Morphine-Methadone Opioid Rotation in Cancer Patients: Analysis of Dose Ratio Predicting Factors

Transcription

1 Vol. 37 No. 6 June 2009 Journal of Pain and Symptom Management 1061 Original Article Morphine-Methadone Opioid Rotation in Cancer Patients: Analysis of Dose Ratio Predicting Factors Miguel Angel Benítez-Rosario, MD, PhD, Antonio Salinas-Martín, MD, Armando Aguirre-Jaime, MSc, Lina Pérez-Méndez, MD, PhD, and Manuel Feria, DSc, PhD Palliative Care Unit (M.A.B.-R., A.S.-M.) and Research Unit (A.A.-J., L.P.-M.), NS Candelaria University Hospital, Canary Health Service; and Department of Pharmacology (M.A.B.-R., M.F.), Faculty of Medicine, University of La Laguna, Tenerife, Spain Abstract The dose ratio that is effective when switching opioid therapy from morphine to methadone in cancer patients varies widely. There are no conclusive data explaining the source of this variability. We analyzed 54 cancer patients undergoing opioid rotation to clarify those factors that influenced the morphine/methadone dose ratio (MMEDR) at Day 10 after the switch. Reasons for switching were uncontrolled pain (10 patients) or side effects (with or without pain, 44 patients). Initial MMEDR was 5:1 or 10:1 (82% or 18% of patients, respectively). Multivariate regression analysis was used to identify the demographic, cancerrelated, and treatment-related variables that were potential predictors of MMEDR. Median previous morphine dose for the entire sample was 220 mg/day (range: 30e1000 mg/day). The stable MMEDR median was 5:1 (range: 2:1e15:1). In the univariate analysis, reasons for opioid rotation, age, and previous morphine doses were associated with MMEDR. Multiple linear regression analysis showed that only the reason for switching (pain vs. side effects; P < 0.001) and previous morphine doses (lower vs. upper to 300 mg/day; P < 0.001) were associated with MMEDR. From this analysis, the MMEDRs for patients rotated for side effects at 300 mg/day or more or less than 300 mg/day of morphine were 9.1:1 or 5.6:1, respectively, and the MMEDRs for those switched for pain at 300 mg/day or more or less than 300 mg/day of morphine were 4.9:1 or 3:1, respectively. Both the reasons for opioid rotation and previous morphine doses are predictive factors and should be used to select the MMEDR more accurately. J Pain Symptom Manage 2009;37:1061e1068. Ó 2009 U.S. Cancer Pain Relief Committee. Published by Elsevier Inc. All rights reserved. Key Words Morphine, methadone, opioid rotation, predicting factors Address correspondence to: Miguel Angel Benítez- Rosario, MD, PhD, Department of Pharmacology, Faculty of Medicine, University of La Laguna, La Laguna, Tenerife, Spain. mabenros@gmail.com Accepted for publication: June 16, Ó 2009 U.S. Cancer Pain Relief Committee Published by Elsevier Inc. All rights reserved. Introduction In the pharmacological treatment of cancer pain, opioids usually provide suitable analgesia for most patients without serious side effects. However, in the course of an advanced cancer disease, the type of opioid and/or the route of /09/$esee front matter doi: /j.jpainsymman

2 1062 Benítez-Rosario et al. Vol. 37 No. 6 June 2009 administration needs to be changed one or more times to improve pain control and/or to reduce opioid toxicity. Opioid rotation, the practice of converting a patient from one opioid to another, is aimed at finding the most favorable balance between analgesia and side effects. 1e4 Inadequate analgesia, opioid neurotoxicity, and nausea and vomiting are the main indications for opioid rotation. Several reports have underlined the effectiveness of opioid rotation to methadone in cancer patients receiving treatment with high doses of other opioids. 1,2,5e7 Methadone is an attractive alternative opioid because of its analgesic efficacy, good oral bioavailability, and lack of known neuroactive metabolites. Unfortunately, the substitution of methadone for another opioid is not simple. Clinical experience shows that methadone has large interindividual variability in pharmacokinetics, a potential for delayed toxicity and, above all, a dose ratio when switching from morphine that varies widely between 16:1 and 2.5:1. 