British Journal of British Journal of Pharmacology (2018) 175 2726–2736 2726

Pharmacology

Themed Section: Emerging Areas of Opioid Pharmacology

REVIEW ARTICLE
Influence of opioids on immune function in
patients with cancer pain: from bench to
bedside
Correspondence Jason W Boland, Senior Clinical Lecturer and Honorary Consultant in Palliative Medicine, Wolfson Palliative Care
Research Centre, Hull York Medical School, University of Hull, Hull HU6 7RX, UK. E-mail: jason.boland@hyms.ac.uk

Received 3 January 2017; Revised 23 May 2017; Accepted 30 May 2017

Jason W Boland1 and A Graham Pockley2

1
Wolfson Palliative Care Research Centre, Hull York Medical School, University of Hull, Hull, UK, and 2John van Geest Cancer Research Centre,
Nottingham Trent University, Nottingham, UK

In patients with cancer, opioids are principally used for the management of acute surgical and chronic cancer-related pain.
However, opioids have many non-analgesic effects, including direct and indirect effects on cancer cells and on anti-tumour
immunity (NK cells, macrophages and T-cells). Direct effects on immune cells are manifested via opioid and non-opioid toll-like
receptors, whereas indirect effects are manifested via the sympathetic nervous system and hypothalamic–pituitary–adrenal axis.
Opioids can also decrease/alter immune cell infiltration into the tumour micro-environment. Animal models have shown that this
is not a class effect, in that morphine and fentanyl suppress NK cell cytotoxicity; buprenorphine does not affect NK cell
cytotoxicity, whereas tramadol increases NK cell cytotoxicity, reducing metastasis. In healthy individuals, morphine suppresses
and fentanyl enhances NK cell cytotoxicity. In patients undergoing surgery, fentanyl decreased and tramadol increased NK cell
cytotoxicity; clinical outcomes were not determined. Meta-analyses of opioid-sparing surgical studies report an association
between improved recurrence-free and/or overall survival with regional/neuraxial anaesthesia compared with systemic opioids. In
patients receiving opioids for non-surgical cancer-related pain, morphine has variable effects on immunity; clinical outcomes were
not assessed. Although there is a potential association between systemic opioid administration and shorter survival in cancer
patients with a prognosis of months to years, studies have not been designed to primarily assess survival, as a consequence of
which causality cannot be apportioned. Pain is immunosuppressive, so analgesia is important. Opioids for cancer-related pain will
continue to be recommended until definitive data on the effects of opioids on clinical outcomes in specific patient groups
becomes available.

LINKED ARTICLES
This article is part of a themed section on Emerging Areas of Opioid Pharmacology. To view the other articles in this section
visit http://onlinelibrary.wiley.com/doi/10.1111/bph.v175.14/issuetoc

Abbreviations
IVME, intravenous morphine-equivalent dose; TLR4, toll-like receptor 4

DOI:10.1111/bph.13903 © 2017 The British Pharmacological Society

 Opioids and immune function in patients with cancer pain

Plain language summary developing, cancer. Many immune cells are involved in anti-
tumour immunity including NK cells, T-cells, mast cells,
Some opioids influence the functioning of the immune dendritic cells and macrophages, as well as soluble immune
system, and this has been shown to affect cancer growth mediators such as cytokines and chemokines (Table 1) (Nguyen
and progression in animal models. Although some opioids et al., 2014; Boland et al., 2014b; Liang et al., 2016). NK cell
affect different aspects of immune function in humans, their activity is a critical endpoint in the immunotoxicological
effect on the cancer or the patient remains unknown. evaluation of pharmaceuticals (EMEA, 2000), and a consensus
However, some of the surgical studies to date report an statement for the use of opioids for alleviating chronic pain
association between improved clinical outcomes with lower in elderly people proposes that the effects of opioids on im-
doses of systemic opioids. Pain suppresses immunity, and mune function should be considered (Pergolizzi et al., 2008).
immunity is important for protecting against cancer. Despite Given that patients with cancer are also at risk of
the uncertainty, adequate control of cancer-related pain is infection, it is important that the anti-infection arm of the
important and opioids will therefore continue to be immune system is maintained at an effective level. However,
recommended for its management. there is evidence that opioids could increase the risk of
infection in patients with cancer (Suzuki et al., 2013; Shao
et al., 2017), as a retrospective study has shown that patients
Opioids and the immune system treated with morphine developed more infections than
those treated with oxycodone (Suzuki et al., 2013). Further-
Opioids are a diverse range of drugs that act on opioid more, the risk of infection has been shown to increase by 2%
receptors. In patients with cancer, opioids are principally for each 10 mg rise in the oral morphine equivalent daily
used for the management of cancer-related pain in the long- dose, with no difference between opioids (Shao et al., 2017).
term, and cancer surgery pain in the shorter term. As with Opioids can influence the establishment and progression
all drugs, opioids can elicit a range of undesired consequences of cancer in many ways, including indirect effects on cancer
(Boland et al., 2013), including the functioning of the growth via angiogenesis and host immunity (Gupta et al.,
immune system (Grace et al., 2015; Wigmore and Farquhar- 2002; Afsharimani et al., 2011; Gach et al., 2011; Koodie
Smith, 2016; Ramaswamy and Langford, 2017). et al., 2014). Opioids also have the potential to directly
The immune system has a crucial role in controlling and influence the growth of cancer cells, as cancer tissue
potentially eradicating cancer cells in individuals with, or overexpresses opioid μ receptors (Mathew et al., 2011;

