Methadone as the Initial Long-Acting Opioid in Children with Advanced Cancer

Abstract

Background:

Methadone is an attractive medication for treating children with advanced cancer with pain as it is the only long-acting opioid available as a liquid. However, it is not frequently used due to concerns about potential toxicities and side effects.

Objective:

Evaluate the efficacy and safety of methadone as the first long-acting opioid in children with advanced cancer.

Design:

Retrospective chart review of 52 consecutive patients referred to Pediatric Supportive Care for pain management started on methadone as their first long-acting opioid. Data collected at baseline, follow-up visits #1 (F1) and #2 (F2) included child and parent-reported outcomes for various physical and psychological symptoms, opioid side effects and other clinical data. Symptoms were rated on a 0 (not at all) to 4 (a lot) scale.

Results:

Pain (mean ± standard deviation [SD]) scored by the child was 3.6 (±0.6)/4 at baseline and 1.8 (±1.1)/4 at F1 (p < 0.0001). Compared to baseline, pain scored by the child at F2 was 1.2 (±1.3)/4 (p < 0.0001). Pain scored by the parent was 3.5 (±0.7)/4 at baseline and 1.4 (±1.3)/4 at F1 (p < 0.0001). Compared to baseline, pain scored by the parent at F2 was 1.0 (±1.2)/4 (p < 0.0001). Thirty-three (70%) patients at F1 and 23 (79%) patients at F2 did not need a change in dose of methadone. No cardiac arrhythmias or opioid neurotoxicity was observed.

Conclusions:

Initiation of methadone was effective and safe as the first long-acting opioid in children with pain.

Introduction

Pain is the most common symptom reported by children with advanced cancer.¹ Methadone is an attractive medication for treating pain, as it is the only long-acting opioid available as a liquid. This allows for precise dosing, and the ability to give to children who cannot reliably swallow tablets. Studies on the use of methadone for pain in children with advanced cancer are primarily limited to case reports,^2–5 potential toxicities,^6,7 or lack statistical analysis of outcomes.^8–10

There is growing evidence for the use of methadone as a first-line long-acting opioid for cancer-related pain in the adult population,^11–13 but in pediatrics, it is often relegated to the long-acting opioid to use after other long-acting opioids have been exhausted. In addition, research addressing the efficacy of methadone for analgesia in children and young adults with advanced cancer using patient-reported outcome (PRO) measures is lacking.

The primary objective of this retrospective study was to determine the efficacy and safety of methadone as the first long-acting opioid in children who have uncontrolled pain while either being opioid-naive or only on immediate-release opioid agonists. Uncontrolled pain often contributes to other burdensome symptoms, and so the secondary objective was to better understand the effect of improved pain control on PRO measures of quality of sleep, fatigue, and psychological symptoms.

Methods

The study was reviewed and approved by the Institutional Review Board at M.D. Anderson Cancer Center (PA16-0171).

Participants

Eligible children who had advanced cancer were referred to the Pediatric Supportive Care Service for management of uncontrolled pain at M.D. Anderson Cancer Center between December 2014 and March 2017 and started on methadone. M.D. Anderson is a National Cancer Institute (NCI)-designated tertiary cancer center. All patients seen in the study had a diagnosis of cancer and were evaluated by a full-time dual board-certified Pediatric and Hospice and Palliative Medicine physician backed up by a team of board-certified Hospice and Palliative medicine physicians. Methadone was initiated at standard analgesic dosing for pediatric patients (0.1 mg/kg PO q12 hours).

Study instruments

We created the Pediatric Symptom Assessment System (PSAS), an adapted version of the validated Edmonton Symptom Assessment System,¹⁴ Memorial Symptom Assessment Scale (MSAS),^15–17 and PediQUEST PQ-MSAS¹⁸ PRO tools, that assesses 11 physical and psychological symptoms. Time to completion is ∼1.5 minutes and the time frame of symptom assessed is the previous 24 hours. Response options for each symptom used a five-Point Likert-type scale (0 = Not at all; 1 = A little bit; 2 = Kind of; 3 = Quite a bit; 4 = A lot). Although there has been success with tools such as the MSAS, the adoption of these tools to a busy daily clinical service is complex. We, therefore, chose to apply some particular symptom assessments and score scaling from previously validated tools to make it more feasible for clinical use. One of the reassuring aspects of using the PSAS is that the 0–4 scoring system to rate the primary outcome, pain, has been previously validated in the MSAS.

Study procedures

Questionnaires were given to a parent/guardian and the child if they were 7 years or older at the time of initial consultation and subsequent clinical encounters. Questions for the parent/guardian and child were similar and written at a Flesch-Kincaid Grade Level of 4 (parent/guardian) and 1 (child). Demographic and clinical information was abstracted from medical records.

