Comparison of the tolerability profile of controlled release oral morphine and oxycodone for cancer pain treatment. An open label randomized controlled trial

(1)

Comparison of the tolerability profile of controlled release oral morphine and oxycodone for cancer pain treatment. An open label randomized controlled trial Ernesto Zecca, Cinzia Brunelli, Paola Bracchi, Giuseppe Biancofiore, Carlo De Sangro, Roberto Bortolussi, Luigi Montanari, Marco Maltoni, Cecilia Moro, Ugo Colonna, Gabriele Finco, Maria Teresa Roy, Vittorio Ferrari, Oscar Alabiso, Giovanni Rosti, Stein Kaasa, Augusto Caraceni

PII: S0885-3924(16)30321-9

DOI: 10.1016/j.jpainsymman.2016.05.030

Reference: JPS 9212

To appear in: Journal of Pain and Symptom Management Received Date: 29 February 2016

Revised Date: 24 May 2016 Accepted Date: 25 May 2016

Please cite this article as: Zecca E, Brunelli C, Bracchi P, Biancofiore G, De Sangro C, Bortolussi R, Montanari L, Maltoni M, Moro C, Colonna U, Finco G, Roy MT, Ferrari V, Alabiso O, Rosti G, Kaasa S, Caraceni A, Comparison of the tolerability profile of controlled release oral morphine and oxycodone for cancer pain treatment. An open label randomized controlled trial, Journal of Pain and Symptom Management (2016), doi: 10.1016/j.jpainsymman.2016.05.030.

This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

(2)

\Original article 16-00137R2

Comparison of the tolerability profile of controlled release oral morphine and oxycodone for cancer pain treatment. An open label randomized controlled trial. Ernesto Zecca1, Cinzia Brunelli1,2 (*), Paola Bracchi1, Giuseppe Biancofiore3, Carlo De Sangro4, Roberto Bortolussi5, Luigi Montanari6, Marco Maltoni7, Cecilia Moro8, Ugo Colonna9, Gabriele Finco10, Maria Teresa Roy11, Vittorio Ferrari12, Oscar Alabiso13, Giovanni Rosti14, Stein Kaasa2,15 and Augusto Caraceni1,2.

(*) Dr. E. Zecca and dr. C. Brunelli are equally first authors of the present article. 1 - Palliative Care, Pain Therapy and Rehabilitation Unit, Fondazione IRCCS Istituto Nazionale dei Tumori, Milano, Italy.

2 - European Palliative Care Research Centre (PRC), Department of Cancer Research and Molecular Medicine, Faculty of Medicine, Norwegian University of Science and Technology (NTNU), Trondheim, Norway.

3 - Emergency Medicine Unit - Istituto Clinico Humanitas Rozzano – Milano, Italy. 4 - Oncology Unit Ospedale San Gennaro – Napoli, Italy.

5 - Palliative Care and Pain Therapy Unit - CRO Aviano National Cancer Institute - Aviano (PN) Italy.

6 - Palliative Care Unit - AUSL della Romagna – Ravenna, Italy.

7 - Palliative Care Unit - Istituto Scientifico Romagnolo per lo Studio e la Cura dei Tumori (IRST) IRCCS - Meldola (Forlì-Cesena), Italy.

8 - Medical Oncology Unit - Ospedale Papa Giovanni XIII - Bergamo, Italy. 9 - Pain And Palliative Medicine Unit - AAS 2 Bassa Friulana-Isontina – Latisana- Udine, Italy.

10 - Department of Medical sciences "M. Aresu" - University of Cagliari - Cagliari, Italy.

11 - Hospice e Cure Palliative - IRCCS Azienda Ospedaliera Universitaria San Martino - IST Istituto Nazionale per la Ricerca sul Cancro - Genova, Italy. 12 - Medical Oncology Unit - A.O. Spedali Civili di Brescia - Brescia, Italy. 13 - Oncology Unit - Ospedale di Novara - Novara, Italy.

14 - Medical Oncology - Ospedale Regionale Treviso - Treviso, Italy.

15 - Department of Oncology - Oslo University Hospital and University of Oslo, - Oslo, Norway.

Correspondence should be addressed to: Dr. Cinzia Brunelli

Palliative Care, Pain Therapy and Rehabilitation Unit Fondazione IRCCS Istituto Nazionale dei Tumori Via Venezian 1, Milano

20133 - Italy Tel. +39-02-23903387 Fax +39-02-23903393 e-mail: cinzia.brunelli@istitutotumori.mi.it http://www.istitutotumori.mi.it/ http://www.ntnu.edu/prc/prc 3 tables 4 figures

M

ANUS

CR

IP

T

AC

CE

PTE

D

(3)

37 references

3446 words, references excluded 6 tables in online appendix ABSTRACT

Context Oxycodone and morphine are recommended as first-choice opioids for moderate/severe cancer pain but evidence about their relative tolerability has significant methodological limitations.

Objectives This study is mainly aimed at comparing the risk of developing adverse events (AEs) with controlled release oral morphine versus oxycodone; secondary aims were comparing their analgesic efficacy and testing heterogeneity in tolerability across different age and renal function subgroups.

Methods An open-label multicentre RCT (EudraCT Number: 2006-003151-21) was carried out in patients with moderate/severe cancer pain. At baseline, 7 and 14 days, patients scored on 0-10 rating scales (0-10NRS) the intensity of pain and of a list of common opioid side effects. The primary endpoint was the percentage of patients reporting an AE (a worsening =2 points on any of the listed side effects); tolerability by subgroups and average follow-up pain intensity were compared through regression models.

Results 187 patients were enrolled (47% of originally planned). ITT analysis (N=185, morphine 94, oxycodone 91) did not show any difference in the risk of developing AEs (risk difference -0.6%, 95%CI -11.0 % to 9.9%) nor in analgesia (0-10NRS pain intensity difference -0.28, 95%CI - 0.83 to 0.27). No evidence of heterogeneity of tolerability across age and renal function patient subgroups emerged.

Conclusion This trial failed to show any difference in tolerability and analgesic

efficacy of morphine and oxycodone as first-line treatment for moderate/severe cancer pain and result interpretation is difficult due to lack of power, potential bias from open-label design and concerns about assay sensitivity. These data, however, can significantly contribute to future meta-analyses comparing WHO-stepIII opioids and is relevant in designing future randomized studies.

KEY WORDS: morphine; oxycodone; pain; cancer; palliative care; adverse drug reactions.

Running Title: “Tolerability of oral morphine versus oxycodone for cancer pain ”

Accepted for publication: May 25, 2016. Tables 1-3 in print.

Tables A1-A6 are online only. INTRODUCTION

Oral morphine has traditionally been considered the WHO ladder step III opioid of first choice for moderate to severe cancer pain (1, 2). An update of the European Association for Palliative Care (EAPC) recommendations on opioids use in cancer pain(3) showed lack of conclusive studies demonstrating superiority in efficacy and/or tolerability of morphine compared to other oral opioids.