Table 1
Role and function of the main immune cells

Cell Role Function Arm

Dendritic cell Antigen presentation Presentation of antigenic peptides in the Innate
context of MHC classes I and II molecules
and the delivery of essential
co-stimulatory molecules
Natural killer cell Anti-tumour Release of cytotoxic molecules Innate
Anti-viral (granzymes and perforin)
Neutrophil Anti-bacterial/fungal Phagocytosis and oxidative burst Innate
Monocyte–macrophage Anti-bacterial/fungal Phagocytosis and oxidative burst Innate
lineage
+
CD4 T-cell Immune coordination/regulation Regulating the activity of Adaptive
other immune cells
Regulatory T-cell Immune regulation/suppression Immune system modulation, Adaptive
+ + +
(CD4 , CD25 , FOXP3 , maintaining tolerance to self-antigens,
low
CD127 , plus other markers) preventing autoimmune disease and
potential barrier to the development
of protective anti-tumour immunity
+
CD8 T-cell Cytotoxicity Eradication of virally infected cells and Adaptive
cancer cells. Induction of apoptosis
by (i) release of cytotoxins
(perforin, granulysin and
granzymes) and (ii) direct
cell–cell contact
B cell Antibody production Antibody production Adaptive
Adapted with permission by British Journal of Cancer from (Boland et al., 2014b); http://www.nature.com/bjc/journal/v111/n5/full/bjc2014384a.
html.

British Journal of Pharmacology (2018) 175 2726–2736 2727

 J W Boland and A G Pockley

Lennon et al., 2012; Nguyen et al., 2014; Singleton et al., opioids can modulate immune function by direct effects on
2014). Overexpression of the μ receptor on cancer cells in immune cells and via indirect effects which involve the CNS
murine models has been shown to increase the growth of and its release of immune mediators (Borner et al., 2008;
lung cancer (Lennon et al., 2014), and decreasing μ receptor Campana et al., 2010; Al-Hashimi et al., 2013).
activity (using μ receptor small hairpin RNA mice or the For opioids to have direct effects on immune cells,
opioid receptor antagonist methylnaltrexone) has been immune cells must express opioid receptors, or opioids must
shown to reduce the growth of lung cancer (Mathew et al., be able to have effects via non-opioid receptors present on
2011). Furthermore, μ receptor overexpression has been asso- immune cells, e.g. toll-like receptor 4 (TLR4) (Xie et al.,
ciated with the development of metastasis in patients with 2017). Although morphine can directly interact with μ recep-
prostate, oesophageal and lung cancer (Zylla et al., 2013; Sin- tors on immune cells in vitro (Borner et al., 2008), the presence
gleton et al., 2014; Zhang et al., 2015). Morphine influences of functional opioid receptors on immune cells, as deter-
the proliferation and survival of cancer cells via direct effects mined using radioligand antibody binding studies, Western
on tumour cell DNA cleavage, MAPK, Src, GRB2-associated blot and PCR analysis continues to be disputed (Williams
binding protein 1 (Gab-1), PI3K, Akt and STAT3 signalling et al., 2007; Borner et al., 2009; Kraus, 2009; Campana et al.,
pathways (Mathew et al., 2011; Lennon et al., 2012; 2014). 2010; Glattard et al., 2010; Langsdorf et al., 2011; Al-Hashimi
The remainder of this review will concentrate on the et al., 2013; 2016). One possibility for the discrepancy is that
cancer-related immune effects of opioids. μ receptor expression depends upon immune cell activation,
in that the expression of μ receptors in T-cells is non-
constitutive (Borner et al., 2008). In activated human T-cells,
Mechanisms of immune effects of μ receptor mRNA is increased to levels about 1% of those in
opioids neurons and can produce functional μ receptors in T-cells
(Borner et al., 2007). Many extracellular signals, including
Numerous mechanisms underlying the influence of opioids the cytokines IL-1, IL-4, IL-6, TNF and IFN-γ, control μ
on immune cells have been described (Figure 1). In vivo, receptor gene transcription and cell surface opioid receptor