Statistical analysis

Demographics and baseline characteristics were summarized for all patients by descriptive statistics such as mean (±standard deviation) and median (interquartile range [IQR]) for continuous variables and frequency and proportion for categorical variables.

Symptom scores for children and parents are summarized by mean (±standard deviation) and median (IQR). Individual pain, fatigue, and insomnia scores are plotted as a function of follow-up time along with the amount of total methadone daily dose. Frequency and proportion of patients who had at least 25% increase in methadone dose and those who weaned off of methadone are calculated along with their 95% exact confidence intervals using the Clopper–Pearson method. Two by two comparisons of physical and psychological symptom scores at baseline, follow-up visit #1 (F1), and follow-up visit #2 (F2) were conducted by the Wilcoxon signed-rank test. Multiple comparisons were adjusted by Bonferroni correction (i.e., the nominal significance level is set to 0.0167). Kappa coefficient evaluated the level of agreement between parents and children concerning physical and psychological symptom scores.

Results

Sixty-two patients were started on a long-acting opioid for pain control during the study period, and 52 patients (84%) were started on methadone under the supervision of the Pediatric Supportive Care service. All patients started on methadone receive a standardized phone follow-up assessment on the first three consecutive days of taking methadone to assess the efficacy and safety of methadone. Adjustments in the dose of methadone and breakthrough immediate-release opioids were made if necessary. At no point during the study period did any patient have opioid-induced neurotoxicities such as excessive somnolence, hallucinations, or myoclonus. No cardiac arrhythmias were detected during the study period. Baseline patient characteristics are summarized in Table 1.

Table 1.

Patient Characteristics and Clinical Aspects of Pain and Opioids Used (n = 52)

Age, median (IQR) years	12.5 (6.8–16.3)
Weight, median (IQR) kg	42 (21–56)
White non-Hispanic, n (%)	28 (54)
Female, n (%)	22 (42)
Diagnosis
Hematologic malignancy, n (%)	11 (21)
Central nervous system tumors, n (%)	5 (10)
Solid tumors, n (%)	36 (69)
Mechanism of pain, n (%)
Nociceptive	25 (48)
Neuropathic	1 (2)
Nociceptive and neuropathic	26 (50)
Opioid before methadone, n (%)
Codeine	6 (12)
Hydrocodone	11 (21)
Hydromorphone	4 (8)
Morphine	27 (52)
Opioid-naive	2 (4)
Oxycodone	2 (4)
MEDD, median (IQR) mg before methadone initiation	38 (13–78)
Methadone initiation
Median (IQR) mg daily total dose	7.8 (5.0–10.0)
Median (IQR) mg/kg/dose	0.1 (0.1–0.1)
Median (IQR) days from BL to F1, (n = 47)	17 (9–26)
Median (IQR) days from BL to F2, (n = 29)	55 (35–64)
Methadone dose change over time
BL to F1 (n = 47), n (%)
No change	33 (70)
Increase	10 (21)
Decrease	4 (9)
BL to F2 (n = 29), n (%)
No change	23 (79)
Increase	1 (4)
Decrease	5 (17)

IQR, interquartile range; MEDD, morphine equivalent daily dose; BL, baseline; F1, follow up #1; F2, follow-up #2.

Forty-seven patients (90%) were assessed at F1. Two patients (4%) died of progressive disease before F1. Twenty-nine patients (62%) were evaluated at F2. Three patients (6%) died between F1 and F2 of progressive disease. Median (IQR) days between baseline evaluation and F1 as well as baseline and F2 are shown in Table 1. There was no standard protocol for obtaining a follow-up electrocardiography (ECG), yet, a majority (38/47 [81%]) of patients had one during the study period. None of these patients demonstrated prolongation of the corrected QT (QTc) interval (≥450 ms).

Clinical characteristics of pain and opioids used are shown in Table 1. Thirty-three patients at F1 (70%) and 23 patients at F2 (79%) did not need a change in their methadone dose. Ten patients at F1 (21%) and one patient at F2 (4%) required an increase in their methadone dose due to inadequate pain control. Few patients at F1 (4, 9%) and F2 (5, 17%) had their methadone decreased, but all were related to the tapering of methadone due to resolution of the original pain syndrome, not due to opioid toxicity.