1,2,5e13 This variability is associated with the extent of previous exposure to opioids, total opioid dose before rotation, and drug interactions. 1,2,5,7,13 Several patient-specific characteristics, treatment-related factors, disease status, and type of pain have been suggested to be influencing opioid responsiveness. 14e22 However, to the best of our knowledge, the influence of these factors on morphine-methadone dose ratio (MMEDR) has not been established. The aim of this study was to explore the influence of the patient, the cancer, and the opioid treatment characteristics on MMEDR in a cohort of advanced cancer patients who were rotated because of severe pain, morphine side effects, or both. Patients and Methods We evaluated potential predictive factors of MMEDR in a cohort of cancer patients who underwent opioid rotation because of uncontrolled pain and/or morphine side effects, specifically neurotoxicity. Uncontrolled pain was considered when patients reported severe pain on a four-level verbal rating scale from none to severe, despite an escalation in morphine doses of more than 300% over two weeks. Neurotoxicity to morphine was considered in the presence of myoclonus, hallucinations, or delirium, or a combination of these. For the objectives of this study, we have only used data from the patients who had a successful opioid rotation with a stable methadone dose (no changes for 72 hours) at Day 10 after switching. All opioid rotations were performed from oral morphine to oral methadone. Successful control of pain was inferred when patients reported no pain or mild pain for at least three consecutive days and the use of none or one supplemental ( rescue ) opioid dose per 24 hours. Recovery from morphine side effects was established when myoclonus and hallucinations were not present, determined by means of patient self-report and clinical examination, or delirium resolved according to the following criteria: normal psychomotor activity, recovery in attention tests, and an increase in Mini-Mental State Examination score to 24 points or more. A 10-day period was selected as enough time to evaluate opioid rotation efficacy on pain control and side effect improvement without methadone toxicity. All patients remained admitted for the entire duration of opioid rotation. Patients receiving anticancer treatment (radiotherapy, chemotherapy, or both) for at least two weeks before switching were excluded from the study. Those with a Karnofsky Performance Index of less than 40% also were excluded because of difficulties in evaluating their response to opioid rotation. Opioid Rotation Protocol Morphine administration was stopped and methadone (administered every eight hours) was commenced. The initial daily methadone dose was usually calculated by using a MMEDR of 5:1. A ratio of 10:1 was used in those patients who, on admission, 1) were receiving a morphine dose of more than 600 mg/day; 2) were anxious, depressed, or delirious, and presented a recent history of a rapid increase in morphine doses to control pain; or 3) had a possible low morphine clearance because of renal impairment (we established a serum creatinine of 1.5 mg/dl or more as a criterion). The starting methadone dose was titrated upward or downward every 48e72 hours by 30%e50%, depending on clinical criteria and patient satisfaction with pain relief and

3 Vol. 37 No. 6 June 2009 Predicting Morphine-Methadone Dose Ratio 1063 side effects. Rescue doses of 10% of the daily methadone dose were given as needed every two hours. Adjuvant drugs normally used for symptom control (e.g., neuroleptics, antiepileptics, corticosteroids, laxatives, and others) were continued at the same dose. Informed consent for opioid rotation was obtained from all patients or, in the case of delirious patients, their relatives. Throughout the rotation, the patients underwent clinical assessment by one or more of the full-time palliative care unit s experienced physicians and attending palliative nurses. Pain intensity and opioid-related adverse effects were assessed on a daily basis. This assessment was only performed for a continuous evaluation of pain and other symptoms, with the aim of adjusting the methadone