Figure 1
Peripheral and central mechanisms of opioid-induced immune suppression. Opioids can have direct effects on immune cells, which express
appropriate receptors such as the μ receptor (MOR) and toll-like receptor 4 (TLR-4). They can also have immunosuppressive effects via central
mechanisms. Acute opioid administration enhances activity in the periaqueductal gray (PAG) matter which activates the sympathetic nervous
system (SNS). The SNS innervates lymphoid organs, such as the spleen, and this activation induces the release of biological amines which suppress
splenic lymphocyte proliferation and NK cell cytotoxicity (Fecho et al., 1996; Irwin et al., 1988). Furthermore, prolonged use of opioids increases
hypothalamic pituitary adrenal (HPA) axis activity and glucocorticoid production, which decrease NK cell cytotoxicity (Fecho et al., 1996; Mellon
and Bayer, 1998). Morphine can also act via dopamine D1 receptors in the nucleus accumbens shell, increasing the release of neuropeptide Y
(NPY) and reducing splenic NK cell cytotoxicity in rodent models (Saurer et al., 2006). Y1R; Y1 receptor. Reproduced from Boland et al.
(2014b) with permission by the British Journal of Cancer; http://www.nature.com/bjc/journal/v111/n5/full/bjc2014384a.html.

2728 British Journal of Pharmacology (2018) 175 2726–2736

 Opioids and immune function in patients with cancer pain

expression in immune cells (Kraus, 2009; Langsdorf et al., 2016; Wang et al., 2016). Incubating blood from patients
2011). TLR4 is involved in innate immune system activation. with gastric or breast cancer with morphine, sufentanil or
In vitro and in silico techniques have shown that many fentanyl in vitro increases the number of regulatory T-cells
commonly used opioids can have direct effects on innate (Gong et al., 2014; Hou et al., 2016). Furthermore, although
immunity via activation of the TLR4 (Hutchinson et al., perioperative treatment of women undergoing breast cancer
2010; Franchi et al., 2012; Grace et al., 2015; Khabbazi et al., resection with sufentanil or fentanyl equally decreased
2016a,b). regulatory T-cell numbers at day 1, numbers were increased
There are two principal pathways via which the central on day 7 (Gong et al., 2014). Given the capacity of regulatory
immunosuppressive effects of morphine can be mediated. T-cells to inhibit anti-tumour immunity, these effects could
These effects have been primarily identified on the basis of therefore be detrimental (Takeuchi and Nishikawa, 2016).
animal model-derived data and might therefore differ between
species (Fecho et al., 1996; Al-Hashimi et al., 2013). Acute mor-
phine administration acts on the periaqueductal gray and sym-
pathetic nervous system (which innervates lymphoid organs) Animal studies: anti-tumour immunity
to release immunosuppressive biological amines that can sup- and tumour growth
press NK cell cytotoxicity (Irwin et al., 1988; Weber and Pert,
1989; Hernandez et al., 1993; Fecho et al., 1996; Gomez-Flores Many studies have described how opioids can affect
and Weber, 2000). An additional arm to the sympathetic ner- immunity, most of which have been based on the acute
vous system pathway is the action of acute morphine/ administration of morphine. However, the duration of opioid
diamorphine via D1 receptors in the nucleus accumbens shell. administration (as well as the opioid itself) is likely to lead to
This increases the release of neuropeptide Y from the sympa- differential effects on immune function (Liang et al., 2016;
thetic nervous system, which in turn acts on peripheral Y1 Xie et al., 2017).
receptors to inhibit splenic NK cell cytotoxicity (Saurer et al., NK cells are key participants in effective tumour
2006, 2009). Prolonged use of opioids increases activity in the immunosurveillance. Suppression of NK cell cytotoxicity cor-
hypothalamic–pituitary–adrenal axis, thereby increasing the relates with increased tumour growth and metastasis in ani-