Oxycodone was suggested as one alternative to morphine in the 2001 EAPC

recommendations and it recently gained increasing popularity in clinical use (4-6). A meta-analysis of five randomized clinical trials (RCTs) (7), comparing controlled release oral oxycodone (CROO) with controlled release oral morphine (CROM) (8-12) indicated that pain control did not differ significantly between the two opioids and

M

ANUS

CR

IP

T

AC

CE

PTE

D

(4)

showed no marked differences in adverse event profile. However, some studies suggested a lower frequency of adverse effects with the use of oxycodone compared to morphine (8, 9, 13). The variable clearance of potentially toxic morphine

metabolites has been considered an explanation for these clinical differences which could be particularly relevant for patients with reduced renal function and in the elderly.

Most RCTs comparing oral morphine and oxycodone, were conducted in patients who had already a favorable response to opioid treatment (7, 14, 15) and therefore they could not conclude whether one of the two drugs can be considered superior to the other when used as “first choice opioid” in daily clinical practice. Two open label RCTs comparing the first line administration of CROM and CROO, failed to demonstrate any differences in tolerability or efficacy but none of them specifically defined the clinical outcome in terms of tolerability (11, 12). Indeed most RCTs comparing different opioids lack methodological quality in the definition, measurement and reporting of AEs as remarked in a recent Cochrane review on opioid adverse effects (16).

Similarly, another systematic review(17) showed a large heterogeneity in the assessment and reporting of AEs among 25 studies on the use of WHO step III opioids for cancer-related pain; as this heterogeneity may influence AEs rates the authors conclude with a urgent need for studies with standardized outcome measures and reporting.

On this basis, we carried out a randomized controlled trial aimed at comparing the risk of developing AEs with CROM vs CROO, in patients with cancer pain not exposed to WHO step III opioids in the previous month. Secondary aims were comparing their efficacy on pain intensity reduction, and testing the heterogeneity in the tolerability of the two drugs across different age and renal function patient groups. MATERIALS AND METHODS

Study design

A two arm open-label, parallel-group, multicentre, superiority RCT with a follow-up duration of 14 days was performed in 14 Italian palliative care and oncology units (EudraCT Number: 2006-003151-21; https://www.clinicaltrialsregister.eu/ctr-search/search?query=2006-003151-21) .

Patient population

Patients with cancer pain were enrolled according to the following inclusion criteria: age = 18 years; previous 24 hours average pain intensity score = 5 on a 0 to 10 numerical rating scale (0-10 NRS); Karnofsky performance status (KPS) score =40; clinical prediction of survival = 1 month; minimum expected follow-up of two weeks at the study centre. Exclusion criteria were: therapy with WHO step III opioids within 30 days of study entry; severe renal impairment; severe hepatic failure; dyspnoea or severe chronic obstructive pulmonary disease; inability to take oral medications; history of psychiatric illness; cerebral metastasis; cognitive impairment; medical history of intolerance to morphine or oxycodone; pregnancy or breast feeding. Administration of hormonotherapy, chemotherapy, analgesic adjuvants (steroids, anticonvulsants and antidepressants) was permitted only if started before study entry, and had to be kept unchanged during the study period.

All patients enrolled provided written informed consent. The trial received

institutional Ethical Committee approval in all participating institutions before patient enrolment and it was conducted in accordance with the Declaration of Helsinki.

M

ANUS

CR

IP

T

AC

CE

PTE

D

(5)

Randomization

Patients were randomized to either CROM or CROO in a 1:1 ratio via a computer-generated block randomization -block size of 4-; assignment was stratified by age (<70 vs =70 years) and ongoing vs not ongoing chemotherapy (ongoing was defined as administered from 15 days before to 15 days after randomization). Randomization was centralized by a trial office available 8 to 16 all working days and randomization sequence was concealed until interventions assignment.

Treatment

The initial daily dose of the study medication was decided by the treating physician, based on patient characteristics, pain intensity and previous analgesic dosage,

according to usual clinical practice. Study medications were taken every 12 hours and the dose was titrated to effect. Dose adjustments during follow-up were encouraged if patients required more than two rescue analgesic doses over 24 hours. Oral immediate release or parenteral morphine, could be prescribed as supplemental analgesics for breakthrough pain. Use of antiemetics and laxatives were permitted for treatment of adverse effects as required.

Study assessments and outcome definition

Patient s were assessed at baseline, at 7 and 14 days thereafter. Demographics and clinical data were recorded at baseline; KPS score, and average pain intensity in the previous 24 hrs (0-10 NRS, 0=”no pain”, 10=”The worst possible pain”)(18) were collected at each visit.

An active surveillance of harms was carried out (18); at each visit patients rated the previous week average intensity of a predefined list of potential opioid AEs: nausea, vomiting, confusion, constipation, somnolence, dry mouth, itching and

hallucinations(16, 17). Patients were interviewed by the treating physician using 0-10 NRS (0=”No symptom”, 10=”The worst possible symptom”) for all symptoms but hallucinations, which were measured as presence/absence.

The study primary outcome was a binary variable indicating AE onset during follow-up. An AE was considered to occur when there was a worsening of at least 2 points, compared to the baseline, of any side effect listed above, at any of the two follow-up assessments; hallucinations worsening was defined as “presence of any episode during follow-up”. Average follow-up pain intensity was the secondary outcome. Statistical design and analysis

The study was planned as a superiority trial; sample size calculations indicated that 180 patients per group (200 allowing for a 10% drop out rate) would be required to detect a difference between the treatment groups of at least 15% in AE onset with a 95% confidence interval (95%CI) estimate ranging ±11%. The study was prematurely stopped after enrolling 187 patients due to slower than expected accrual rate. No interim analyses were carried out; no post-hoc power calculations were performed and 95%CIs were used when interpreting estimates precision(19, 20).

The ITT population included all randomized patients who received at least one dose of the allocated study drug. Worst case imputation (21) was applied: patients who prematurely discontinued the study for any reason - both before visit on day 7 and day 14- were imputed a “worsening”; complete case and available data analyses

(respectively defined as “both” and “at-least-one” non missing follow-up

M

ANUS

CR

IP

T

AC

CE

PTE

D

(6)

assessments) without any imputation were carried out as sensitivity analyses. For secondary efficacy endpoints, available data analysis with no imputation was applied. In the overall study population the between group risk difference (RD) of developing AEs and its 95%CI were estimated by a logistic regression model; in order to rule out any potential centre-bias, the main outcome analysis was also adjusted by centre. Statistical tests for interaction in multivariable logistic regression models were employed in order to assess heterogeneity of treatment effect across different age (age<70 vs. age=70 years) and renal function (estimated glomerular filtration rate (e-GFR) <60ml/min vs =60 ml/min) groups, as pre-declared in the protocol; e-GFR was calculate from blood creatinine, age, gender and race using the CKD-EPI formula(22). Tolerability data were also described separately for each single AE, calculating AE occurrence as a worsening = 2 points between baseline and follow-up average score (average of day 7 and day 14). Efficacy in the two treatment groups was compared by ANCOVA summary measure approach as suggested by Frison(23); the method consists in averaging pain intensity scores collected at days 7 and 14 for each patient (average follow-up pain intensity score) and using it as dependent variable in an ANCOVA model adjusted by baseline pain intensity; this allowed to estimate the between group “average follow-up pain intensity score” difference which constitute the secondary outcome. Opioid dosages were reported as oral morphine equivalent daily dose (MEDD) mg, converted using a 1.5:1 ratio between CROM and CROO. Between group difference of the opioid escalation index, (i.e. the mean daily percentage increase in opioid dosage) was also reported. Analyses adjusted for age (<70yrs vs. =70 yrs), renal function (e-GFR <60ml/min vs =60 ml/min) , gender and ongoing chemotherapy were also reported for the main tolerability and efficacy outcomes. All statistical analyses were performed using Stata 13 (StataCorp, College Station, TX); raw and adjusted RD were estimated the post-estimation command adjrr (24).