doses, and it was not included in data analysis. Data Collection The following data were collected for each patient: age, gender, primary tumor site, presence of bone and/or hepatic metastases, renal function, hematological-biochemical parameters (hemoglobin, lymphocyte count, serum protein levels, liver transaminases), Karnofsky Index, time of exposure to morphine, morphine dose at time of substitution, reason for opioid rotation, and adjuvant drugs used during opioid switching. Data Analysis For the purpose of the study, we considered the MMEDR at Day 10 as the outcome. To determine which variables were candidates for MMEDR explanatory factors in multiple linear regression modeling, their association with MMEDR was previously explored in a univariate way by Mann-Whitney U rank difference and Spearman rank correlation tests. Previous oral morphine doses were considered, for the analysis, as both continuous and categorical variables (#90 vs. >90 mg/day and #300 vs. >300 mg/day, following two of the cutoff points suggested by Ripamonti et al. 7 ). Moreover, patients were grouped according to the reason for rotation (pain vs. side effects). The pain group included the patients rotated for uncontrolled pain. The side effect group included those who were rotated for morphine neurotoxicity in the presence or the absence of pain. MMEDR at Day 10 was logarithmically transformed to fulfill the conditions for linear modeling. The results of this modeling were then antilogarithmically transformed to calculate the predicted MMEDR, expressed in its habitual values. All the tests were performed at 0.05 significance levels. SPSS version 13.0 (SPSS Inc., Chicago, IL) was used for statistical analysis. Results Between June 2000 and 2005, 980 advanced cancer patients were admitted to the Palliative Care Unit, Hospital La Candelaria, because of severe symptom distress requiring medical intervention. Seventy-four of the 980 patients underwent opioid rotation from morphine to methadone. Fifty-four (72.9%) achieved adequate control of pain and morphine side effects with stable methadone doses at Day 10 after switching, and were included in the study. Tables 1 and 2 summarize the most relevant characteristics of these patients and reasons for opioid rotation, respectively. Median previous morphine dose for the whole sample was 220 mg/day (range: 30e1000 mg/day). Patients rotated for pain had a significantly higher morphine dose than those rotated for side effects (475 mg/ day; range: 180e800 vs. 137 mg/day; range: 30e1000, P ¼ 0.02), and initial MMEDRs of 5:1 and 10:1 were applied in 44 and 10 Characteristics Table 1 Patient Characteristics Summary Sample size 54 Age (years) (median, range) 64 (31, 84) Gender (male/female) 31 (57%)/3 (43%) Primary tumor Genitourinary 17 (32%) Gastrointestinal 13 (24%) Lung 12 (22%) Breast 4 (7%) Skin 2 (4%) Others 6 (11%) Metastases Bone 27 (50%) Hepatic 27 (50%) Karnofsky Index (median, 60 (40, 90) range) Opioid treatment duration 44 (20, 240) (days) (median, range) Previous morphine doses 220 (30, 1000) (mg/day) (median, range)

4 1064 Benítez-Rosario et al. Vol. 37 No. 6 June 2009 Reasons Table 2 Reasons for Rotation and Pain Level Baseline Pain Level a No. of Patients (n ¼ 54) Pain 3 10 Side effects pain 44 Delirium 0 13 Myoclonus and/or 2 10 hallucinations plus pain Delirium plus b 9 myoclonus plus pain Myoclonus and/or hallucinations 0 12 a According to a verbal pain scale from 0 (no pain) to 3 (severe). b Pain intensity not assessed because of delirium. patients, respectively. The stable methadone doses ranged from 6 to 180 mg daily (median: 45 mg/day). There was a positive correlation between previous morphine and final methadone doses (Spearman rho ¼ 0.764; P < ). Median MMEDR was 5:1 and ranged from 2:1 to 15:1. Table 3 shows the results of the correlation estimation between age, hematological and biochemical parameters, and length of opioid treatment. None of the variables was significantly associated with MMEDR. Figure 1 shows that MMEDR was associated with previous morphine doses. However, the correlation between these variables was weak (Spearman rho ¼ 0.392; P ¼ ) and, for this reason, previous morphine