Adjuvants are drugs whose primary indication is not analgesia, but when used in specific pain situations may be effective as analgesics or co-analgesics. They include drugs used as antineuropathic medications (antidepressants, anticonvulsants and systemic local anaesthetics) and also steroids, bisphosphonates, ketamine, and clonidine.  Non-steroidal anti-inflammatory medications are also considered here. Adjuvants may be used in situations where the pain is not fully responsive to opioids, as 'step 1' analgesics in the World Health Organization analgesic ladder approach to pain management, and most commonly, in conjunction with opioids in treating neuropathic or opioid-resistant pain.
Neuropathic pain is a considerable problem for palliative care patients.  Neuropathic pain may be due directly to cancer, to the treatment of cancer, or to non-cancer causes. However the evidence for most adjuvants for neuropathic pain is limited, and mainly relates to non-malignant chronic conditions like diabetic neuropathy or post-herpetic neuralgia, often from quite small studies. The analgesic effect of any adjuvant is less in malignant pain than in non-malignant pain. 
What is known
A systematic review  supports their role in treatment of neuropathic pain. Both tricyclic antidepressants and the serotonin–noradrenaline reuptake inhibitor, venlafaxine, were effective, with NNTs (number needed to treat) of around 3. Duloxetine may also be effective, and has been studied in non-malignant pain syndromes, and the evidence is not strong.  Tricyclics were not shown to be effective for treating the neuropathic pain of HIV, and nor is there sufficient evidence to support serotonin-specific reuptake inhibitors (SSRIs) or St John’s Wort for use in neuropathic pain.
A systematic review concluded that evidence for most anticonvulsants in cancer pain is lacking.  Only gabapentin has been specifically studied in cancer pain, and a systematic review suggests that it is effective, although these findings are derived from a single study.  The evidence for carbamazepine has also been separately reviewed, and supports its role in treating non-malignant pain, although the numbers are very small, and cancer pain was not included.  The newer agent, pregabalin, has not been studied in cancer, but in a systematic review of studies in non-malignant pain NNTs ranged from 3 – 11.  Lamotrigine has been shown not to be effective as an adjuvant. 
Combining antidepressants or anticonvulsants with opioids
Opioids have efficacy in the treatment of neuropathic pain.  A recent systematic review studied the strategy of combining opioids with non-opioid adjuvants and found that the combination was effective within 4 – 8 days, however the magnitude of the improvement is likely to be small (1 point on a 1-10 pain scale), and the rate of adverse events is increased.  The studies that were included in this review were short – fourteen days was the longest duration – and confirmed that opioids were likely to be effective on their own, and also that a significant placebo effect should be anticipated.
Systemic local anaesthetics
A systematic review  suggested that lignocaine and its oral analogues (eg, mexiletine) were as effective as other analgesics in trials where they were compared with carbamazepine, gabapentin, or morphine, although the data were limited. Systemic local anaesthetics appeared to be safe when used in the context of the clinical trials.
Its role as an adjuvant has been reviewed [8,24] but the evidence is limited due to very small studies. There is inadequate and contradictory data from controlled trials to demonstrate effectiveness, and its use in palliative care is therefore still controversial, until further evidence becomes available.
Non-steroidal anti-inflammatory drugs
An evidence-based report on the management of cancer pain reviewed studies on non-steroidal anti-inflammatory medications, and found that there was evidence that NSAIDs are efficacious, but could not find any differences between different NSAIDS. [18,26] There was also no difference between the efficacy of NSAIDS with and without opioids. Side effects of NSAIDs are potentially significant, including gastric ulceration. The clinical benefit of drugs for gastric protection when treating with NSAIDs is not clear, and it is not known which drugs are most effective in this role.  Recent evidence-based guidelines from the EAPC recommend the use of NSAIDs to improve analgesia or as opioid sparing agents, however the adverse effect profile is also highlighted, and suggest that paracetamol should therefore be preferred on this basis. 
In the management of metastatic bone pain, systematic reviews support the effectiveness of palliative radiotherapy  and bisphosphonates.  Whilst bisphosphonates have no immediate analgesic effect, they are recommended as an adjuvant where the response to radiotherapy and analgesia are inadequate. There is also evidence to support their role in pain management in the treatment of multiple myeloma,  advanced prostate cancer with bone metastases,  and for prevention of skeletal problems in women with advanced breast cancer and clinically evident bone metastases.  A systematic review of studies of calcitonin in bone pain found no evidence to support its use as an adjuvant treatment in bone pain.  Whether schedules involving combinations of external beam radiotherapy and bisphosphonates improves outcomes is an issue which is being studied. 
External beam radiotherapy is the mainstay of treatment of painful bone metastases.  Single fraction radiotherapy to palliate painful bone metastases is as effective as multiple fractions for controlling pain.  Systematic reviews show that more patients who have had single fraction radiotherapy may require retreatment for pain, and there is a non-significant trend to more fractures. [18,26] When there are multifocal painful bone metastases, treatment with radioisotopes such as strontium-89 or samarium-153 can produce sustained analgesia, although they are associated with bone marrow toxicities. [27-30]
A systematic review has examined studies of topical capsaicin in painful neuropathic conditions, and concluded that it may be helpful, although the evidence is varied not robust. This agent is associated with local skin irritation and burning, but may be considered in patients who do not tolerate other options. 
There is some emerging evidence that antipsychotics may provide analgesia in both acute and chronic pain, with a number-needed-to-treat of 2.6, although the findings are mixed. Extra-pyramidal side-effects were noted in these studies. 
What it means in practice
Adjuvants should be added when neuropathic or other pain is not fully responsive to opioids. Antidepressants and anticonvulsants provide a small improvement in analgesia, however this is at the cost of an increase in adverse events. The choice of a specific adjuvant relates to clinical context and to the specific side-effect profile of the treatment combination and patient characteristics.
The role of gabapentin in neuropathic pain is evolving. There is evidence to support its efficacy in cancer pain, and recent evidence-based guidelines now recommend it as one of the first line treatments for neuropathic pain. [33-34,36] However, it does not have a listing for treating pain in the current PBS. Tricyclics and venlafaxine are alternative recommended first line drugs from the antidepressant class.
The antidepressant secondary amine tricyclics (nortriptyline, despiramine) have less anticholinergic side effects and are often better tolerated than the frequently used tertiary amines, such as amitriptyline.  However, there is strong evidence to support the use of amitriptyline, and it is recommended as an adjuvant by the EAPC.  Antipsychotics, including antypical antipsychotics, have a potential role as adjuvant analgesics, although the evidence is not strong.
For bone pain due to metastases, external beam radiotherapy is an effective treatment and may, if possible, be offered first line along with opioid analgesia. Bisphosphonates may be used if the therapeutic response is inadequate, but their analgesic effect is not immediate. There is no role for calcitonin in managing bone pain. Radioisotope treatment may be highly effective for multiple sites of bone pain, but is associated with the potential for significant marrow toxicity.
In considering a radiotherapy fractionation schedule the palliative goals, prognosis, and whether the problem is uncomplicated, should be taken into account. Whilst patients who have had single fraction (hypofractionated) radiotherapy may require re-treatment more often than those treated with other schedules, overall, shifting to hypofractionated schedules reduces the burden on patients and the workload of radiotherapy units  and the practice is supported by evidence. 
NSAIDs are effective analgesics for musculoskeletal pain in cancer. No differences have been shown between different NSAIDs. Recommended doses produce close to the maximum benefit, whilst side effects increase linearly with dose. There is therefore little clinical benefit to be gained from using high dose non-steroidals.