release of immunosuppressive glucocorticoids and decreasing mal models (Gaspani et al., 2002; Shavit et al., 2004; Franchi
T-cell function and NK cell cytotoxicity. As pain/stress activates et al., 2007). The literature generally reports that morphine
this axis, opioids used as analgesics might be protective (Yeager et al., 1995; Franchi et al., 2007) and fentanyl
(Hernandez et al., 1993; Mellon and Bayer, 1998; Borman (Martucci et al., 2004; Shavit et al., 2004; Franchi et al.,
et al., 2009; Zhang et al., 2011). 2007) decrease NK cell cytotoxicity, that buprenorphine
As well as having direct and indirect effects on immune does not affect NK cell cytotoxicity (Martucci et al., 2004;
cell activation, μ receptor activation can inhibit NK cell Franchi et al., 2007) and that tramadol enhances it
migration. Fewer NK cells are in the tumours of wild-type (Sacerdote et al., 2000; Gaspani et al., 2002) (Table 2). How-
compared to μ receptor knockout mice (Boehncke et al., ever, effects on NK cell cytotoxicity are likely to be dose-
2011). Morphine has been shown to reduce leukocyte dependent, as low-dose morphine has been shown to stimu-
migration in a murine lung tumour model (Koodie et al., late cytotoxicity and high-dose morphine to induce an initial
2014), and NK cell infiltration into breast cancer tissue is increase in cytotoxicity followed by a suppression of cytotox-
decreased in women who have received more perioperative icity in pigs (Borman et al., 2009).
systemic opioids (Desmond et al., 2015). For effective anti- The differential effects of opioids on immune function
tumour immunity, immune cells need to be in the tumour and the development of tolerance are illustrated by a series
micro-environment, and a lack of immune cell infiltration of murine experiments by Martucci et al. (2004). In these
could be detrimental to the establishment of protective studies, single dose fentanyl only decreased lymphocyte pro-
anti-tumour immunity. However, morphine has been shown liferation, whereas chronic fentanyl reduced lymphocyte
to be protective in a murine Lewis lung carcinoma model by proliferation and splenic NK cell cytotoxicity, IL-2 and IFN-γ
reducing immune cell infiltration and angiogenesis (Koodie production at 24 h. NK cell cytotoxicity recovered by day 3,
et al., 2014). Morphine can also elicit anti-tumour effects that whereas lymphocyte proliferation, IL-2 and IFN-γ production
are mediated via modulation of paracrine communication only recovered by day 7. After the 7 day infusion, the admin-
between cancer cells and non-malignant cells in the tumour istration of twice the dose of fentanyl for 1 or 3 days to these
micro-environment (Afsharimani et al., 2014) and can mice did not affect any immune parameter, thereby suggest-
prevent the pro-angiogenic interaction between macro- ing that once developed, increasing the dose did not over-
phages and breast cancer cells (Khabbazi et al., 2016a,b). come immunological non-responsiveness (Martucci et al.,
Furthermore, morphine modulates cells in the tumour 2004). In the same study, buprenorphine had no effect on
micro-environment via many mechanisms, some of which any of the measured parameters at any time point (Martucci
increase and others decrease tumour aggressiveness et al., 2004).
(Khabbazi et al., 2015). The effect of the timing of morphine administration on
However, it should be noted that not all immune cells are cancer progression has been studied in a murine model of
beneficial to the anti-tumour response. Immunosuppressive breast adenocarcinoma. If given before the development of
‘regulatory’ T-cell populations can suppress the immune re- cancer, morphine had no effect on cancer growth, whereas
sponse (Takeuchi and Nishikawa, 2016), and meta-analyses morphine increased the progression of established cancer
have reported that increased regulatory T-cell infiltration into and decreased survival via interactions with tumour μ recep-
breast cancers is associated with a poorer survival (Shou et al., tors (Nguyen et al., 2014).

