A Comparative Study of Opioid Switching to Methadone for Cancer Pain Control in Successful and Unsuccessful Cases

Introduction

Opioid switching from other opioids to methadone often gives good control of refractory cancer pain.^1,2 As there have been few clinical reports^3–5 in Japan, only oral administration of methadone has been approved for patients with cancer pain since March 2013. Methadone exhibits complex pharmacokinetics with interindividual differences and rare but serious adverse effects. Then the Ministry of Health in Japan strictly limits administration of methadone. The refractory cancer pain for indication of methadone is uncontrolled by other strong opioids, which were prescribed oral morphine equivalent daily dose (MEDD) about 60 mg or more, and there are probable difficulties in the relief of pain even if other opioids are increased. But as methadone switching failed in some cases, we evaluate the differences in clinical features of successful cases (SCs) and unsuccessful cases (UCs) in switching to methadone.

Patients and Methods

Switching to oral methadone from other strong opioids was performed in 87 patients at Takarazuka Municipal Hospital between April 2013 and March 2017. All the patients had histological diagnoses of malignancy.

Among the 87 cases, 7 were excluded from the analysis because methadone administration was stopped due to vomiting or self-cessation within six days after switching. Seven days are required to evaluate the success of dose titration. The clinical features of the remaining 80 cases were evaluated. In all cases (14 outpatients and 66 inpatients), methadone was initiated from other opioids due to refractory pain using stop-and-go switching, in which the current opioid is immediately replaced by methadone. As indicated in Table 1, the starting dose of methadone was 15, 30, or 45 mg/day depending on the previous opioid dose, which was calculated as MEDD of the opiate, with a further appropriate increase or decrease of the dose for each case. MEDD was calculated as prescribed by the National Comprehensive Cancer Network. ⁶ The calculated dose of methadone was divided into three administrations per day. Previously administered adjuvant drugs (corticosteroids, neuroleptics, and antiepileptics) and nonopioid analgesics for relief of pain were, in principle, continued at the same doses during the titration. Anticancer therapy was also continued in a few cases.

As instructed by the Ministry of Welfare in Japan, we did not increase the methadone dose for one week after the first dose and did not use methadone as a rescue dose because it takes about one week for the blood concentration of methadone to stabilize. A short-acting opioid was used for rescue administration (Table 1). Eastern Cooperative Oncology Group (ECOG) performance status (PS) scores ⁷ and palliative prognostic index (PPI) scores ⁸ were assessed before switching. Pain was assessed and recorded by the nurse using a numerical rating score (NRS), in which patients were asked to rate their pain on a scale from 0 to 10. ⁹ In the Japanese definition, the success of dose titration is determined on the day after methadone has been administered for more than six days at the same dose with acceptable adverse effects. The efficacy parameters are as follows: rescue administration has been used less than twice in a day and NRS has decreased from baseline. Based upon this definition, cases were sorted into unsuccessful (UCs) and successful (SCs) switching to methadone.

Data are reported as means with confidence intervals (CIs). Comparisons between two groups were made by Wilcoxon signed rank test. p < 0.05 was considered significant. Data were collected and analyzed using appropriate software.

Results

Among 80 cases treated with methadone for seven days or more, there were 70 SCs (13 outpatients and 57 inpatients) and 10 UCs (1 outpatient and 9 inpatients). All 10 UCs could not take methadone subsequently due to rapid progression of illness and then other opioids were given parenterally. According to the Japanese definition, it was judged that pain control by methadone had failed due to lack of analgesia at the time when they could not take methadone. The average NRS for pain was unchanged from 7.0 (CI: 5.8–8.2) at the time of the start of methadone administration to 7.0 (CI: 5.8–8.2) at the time when they could not take methadone.