Child and parent-reported outcomes are shown in Table 2. Pain improved significantly between baseline and F1 as scored by the child (p < 0.0001) and the parent (p < 0.0001); pain continued to improve significantly between baseline and F2 as scored by the child (p < 0.0001) and the parent (p < 0.0001). Insomnia improved significantly between baseline and F1 as scored by the child (p < 0.001) and the parent (p < 0.001); insomnia continued to improve significantly between baseline and F2 as scored by the child (p < 0.005) and the parent (p < 0.001). Fatigue improved significantly between baseline and F2 as scored by the child (p = 0.05) and the parent (p = 0.01). There was no significant change in psychological symptoms of nervousness, sadness, or irritability over time. At baseline, 46 parents were evaluable as 6 did not have a parent present; 33 children were evaluable as 14 were <7 years of age, 2 families requested not to wake the child, 2 children had autism, and 1 child had too much pain to complete. At F1, 43 parents were evaluable as 4 did not have a parent present; 30 children were evaluable as 13 were <7 years of age, 3 families requested not to wake the child, and 1 had autism. At F2, 26 parents were evaluable as 3 did not have a parent present; 20 children were evaluable as 9 were <7 years of age.

Table 2.

Child- and Parent-Reported Outcomes (0 [Not at All] to 4 [A Lot])

Child	BL Mean ± SD (n = 33)	F1 Mean ± SD (n = 30)	p-Value BL vs. F1	F2 Mean ± SD (n = 20)	p-Value BL vs. F2
Pain	3.59 ± 0.61	1.80 ± 1.13	<0.0001	1.15 ± 1.27	<0.0001
Fatigue	1.81 ± 1.26	1.53 ± 1.25	0.22	1.45 ± 1.28	0.05
Nervous	0.72 ± 0.89	0.77 ± 1.04	0.76	1.05 ± 1.28	0.23
Nausea	0.78 ± 1.01	0.93 ± 1.31	0.31	1.00 ± 1.08	0.75
Insomnia	2.50 ± 1.08	1.07 ± 1.23	<0.001	1.05 ± 1.10	0.005
Dyspnea	0.34 ± 0.79	0.43 ± 0.86	0.38	0.55 ± 0.76	0.12
Diarrhea	0.22 ± 0.79	0.37 ± 0.81	0.47	0.30 ± 0.57	0.41
Sadness	0.53 ± 0.72	0.63 ± 1.00	0.82	1.15 ± 1.14	0.11
Constipation	1.13 ± 1.26	0.70 ± 1.06	0.35	0.45 ± 0.69	0.06
Irritable	0.77 ± 1.02	0.87 ± 1.31	0.66	0.75 ± 0.64	0.56
Appetite	3.19 ± 1.08	3.37 ± 0.96	0.60	2.80 ± 1.20	0.09

Parent	BL Mean ± SD (n = 46)	F1 Mean ± SD (n = 43)	p-Value BL vs. F1	F2 Mean ± SD (n = 26)	p-Value BL vs. F2
Pain	3.46 ± 0.66	1.35 ± 1.25	<0.0001	1.04 ± 1.16	<0.0001
Fatigue	1.93 ± 1.06	1.60 ± 1.18	0.09	1.41 ± 1.01	0.01
Nervous	1.26 ± 1.20	1.00 ± 1.20	0.34	0.89 ± 0.93	0.31
Nausea	0.87 ± 1.20	1.02 ± 1.34	0.53	1.04 ± 1.32	0.85
Insomnia	2.50 ± 1.05	1.07 ± 1.12	<0.001	1.26 ± 1.23	<0.001
Dyspnea	0.35 ± 0.90	0.58 ± 1.10	0.24	0.67 ± 1.14	0.03
Diarrhea	0.17 ± 0.68	0.19 ± 0.70	0.92	0.26 ± 0.76	1.00
Sadness	1.02 ± 1.18	0.77 ± 1.23	0.32	0.85 ± 0.99	0.86
Constipation	1.17 ± 1.18	0.98 ± 1.32	0.44	0.70 ± 1.10	0.48
Irritable	1.30 ± 1.28	1.35 ± 1.27	0.83	1.04 ± 1.02	0.60
Appetite	1.63 ± 1.04	3.47 ± 1.20	0.41	2.89 ± 1.31	0.08

SD, standard deviation.

Values in bold and italics are statistically significant.

The level of agreement between the child and parent was determined by the Kappa coefficient and is shown in Table 3. Kappa coefficient for pain was 0.58 (p < 0.0001) indicating moderate agreement between parents and children. Sadness had a kappa coefficient of 0.30 (p < 0.07) suggesting fair agreement. The remainder of the symptoms had a poor agreement between the parent and the child.

Table 3.