Additional data were provided as online Appendix with tables reporting tolerability and efficacy outcomes by treatment arm, assessment time and covariate subgroups (age and renal impairment).

RESULTS

From Sept14, 2006 to Dec 21, 2007, 187 eligible patients were randomly allocated either to CROM (95 patients) or CROO (92 patients) as shown in Figure1.

Participating centres enrolled a median of 11 patients (range 3-44). Table 1 reports similar background demographic and clinical characteristics for the two groups with the exception of a slightly higher amount of patients with KPS=70 in the CROO group.

Study compliance description

One patient in each arm did not receive the allocated intervention due to consent withdrawal, and 74 vs. 72 patients, 78% in both groups, completed the study (Figure 1). All hospitalizations and deaths (seven patients) were due to disease progression; protocol violations were due to cognitive impairment (two patients) and inability to take oral medications (two patients); one patient failed to fill-in baseline main outcome evaluation. Nine patients (three and six respectively with CROM and CROO) discontinued treatment due to AEs (Table 2); in four patients (two in both groups) the AEs were deemed “not related” with the study drug and in five cases (one with CROM and four with CROO) the relationship with the study drug was classified as “certain or probable”. The two patients who did not receive the allocated

intervention were excluded from the ITT analysis on the main outcome (Figure 1).

M

ANUS

CR

IP

T

AC

CE

PTE

D

(7)

Primary tolerability endpoint in the overall sample, by age and renal function.

The percentages of patients reporting an AE during treatment were 84.0% and 84.6% in the CROM and in the CROO groups respectively, unadjusted RD -0.6% (95%CI from -11.0% to 9.9%) (Figure 2, overall sample). Risk difference adjusted by centre of was not substantially different from the unadjusted estimate (RD -1.0%, 95%CI from -11.8% to 9.8% - data not reported in figure). Also sensitivity analyses for missing data imputation assumption in unadjusted analysis provided similar results: a RD of -0.6% (95%CI from -13.5% to 12.3%) with complete case analysis (N=74 and N=72 in CROM and CROO groups) and a RD of -1.5% (95%CI from -13.8% to 10.7%) with available data analysis (N=88 and N=85).

When examining the two models respectively adjusted by age (<70yrs vs. =70 yrs), and by renal function (e-GFR <60ml/min vs =60 ml/min), no evidence of

heterogeneity of treatment effect across subgroups was found: none of the two interaction parameters (treatment by age and treatment by renal function) resulted significant (p=0.11 for age and p=0.63 for renal function) and, consistently, 95%CIs of adjusted RDs both contained zero (Figure 2, renal impairment and age). RD was -6.8% (95%CI -21.2% to 7.5%) vs 15% (95%CI -3.3% to 33%) respectively in younger and older patients (Figure 2) and -6.2% (95%CI -18.5% to 5.9%) vs 1% (95%CI -24% to 26%) in patients with normal and moderately impaired renal function.

A final model adjusting for age, renal function, gender and ongoing chemotherapy on ITT population, provided slightly higher still non-significant adjusted RD (-4.6% , 95%CI from -15.3% to 6.2% - data not reported in figure). Due to missing in covariates, sample size for this last analysis is 83 and 84 respectively in CROM and CROO group.

Detailed adverse effect profile

Separate results for each AE are shown in Figure 3. Some symptoms were already present at baseline with no marked differences between the two arms (Figure 3-A). Figure 3-B shows the percentages of patients reporting baseline to follow-up worsening of at least two points by treatment group: worsening of nausea (15% vs. 21%, RD 95%CI -17% to 5%), constipation (25% vs. 35%, 95%CI -23% to 4%) and dry mouth (16% vs. 22%, RD 95%CI -18% to 5%) were numerically more frequent with CROO, whereas somnolence (35% vs. 32%, 95%CI RD -10% to 17%) with CROM. No patients experienced hallucination at baseline. Seven patients (four in CROM and three in CROO groups) had visual hallucinations which resolved spontaneously and did not require to stop treatment except in one patient (data not reported in Figure). In both groups, about two thirds of patients were prescribed laxatives, while antiemetics were used less frequently (around 40%) (Figure 4). Pain relief and opioid dosage

Average follow-up pain intensity score was 3.3 vs. 3.5 on a 0-10 NRS respectively for CROM and CROO, with a between group difference of –0.3 (95%CI -0.8 to 0.3) after adjusting for baseline pain intensity (Table 3). When adjusting also for age, renal function, gender and ongoing chemotherapy, as well as for baseline pain intensity, follow-up pain intensity difference between the two drugs was substantially the same as the one estimated unadjusted analysis (-0.4 and 95%CI -1.0 to 0.2, data not

reported in table); due to missing in covariates, sample size for this last analysis is 77 and 78 respectively in CROM and CROO groups. Average MEDD prescribed at

M

ANUS

CR

IP

T

AC

CE

PTE

D

(8)

baseline were comparable between CROM and CROO (30.9 mg and 36.6 mg

respectively); end of study average MEDD, including around the clock and as needed dosages, (53.9 mg for CROM and 69.8 mg for CROO) point to similar opioid dose escalation for the two drugs, with a between group difference of -1.8% (95%CI -5.4% to 1.7%) (Table 3).

DISCUSSION

The results of this open label multicenter RCT fail to show any difference in the risk of developing AEs and in analgesic effect between CROM to CROO when the used as first line WHO step III opioid for cancer pain.

To our knowledge this is the first study addressing tolerability as primary outcome in comparing CROO and CROM as first line opioid for cancer pain and also the first using an active surveillance of harms (7, 16). Heiskanen and Kalso (8) showed similar frequencies of AEs with the two drugs, although vomiting was more frequent

(p<0.01) with morphine, and constipation more common with oxycodone (p<0.01). Mucci-LoRusso et al. (9) reported side effects in 83% for oxycodone and 75% for morphine with a statistically significant lower incidence regarding itching with oxycodone (p<0.04) and 2 cases of hallucinations with morphine. No significant differences in adverse effects were reported by other authors (8-12).

Our study also shows that some symptoms were already present at baseline (Figure 3-A); the concurrent effect of antineoplastic therapies can cause side effects per se, by modifying the kinetics of opioids and/or adjuvant antiemetic medications; advanced disease and comorbidities may also concur to the etiology of several symptoms such as nausea or vomiting.. This confirms the importance of an active surveillance of harms and closely reflects clinical practice when adverse effects can have a multifactor aetiology(16).

