Effect of Combination Treatment or Monotherapy in Pain Intensity Reduction in Cancer Patients with Neuropathic Pain

Research Proposal

Title

A Phase III double-blind Randomised Controlled Trial of the effect of amitriptyline and gabapentin combination versus amitriptyline alone or gabapentin alone or Placebo as a first line treatment on pain intensity reduction, quality of life and adverse effects in cancer patients with neuropathic pain.

Introduction

Neuropathic pain (NP) is very common both in cancer patients with advanced disease and in patients who received treatment with a curative intent [1]. The prevalence of NP in general population, ranges from 0.9% to 17.9% [2, 3]. A cohort which included 358 patients with cancer demonstrated that 37.6% of patients with pain and 20.4% of all patients experienced either pure or mixed NP [4].  This was consistent with a systematic review of 13,683 cancer patients which showed that approximately 40% of pains are of mixed or pure neuropathic aetiology [5]. The above evidence suggest that the burden of NP in cancer patients is significant and constitutes and major public health concern [1]. Currently, there is no clear evidence on the efficacy of the gabapentin and amitriptyline in cancer patients with NP mainly due to the methodological limitations of the previous studies.

Aims of the research

It is known that both gabapentin and amitriptyline reduce the pain intensity in patients with NP [6, 7]. Two Cochrane systematic reviews demonstrated promising evidence on the effectiveness of these drugs as monotherapy in NP and quality of life [6, 7]. Nevertheless, there are weak evidence on the effectiveness of the combination of amitriptyline and gabapentin in patients with NP according to a recent metanalysis [8]. These findings are attributed to the poor methodological quality and heterogeneity of previous studies which made the interpretation of the outcomes difficult. Thus, the aim of this clinical trial is to investigate the effect of combination treatment or monotherapy in pain intensity reduction, quality of life and adverse effects in cancer patients with NP and to overcome the methodological limitations which have made the interpretation of the results of the previous trials difficult.

The objectives of this trial are to assess the effectiveness of amitriptyline and gabapentin vs gabapentin vs amitriptyline vs placebo in: a) pain intensity reduction b) quality of life and c) adverse effects.

Literature review

A Cochrane systematic review by Moore and colleagues on the analgesic efficacy and adverse effects of gabapentin in chronic NP in general population using randomised double-blind clinical trials showed that 35% of the patients achieved 50% pain reduction which is considered to be associated with important physical and mental benefits as well as improvement on the quality of life compared to 21% for placebo [7]. In addition, a systematic review which evaluated the efficacy of amitriptyline compared to placebo or another active treatment in chronic NP using randomised, double-blind trials revealed modest data quality because the majority of studies were at high risk of bias due to small sample size [6]. The authors concluded that although there is no good evidence of a lack of effect, the benefit of amitriptyline might have been overestimated [6].

The findings of the above systematic reviews are characterised by certain limitations. The main weaknesses of these trials were the inadequate descriptions of randomisation, inadequate report of sequence generation and allocation concealment, small number of patients, short duration trials and losses to follow up which might have introduced selection bias and random error [7]. In addition, there was both methodological and clinical heterogeneity among the studies. For instance, not all studies performed an intention-to-treat analysis and there was a variation in the types of the NPs and grading of pain. These differences might have introduced selection bias and overestimate or underestimate the outcomes [9]. Furthermore, with regards to the combination trials, they were not well-designed and their results varied  from no added benefit to equivocal or inferior tolerability [9].

Thus the major obstacle is the methodological quality of the conducted studies, which created gaps in the interpretation of the results. In addition there are no good quality randomised controlled trials assessing the effect of the combination of gabapentin and amitriptyline versus monotherapy.