Agreement in Symptom Reporting Between Parent and Child at Follow-Up #1

Symptom	Kappa coefficient	p-Value
Pain	0.58	<0.0001
Diarrhea	0.45	<0.0001
Sadness	0.30	0.07
Appetite	0.01	0.92
Dyspnea	0.00	0.62
Fatigue	−0.02	0.75
Insomnia	−0.02	0.76
Nervous	−0.03	0.49
Irritable	−0.03	0.06
Nausea	−0.04	0.43
Constipation	−0.07	0.06

Discussion

This chart review of 52 consecutive children found that methadone was highly effective and safe as the first long-acting opioid in for pain due to advanced cancer. The most severe symptoms in these children were pain, fatigue, and insomnia. It is not surprising that methadone significantly improved pain but the use of methadone as the first long-acting opioid in pediatrics is rare,¹⁹ likely due to the concerns about the complex pharmacokinetics and pharmacodynamics.^20,21 Although a small number of patients were evaluated, no patient demonstrated cardiac arrhythmias or opioid-induced neurotoxicity at any point during the study period. While some adult literature reports that methadone requires more frequent titration than other long-acting opioids²⁰ we found that most children did not need an increase in their methadone; these findings should be reassuring to clinicians who may hesitate to use methadone.

Insomnia can have a dramatic effect on the quality of life of a child,^22–24 and children with pain have difficulty with sleep.^25–27 We found significant improvement in insomnia reported scores by the child and parent at F1 that was sustained through F2. Fatigue is related to insomnia,²⁸ but is a more indolent symptom. We found no significant difference in reported levels of fatigue at F1, but it was significantly improved at F2. No specific therapies such as stimulants or steroids were used in any of the children to treat their fatigue. This highlights the complex interaction among pain, insomnia, and fatigue and more research is needed to understand better the factors that may directly influence these burdensome symptoms.

Many clinicians have reservations using methadone due to its potential side effects,²⁰ including prolongation of the QTc interval. Our study did not have any patients with cardiac arrhythmias. While the frequency of prolongation of the QTc interval and subsequent cardiac arrhythmias in children given methadone is low,⁶ it would be prudent to obtain a baseline electrocardiogram and repeat if there are electrolyte abnormalities or significant increases in methadone. Our study did not find any opioid-induced neurotoxicities; furthermore, parent and PROs did not show an improvement in fatigue that one might expect to see if the children had excessive somnolence from methadone toxicity.

It was unexpected that there was no significant improvement in the psychological scores of nervousness, sadness, or irritability as pain, insomnia, and fatigue improved. It is important, however, to highlight that the severity of the psychological symptoms at baseline was low, and so, the margin for clinically significant improvement in these scores is limited.

The level of agreement between the parent and child for most symptoms was poor, and so, it is advisable to advocate for the child to have a voice in the conversation about his/her state of health and not rely solely on the parent. Our data, and that of other authors,²⁹ suggest that the child's reported outcome should continue to be the gold standard of symptom assessment.

Strengths of the study include a good size case series demonstrating the analgesic efficacy of methadone and the absence of side effects as determined both by phone follow-up and in the child and parent-reported outcomes. There are several limitations to this study. The retrospective nature of the study limits the consistency of clinical practice when starting methadone, and monitoring for its safety, as evidenced that only 81% of patients had follow-up ECGs. Another limitation of the study is that it was conducted in a tertiary institution with access to a board-certified specialist as well as resources to provide diligent phone follow-up. Efficacy and monitoring of safety may be different in other settings with fewer resources.

Conclusions

Methadone as the first long-acting opioid in children with advanced cancer is both effective and safe. More research is necessary to understand better how to most effectively screen, evaluate, and treat psychological symptoms in children with advanced cancer who have significant pain.

Footnotes

Author Disclosure Statement

No competing financial interests exist.

References

Wolfe

, Grier

, Klar

, et al.: Symptoms and suffering at the end of life in children with cancer. N Engl J Med, 2000; 342:326–333.

Rasmussen

, Lundberg

, Jespersen

, Hasle

: Extreme doses of intravenous methadone for severe pain in two children with cancer. Pediatr Blood Cancer, 2015; 62:1087–1090.

Berde

, Beyer

, Bournaki

, et al.: Comparison of morphine and methadone for prevention of postoperative pain in 3- to 7-year-old children. J Pediatr, 1991; 119(1 Pt 1):136–141.

Shir

, Shenkman

, Shavelson

, et al.: Oral methadone for the treatment of severe pain in hospitalized children: A report of five cases. Clin J Pain, 1998; 14:350–353.

Sabatowski

, Kasper

, Radbruch

: Patient-controlled analgesia with intravenous L-methadone in a child with cancer pain refractory to high-dose morphine. J Pain Symptom Manage, 2002; 23:3–5.