British Journal of Pharmacology (2018) 175 2726–2736 2729

 J W Boland and A G Pockley

Table 2
Summary of the effects of opioids on NK cell cytotoxicity and cancer outcomes in the different models

Animal Healthy volunteer Human cancer surgery

Opioid NK cell cytotoxicity Cancer NK cell cytotoxicity NK cell cytotoxicity Clinical cancer outcomes
Morphine Decrease/no effect on Increase Decrease No effect/prolonged Not assessed
suppressive effects of post-operative suppression
surgery
Fentanyl Decrease/no effect on Increase Increase Prolonged post-operative Not assessed
suppressive effects of suppression
surgery
Buprenorphine No effect/reversed the No effect Not assessed No effect/reversed the Not assessed
suppressive effects of suppressive effects of
surgery surgery
Tramadol Increase/reversed the Decrease Not assessed Increase/reversed the Not assessed
suppressive effects of suppressive effects of
surgery surgery

Several surgery-based studies have assessed how different et al., 2002). For potentially immunosuppressive opioids,
opioids, and the timing of opioid administration, affect there could be a balance between the immunosuppressive
cancer progression. In a series of rodent experiments, Franchi effect of the opioid and the reduction of putative immuno-
and co-workers showed that morphine and fentanyl suppression of pain (Figure 2; Page, 2005).
decreased NK cell cytotoxicity and increased MADB106 cell- A recent study using mouse models of breast cancer has
derived lung metastases, whereas buprenorphine did not shown that perioperative morphine did not affect tumour
(Franchi et al., 2007). Surgery alone decreased NK cell cytotox- growth, peripheral blood or tumour-infiltrating immune cells
icity and increased MADB106 lung metastases. In the surgical (although NK cells were not assessed), even when surgery was
model, perioperative buprenorphine reversed the suppressive performed (Doornebal et al., 2015). The difference between this
effects of surgery on NK cell cytotoxicity and MADB106 lung and the other presented studies could be that the tumours in
metastases, whereas morphine and fentanyl did not (Franchi this study are grown in vivo and transplanted from mouse to
et al., 2007). Decreased NK cell cytotoxicity correlated with mouse without being cultured on plastic, the orthotopic rather
high corticosterone levels, thereby suggesting the involvement than intravenous inoculation of cancer cells, female mice being
of the hypothalamic–pituitary–adrenal axis (Franchi et al., used (different species and genders could respond differently)
2007). Fentanyl has been shown to suppress NK cell cytotoxic- and the 14 day dosing schedule of morphine which was initi-
ity in rats, and this suppression correlates to increased tumour ated once mammary tumours had developed, or 1 day after
load when fentanyl is administered close to the time of mastectomy (Afsharimani et al., 2015; Doornebal et al., 2015).
MADB106 tumour cell inoculation, but not when fentanyl
was administered 6 h before tumour inoculation (Shavit et al.,
2004). Thus, the acute administration of morphine and fenta- Healthy volunteer studies
nyl around the time of tumour resection in animals could be
detrimental to immune function and cancer outcomes. Several studies have assessed the effect of opioids in healthy
In another series of rodent experiments, Gaspani et al. volunteers (Table 2). These have the advantage of being able
demonstrated that tramadol increased, and morphine
decreased NK cell cytotoxicity in non-operated rats (Gaspani
et al., 2002). Surgery reduced NK cell cytotoxicity, and this
reduction in cytotoxicity correlated with increased numbers
of MADB106 cell-derived lung metastases. Morphine did
not influence surgery-induced NK cell cytotoxicity suppres-
sion, whereas tramadol prevented this and reduced lung me-
tastasis (Gaspani et al., 2002). The (+) enantiomer of tramadol
inhibits 5-HT (serotonin) uptake and is immunostimulatory,
and this effect is inhibited by the 5-HT receptor antagonist
metergoline (Sacerdote et al., 1999). It has also been shown
Figure 2
that an increase in serotoninergic tone stimulates NK cell cy-
Opioids, pain, immunity and cancer. Pain is immunosuppressive and
totoxicity (Mossner and Lesch, 1998). Thus, the effect of
may worsen outcomes in rodent models of cancer (Page, 2003; Page
tramadol on NK cell cytotoxicity following surgery might be et al., 2001). By reducing pain, opioids might have a beneficial effect
the summation of stimulation of NK activity, which is also on immune function and cancer (Gaspani et al., 2002; Page, 2005).
present in non-operated rats, due to 5-HT (serotonin) re- However, opioids which suppress immune function may decrease
uptake inhibition, and the reduction of surgical pain which anti-tumour immunity and promote the development of cancer;
is itself immunosuppressive (Page et al., 2001; 2003; Gaspani the balance of these effects is critical (Shavit et al., 2004).