Rationale for the study and justification of the proposed method

A multicentre randomised double-blind controlled trial will be conducted in Cancer Centres and Palliative Care Hospices in the United Kingdom. This design is an efficient way of establishing the effect of amitriptyline and gabapentin combination versus amitriptyline and placebo, gabapentin and placebo or placebo and placebo on pain intensity reduction and quality of life in cancer patients with NP for the following reasons: Firstly, the use of multicentre design will allow us to obtain a large sample size in order to detect a clinically significant difference. The prevalence of NP in cancer patients  is of the order of  40% which will enable us to achieve the estimated sample size [5]. Secondly, the selection of a RCT allow us to establish the potential postulate effect of the above drugs. RCT has the advantage of the best representation of the study population because all the participants have equal possibility of being selected, they are comparable at the start of the trial, as well as selection and information bias are minimised. The control group is necessary to establish a temporal effect. Thus, RCT strengthens the internal validity of the study and the validity of the statistical test is guaranteed. Furthermore, known and unknown confounders have equal possibility of being present in all treatment arms which ensures that any differences in the outcomes can be attributed to the treatments alone. The primary outcome is pain intensity reduction and the secondary outcomes are quality of life and adverse events within 90 days after randomisation.

Study Design and Methodology

Study Population: The study population is based in the United Kingdom cities of Nottingham, Southampton, Cardiff, Edinburgh and Glasgow where there are large cancer centres and hospices. Considering the size of the study, the inclusion criteria as well as the intention to treat analyses, the results can be applicable to every cancer patient with NP.

Methods: Consecutive patients who are referred to the oncology or palliative care departments either as outpatients or inpatients and meet the entry criteria are included in the study. Those who are excluded or refuse to participate are listed in a log registry with the reasons for exclusion and their baseline characteristics.

Inclusion criteria: All patients who are admitted to investigator’s hospital or hospice between February 2019 and February 2022, over 18 years old with cancer-related and/or anticancer treatment-related (surgical, chemotherapy or radiotherapy) NP, which was defined using the Leeds Assessment of Neuropathic Symptoms and Signs will be included in the RCT. Pre-existing analgesia will be continued throughout the trial.

Exclusion Criteria: Patients with contraindications to amitriptyline and gabapentin such as arrhythmias, recent myocardial infraction, severe liver disease, porphyria, hypersensitivity reactions or current use of the above drugs due to NP or other indications, or limited mental capacity (for the needs of the informed consent)  will be excluded from the study.

Randomization procedures: Patients who are eligible will be randomly assigned in the four arms of the trial through 24-hour computer generated randomization via telephone by the University of Nottingham Coordinating Centre, after the patient is consented. The allocation sequence that will be used is the stratified randomisation ensuring balance between treatment groups and factors such as the type and cause of NP, cancer type, medications and anticancer treatment. Both participants and the investigators will be concealed until the interventions are assigned. Independent clinical trial pharmaceutical company assigns a unique code in each medication box stored in the oncology or palliative care departments. Trial information leaflets and consent forms for the patients as well as outcome forms will be included into each treatment pack. After randomisation participants still have the right to withdraw their consent.

Blinding: Study participants and investigators assessing the outcomes will be unaware of which treatment was allocated. Gabapentin, amitriptyline and placebo are identical in physical features (size, shape, colour, sheen and texture). Staff unconnected with the RCT will compare the three interventions including placebo, to assess whether the drugs can be identified. Treatments are known by a random number, different in control and treatment groups, which will be recorded by a unique patient code that remains with them during the whole duration of the trial. If unblinding is required due to medical reasons after randomization an independent investigator will be able to unblind the allocated drug.

Interventions: As far as the recommended dose of gabapentin is concerned, 300 mg once daily on day 1, 300 mg twice daily on day 2 and 300 mg three times daily on day 3 orally [10]. The dose of gabapentin can be increased by 100 mg three times a day every 3 days to a maximum of 1200 mg three times a day[10].  The amitriptyline recommended initial dose is 10 mg in the evening orally and can be increased as necessary to a maximum of 150 mg daily in divided doses every 3 – 7 days as tolerated [11].