doses expressed as a continuous variable were excluded from the multivariate analysis. The analysis of the association of MMEDR with other patient and treatment characteristics, Table 3 Association of MMEDR with Patients and Opioid Treatment Characteristics Characteristics P-value a Age (years) 0.64 Hemoglobin (g/l) 0.81 Lymphocyte count 0.66 Urea (mg/dl) 0.74 Creatinine (mg/dl) 0.08 ALT (IU/L) 0.92 AST (IU/L) 0.85 GGT (IU/L) 0.95 Total protein (g/l) 0.59 Length of opioid treatment (days) 0.71 ALT ¼ alanine transaminase; AST ¼ aspartate transaminase; GGT ¼ gamma glutamyl transpeptidase. a Estimated by Spearman rank correlation test. Fig. 1. Correlation between MMEDR on Day 10 and the morphine dose administered before the switch. Each dot represents one patient (n ¼ 54 patients). Spearman rho is correlation coefficient. Solid and dashed lines represent best-fit straight line and 95% confidence intervals, respectively. expressed as categorical variables, is summarized in Table 4. MMEDR was not associated with gender, the presence of hepatic and bone metastases, the use of adjuvant drugs, or the Table 4 Differences in MMEDR According to Patient and Treatment Characteristics Characteristics a MMEDR (median ranges) P-value b Gender Male (31) 5: 1 (10:1e2:1) 0.19 Female (23) 5: 1 (16:1e2:1) Age (years) #65 (30) 5:1 (14:1e2:1) 0.04 >65 (24) 5:1 (16:1e2:1) Hepatic metastases Yes (27) 5:1 (16:1e2:1) No (27) 5:1 (10:1e3:1) 0.60 Bone metastases Yes (27) 5:1 (15:1e2:1) No (27) 5:1 (16:1e2:1) 0.96 Metabolism inducers Yes (30) 5:1 (16:1e2:1) No (24) 5:1 (10:1e3:1) 0.60 Steroid treatment Yes (23) 5:1 (16:1e2:1) No (31) 5:1 (15:1e2:1) 0.36 Rotation reason Pain (10) 5:1 (10:1e2:1) Side effects (44) 5:1 (16:1e2:1) 0.01 Previous morphine doses (mg/day) #90 (11) 5:1 (12:1e2:1) 0.07 >90 (43) 5:1 (16:1e2:1) #300 (31) 5:1 (12:1e2:1) 0.02 >300 (23) 5:1 (15:1e3:1) a Numbers in parentheses are the number of patients. b Mann-Whitney U rank test.

5 Vol. 37 No. 6 June 2009 Predicting Morphine-Methadone Dose Ratio 1065 co-administration of methadone metabolism inducers. Reasons for opioid rotation, previous morphine doses and age were significantly associated with MMEDR and were included as potential predictive factors in the multivariate linear regression analysis. The results of the multivariate linear regression analysis (Table 5) show that only reasons for rotation (pain vs. side effects) and previous morphine doses (lower vs. upper to 300 mg/ day) were associated with MMEDR. These variables explained up to 93% of the MMEDR variability. Applying the results of the final regression model, after antilogarithmic transformation, the predicted dose ratio between oral morphine and oral methadone, with regard to the aforementioned predictive factors, can be obtained from the mathematical expression MMEDR ¼ RRþ0.21PMD. In the expression, RR is the reason for rotation, codified as 1 for pain and 2 for side effects, and PMD is the previous morphine dose, codified as 1 for 300 mg/day or less and 2 for more than 300 mg/day. Table 6 describes the predicted MMEDR in connection with the four possible combinations of the predictive factors. It is noteworthy that the highest predicted MMEDR (9.1:1) is associated with side effects as the reason for rotation and previous morphine of more than 300 mg/day, whereas the lowest MMEDR (3:1) is associated with pain as the reason for rotation and previous morphine of 300 mg/day or less. Discussion In the current study, we analyzed several potential predictive factors of the MMEDR Table 5 Results of Multivariate Linear Association Analysis (Enter Method) for MMEDR at Day 10 as Dependent Variable Factor b (95% CI) P-value Age 0.01 ( 0.04, 0.004) 0.97 Previous morphine doses 0.21 (0.11, 0.30) (lower vs. higher than 300 mg/day) Reasons for rotation (pain 0.27 (0.17, 0.38) vs. side effects) Adjusted R 2 ¼ 0.93 b ¼ standardized regression coefficient; CI ¼ confidence interval. Constant was removed in the model, because it was not statistically significant. Table 6 MMEDR Predicted from Multivariate Linear Regression Analysis According to Predicting Factors a Reason for Rotation and Previous Morphine Doses MMEDR 95% CI Side effects and >300 