The primary cancers, the classified type of pain, and the previously prescribed opioids before switching to methadone are indicated in Table 2. The average MEDD before methadone administration in the SCs was 150 mg (range, 40–660 mg) and the starting methadone dose was 10 mg in 2 cases (old age and taking multiple drugs), 15 mg in 48 cases, 20 mg in 2 cases, 30 mg in 15 cases, and 45 mg in 3 cases. The average time from the start of methadone administration to completion of switching was 14.7 days (CI: 12.2–17.3). The methadone dose in the completed titration was the same as the starting dose in 28 cases, and was decreased from the starting dose due to somnolence or increased due to uncontrolled pain in 7 and 35 cases, respectively. The average NRS for pain decreased from 6.8 (range: 3–10) to 2.0 (range: 0–6). No serious effects such as QT interval prolongation or respiratory depression were observed.

A comparison of the clinical features of the SC and UC groups is shown in Table 2. The number of days alive after the start of methadone administration was significantly greater in the SC group (87.1 days vs. 19 days, p < 0.0001). There were no significant differences in mean age (SCs: 69, UCs: 67), NRS (SCs: 6.8, UCs: 7.0), average MEDD before methadone administration (SCs: 150.4, UCs: 146.5), and ECOG PS scores (SCs: 2.69, UCs: 2.8) between the two groups. PPI scores tended to be lower in the SC group, but without significance (2.75 vs. 4.5, p = 0.05462).

Discussion

Switching from other opioids to methadone has been examined in two randomized clinical trials.^10,11 In summary, there were no significant clinical advantages of stop-and-go switching over three-day switching, in which the dose of the current opioid is reduced stepwise by one-third every day and substituted with one-third of the equianalgesic dose of methadone for three days. ¹⁰ In Japan, the Ministry of Welfare recommends stop-and-go switching due to its simplicity, as given in Table 1.

Among 345 substitutions for methadone from other opioids in an acute palliative care unit, Mercadante ¹² found that 267 were considered successful, 51 failed because pain relief was not obtained, and 27 patients died during admission. Although the factors that determined the success or failure were not reported, it was inferred that patients who failed in switching to methadone were severely ill, contrary to expectation. Other studies have also reported that discontinuation of methadone administration after switching was mostly due to acute worsening.

In our patients, only the number of days alive from the start of methadone administration was found to differ significantly between cases with successful and unsuccessful switching. Therefore, failure of switching to methadone does not seem to be based on the severity of illness at the time of switching, but on rapid progression of illness and a poor prognosis after switching. PPI scores also tended to be lower in cases with successful switching, which suggests the difficulty of methadone induction in the patient at the end of life and a need to assess whether patients will be able to continue to take methadone after it is started. Recently, Reddy et al. have reported that there was no significant difference in overall survival among cancer patients taking methadone compared with other opioids. ¹³ However, oral methadone may not be appropriate for cases with a poor prognosis, although an accurate estimation of prognosis is difficult. Considering that mentioned, early switching from other opioids to oral methadone is desirable for relief of refractory cancer pain, at a time when the other opioids are suspected to not be controlling the pain, even if the dose is increased.

This study has several limitations. First, methadone has only been used for five years in Japan, and the Ministry of Welfare still does not allow use of parental methadone. Thus, only oral methadone can be used in Japan, and switching from other opioids to methadone could only be performed using oral methadone. Second, the study was performed as a retrospective analysis. A prospective study is necessary, but our data are closely recorded due to the directive from the Ministry of Welfare in Japan. Third, it might be necessary to define prognostic time points at which patients were switched to methadone and to compare pain scores among these groups. However, because of the short time since the approval of oral methadone, such clinical research is unlikely to be performed in the near term. Finally, the study was performed by a team of palliative professionals, and the findings may not be extendible to other settings.

This study has the advantage of including only patients selected by a checklist and treated by stop-and-go switching using the Japanese opioid conversion ratio and the Japanese definition of completed titration. Thus, the data are reliable for assessment of the usefulness of switching to methadone. Based on the results of this comparative retrospective study, we conclude that early switching to methadone for cancer pain control may be useful for patients with advanced cancer pain.