Size of study: The study design was calculated on the basis of the primary outcome measure. The level of statistical significance and power that will be used is 5% and 80%, respectively. To detect a 15% improvement in the reduction of pain while receiving the above interventions, the sample was estimated to be 148 patients in each arm using EpiInfo software version 7.2. The pain intensity reduction which is associated with improved physical function, mood and quality of life is at the level of 50%. Although losses to follow up is unlikely due to the short follow-up period and the regular clinical reviews we adjusted our calculations taking into account potential loss to follow up. The loss to follow up according is expected to be the maximum 2%. Adjusting for the expected loss to follow-up the total sample size need to be increase to 604 patients. The cancer centres of Southampton, Nottingham, Cardiff, Glasgow and Edinburgh constitute tertiary cancer centres which provide specialist services to their local population of approximately 2 million residents each. Taking into account the prevalence of NP in cancer patients as well as the calculated sample size, the recruitment of the study population can be performed within the timeframe.

Data requirements: The sub-investigators and investigators will record the primary and secondary outcomes as well as the patient’s condition, the adverse effects according to the Good Clinical Practice framework. The pain intensity will be measured using the short-form McGill Pain Questionnaire (SF-MPQ) [12, 13]. It comprise of 11 descriptors of the sensory and 4 descriptors of the affective dimension of pain experience. Each descriptor is ranked on an intensity scale of 0 = none, 1 = mild, 2 = moderate, 3 = severe[13]. The depression and anxiety will be measure using the Hospital Anxiety and Depression Score tool which is a patient-administered anxiety and depression screening tool for use in non-psychiatric patients [14]. This questionnaire includes 14 elements, which evaluate the emotional and cognitive aspects of their condition. Each element is rated from 0 to 3 for a combined maximum of 21 for each element, with higher scores suggesting a higher symptom burden. The quality of life will be assessed using the EuroQoL tool which is a self-administered questionnaire of the current quality of life consisting of 5 questions, each rated with a score of 0 to 2 points, with higher score reflecting better quality of life [15]. The side effects will be evaluated by utilising the National Cancer Institute Common Toxicity Criteria for Adverse Events version 5.0. Each side effect is graded on a scale of 1 to 5, with higher grades suggesting greater severity [16]. The outcome forms will be completed at baseline and then weekly for the first 4 weeks and then every 3 weeks until the completion of the follow-up.

Follow up: The end of follow-up for each patient will be either discharge, or withdrawal from the study due to side effects or 90 days after randomization. The total duration of the study in order to detect a difference among the four groups will be from February 2019 to February 2022. To minimise losses to follow-up, clinical assessments will be sought by contacting the patients on regular intervals.

Analysis: The variables that will be used are pain intensity reduction which is a binary variable, quality of life, anxiety and depression and adverse effects will be treated as a categorical variables. Preliminary analysis of baseline comparability will be done to examine the distribution of each variable and identify outliers and missing data. Comparison among the four treatments groups regarding primary and secondary outcomes, as well as exploratory analysis for potential confounders, will be performed using cross-tabulations, chi square test and calculating percentages.  Crude analysis using logistic regression method will be performed to examine the effect of gabapentin and amitriptyline combination, single agents alone or placebo on the primary outcome on an intention to treat analysis. Comparisons will be made with the control group which is the placebo arm.

The measures of effect that will be used are the hazard ratio and the number needed to treat together with the 95% confidence intervals and the p-value.  Multivariate analysis will be conducted to adjust for differences among the four groups, given that they are confounders, and try to evaluate the effectiveness of the intervention under study on the primary outcome. Subgroup analysis will be done according to the type of NP (peripheral vs central). All analysis will be carried out by STATA statistical software, version 11.

Validity and reliability of the proposed study

Validity of questionnaires:

The validity of the Leeds Assessment of Neuropathic Symptoms and Signs scale, the Short Form McGill Pain Questionnaire, the EuroQoL health-related quality of life tool, and the Hospital Anxiety and Depression Score, have been found to be very good on published validation studies [14, 15, 17-19].The outcome form as well as the whole procedure will be evaluated during the pilot study between July and September 2018, in patients similar to the reference population, where the kappa statistic agreement between the investigators and trained sub-investigators will be examined. A manual of procedures and guidelines for safety reporting, and informed consent will be given to all investigators by the coordinating centre, which includes how the eligibility criteria and clinical variables are determined. Instructions will be given about the completion of outcome form as well as how and where to be sent. Training sessions will be provided to investigators about the use of the telephone randomization process. The responsible authority for training is the coordinating centre.