2730 British Journal of Pharmacology (2018) 175 2726–2736

 Opioids and immune function in patients with cancer pain

to evaluate the effect of opioids on both direct and indirect numbers had not returned to baseline at 5 days after surgery.
opioid-immune pathways in humans, as opposed to in vitro In the flurbiprofen group, NK cell numbers at 2 h and T and
studies which can only assess direct pathways, albeit in a very NK cell numbers at 1 day were higher, despite similar levels of
controlled way (Boland et al., 2014a). However, it should be analgesia (Shen et al., 2014). In a randomized controlled trial
noted that healthy volunteer study results cannot necessarily of 25 patients undergoing neck surgery, fentanyl suppressed
be extrapolated to patients with cancer, as the inflammatory NK cell cytotoxicity more than flurbiprofen on day 1, but not
and immune cell activation context are different. day 2 post-operatively (Narahara et al., 2013). The difference
In a volunteer study, morphine was shown to decrease NK could either be a beneficial effect of the flurbiprofen or a
cell cytotoxicity; with higher doses, it took over 24 h for the negative effect of the opioid (Hooijmans et al., 2015; Boland
NK cell cytotoxicity to normalize after the morphine was and Bennett, 2016a).
stopped (Yeager et al., 1995). In two volunteer studies, acute Although NK cell cytotoxicity correlates with tumour
administration of intravenous fentanyl increased NK cell growth and metastasis in animal models, cancer outcomes
cytotoxicity, an effect which resulted from an increase in have not been assessed in human cancer surgery settings
the proportion of NK cells in the peripheral blood, rather (Table 2). Furthermore, the magnitude of change in NK cell
than an increase in the cytotoxicity of individual NK cells cytotoxicity which is needed to produce a clinically relevant
(Jacobs et al., 1999; Yeager et al., 2002). Although these effect is unknown. It is therefore difficult to definitively
studies suggest that morphine suppresses NK cell cytotoxicity attribute a decrease in NK cell cytotoxicity to poorer
in healthy subjects, with a more prolonged suppression at a outcomes in patients with cancer.
higher dose, and that fentanyl increases NK cell cytotoxicity,
we cannot extrapolate healthy volunteer studies to the
surgical/clinical setting. Systemic opioid sparing surgical studies
Many studies have assessed the effects of regional analgesia
Human cancer surgical studies versus systemic opioids on clinical outcomes. In general,
these compare clinical outcomes between pain control with
There is preclinical evidence that the immune response at the regional/neuraxial analgesia and systemic opioids in patients
time of surgical removal of cancer is critical, that some having cancer surgery.
opioids can be detrimental to the immune response and that Meta-analyses for the effect of neuraxial blockade on
this can promote cancer progression (Gaspani et al., 2002; cancer surgery have generated mixed findings from the
Shavit et al., 2004; Byrne et al., 2016). As in vitro conditions of heterogeneous and mostly retrospective studies that were
most studies do not reproduce the biology of cancer in vivo, included. One concluded that there was no advantage for
and that there are differences between opioids (Borner et al., overall or progression-free survival (Cakmakkaya et al.,
2013), immune response and clinical outcomes to different 2014), whereas another suggested there might be a benefit
opioids need to be studied in humans undergoing surgical for neuraxial blockade for prostate cancer surgery, but not
resection of cancer. During surgery, opioids are just one of for colonic cancer (Pei et al., 2014). The most recent and
many factors that might affect cancer progression/metastasis largest meta-analysis (incorporating findings from 21 studies)
(Byrne et al., 2016). Opioids also influence cancer growth in reported an association between improved recurrence-free
animal models via numerous mechanisms, with the potential and overall survival with neuraxial anaesthesia (compared
influence on anti-tumour immunity being just one of these to general anaesthetic alone), especially in patients having
(Afsharimani et al., 2011; Gach et al., 2011; Jaura et al., 2014). colorectal cancer surgery (Weng et al., 2016).
In patients undergoing surgery, surgical stress variably Many reviews have considered the various perioperative
reduces NK cell cytotoxicity, and different opioids (and the factors which might influence cancer-related outcomes. They
dose used) have different effects on NK cell cytotoxicity. In suggest that although preclinical studies indicate a benefit of
30 patients undergoing surgery for uterine carcinoma to regional anaesthesia and stress response reduction in cancer
whom were administered fentanyl, anaesthetic medications formation, there is no clear association between regional/
and either morphine or tramadol, NK cell cytotoxicity was neuraxial anaesthesia for cancer surgery and tumour
not affected by the surgery or by morphine (trend for recurrence and cancer-related survival benefit from the
inhibition only). However, tramadol increased NK cell heterogeneous and mostly retrospective studies. However,
cytotoxicity at 2 h post-treatment, despite similar analgesia there might be an association with improved overall survival
(Sacerdote et al., 2000). In a randomized controlled trial of (Sun et al., 2015; Vogelaar et al., 2016). This needs confirma-
40 patients undergoing surgery for malignant or benign tion and causality explored in prospective studies.
conditions, post-operative NK cell cytotoxicity was reduced, A randomized controlled trial of 503 patients undergoing
with high-dose fentanyl delaying the post-operative recovery abdominal surgery for cancer resection compared the effect of
of NK cell cytotoxicity (Beilin et al., 1996). The long-term general anaesthesia with either epidural analgesia or post-
impact or overall outcome of different opioids and doses were operative systemic opioids (median 3 day morphine dose:
not determined. 0 mg epidural group; 107 mg opioid group) and showed no
The influence of morphine with or without flurbiprofen difference in cancer recurrence and mortality at 2–3 years
(the flurbiprofen group consumed less morphine) on immune between the two groups (Myles et al., 2011). In a prospective
status has been evaluated in 60 patients undergoing surgery cohort study of 34 188 cancer survivors who had resections
for gastric cancer. In the morphine-treated patients, T and NK for early-stage breast cancer, opioid prescriptions were not
cell numbers decreased at 2 h after incision, and NK cell associated with breast cancer recurrence (Cronin-Fenton