mg/day 9.1:1 22.9:1, 3.6:1 Pain and >300 mg/day 4.9:1 9.5:1, 2.5:1 Side effects and #300 mg/day 5.6:1 11.5:1, 2.8:1 Pain and #300 mg/day 3:1 4.8:1, 1.9:1 CI ¼ confidence interval. a Results are expressed in the habitual MMEDR scale after the antilogarithmic transformation of the final adjusted model. achieved during successful opioid rotation in cancer patients. The multivariate linear regression analysis showed that the reason for opioid rotation and previous morphine dose were significantly associated with MMEDR at Day 10 post-switching. In contrast, neither age, gender, several characteristics of cancer, and its treatment (such as the duration of previous exposure to opioids or the use of adjuvant drugs), nor hematological and biochemical parameters were associated with MMEDR. Univariate analysis of our data showed that only age, reason for rotation, and previous morphine dose were significantly associated with MMEDR. With regard to age, most of the studies suggest an inverse association between patient age and opioid dose, 15,18,20,23,24 although two studies failed to confirm this. 17,25 To the best of our knowledge, no research has been carried out on the influence of age on the conversion ratio from morphine to methadone. Both opioids are affected differently by the pharmacodynamic and pharmacokinetic changes associated with age. Thus, although morphine requirements seem to be inversely related to age, 18 the therapeutic response to methadone appears to be age-independent; 26,27 therefore, it is difficult to anticipate how MMEDR is affected by age. In our series, age lost significance as an independent predictor of MMEDR when it was introduced in the multiple regression analysis. Nevertheless, we are aware of the limitations of ruling out an association between age and MMEDR because of our small sample size. According to published data, the dose ratio between morphine and methadone ranges from 16:1 to about 2.5:1. 6e8 This variability has led to the proposal of several approaches

6 1066 Benítez-Rosario et al. Vol. 37 No. 6 June 2009 about how to do the rotation to methadone that, essentially, take into account previous morphine doses to select the most appropriate conversion ratio. Lawlor et al., 6 starting from a fixed 10:1 MMEDR, reported a widely variable (16.27:1e5.98:1) dose ratio following post-rotation stabilization. Data from one prospective study showed that, starting on a fixed 5:1 MMEDR, up to 71% of rotated patients need a change in the methadone dose during the first three days of rotation. 9 Using a different strategy, that is, starting the rotation on a variable (8:1e4:1) MMEDR, selected according to a range of previous morphine doses from 30 to 300 mg/day, Ripamonti et al. 7 achieved a final equianalgesic dose ratio ranging from 14.3:1 to 2.5:1. Finally, using an initial 12:1e4:1 MMEDR, Mercadante et al. 9 reported the need to increase the methadone dose in 50% of the patients. Thus, the final effective MMEDR, that is, the one which completely restores pain control or solves a side effect, always varies widely, and this is true regardless of the method of choosing the initial conversion ratio, whether it is more or less empirical or it takes previous morphine doses into account. The aforementioned data, together with present results, show that selecting a certain initial dose ratio, based only on previous morphine doses, is useful at best as a conservative strategy, to reduce the risk of methadone overdosing, but it cannot predict the effective final MMEDR. As regards the reasons for rotation, our multivariate linear regression model predicted quite different MMEDR according to whether patients were rotated for side effects, in the presence or the absence of pain, or for uncontrolled pain alone. This fact was especially evident when switching took place from previous morphine doses over 300 mg/day (9.1:1, when the reason for rotation was side effects, and 4.9:1, when it was pain), but it was also noticeable with previous morphine doses of less than 300 mg/day (5.6:1 vs. 3:1, for side effects or pain as reason for rotation, respectively). To the best of our knowledge, no other study has described this different