adverse events in oxycodone and morphine groups
| Oxycodone (n = 8) | Morphine (n = 9) | ||
|---|---|---|---|
| Nausea (2 or 3) [n (%)] | 60 min | 0 (0) | 0 (0) |
| 120 min | 0 (0) | 0 (0) | |
| Drowsiness (2 or 3) [n (%)] | 60 min | 0 (0) | 2 (22.2) |
| 120 min | 0 (0) | 4 (44.4) | |
| Confusion (2 or 3) [n (%)] | 60 min | 0 (0) | 0 (0) |
| 120 min | 0 (0) | 0 (0) | |
| Oxycodone (n = 8) | Morphine (n = 9) | ||
|---|---|---|---|
| Nausea (2 or 3) [n (%)] | 60 min | 0 (0) | 0 (0) |
| 120 min | 0 (0) | 0 (0) | |
| Drowsiness (2 or 3) [n (%)] | 60 min | 0 (0) | 2 (22.2) |
| 120 min | 0 (0) | 4 (44.4) | |
| Confusion (2 or 3) [n (%)] | 60 min | 0 (0) | 0 (0) |
| 120 min | 0 (0) | 0 (0) | |
Adverse events in oxycodone and morphine groups
| Oxycodone (n = 8) | Morphine (n = 9) | ||
|---|---|---|---|
| Nausea (2 or 3) [n (%)] | 60 min | 0 (0) | 0 (0) |
| 120 min | 0 (0) | 0 (0) | |
| Drowsiness (2 or 3) [n (%)] | 60 min | 0 (0) | 2 (22.2) |
| 120 min | 0 (0) | 4 (44.4) | |
| Confusion (2 or 3) [n (%)] | 60 min | 0 (0) | 0 (0) |
| 120 min | 0 (0) | 0 (0) | |
| Oxycodone (n = 8) | Morphine (n = 9) | ||
|---|---|---|---|
| Nausea (2 or 3) [n (%)] | 60 min | 0 (0) | 0 (0) |
| 120 min | 0 (0) | 0 (0) | |
| Drowsiness (2 or 3) [n (%)] | 60 min | 0 (0) | 2 (22.2) |
| 120 min | 0 (0) | 4 (44.4) | |
| Confusion (2 or 3) [n (%)] | 60 min | 0 (0) | 0 (0) |
| 120 min | 0 (0) | 0 (0) | |
Effects of study medications on respiration
| Baseline | 60 min | 120 min | ||
|---|---|---|---|---|
| Oxycodone | RR (mean ± SD, /min) | 14.13 ± 3.18 | 13.75 ± 2.95 | 14.38 ± 3.53 |
| SpO2 (mean ± SD, %) | 96.25 ± 2.25 | 96.25 ± 1.17 | 96.00 ± 1.60 | |
| Morphine | RR (mean ± SD, /min) | 15.56 ± 3.95 | 17.33 ± 2.79 | 16.78 ± 4.05 |
| SpO2 (mean ± SD, %) | 96.00 ± 1.50 | 96.44 ± 2.56 | 96.22 ± 2.49 |
| Baseline | 60 min | 120 min | ||
|---|---|---|---|---|
| Oxycodone | RR (mean ± SD, /min) | 14.13 ± 3.18 | 13.75 ± 2.95 | 14.38 ± 3.53 |
| SpO2 (mean ± SD, %) | 96.25 ± 2.25 | 96.25 ± 1.17 | 96.00 ± 1.60 | |
| Morphine | RR (mean ± SD, /min) | 15.56 ± 3.95 | 17.33 ± 2.79 | 16.78 ± 4.05 |
| SpO2 (mean ± SD, %) | 96.00 ± 1.50 | 96.44 ± 2.56 | 96.22 ± 2.49 |
RR, respiratory rate.
Effects of study medications on respiration
| Baseline | 60 min | 120 min | ||
|---|---|---|---|---|
| Oxycodone | RR (mean ± SD, /min) | 14.13 ± 3.18 | 13.75 ± 2.95 | 14.38 ± 3.53 |
| SpO2 (mean ± SD, %) | 96.25 ± 2.25 | 96.25 ± 1.17 | 96.00 ± 1.60 | |
| Morphine | RR (mean ± SD, /min) | 15.56 ± 3.95 | 17.33 ± 2.79 | 16.78 ± 4.05 |
| SpO2 (mean ± SD, %) | 96.00 ± 1.50 | 96.44 ± 2.56 | 96.22 ± 2.49 |
| Baseline | 60 min | 120 min | ||
|---|---|---|---|---|
| Oxycodone | RR (mean ± SD, /min) | 14.13 ± 3.18 | 13.75 ± 2.95 | 14.38 ± 3.53 |
| SpO2 (mean ± SD, %) | 96.25 ± 2.25 | 96.25 ± 1.17 | 96.00 ± 1.60 | |
| Morphine | RR (mean ± SD, /min) | 15.56 ± 3.95 | 17.33 ± 2.79 | 16.78 ± 4.05 |
| SpO2 (mean ± SD, %) | 96.00 ± 1.50 | 96.44 ± 2.56 | 96.22 ± 2.49 |
RR, respiratory rate.