The rate of AE worsening, around 85% in both groups, confirms previous data (25) and may have different explanations: active surveillance of potential AEs, low cut off used for the definition of AE (i.e. any symptom worsening =2 on any of the two follow-up assessments) and a broad definition of the tolerability outcome which did not consider the relationship of the symptom to the analgesic treatment. Yet, only 5 patients dropped out because of AEs judged as potentially related to the study drug. The use of opioids in patients with reduced renal function has recently been appraised in a systematic review(26) showing evidence of morphine metabolytes accumulation. Case series and pharmacodynamic models based on humane experimental data support possibility of an enhanced efficacy at lower morphine doses and increased risk of side effects(26, 27). At the same time the clinical impact of renal impairment in affecting different opioids toxicity is poorly documented(26).Also oxycodone has active metabolytes but their clinical effect and the eventual impact of renal

impairment are even less documented(26, 28) The effect of age on opiods

pharmacodynamics is also known(29). Elderly have been shown to be sensitive to relatively lower doses of morphine probably due to a combiniation of dynamics and kinetics factors(30).In our study, AEs were numerically more frequent with CROM than with CROO among elderly patients while no relevant differences emerged for patients with moderate renal impairment (Figure 2); also due to the reduced number of patients in the stratified groups, RD estimates by subgroup were imprecise (Figure 2), which advocates for further investigation of this debated issue.

This study has some limitations. It was stopped early because of slow recruitment rate but, since the reason for stopping was independent of trial findings, it is unlikely that bias was introduced in the study results(19). Conversely, a decrease in the estimates

M

ANUS

CR

IP

T

AC

CE

PTE

D

(9)

precision occurred because of the sample size reduction; nonetheless the sample size reached in our study is the second largest in RCTs comparing CROM to CROO(7). Patient attrition, 22% in both groups, was another potential limitation but it was lower than what commonly found in studies performed in palliative care(31) and worst case imputation, applied according to EMA indication for responder analyses reporting on AEs (21), allowed to carry out ITT analysis on 99% of enrolled patients. A further potential limitation of the study is given by the choice of an unidimensional main outcome summarizing effect on multiple symptoms. Although this may have had an impact on assay sensitivity of the trial, secondary tolerability analyses confirmed no statistically nor clinically meaningful differences on any of the AEs when analysed separately. Finally, the open-label design may have resulted in a biased estimate of the study endpoints; it is possible that knowledge of the treatment received has systematically altered patient-reported outcomes because of likely differences in the way morphine and oxycodone are perceived: morphine may be more known than oxycodone and more often associated with the negative attitudes surrounding the use of strong opioids. However this mirrors daily clinical practice in which patients do actually know the opioid they take. The choice of the open label design was also aimed at avoiding that complexity of the blinding procedures could interfere with normal practice and decreased the number of patients enrolled.

Rigorous clinical trials are an essential step in the development of evidence based palliative care, but difficulties in running clinical studies in this setting are well known and only partially overcome(32, 33).Although we designed our study to reflect clinical practice as much as possible we did not reach the desired sample size mainly because a number of potential participating centers actually did not take part into the study and because some smaller ones enrolled only a limited number of patients. This seems common in other recently published RCT(34, 35) and indeed several

consideration may apply in designing studies with shorter follow up periods and in enhancing the skills in screening the population of interest; however difficulties are also related to patient who are lost to follow up for reasons unrelated to the study conditions(31) and to the gate-keeping attitude of some health care providers who are often reluctant to refer advanced patients for research protocols(36). Although our results and results from previous studies, were inconclusive in showing superiority of any of the two drugs over the other, the evidence now available justifies the

hypothesis of equivalence between morphine and oxycodone for moderate to severe pain to be tested in future studies. Nonetheless, in the hypothesis of 85% AE

occurrence as in the present study, properly designed equivalence studies (alpha level 0.05, power 0.8 ) aimed at estimating between arm AE risk difference within an equivalence limit of 5% would require sample sizes of 2146 patients(37). While this is not to be considered a post hoc power calculation, it adds values to the publication of inconclusive results to encourage subsequent execution of data meta-analyses.

CONCLUSION

Our study does not bring conclusive results on the relative tolerability and efficacy of CROM and CROO mainly due to the insufficient sample size reached, to the potential bias introduced by the open-label design and to the difficulty in evaluating assay sensitivity of the trial. Also the hypothesis of a lower risk of toxicity with oxycodone among elderly patients and among those with moderate renal impairment warrants further investigations. These data, however, can significantly contribute to future meta-analyses comparing WHO-stepIII opioids and is relevant in designing future

M

ANUS

CR

IP

T

AC

CE

PTE

D

(10)

studies aimed at testing equivalence in efficacy and tolerability between the two drugs, potentially addressing specific side effects.

ACKNOWLEDGEMENTS

The study was supported in part by Mundipharma Pharmaceuticals, by Floriani Foundation - Milan (del.CDA 22/11/12) and by AIRC - Associazione Italiana per la Ricerca sul Cancro – (IG15314). The funders of the study had no role in study design, data collection, analysis and interpretation nor in writing of the report.

DISCLOSURE OF POTENTIAL CONFLICT OF INTEREST

EZ has received honoraria from Molteni Farmaceutici, Italfarmaco and Amgen. CB has served as a consultant for Molteni Farmaceutici and Mundipharma

Pharmaceuticals. PB has received honoraria from Molteni Farmaceutici and Italfarmaco. LM has served as a consultant or on advisory boards for Grunenthal, Angelini Pharma, Italfarmaco. GF has served as a consultant or on advisory boards for Janssen and Grunenthal and has participated in the speakers' bureau of Grunenthal, Pfizer and Jansen; he has had travel/accommodations expenses covered by

Grunenthal, Jansen, Pfizer, Astellas, Pharma and Abbvie. MTR has had

travel/accommodations expenses covered by Molteni Farmaceutici, Prostrakan, TEVA and Mundipharma Pharmaceuticals. GR has received honoraria from MSD, Italfarmaco and Teva; he has participated in the speakers' bureau of Teva, Italfarmaco and MSD. AC has received honoraria from Pfizer, Menarini and Grunenthal and has served as a consultant or on advisory boards for Helsin ; his organization has received research support from Amgen, TEVA Prostrakan and GW Pharmaceuticals. The remaining authors did not declared any potential conflict of interest.

REFERENCES

1. World Health Organization, ed. Cancer Pain Relief, Second Edition, with a Guide to Opioid Availability. Geneva: ; 1996.

2. Hanks GW, Conno F, Cherny N, et al. Morphine and alternative opioids in cancer pain: The EAPC recommendations. Br J Cancer. Scotland: Cancer Research

Campaign; 2001;84:587-593.

3. Caraceni A, Hanks G, Kaasa S, et al. Use of opioid analgesics in the treatment of cancer pain: Evidence-based recommendations from the EAPC. Lancet Oncol. England: Elsevier Ltd; 2012;13:e58-68.

4. Hamunen K, Paakkari P, Kalso E. Trends in opioid consumption in the nordic countries 2002–2006. European Journal of Pain. Elsevier; 2009;13:954-962.

5. Caraceni AT, Brunelli C, Rocco P, Minghetti P. Trends in opioid analgesics sales to community pharmacies and hospitals in italy (2000-2010). Minerva Anestesiol. Italy:2013;79:906-914.

6. International Narcotics Control Board. Technical Reports 2014. Available at:

http://www.incb.org/incb/en/narcotic-drugs/Technical_Reports/narcotic_drugs_reports.html. Accessed 4/25/2016, 2015. 7. Schmidt-Hansen M, Bennett MI, Arnold S, Bromham N, Hilgart JS. Oxycodone for cancer-related pain. Cochrane Database Syst Rev. England:2015;2:CD003870.