behavior of MMEDR depending on reasons for switching. However, our results must be taken with care, because the number of patients switched to methadone for uncontrolled pain was comparatively lower. When performing an opioid rotation, the wide variability in MMEDR could be attributed to morphine, to methadone, or to a complex interaction between the two, in the context of a difficult pain and a progressing cancer disease. Regarding morphine, it has been suggested that the clearance of morphine- 3-glucuronide, a metabolite of morphine that could have proalgesic activity facilitated by opioid rotation, could play a role in the lower post-switching requirements of methadone. 28 Furthermore, chronic opioid use has been reported to induce a hyperalgesic state, both in cancer and noncancer chronic pain. 29,30 Although no definitive explanation has been advanced for the morphine-induced hyperalgesia, several mechanisms of central sensitization putatively involved in this phenomenon have been identified (for a review, see Ossipov et al. [2004] 31 and Koppert [2007] 32 ). Whatever is the mechanism of chronic morphine hyperalgesia, it would presumably alter the potency relation between morphine and methadone and, consequently, their relative dose ratio. With regard to methadone, its involvement in dose ratio variability has been attributed to several factors, mostly related with its pharmacodynamic and pharmacokinetic characteristics. Among them are the presence of an incomplete cross-tolerance with previous opioids, which unexpectedly makes methadone more potent than anticipated, 6,7,33 a nonopioid antagonism of N-methyl-D-aspartate receptor by methadone, 34 and its pharmacokinetics, characterized by large individual variation. 35 Obviously, the aforementioned mechanisms are unlikely to be proved in the palliative setting but, conceivably, a combination of several of them could explain both the high potency and efficacy of methadone in patients under chronic morphine treatment, and the wide dose ratio variability when performing an opioid rotation from morphine to methadone. In the meantime, in palliative care, the decisions about MMEDR must be taken by using clinical criteria and all the information available. The European Working Group of the European Association for Palliative Care, 8 using data obtained from a univariate analysis of a small sample of patients, recommends selecting a variable dose ratio, from 12:1 to 4:1 and higher, according to previous oral morphine doses from

7 Vol. 37 No. 6 June 2009 Predicting Morphine-Methadone Dose Ratio 1067 more than 300 to 30 mg/day, respectively. Our data suggest that it should be possible to refine the aforementioned recommendations by considering, in addition, the reason for rotation (Table 6). We are aware of the limitations of this study, especially those related to the small size of our sample, but in the palliative setting, it is difficult to conduct prospective opioid rotation studies in large and homogeneous populations. Further research is needed to ascertain the reliability of the MMEDR predicted from our study. References 1. Mercadante S. Opioid rotation for cancer pain: rationale and clinical aspects. Cancer 1999;86: 1856e Mancini I, Lossignol DA, Body JJ. Opioid switch to oral methadone in cancer pain. Curr Opin Oncol 2000;12:308e Quigley C. Opioid switching to improve pain relief and drug tolerability. Cochrane Database Syst Rev 2004;3. CD Morita T, Takigawa C, Onishi H, et al. Japan Pain, Rehabilitation, Palliative Medicine, and Psycho-Oncology (PRPP) Study Group. Opioid rotation from morphine to fentanyl in delirious cancer patients: an open-label trial. J Pain Symptom Manage 2005;30:96e Ripamonti C, Bianchi M. The use of methadone for cancer pain. Hematol Oncol Clin North Am 2002;16:543e Lawlor PG, Turner KS, Hanson J, Bruera ED. Dose ratio between morphine and methadone in patients with cancer pain. A retrospective study. Cancer 1998;82:1167e Ripamonti C, Groff L, Brunelli C, et al. Switching from morphine to oral methadone in treating cancer pain: what is the equianalgesic dose ratio? J Clin Oncol 1998;16:3216e Cherny N, Ripamonti C, Pereira J, et al. Strategies to manage