Methods against bias:

Selection bias is minimised by the randomised assignment and concealment allocation of treatments through telephone randomisation centre according to stratification method, before treatments will be assigned. Furthermore this strategy will minimise the intercenter variability. Subjects are also checked for consent and eligibility before the random allocation. Observer bias is minimised by using the double blinding method. However, the possible side-effects of gabapentin are different from those in amitriptyline and placebo which may cause unmasking. This can be avoided using unique codes for each patient.

Considering bias in the study management, the research staff must be well-trained for particular tasks to ensure that assessments are performed identically on every patient. The reliability of the outcome forms will be checked in the pilot study. Pain intensity reduction as primary outcome can potentially lead to bias since this may introduce information bias and misclassification. Intention to treat analysis is used to minimise bias introduced by post randomization withdrawals and losses to follow up and that reflects practice in real life. The subgroup analysis involves loss of statistical power. The clinical trial will be registered to the Controlled Clinical Trials Registry to avoid publication bias and will follow the CONSORT guidelines for reporting. In addition, randomisation methods contributes to the comparability of the intervention groups. To reduce the risk of Type I error we use two outcome measures and obtain sufficient sample size. Finally, confounding is controlled by randomisation and preliminary statistical analysis will identify uncontrolled or residual confounding.

Ethical issues

The study protocol is approved by the Bioethics Commission of the University of Nottingham as well as by the participating hospital’s ethics committees in Southampton, Cardiff, Edinburgh and Glasgow and all subjects will give a written informed consent. All data extracted will be anonymised and protected before analysed. Quantitative and qualitative data will be stored on PCs, backed up on secure servers and password protected. To ensure data security and high availability, the involved institutes have an intranet hosted on secure virtual servers, housed in physically secure rooms across multiple sites. The local trust guidelines and policies will be followed with regards to data security and protection. Finally, the random assignment of drugs under investigation is ethical because in previous systematic reviews due to the poor methodological quality of the trials it was difficult to interpret the results. In addition, placebo is deemed ethical because all the participants have the right to withdraw their consent from the trial at any time to pursue other treatment and rescue analgesia will be provided during the study.

Presentation of results

The first results will be presented to the coordinating centre, then in the National Cancer Research Institute and European Society for Medical Oncology conferences either as posters or oral presentations. The final results will be submitted for publication in a high impact journal, in the Cancer Research UK website and in the www.clinicaltrials.gov database.

Dissemination of findings

We will increase awareness of this trial and provide wide access to our data within the palliative care, oncology and general medicine fields. Open access peer-reviewed publications in high impact academic journals will communicate the results to other researchers, clinicians and stake holders involved in pain management. We expect two high impact publications from this project. Additionally, preliminary results will be presented in National Cancer Research Institute and European Society for Medical Oncology conferences either as posters or oral presentations.

We will promote public awareness and understanding of the advances in the management of NP. The trials team, frequently lectures at charity events raising awareness of palliative care. The trial’s aims and findings will also be summarised in the Cancer Research UK and University of Nottingham websites so everyone can be informed about and reflect on the outcomes.

Costs

Standard care such as history, physical examination and clinical visits, the prescription of of Amitriptyline, Gabapentin, Placebo, salaries of research nurses, physicians, statisticians and epidemiologists including principles investigators, as well as office and clinical supplies, will be funded by the Cancer Research UK research grant. The University of Nottingham and Nottingham University Hospitals NHS Trust and Nottingham Hospital Charities will act as co-sponsors.

Thus, the funding of £674,850 requested for three years is based on:

Overall cost of salaries: Trial Personnel: a)Trial coordinator:48h/week x 36 months x £28,000 per annum, b) Follow-up coordinator:48h/week x 36 months £26,000 per annum, c) Data assistant: 48h/week x 36 months £18,500 per annum, d)Database manager:48h/week x 36 months £30,000 per annum, e)Administrator:48h/week x 36 months £27,000 per annum, f)Staff for the telephone randomisation center:48h/week x 36 months £25,000 per annum.