8. Heiskanen T, Kalso E. Controlled-release oxycodone and morphine in cancer related pain. Pain. Elsevier; 1997;73:37-45.

9. MucciLoRusso P, Berman BS, Silberstein PT, et al. Controlled release oxycodone compared with controlled release morphine in the treatment of cancer pain: A

M

ANUS

CR

IP

T

AC

CE

PTE

D

(11)

randomized, double blind, parallel group study. European Journal of Pain. Wiley Online Library; 1998;2:239-249.

10. Bruera E, Belzile M, Pituskin E, et al. Randomized, double-blind, cross-over trial comparing safety and efficacy of oral release oxycodone with controlled-release morphine in patients with cancer pain. J Clin Oncol. UNITED

STATES:1998;16:3222-3229.

11. Mercadante S, Tirelli W, David F, et al. Morphine versus oxycodone in pancreatic cancer pain: A randomized controlled study. Clin J Pain. LWW; 2010;26:794-797. 12. Riley J, Branford R, Droney J, et al. Morphine or oxycodone for cancer-related pain? A randomized, open-label, controlled trial. J Pain Symptom Manage. United States: . Published by Elsevier Inc; 2015;49:161-172.

13. Lauretti G, Oliveira G, Pereira N. Comparison of sustained-release morphine with sustained-release oxycodone in advanced cancer patients. Br J Cancer. Nature

Publishing Group; 2003;89:2027-2030.

14. Caraceni A, Pigni A, Brunelli C. Is oral morphine still the first choice opioid for moderate to severe cancer pain? A systematic review within the european palliative care research collaborative guidelines project. Palliat Med. England:2011;25:402-409. 15. Pigni A, Brunelli C, Caraceni A. The role of hydromorphone in cancer pain treatment: A systematic review. Palliat Med. England:2011;25:471-477.

16. Wiffen PJ, Derry S, Moore RA. Impact of morphine, fentanyl, oxycodone or codeine on patient consciousness, appetite and thirst when used to treat cancer pain. Cochrane Database Syst Rev. England:2014;5:CD011056.

17. Oosten AW, Oldenmenger WH, Mathijssen RH, van der Rijt, Carin CD. A systematic review of prospective studies reporting adverse events of commonly used opioids for cancer-related pain: A call for the use of standardized outcome measures. The Journal of Pain. Elsevier; 2015.

18. Kaasa S, Apolone G, Klepstad P, et al. Expert conference on cancer pain assessment and classification--the need for international consensus: Working

proposals on international standards. BMJ Support Palliat Care. England:2011;1:281-287.

19. Schulz KF, Altman DG, Moher D, CONSORT Group. CONSORT 2010 statement: Updated guidelines for reporting parallel group randomised trials. BMC Med. England:2010;8:18-7015-8-18.

20. Goodman SN, Berlin JA. The use of predicted confidence intervals when planning experiments and the misuse of power when interpreting results. Ann Intern Med. Am Coll Physicians; 1994;121:200-206.

21. European Medicines Agency. Guideline on Missing Data in Confirmatory Clinical Trials. Available at:

http://www.ema.europa.eu/docs/en_GB/document_library/Scientific_guideline/2010/ 09/WC500096793.pdf. Accessed 4/25/2016, 2010.

22. Levey AS, Stevens LA, Schmid CH, et al. A new equation to estimate glomerular filtration rate. Ann Intern Med. Am Coll Physicians; 2009;150:604-612.

23. Frison L, Pocock SJ. Repeated measures in clinical trials: Analysis using mean summary statistics and its implications for design. Stat Med. Wiley Online Library; 1992;11:1685-1704.

24. Norton EC, Miller MM, Kleinman LC. Computing adjusted risk ratios and risk differences in stata. Stata Journal. StataCorp LP; 2013;13:492-509.

25. Reid CM, Martin RM, Sterne JA, Davies AN, Hanks GW. Oxycodone for cancer-related pain: Meta-analysis of randomized controlled trials. Arch Intern Med.

American Medical Association; 2006;166:837-843.

M

ANUS

CR

IP

T

AC

CE

PTE

D

(12)

26. King S, Forbes K, Hanks GW, Ferro CJ, Chambers EJ. A systematic review of the use of opioid medication for those with moderate to severe cancer pain and renal impairment: A european palliative care research collaborative opioid guidelines project. Palliat Med. England:2011;25:525-552.

27. Sverrisdóttir E, Lund TM, Olesen AE, Drewes AM, Christrup LL, Kreilgaard M. A review of morphine and morphine-6-glucuronide’s pharmacokinetic–

pharmacodynamic relationships in experimental and clinical pain. European Journal of Pharmaceutical Sciences. Elsevier; 2015;74:45-62.

28. Samer CF, Daali Y, Wagner M, et al. The effects of CYP2D6 and CYP3A activities on the pharmacokinetics of immediate release oxycodone. Br J Pharmacol. Wiley Online Library; 2010;160:907-918.

29. Scott JC, Stanski DR. Decreased fentanyl and alfentanil dose requirements with age. A simultaneous pharmacokinetic and pharmacodynamic evaluation. J Pharmacol Exp Ther. UNITED STATES:1987;240:159-166.

30. Wilder-Smith OH. Opioid use in the elderly. European Journal of Pain. Wiley Online Library; 2005;9:137-140.

31. Currow DC, Plummer JL, Kutner JS, Samsa GP, Abernethy AP. Analyzing phase III studies in hospice/palliative care. A solution that sits between intention-to-treat and per protocol analyses: The palliative-modified ITT analysis. J Pain Symptom Manage. Elsevier; 2012;44:595-603.

32. Jordhoy MS, Kaasa S, Fayers P, Ovreness T, Underland G, Ahlner-Elmqvist M. Challenges in palliative care research; recruitment, attrition and compliance:

Experience from a randomized controlled trial. Palliat Med. ENGLAND:1999;13:299-310.

33. Stone PC, Gwilliam B, Keeley V, Todd C, Kelly LC, Barclay S. Factors affecting recruitment to an observational multicentre palliative care study. BMJ Support Palliat Care. England:2013;3:318-323.

34. Corli O, Floriani I, Roberto A, et al. Are strong opioids equally effective and safe in the treatment of chronic cancer pain? A multicenter randomized phase IV 'real life' trial on the variability of response to opioids. Ann Oncol. . Published by Oxford University Press on behalf of the European Society for Medical Oncology; 2016. 35. Bandieri E, Chiarolanza A, Luppi M, Magrini N, Marata AM, Ripamonti C. Prescription of opioids in italy: Everything, but the morphine. Ann Oncol. England:2009;20:961-962.

36. Kars MC, van Thiel GJ, van der Graaf R, Moors M, de Graeff A, van Delden JJ. A systematic review of reasons for gatekeeping in palliative care research. Palliat Med. 2015.