Madden

, Park

, Liu

, Bruera

: The frequency of QTc prolongation among pediatric and young adult patients receiving methadone for cancer pain. Pediatr Blood Cancer, 2017 [Epub ahead of print]; DOI: 10.1002/pbc.26614.

Anghelescu

, Patel

, Mahoney

, et al.: Methadone prolongs cardiac conduction in young patients with cancer-related pain. J Opioid Manag, 2016; 12:131–138.

Anghelescu

, Faughnan

, Hankins

, et al.: Methadone use in children and young adults at a cancer center: A retrospective study. J Opioid Manag, 2011; 7:353–361.

Davies

, DeVlaming

, Haines

: Methadone analgesia for children with advanced cancer. Pediatr Blood Cancer, 2008; 51:393–397.

10.

Martinson

, Nixon

, Geis

, et al.: Nursing care in childhood cancer: Methadone. Am J Nurs, 1982; 82:432–435.

11.

Parsons

, de la Cruz

, El Osta

, et al.: Methadone initiation and rotation in the outpatient setting for patients with cancer pain. Cancer, 2010; 116:520–528.

12.

Bruera

, Palmer

, Bosnjak

, et al.: Methadone versus morphine as a first-line strong opioid for cancer pain: A randomized, double-blind study. J Clin Oncol, 2004; 22:185–192.

13.

Mercadante

, Porzio

, Ferrera

, et al.: Sustained-release oral morphine versus transdermal fentanyl and oral methadone in cancer pain management. Eur J Pain, 2008; 12:1040–1046.

14.

Chang

, Hwang

, Feuerman

: Validation of the Edmonton symptom assessment scale. Cancer, 2000; 88:2164–2171.

15.

Drake

, Frost

, Collins

: The symptoms of dying children. J Pain Symptom Manage, 2003; 26:594–603.

16.

Collins

, Byrnes

, Dunkel

, et al.: The measurement of symptoms in children with cancer. J Pain Symptom Manage, 2000; 19:363–377.

17.

Collins

, Devine

, Dick

, et al.: The measurement of symptoms in young children with cancer: The validation of the Memorial Symptom Assessment Scale in children aged 7–12. J Pain Symptom Manage, 2002; 23:10–16.

18.

Wolfe

, Orellana

, Cook

, et al.: Improving the care of children with advanced cancer by using an electronic patient-reported feedback intervention: Results from the PediQUEST randomized controlled trial. J Clin Oncol, 2014; 32:1119–1126.

19.

Fife

, Postier

, Flood

, Friedrichsdorf

: Methadone conversion in infants and children: Retrospective cohort study of 199 pediatric inpatients. J Opioid Manag, 2016; 12:123–130.

20.

Roth

, Davies

, Friebert

, et al.: Attitudes and practices of pediatric oncologists regarding methadone use in the treatment of cancer-related pain: Results of a North American Survey. J Pediatr Hematol Oncol, 2013; 35:103–107.

21.

Shaiova

, Berger

, Blinderman

, et al.: Consensus guideline on parenteral methadone use in pain and palliative care. Palliat Support Care, 2008; 6:165–176.

22.

LaPlant

, Adams

, Haftel

, Chervin

: Insomnia and quality of life in children referred for limb pain. J Rheumatol, 2007; 34:2486–2490.

23.

Palermo

, Fonareva

, Janosy

: Sleep quality and efficiency in adolescents with chronic pain: Relationship with activity limitations and health-related quality of life. Behav Sleep Med, 2008; 6:234–250.

24.

Palermo

, Kiska

: Subjective sleep disturbances in adolescents with chronic pain: Relationship to daily functioning and quality of life. J Pain, 2005; 6:201–207.

25.

Gilman

, Palermo

, Kabbouche

, et al.: Primary headache and sleep disturbances in adolescents. Headache, 2007; 47:1189–1194.

26.

Haim

, Pillar

, Pecht

, et al.: Sleep patterns in children and adolescents with functional recurrent abdominal pain: Objective versus subjective assessment. Acta Paediatr, 2004; 93:677–680.

27.

Palermo

, Wilson

, Lewandowski

, et al.: Behavioral and psychosocial factors associated with insomnia in adolescents with chronic pain. Pain, 2011; 152:89–94.

28.

Owens

, Mindell

: Pediatric insomnia. Pediatr Clin North Am, 2011; 58:555–569.

29.

Zhukovsky

, Rozmus

, Robert

, et al.: Symptom profiles in children with advanced cancer: Patient, family caregiver, and oncologist ratings. Cancer, 2015; 121:4080–4087.