the adverse effects of oral morphine: an evidence-based report. J Clin Oncol 2001;19: 2542e Mercadante S, Cassucio A, Fulfaro F, et al. Switching from morphine to methadone to improve analgesia and tolerability in cancer patients: a prospective study. J Clin Oncol 2001;19:2898e Bruera E, Sweeney C. Methadone use in cancer patients with pain: a review. J Palliat Med 2002;5: 127e Indelicato RA, Portenoy RK. Opioid rotation in the management of refractory cancer pain. J Clin Oncol 2003;21(Suppl 9):87e Pereira J, Lawlor P, Vigano A, Dorgan M, Bruera E. Equianalgesic dose ratios for opioids. A critical review and proposals for long-term dosing. J Pain Symptom Manage 2001;22:672e Soares LGL. Methadone for cancer pain: what have we learned from clinical studies? Am J Hosp Palliat Med 2005;22:223e Kaiko RF, Wallenstein SL, Rogers AG, Houde RW. Sources of variation in analgesic responses in cancer patients with chronic pain receiving morphine. Pain 1983;15:191e Hall S, Gallagher RM, Gracely E, Knowlton C, Wescules D. The terminal cancer patient. Effects of age, gender, and primary tumor site on opioid dose. Pain Med 2003;4:125e Cepeda MS, Carr DB. Women experience more pain and require more morphine than men to achieve a similar degree of analgesia. Anesth Analg 2003;97:1464e Mercadante S, Cassucio A, Pumo S, Fulfaro F. Factors influencing the opioid response in advanced cancer patients with pain followed at home: the effects of age and gender. Support Care Cancer 2000;8:123e Viganó A, Bruera E, Suarez-Almazor ME. Age, pain intensity, and opioid dose in patients with advanced cancer. Cancer 1998;83:1244e Unruh AM. Gender variations in clinical pain experience. Pain 1996;65:123e Rees WD. Opioid needs of terminal care patients: variations with age and primary site. Clin Oncol (R Coll Radiol) 1990;2:79e Klepstad P, Dale O, Kaasa S, et al. Influences on serum concentrations of morphine, M6G and M3G during routine clinical drug monitoring: a prospective survey in 300 adult cancer patients. Acta Anaesthesiol Scand 2003;47:725e Hanks GW, Reid C. Contribution to variability in response to opioids. Support Care Cancer 2005; 13:145e Ventafridda V, Oliveri E, Caraceni A, et al. A retrospective study on the use of oral morphine in cancer pain. J Pain Symptom Manage 1987;2:77e Bercovitch M, Waller A, Adunsky A. High dose morphine use in the hospice setting. A database survey of patient characteristics and effect of life expectancy. Cancer 1999;86:871e Ventafridda V, Saita L, Barletta L, Sbanotto A, De Conno F. Clinical observations on controlledrelease morphine in cancer pain. J Pain Symptom Manage 1989;4:124e Mercadante S, Cassucio A, Agnello A, Barresi L. Methadone response in advanced cancer patients

8 1068 Benítez-Rosario et al. Vol. 37 No. 6 June 2009 followed at home. J Pain Symptom Manage 1999;18: 188e De Conno F, Groff L, Brunelli C, et al. Clinical experience with oral methadone administration in the treatment of pain in 196 advanced cancer patients. J Clin Oncol 1996;14:2836e Daeninck PJ, Bruera E. Opioid use in cancer pain. Is a more liberal approach enhancing toxicity? Acta Anaesthesiol Scand 1999;43:924e Mercadante S, Ferrera P, Villari P, Arcuri E. Hyperalgesia: an emerging iatrogenic syndrome. J Pain Symptom Manage 2003;26:769e Chu LF, Clark DJ, Angst MS. Opioid tolerance and hyperalgesia in chronic pain patients after one month of oral morphine therapy: a preliminary prospective study. J Pain 2006;7:43e Ossipov MH, Lai J, King TW, et al. Antinociceptive and nociceptive actions of opioids. J Neurobiol 2004;61:126e Koppert W. Opioid-induced hyperalgesiad pathophysiology and clinical relevance. Acute Pain 2007;9:21e Ripamonti C, DeConno F, Groff L, et al. Equianalgesic dose/ratio between methadone and other opioid agonists in cancer pain: comparison of two clinical experiences. Ann Oncol 1998;9:79e Gorman AL, Elliott KJ, Inturrisi CE. The D- and L-isomers of methadone bind to the non-competitive site on the N-methyl-D-aspartate (NMDA) receptor in rat forebrain and spinal cord. Neurosci Lett 1997;223:5e Eap CB, Buclin T, Baumann P. Interindividual variability of clinical pharmacokinetics of methadone. Clin Pharmacokinet 2002;41:1153e1193.