British Journal of Pharmacology (2018) 175 2726–2736 2731

 J W Boland and A G Pockley

et al., 2015). This echoes previous studies indicating that described an association between strong systemic opioid use
opioids do not increase de novo cancer risk. or increasing dose and shorter survival. However, these
As surgical stress response and increased glucocorticoids studies did not have survival as a primary, appropriately
can cause immune suppression, some of the benefits from powered, endpoint. Furthermore, they were limited, in that
regional anaesthesia may result from better analgesia they included variable populations, starting points for opioid
(reducing pain-associated immune suppression), along with use, durations of opioid administration and from when
attenuation of the surgical stress response, decreases in levels survival was measured. The main confounding factor is that
of endogenous opioids and the lower doses of systemic opi- the control groups were not directly matched (i.e. not
oids (Al-Hashimi et al., 2013; Juneja, 2014; Byrne et al., 2016). patients who had refractory severe symptoms but did not
Data on the influence of opioid-sparing techniques for choose opioids) and that greater analgesic requirements and
the surgical resection of cancer are mixed and are primarily shorter survival are likely to be mediated by painful progres-
derived from retrospective studies. Although the most recent sive cancer. As a consequence, these studies cannot show
meta-analysis reported an association between improved causality, only associations (Boland et al., 2015).
recurrence-free and overall survival with neuraxial anaesthe- Pharmacogenetic factors are also important. In 2039
sia (Weng et al., 2016), more randomized controlled trials women with breast cancer, the A118G μ receptor polymor-
are needed in different cancers to elucidate causation, some phism, which confers a reduced receptor response to opioids,
of which are underway (Buggy et al., 2015; Byrne et al., was associated with increased cancer-related survival in
2016; Connolly and Buggy, 2016). In the meantime, there is invasive breast cancer (Bortsov et al., 2012). Although, opioid
consensus that there is currently insufficient evidence to consumption was not recorded, patients with the A118G
change perioperative practice until the results of definitive polymorphism need a higher morphine dose for analgesia.
randomized controlled trials are available (Buggy et al., It is thus likely that the A118G patients received more opioids
2015; Byrne et al., 2016; Connolly and Buggy, 2016). and had increased cancer-related survival (Bortsov et al.,
2012). Furthermore, Chinese people with the A118G μ
receptor polymorphism had a lower incidence of oesophageal
cancer (Wang et al., 2013). However, this might be cancer-
type dependent, as North-eastern Polish patients with the
Effects of opioids on anti-tumour A118G allele are more likely to develop breast cancer
immunity and survival in patients with (Cieslinska et al., 2015).
cancer
Many patients who are not undergoing surgery receive long- Future studies
term opioids for cancer-related pain, which is a different
scenario in terms of the immune effects of opioids and their The variable findings from in vitro, animal or healthy volun-
clinical consequences. teers cannot be extrapolated to clinical settings, between
Effects of opioids on immunological parameters that are the different clinical settings or between opioids (Juneja,
relevant to anti-tumour immune potential in patients with 2014; Ramaswamy and Langford, 2017). Therefore, appropri-
cancer have been reviewed previously (Boland et al., 2014b). ate prospective studies in each of the clinical settings in
Five human studies which assessed the immune effects of which they are used are required. These need to focus on
morphine in patients with cancer showed variable effects on long-term clinical studies and be undertaken in patients
immunological end points. Given that none of these studies needing long-term continuous and intermittent opioids for
measured the clinical effects, it is not possible to know the non-surgical cancer-related pain, in specific cancer types
clinical significance of these (Boland et al., 2014b). (Xie et al., 2017). They should also elucidate mechanisms by
Opioids, via immune (and non-immune) effects, can analysing the tumours for mitogenic activity, immune cell
influence cancer progression in animal studies and other status in circulation and tumours, pain, symptoms and
patient groups. The effect of opioids on prognosis in patients survival-related outcomes. In vitro, animal and healthy
with cancer not undergoing surgery has been systematically volunteer studies should principally be used to explore
reviewed (Boland et al., 2015). In the 13 studies of patients mechanisms, as the environments can be carefully
in the last days/weeks of life, there were mixed effects of controlled. These should use clinically relevant conditions,
opioids on survival. These studies were short term with poor including opioid concentrations, in the experimental design
methodology, sometimes only including patients with a very (Boland and Pockley, 2016b). Furthermore, subset analysis
limited life expectancy such as from hospice admission, or of NK cells is needed (as CD56bright and CD56dull NK cells
only measured opioids that were taken in the last day(s) of life exhibit different functionality) (Tabellini et al., 2014), and
(Boland et al., 2015). The best quality end-of-life study, a pharmacogenomics, such as the A118G μ receptor polymor-
secondary data analysis, showed that an intravenous phism, should be taken into account.
morphine-equivalent dose (IVME) above 20 mg·day 1 was as- Patients taking opioids for cancer-related pain could be on
sociated with a shorter survival compared with ≤17 mg·day 1 opioids for a long time; thus, tolerance could develop. Given
(the lowest dose group); mean survival: 27 days for patients that they might also be on opioids intermittently, it is impor-
on ≤17 mg·day 1 IVME versus 12 days for patients on tant to know if regular or intermittent opioids are more
20–25 mg·day 1 IVME (Portenoy et al., 2006). Studies in immune protective and how this influences clinical out-
patients with a longer prognosis (months to years) tended comes. As pain is immunosuppressive, the effects of pain in
to be larger and of better quality. Six out of seven studies patients and how opioids influence immunity and clinical