37. Julious SA, Campbell MJ. Tutorial in biostatistics: Sample sizes for parallel group clinical trials with binary data. Stat Med. Wiley Online Library; 2012;31:2904-2936. FIGURE LEGENDS

2 Karnofsky Performance Status

40-50 10 (11) 14 (16) 60-70 26 (28) 32 (36) 80-90 52 (56) 40 (45) 100 5 (5) 4 (4) Ongoing chemotherapy Yes 39 (41) 36 (39) No 56 (59) 56 (61)

Creatinine blood level (mg/dl)

Mean (SD) 0.93 (0.35) 0.93 (0.37)

Range 0.40 – 2.97 0.30 - 2.34

eGFR (ml/min)

Mean (SD) 81.9 (22.3) 83.5 (24.2)

Mini Mental State Examination

Mean (SD) 27.4 (2.5) 27.3 (3.1)

Anatomical pain site c

Lower back 11 (12) 25 (29)

Abdomen 20 (21) 21 (24)

Lower limb 7 (7) 3(4)

Thorax 25 (25) 11 (13)

ATC analgesic medication in the previous 24 h

None 5 (5) 4 (4)

WHO stepI (NSAIDs d / paracetamol) 15 (16) 11 (12)

WHO step II

Codeine 25(26) 33 (36)

Tramadol 50 (53) 44 (48)

(15)

M

AN

US

CR

IP

T

AC

CE

PT

ED

3 Tramadol - Mean (SD) 222.7 (87.5) 265.0 (137.9)

Rescue medication assumed in the previous 24 h

Yes 68 (72) 51 (56)

No 27 (28) 41 (44)

Adjuvant analgesic drugs c

Steroids 39 (41) 31 (34)

Anticonvulsants 13 (14) 9 (10)

Antidepressants 8 (8) 4 (5)

a

Digestive tract, liver, pancreas

b

Ovary, prostate, kidney, uterus, bladder, vulva

c

Multiple responses were possible

d

(16)

M

AN

US

CR

IP

T

AC

CE

PT

ED

1

Table 2 Patients withdrawn due to AE during the study period

Patient number Gender Age (years) Primary tumour SYMPTOMS at baseline KPS at baseline Previous opioid (doses) STUDY DRUG AE Relationship with study drug Daily dose at discontinuation Days of treatment before withdrawal 1 M 72 Unknown Pain 6/10 Somnolence 10/10 Confusion 2/10 Constipation 5/10 40 Tramadol 450 mg MORPHINE Confusion

hallucination NOT RELATED 20 mg 7 days

2 F 59 HCC Pain 8/10 Nausea 1/10 Somnolence 2/10 Xerostomia 5/10 90 Codeine 60 mg MORPHINE Nausea

Vomiting CERTAIN 20 mg 1 days

3 F 39 Colorectal Pain 5/10 Xerostomia 5/10 Stipsis 5/10 60 Codeine 120 mg MORPHINE Nausea

Vomiting NOT RELATED 40 mg 5 days

4 F 44 Colorectal Pain 10/10

Somnolence 2/10 40

Tramadol

300 mg OXYCODONE Somnolence PROBABLE 20 mg 2 days

5 M 72 Colorectal Pain 5/10 Nausea 5/10 Confusion 4/10 Somnolence 4/10 80 Tramadol 900 mg OXYCODONE Nausea

Vomiting NOT RELATED 20 mg 4 days

6 F 40 Ovary Pain 6/10 90 Tramadol

300 mg OXYCODONE Somnolence CERTAIN 20 mg 1 days

7 M 78 Colorectal

Pain 10/10 Stipsis 7/10 Xerostomia 10/10

40 Tramadol

300 mg OXYCODONE Somnolence PROBABLE 20 mg 5 days

8 M 68 Lung Pain 7/10 Somnolence 2/10 Constipation 4/10 Xerostomia 4/10 60 Tramadol 400 mg OXYCODONE Nausea Vomiting Somnolence PROBABLE 30 mg 7 Days 9 F 58 Breast Pain 8/10 Vomiting 2/10 80 Codeine 60 mg OXYCODONE Vomiting

(17)

M

AN

US

CR

IP

T

AC

CE

PT

ED

1 Table 3 Pain intensity and opioid dosages during follow-up.

CROM CROO Mean (95% CI) Mean (95% CI) Between group difference (95% CI) PAIN INTENSITY (0-10 NRS) N=88 N=85

Baseline pain intensity score 7.2 (6.9-7.5)

7.1 (6.8-7.4) Average follow-up pain intensity scores a 3.3

(2.9-3.6)

3.5 (3.1-3.9)

– 0.3 b (– 0.8 - 0.3) DAILY OPIOID DOSAGES (mg) N=89 N=84

MEDDc prescribed at baseline 30.9 (27.7-34.1)

36.6 (33.6-39.6) Average follow-up PRNd morphine 3.8

(2.3-5.3)

3.5 (1.7-5.3) End of study MEDDc 53.9

(44.4-63.4)

69.8 (58.3-81.4) OPIOID ESCALATION INDEX e 6.5%

(4.7% - 8.3%) 8.3% (5.1% - 11.5%) – 1.8% (–5.4% - 1.7%) a

Average of pain intensity scores at days 7 and 14

b

Adjusted for baseline pain intensity

c

MEDD: oral morphine equivalent daily dose

d

PRN: pro re nata, as needed

CE

PT

ED

1

TABLE A1: Adverse event occurrence (*) by treatment arm and time. N=187

DAY 7 DAY 14

MORPHINE OXYCODONE MORPHINE OXICODONE

NAUSEA NO 72 67 61 55 YES 16 18 14 17 missing 7 7 20 20 VOMITING NO 78 69 66 64 YES 10 15 9 8 missing 7 8 20 20 CONFUSION NO 70 71 67 61 YES 18 14 8 11 missing 7 7 20 20 HITCHING NO 82 78 72 68 YES 6 7 3 4 missing 7 7 20 20 CONSTIPATION NO 62 54 52 42 YES 25 31 23 30 missing 8 7 20 20 SOMNOLENCE NO 58 58 49 47 YES 30 27 26 25 missing 7 7 20 20 DRY MOUTH NO 67 65 61 56 YES 21 20 14 16 missing 7 7 20 20 HALLUCINATIONS(**) NO 85 83 72 71 YES 2 2 2 1 missing 8 7 21 20 (*)

Baseline follow-up worsening ≥ 2 on a 0-10 Numerical Rating Scale

(**)

(23)

M

AN

US

CR

IP

T

AC

CE

PT

ED

1

TABLE A2: Adverse event occurrence (*) by treatment arm, time and age. N=187.

