Palliative treatment of cancer-related pain
Health technology assessment
|Updated
Summary
Background
The majority of cancer patients with advanced disease have pain. Pain severely impairs health-related quality of life and is the most feared symptom in cancer patients. The quality of palliative treatment of cancer-related pain depends both on knowledge and competence of the health care workers involved, as well as on organisational aspects including a well-developed health services and well-functioning collaboration among health personnel at the different treatment levels. International research as well as national reports emphasize the potential for improving pain therapy for cancer patients in the palliative phase by means of improving all three factors: knowledge, competence and health services organisation.
Objective
The main objective was to conduct a systematic review to evaluate the best available evidence on the analgesic efficacy of medication and radiotherapy in the palliative cancer patient. In addition, economical, organisational, ethical and juridical aspects of these treatment methods, with particular reference to the Norwegian healthcare system, were evaluated.
The report focused on pain caused by the cancer disease itself, i.e. pain due to tumour growth.
The main outcome of the systematic review was pain relief, whereas treatment-related side effects were secondary outcomes.
Methods
An interdisciplinary review team performed a health technology assessment according to internationally approved principles by doing a systematic review of the published literature.
This systematic review was based on two evidence reports from Agency for Healthcare Results and Quality (AHRQ, USA) focusing on the management of cancer pain published in 2001 and 2002, ten Cochrane reviews covering the topic, and an additional 47 Medlinerecorded studies reporting on cancer pain treatment published between 2001 and 2003 (not included in AHRQ). We selected randomized controlled trials (RCTs) and meta-analyses reporting on cancer pain treatment, however comparative case series were included for treatments where studies of higher evidence level were lacking (switch the opioid and/or switch the route, pain palliation of children, invasive procedures). Two reviewers assessed 172 each publication according to criteria defined by protocol for evaluating relevance, quality, and validity.
The two AHRQ reports and the 10 Cochrane reviews include a total of 258 relevant RCTs (of which 253 RCTs were published no later than 2000, the year before our own literature search and assessment). The references to these studies are shown in Appendix 3.
The literature identified in the time period 2001-2003 was assessed stepwise. Among 1326 abstracts identified, 85 were read as full text articles, and 47 studies were critically assessed as relevant publications (one meta-analysis, 37 RCTs and 9 case series). The relevant studies were systematised in three subgroups according to the quality of the study design in question and the validity of the results: high, moderate and low. In order to be rated as acceptable (high or moderate quality), studies had to: i) report effect estimates of pain relief assessed with a validated and widely acknowledged pain assessment tool (defined as a method able to convert a subjective pain experience to a unit of measure on a scale, for example VAS/NRS scale), ii) systematically report pain relief in the whole patient group (independent of whether pain is a primary or secondary outcome in the study), and iii) simultaneously report the consumption or change in consumption of analgesic (as pain relief is the sum of pain intensity and analgesic consumption). Of the 47 included studies, 30 studies (one meta-analysis, 24 RCTs and 5 case series) were rated as high and moderate (acceptable) quality, the remaining 17 studies as low quality. Evidence tables of studies of high and moderate quality and characteristics of studies of low quality are shown in Appendix 5 (in English).
The collected documentation (from AHRQ reports, Cochrane reviews and our own literature assessment) was a total of 300 studies, which were synthesized using qualitative methodology.
Main results
The results were summarized according to the following ten treatment categories:
Opioid analgesics
Opioids are effective in relieving moderate to severe cancer pain. No specific opioid drug has been shown to have greater analgesic efficacy, and the side effects profile of the different drugs have not been shown to be different. Morphine is generally accepted the drug of choice, 173 however, its position is being challenged by the introduction of other opioids such as oxycodone and hydromorphone. There are data indicating that the combination oxycodonemorphine is more effective than morphine alone. No differences in analgesic efficacy of different formulations (e.g. immediate-release versus sustained-release) and routes of administration (e.g. oral versus rectal) have been demonstrated. Still, some formulations and routes of administration have advantages in certain patient groups. There are data indicating that oral-transmucosal fentanyl is more effective for breakthrough pain than immediate-release morphine. Switching opioids and/or administration route is an adequate clinical manoeuvre for improving pain palliation or reducing side effects both in adults and children.
Non-opioid analgesics
Non-steroidal anti-inflammatory drugs (NSAIDs) are effective in relieving moderate cancer pain. There has not been shown any difference in analgesic efficacy or side effects of different drugs. The analgesic efficiency of the combination NSAIDs and weak opioids is not different from NSAIDs alone. The co-administration of NSAIDs to strong opioids has an opioid dosesparing effect, but there is no evidence that this combination reduces side effects due to opioid use.