Consumables: Manual of procedures and guidelines and other forms, Gabapentin, Amitriptyline and Placebo cost: £50,000.

Non-Staff direct cost: The conduct of pilot study, training sessions, assembly meetings, conferences attendance: £45,000.

Logistics: Phone calls, telephone randomization process, travel expenses: £55,000.

Overhead (10%): £61,350

Timescale

	Month 3	Year 1	Year 2	Year 3	Year 4
Pilot study	x
Collect data	x	x	x	x
Code and process data	x	x	x	x	x
Analyse data	x				x
Computer interface					x
Prepare tables					x
Prepare figures					x
Write results draft 1					x
Revise results draft 2					x
Revise results draft 3					x
Obtain feedback and criticism	x				x

 

References

1. Goyal, A. and S. Bhatnagar, Neuropathic pain in cancer. Ann Palliat Med, 2014. 3(1): p. 1-3.

2. Smith, E.M.L., et al., Cancer Treatment-Related Neuropathic Pain Syndromes—Epidemiology and Treatment: An Update. Current Pain and Headache Reports, 2014. 18(11): p. 459.

3. van Hecke, O., et al., Neuropathic pain in the general population: a systematic review of epidemiological studies. Pain, 2014. 155(4): p. 654-62.

4. Perez, C., et al., Prevalence of pain and relative diagnostic performance of screening tools for neuropathic pain in cancer patients: A cross-sectional study. Eur J Pain, 2015. 19(6): p. 752-61.

5. Bennett, M.I., et al., Prevalence and aetiology of neuropathic pain in cancer patients: a systematic review. Pain, 2012. 153(2): p. 359-65.

6. Moore, R.A., et al., Amitriptyline for neuropathic pain in adults. Cochrane Database Syst Rev, 2015(7): p. Cd008242.

7. Moore, R.A., et al., Gabapentin for chronic neuropathic pain and fibromyalgia in adults. Cochrane Database Syst Rev, 2011(3): p. Cd007938.

8. Finnerup, N.B., et al., Pharmacotherapy for neuropathic pain in adults: a systematic review and meta-analysis. Lancet Neurol, 2015. 14(2): p. 162-73.

9. Gilron, I., Methodological issues associated with clinical trials in neuropathic pain. Expert Review of Clinical Pharmacology, 2016. 9(11): p. 1399-1402.

10. Gabapentin. 12/05/2017 24/04/2018]; Available from: https://www.medicines.org.uk/emc/product/4122/smpc#POSOLOGY.

11. Amitriptyline. 04/07/2017; Available from: https://www.medicines.org.uk/emc/product/6017/smpc#POSOLOGY.

12. Dudgeon, D., R.F. Raubertas, and S.N. Rosenthal, The short-form McGill Pain Questionnaire in chronic cancer pain. J Pain Symptom Manage, 1993. 8(4): p. 191-5.

13. Melzack, R., The short-form McGill Pain Questionnaire. Pain, 1987. 30(2): p. 191-7.

14. Bjelland, I., et al., The validity of the Hospital Anxiety and Depression Scale. An updated literature review. J Psychosom Res, 2002. 52(2): p. 69-77.

15. Hwang, S., et al., Validity, responsiveness and reliability of EuroQOL (EQ5D) in veterans with cancer pain. The Journal of Pain, 2005. 6(3): p. S75.

16. NCI Common Terminology Criteria for Adverse Events (CTCAE)  v5.0. 27/11/2017 26/07/2018]; Available from: https://evs.nci.nih.gov/ftp1/CTCAE/About.html.

17. Yucel, A., et al., Results of the Leeds assessment of neuropathic symptoms and signs pain scale in Turkey: a validation study. J Pain, 2004. 5(8): p. 427-32.

18. Melzack, R., The short-form McGill pain questionnaire. Pain, 1987. 30(2): p. 191-197.

19. Wright, K.D., G.J.G. Asmundson, and D.R. McCreary, Factorial validity of the short-form McGill pain questionnaire (SF-MPQ). European Journal of Pain, 2001. 5(3): p. 279-284.