Discussion
To the best of our knowledge, this is the first study comparing the efficacy of oxycodone and morphine for relieving dyspnoea in patients with cancer. Unfortunately, we could include only 17 patients and terminated the study prematurely due to low accrual. However, our findings prompted us to make some suggestions. First, although our study does not have enough statistical power, the efficacy of oxycodone to relieve dyspnoea in patients with cancer may similar to morphine. To evaluate the true efficacy of oxycodone for cancer dyspnoea, placebo control design is appropriate. However, we think it is unethical to conduct placebo control study in patients with cancer dyspnoea because morphine has already been established as a standard pharmacological treatment for dyspnoea in patients with cancer (10–13). Therefore, we conducted the study evaluating the non-inferiority of oxycodone to morphine in relieving cancer dyspnoea. In this study, the absolute reduction in dyspnoea intensity evaluated by NRS were 1.00–1.75 in both groups at 60 and 120 min after administration, and the differences between groups did not exceed one point which was the clinically meaningful difference at either evaluation time points (22,23). Moreover, the rate of responders to oxycodone and morphine were similar at 120 min. Therefore, oxycodone may be useful as a pharmacological treatment for dyspnoea in patients with cancer. Second, oxycodone seems to be at least as safe as morphine when used for relieving dyspnoea in patients with cancer. During the study period, no patient in the oxycodone group developed moderate/severe opioid-related adverse events (nausea, drowsiness or confusion). In addition, the respiratory rate and SpO2 were maintained after the administration of oxycodone during study period in a manner similar to that in the morphine group. No evidence exists for morphine, at a therapeutic dose for dyspnoea, to cause significant or clinically relevant respiratory adverse effects (28,29). Therefore, oxycodone may be useful for relieving dyspnoea in patients with cancer with a safety profile similar to that of morphine, at a therapeutic dose for dyspnoea.
There are several limitations in this study. First, due to inadequate statistical power because of few enroled patients, our results cannot be conclusively interpreted. Future studies should enrol adequately powered samples and complete the study to confirm the efficacy of oxycodone for relieving dyspnoea in patients with cancer. Second, due to the lack of an established therapeutic dose of opioid for dyspnoea in patients with cancer, the dose we used in this study may have corresponded to underdosing. Moreover, although we used a conversion ratio between oxycodone and morphine for analgesic effect, the conversion ratio for relieving dyspnoea may be different. This uncertainty may have affected the results of our study. Therefore, a dose-finding study may help to standardize the therapeutic dose of opioids for dyspnoea in the future. Third, this study was not conducted as blinded manner because of feasibility considerations under our clinical and research circumstances. Because dyspnoea is a subjective symptom (1), the unblinded nature may have affected the results. Fourth, we only included patients who had already used oxycodone as a regular opioid analgesic for pain. This is because whether we should use morphine or oxycodone for dyspnoea in cancer patients who has already used regular oxycodone is one of the controversies in daily practice among Japanese palliative care physicians. Baseline oxycodone use may contribute participants to make some tolerance to oxycodone and this may influence the responsivity to study medications. Moreover, this may be resulted in selection bias and our results may not be generalized to opioid naïve patients. Finally, we could not evaluate how many patients had been screened for eligibility. This insufficiency in organizing this study may contribute to the result of premature termination and possible selection bias.
In conclusion, we could not show the non-inferiority of oxycodone for relieving dyspnoea in patients with cancer compared with morphine. However, oxycodone may be effective to alleviate dyspnoea in patients with cancer without posing significant safety concerns. To confirm the role of oxycodone in relieving dyspnoea in patients with cancer, an adequately powered double-blinded control trial is needed in the future.
Acknowledgements
The authors thank Tsukuba Clinical Research and Development Organization [T-CReDO: former Critical Path Research Education Integrated Leading (CREIL) Center] of University of Tsukuba for their support in writing protocol and managing the data of this study.
Conflict of interest statement
None declared.
Author contributions
Conception and design: Takashi Yamaguchi, Satoru Tsuneto, Yukiko Wagatsuma, Yoshiyuki Kizawa
Collection and assembly of data: Yoshinobu Matsuda, Hiromichi Matsuoka, Takayuki Hisanaga, Iwao Osaka, Hiroaki Watanabe, Isseki Maeda, Kengo Imai.
Data analysis and interpretation: Takashi Yamaguchi, Isseki Maeda, Yukiko Wagatsuma.
Manuscript writing: All authors
Final approval of manuscript: All authors
Financial/nonfinancial disclosures
Yoshiyuki Kizawa receives personal fees from Shionogi Pharmaceutical Co. Ltd., outside this study. Other authors declare that there is no conflict of interest.
Funding
This study was supported by a Grant for Research Advancement on Palliative Medicine, Japanese Society for Palliative Medicine (2011-111).