DAY 7 DAY 14

AGE<70 years AGE≥70 years AGE<70 years AGE≥70 years

MORPHINE OXYCODONE MORPHINE OXYCODONE MORPHINE OXYCODONE MORPHINE OXYCODONE

NAUSEA NO ₅₄ ₄₇ ₁₈ ₂₀ ₄₇ ₄₀ ₁₄ ₁₅ YES ₁₄ ₁₅ ₂ ₃ ₁₁ ₁₃ ₃ ₄ missing ₅ ₄ ₂ ₃ ₁₅ ₁₃ ₅ ₇ VOMITING NO ₆₀ ₅₀ ₁₈ ₁₉ ₅₂ ₄₉ ₁₄ ₁₅ YES ₈ ₁₁ ₂ ₄ ₆ ₄ ₃ ₄ missing ₅ ₅ ₂ ₃ ₁₅ ₁₃ ₅ ₇ CONFUSION NO ₅₄ ₅₁ ₁₆ ₂₀ ₅₃ ₄₅ ₁₄ ₁₆ YES ₁₄ ₁₁ ₄ ₃ ₅ ₈ ₃ ₃ missing ₅ ₄ ₂ ₃ ₁₅ ₁₃ ₅ ₇ HITCHING NO ₆₃ ₅₈ ₁₉ ₂₀ ₅₆ ₅₁ ₁₆ ₁₇ YES ₅ ₄ ₁ ₃ ₂ ₂ ₁ ₂ missing ₅ ₄ ₂ ₃ ₁₅ ₁₃ ₅ ₇ CONSTIPATION NO ₅₀ ₄₀ ₁₂ ₁₄ ₄₃ ₃₂ ₉ ₁₀ YES ₁₇ ₂₂ ₈ ₉ ₁₅ ₂₁ ₈ ₉ missing ₆ ₄ ₂ ₃ ₁₅ ₁₃ ₅ ₇ SOMNOLENCE NO ₄₅ ₄₂ ₁₃ ₁₆ ₃₉ ₃₃ ₁₀ ₁₄ YES ₂₃ ₂₀ ₇ ₇ ₁₉ ₂₀ ₇ ₅ missing ₅ ₄ ₂ ₃ ₁₅ ₁₃ ₅ ₇ DRY MOUTH NO ₅₄ ₄₇ ₁₃ ₁₈ ₄₇ ₄₃ ₁₄ ₁₃ YES ₁₄ ₁₅ ₇ ₅ ₁₁ ₁₀ ₃ ₆ missing ₅ ₄ ₂ ₃ ₁₅ ₁₃ ₅ ₇ HALLUCINATIONS( **) NO 67 61 18 22 55 52 17 19 YES ₀ ₁ ₂ ₁ ₂ ₁ missing ₆ ₄ ₂ ₃ ₁₆ ₁₃ ₅ ₇ (*)

Baseline follow-up worsening ≥ 2 on a 0-10 Numerical Rating Scale (**)

(24)

M

AN

US

CR

IP

T

AC

CE

PT

ED

1

TABLE A3: Adverse event occurrence (*) by treatment arm, time and estimated the glomerular filtration rate (e-GFR). N=169.

DAY 7 DAY 14

e-GFR≥60 ml/min e-GFR<60 ml/min e-GFR≥60 ml/min e-GFR<60 ml/min

MORPHINE OXYCODONE MORPHINE OXYCODONE MORPHINE OXYCODONE MORPHINE OXYCODONE

NAUSEA NO ₅₃ ₅₁ ₁₁ ₉ ₄₅ ₄₁ ₁₀ ₈ YES ₁₂ ₁₄ ₂ ₄ ₁₁ ₁₆ ₂ ₁ missing ₄ ₆ ₂ ₁ ₁₃ ₁₄ ₃ ₅ VOMITING NO ₅₇ ₅₃ ₁₁ ₉ ₄₉ ₄₉ ₁₀ ₉ YES ₈ ₁₁ ₂ ₄ ₇ ₈ ₂ ₀ missing ₄ ₇ ₂ ₁ ₁₃ ₁₄ ₃ ₅ CONFUSION NO ₅₁ ₅₃ ₁₁ ₁₂ ₅₀ ₄₉ ₁₁ ₈ YES ₁₄ ₁₂ ₂ ₁ ₆ ₈ ₁ ₁ missing ₄ ₆ ₂ ₁ ₁₃ ₁₄ ₃ ₅ HITCHING NO ₅₉ ₅₉ ₁₃ ₁₂ ₅₄ ₅₃ ₁₁ ₉ YES ₆ ₆ ₀ ₁ ₂ ₄ ₁ ₀ missing ₄ ₆ ₂ ₁ ₁₃ ₁₄ ₃ ₅ CONSTIPATION NO ₄₇ ₄₀ ₇ ₇ ₄₀ ₃₃ ₇ ₄ YES ₁₇ ₂₅ ₆ ₆ ₁₆ ₂₄ ₅ ₅ missing ₅ ₆ ₂ ₁ ₁₃ ₁₄ ₃ ₅ SOMNOLENCE NO ₄₂ ₄₄ ₉ ₈ ₃₆ ₃₇ ₇ ₅ YES ₂₃ ₂₁ ₄ ₅ ₂₀ ₂₀ ₅ ₄ missing ₄ ₆ ₂ ₁ ₁₃ ₁₄ ₃ ₅ DRY MOUTH NO ₅₃ ₄₉ ₈ ₁₀ ₄₅ ₄₅ ₁₁ ₅ YES ₁₂ ₁₆ ₅ ₃ ₁₁ ₁₂ ₁ ₄ missing ₄ ₆ ₂ ₁ ₁₃ ₁₄ ₃ ₅ HALLUCINATIONS (**) NO 64 63 11 13 54 56 12 9 YES ₀ ₂ ₂ ₀ ₁ ₁ missing ₅ ₆ ₂ ₁ ₁₄ ₁₄ ₃ ₅ (*)

Baseline to follow-up worsening ≥ 2 on a 0-10 Numerical Rating Scale (**)

(25)

M

AN

US

CR

IP

T

AC

CE

PT

ED

1

TABLE A4 Efficacy and tolerability outcome intensities (0-10 NRS) by treatment arm and assessment time.

OUTCOMES EFFICACY

BASELINE DAY 7 DAY 14

PAIN MORPHINE mean 7.18 3.49 3.01 sd 1.31 2.07 1.96 N 95 88 75 OXYCODONE mean 7.16 3.55 3.50 sd 1.43 2.01 2.41 N 92 85 72 TOLERABILITY

NAUSEA MORPHINE mean 1.48 1.34 1.30 sd 2.23 1.80 1.52 N 94 89 76 OXYCODONE mean 1.20 1.68 1.69 sd 1.82 2.19 2.05 N 92 85 72

VOMITING MORPHINE mean 0.78 0.72 0.80 sd 1.81 1.41 1.55 N 94 89 76 OXYCODONE mean 0.62 1.05 0.74 sd 1.76 1.98 1.40 N 92 84 72

CONFUSION MORPHINE mean 0.44 1.09 0.66 sd 1.10 1.98 1.23 N 94 89 76 OXYCODONE mean 0.73 0.95 0.99 sd 1.43 1.68 1.80 N 92 85 72

HITCHING MORPHINE mean 0.21 0.35 0.22 sd 0.87 1.05 0.72 N 94 89 76 OXYCODONE mean 0.28 0.38 0.28 sd 1.19 1 0.81 N 92 85 72

(26)

M

AN

US

CR

IP

T

AC

CE

PT

ED

2

CONSTIPATION MORPHINE mean 2.80 2.90 2.83 sd 2.98 2.66 2.71 N 93 89 76 OXYCODONE mean 2.43 3.33 2.93 sd 2.85 2.97 2.57 N 92 85 72

SOMNOLENCE MORPHINE mean 1.33 2.34 2.12 sd 1.99 2.41 2.27 N 94 89 76 OXYCODONE mean 1.21 2.18 2.04 sd 1.80 2.07 2.36 N 92 85 72