The documentation of the analgesic efficacy of paracetamol on cancer-related pain is sparse.
Most studies have analysed paracetamol in combination with other analgesics. Available data do not give any information of the analgesic effect that paracetamol provides on its own.
Paracetamol´s pain relieving effect has, however, been documented in other pain conditions than cancer and is as effective as NSAIDs.
Adjuvant analgesics
It is not possible from the available documentation to conclude whether adjuvant analgesics give an additive pain relief gain beyond the treatment effect obtained with standard analgesics. No adjuvant drug has been shown to have greater analgesic efficacy than others.
Based on the existing data it is not possible to conclude whether analgesic efficacy is different among specific drug groups (e.g. antidepressive agents, anticonvulsants, corticosteroids or NMDA-antagonists). Pain relieving effect of adjuvant analgesics has, however, been documented in other pain conditions than cancer.
Chemotherapy
It is unclear whether palliative chemotherapy causes pain relief beyond the analgesic effect obtained with standard analgesics. There has not been shown any difference in analgesic efficacy between various chemotherapeutic agents or chemotherapeutic regimens.
Combination regimens with chemotherapeutic agents and corticosteroids or external radiation therapy seem to have better analgesic efficiency on certain cancer diseases than regimens without chemotherapeutic agents.
Hormones
It is unclear whether hormonal therapy causes pain relief beyond the analgesic effect obtained with standard analgesics. There has not been shown any difference in analgesic efficacy between different hormones/hormonal regimens or between hormonal regimens versus other medications.
Bisphosphonates
There is evidence to support that bisphosphonates provide an additive analgesic effect beyond the treatment effect obtained with standard analgesics. Current data indicate that bisphosphonates have a weak or moderate effect on bone pain after 4-12 weeks of medication.
Whether certain bisphosphonates are more effective in reducing bone pain than others is not clarified.
Intraspinal and epidural analgesics
Intraspinal and epidural analgesics are both effective in relieving cancer pain in selected patients. Available documentation does not clarify whether this route of administration is more effective in relieving pain than other administration routes. There is no evidence of any difference in analgesic efficacy between various intraspinal/epidural analgesics or combinations and various categories of administration route (intraspinal versus epidural, continuous infusion versus bolus doses, catheter versus implantable pump). Pain relief by intraspinal and epidural administration of analgesics is feasible for most cancer patients.
Neurolytic blockades
Treatment with neurolytic blockades is effective in relieving pain due to cancer in the area of the pancreatic gland. Available documentation does not clarify whether neurolytic blockades are more effective in relieving pain than conventional pain treatment with opioids and 175 NSAIDs. Also, current data are inconclusive regarding possible effect differences between various blockade techniques or between various medications (volume and concentration).
External radiation therapy
External radiotherapy is effective in relieving cancer pain due to bone metastases. Available documentation, however, does not clarify whether this treatment effect is obtained without any change in the consumption of analgesics. Current data is too restricted for making conclusions about possible effect differences between external radiation therapy and radionuclides or medications. Data indicate that single dose (unfractionated) radiation is as effective as fractionated radiation on bone pain. That is, short courses of palliative radiotherapy with higher doses yield results that are similar to those of longer courses that provide a lower dose per treatment. The re-treatment rate and pathological fractures seem to be more frequent in patients receiving unfractionated radiation. The minimal total dose of radiation that relieves pain has not yet been determined.
Radionuclides
Radionuclide therapy is effective in relieving cancer pain due to bone metastases. Evidence is, however, too sparse to conclude whether pain reduction after radionuclide therapy is a result also of simultaneous changes in consumption of analgesics. Radionuclide therapy is as effective as external radiation therapy. Available data are inconclusive regarding possible effect differences between radionuclide therapy and bisphosphonates, as well as between different radionuclides.
Comments RCTs of cancer pain control often enrol few subjects, have low methodological quality and involve heterogeneous interventions and outcome measures assessed by different pain assessment tools. This review demonstrates that a significant body of research has failed to produce a clear answer to key questions in the management of cancer pain. The heterogeneity of existing trials precludes meta-analyses for most sub questions. Future trials of cancer pain relief would benefit from common criteria for the reporting of pain response, thus improving the comparability and thereby allowing for meta-analyses.