DRY MOUTH MORPHINE

mean 2.66 3.11 3.11 sd 3.14 3.29 3.15 N 94 89 76 OXYCODONE mean 2.17 2.51 2.33 sd 2.60 2.58 2.47 N 92 85 72

(27)

M

AN

US

CR

IP

T

AC

CE

PT

ED

1

TABLE A5 Efficacy and tolerability outcome intensities (0-10 NRS) by treatment arm, time and age. OUTCOMES

AGE<70 YEARS AGE≥70 YEARS

EFFICACY

BASELINE DAY 7 DAY 14 BASELINE DAY 7 DAY 14

PAIN MORPHINE mean 7.21 3.41 3.07 7.09 3.75 2.83 sd 1.21 2.05 2.01 1.63 2.20 1.82 N 73 68 57 22 20 18 OXYCODONE mean 7.12 3.47 3.45 7.27 3.78 3.63 sd 1.38 2.18 2.58 1.59 1.48 1.92 N 66 62 53 26 23 19 TOLERABILITY

NAUSEA MORPHINE mean 1.49 1.56 1.38 1.43 0.62 1.06 sd 2.30 1.94 1.52 1.99 0.92 1.55 N 73 68 58 21 21 18 OXYCODONE mean 1.15 1.82 1.75 1.31 1.30 1.53 sd 1.87 2.21 2.06 1.72 2.14 2.06 N 66 62 53 26 23 19

VOMITING MORPHINE mean 0.84 0.74 0.74 0.57 0.67 1 sd 1.89 1.45 1.38 1.54 1.28 2.03 N 73 68 58 21 21 18 OXYCODONE mean 0.70 1.07 0.72 0.42 1.00 0.79 sd 1.82 2.01 1.39 1.60 1.95 1.47 N 66 61 53 26 23 19

CONFUSION MORPHINE mean 0.33 0.94 0.57 0.81 1.57 0.94 sd 0.96 1.79 1.13 1.47 2.50 1.51 N 73 68 58 21 21 18 OXYCODONE mean 0.65 0.84 0.91 0.92 1.26 1.21 sd 1.45 1.46 1.73 1.38 2.18 1.99 N 66 62 53 26 23 19

(28)

M

AN

US

CR

IP

T

AC

CE

PT

ED

2

HITCHING MORPHINE mean 0.25 0.35 0.14 0.10 0.33 0.50 sd 0.97 1.09 0.44 0.30 0.91 1.25 N 73 68 58 21 21 18 OXYCODONE mean 0.30 0.32 0.13 0.23 0.52 0.68 sd 1.35 0.95 0.52 0.65 1.12 1.25 N 66 62 53 26 23 19

CONSTIPATION MORPHINE mean 2.78 2.74 2.72 2.86 3.43 3.17 sd 3.10 2.86 2.81 2.57 1.86 2.38 N 72 68 58 21 21 18 OXYCODONE mean 2.36 3.16 2.68 2.62 3.78 3.63 sd 3.09 3.15 2.65 2.19 2.45 2.27 N 66 62 53 26 23 19

SOMNOLENCE MORPHINE mean 1.23 2.19 2.07 1.67 2.81 2.28 sd 1.84 2.32 2.27 2.46 2.69 2.35 N 73 68 58 21 21 18 OXYCODONE mean 1.11 1.98 1.96 1.46 2.70 2.26 sd 1.90 2.12 2.37 1.53 1.87 2.40 N 66 62 53 26 23 19

DRY MOUTH MORPHINE

mean 2.73 3.01 3.17 2.43 3.43 2.89 sd 3.26 3.29 3.14 2.75 3.34 3.27 N 73 68 58 21 21 18 OXYCODONE mean 1.98 2.48 2.13 2.65 2.57 2.89 sd 2.48 2.63 2.43 2.88 2.48 2.58 N 66 62 53 26 23 19

(29)

M

AN

US

CR

IP

T

AC

CE

PT

ED

1

TABLE A6 Efficacy and tolerability outcome intensities (0-10 NRS) by treatment arm, time and estimated the glomerular filtration rate (e-GFR).

OUTCOMES e-GFR≥60 ml/min e-GFR<60 ml/min

EFFICACY

PAIN MORPHINE mean 7.22 3.35 2.86 6.67 3.50 3.17 sd 1.30 1.84 1.58 1.18 3.12 2.86 N 69 65 56 15 12 12 OXYCODONE mean 6.97 3.49 3.40 7.71 4.23 4.22 sd 1.48 2.02 2.48 0.91 2.20 2.39 N 71 65 57 14 13 9 TOLERABILITY

NAUSEA MORPHINE mean 1.57 1.46 1.46 1.20 0.77 0.67 sd 2.38 1.88 1.63 1.57 1.30 1.07 N 69 65 56 15 13 12 OXYCODONE mean 1.28 1.77 1.93 0.71 1.85 1.22 sd 1.90 2.24 2.09 1.64 2.38 2.05 N 71 65 57 14 13 9

VOMITING MORPHINE mean 0.93 0.75 0.91 0.40 0.85 0.58 sd 1.97 1.45 1.64 1.55 1.57 1.51 N 69 65 56 15 13 12 OXYCODONE mean 0.66 1.06 0.82 0.71 1.54 0.67 sd 1.76 2.01 1.43 2.16 2.22 1.66 N 71 64 57 14 13 9

CONFUSION MORPHINE mean 0.38 0.95 0.61 0.53 1.23 0.50 sd 1.14 1.60 1.06 0.92 2.98 1.24 N 69 65 56 15 13 12 OXYCODONE mean 0.79 0.97 1.04 0.36 0.77 0.78 sd 1.52 1.79 1.95 0.74 1.17 1.20 N 71 65 57 14 13 9

(30)

M

AN

US

CR

IP

T

AC

CE

PT

ED

2

HITCHING MORPHINE mean 0.22 0.46 0.18 0.33 0.08 0.58 sd 0.89 1.20 0.47 1.05 0.28 1.51 N 69 65 56 15 13 12 OXYCODONE mean 0.37 0.42 0.33 0 0.38 0.11 sd 1.34 1.07 0.89 0 0.87 0.33 N 71 65 57 14 13 9

CONSTIPATION MORPHINE mean 2.68 2.68 2.63 3.33 4 3.92 sd 3.01 2.62 2.68 3.37 2.94 3.18 N 68 65 56 15 13 12 OXYCODONE mean 2.51 3.60 3.02 1.71 2.46 3.33 sd 2.90 3.12 2.64 2.84 2.44 2.65 N 71 65 57 14 13 9

SOMNOLENCE MORPHINE mean 1.26 2.26 2.13 1.87 2.92 2.50 sd 1.91 2.30 2.19 2.45 2.66 2.39 N 69 65 56 15 13 12 OXYCODONE mean 1.23 2.34 2.11 0.86 1.77 2.33 sd 1.78 2.23 2.44 1.75 1.36 2.50 N 71 65 57 14 13 9

DRY MOUTH MORPHINE

mean 2.52 3.02 3.07 3.40 3.38 2.83 sd 3.05 3.20 3.24 3.36 2.90 2.33 N 69 65 56 15 13 12 OXYCODONE mean 2.35 2.72 2.37 1.29 1.69 2.78 sd 2.78 2.73 2.52 1.73 2.02 2.68 N 71 65 57 14 13 9