2732 British Journal of Pharmacology (2018) 175 2726–2736

 Opioids and immune function in patients with cancer pain

outcomes in these populations need further study. Appropri- Alexander SPH, Fabbro D, Kelly E, Marrion N, Peters JA, Benson HE
ately designed and powered studies assessing clinical out- et al. (2015b). The Concise Guide to PHARMACOLOGY 2015/16:
comes of opioid use in patients with cancer are required to Catalytic receptors. Br J Pharmacol 172: 5979–6023.
inform the optimal use of opioids in these patient groups. Al-Hashimi M, McDonald J, Thompson JP, Lambert DG (2016).
Evidence for nociceptin/orphanin FQ (NOP) but not micro (MOP),
delta (DOP) or kappa (KOP) opioid receptor mRNA in whole human
Conclusion blood. Br J Anaesth 116: 423–429.

Al-Hashimi M, Scott SW, Thompson JP, Lambert DG (2013). Opioids

Evidence from preclinical, healthy volunteer, clinical and and immune modulation: more questions than answers. Br J Anaesth
surgical models suggests that different opioids variably influ- 111: 80–88.
ence protective anti-tumour immunity. There are discrepan-
cies in the results of these studies, which might be partly Beilin B, Shavit Y, Hart J, Mordashov B, Cohn S, Notti I et al. (1996).
Effects of anesthesia based on large versus small doses of fentanyl on
explained by differing methodologies, species used, opioid
natural killer cell cytotoxicity in the perioperative period. Anesth
used and the dose and duration of administration. However,
Analg 82: 492–497.
in general, the literature reports that morphine and fentanyl
decrease NK cell cytotoxicity in vivo (although short-term Boehncke S, Hardt K, Schadendorf D, Henschler R, Boehncke WH,
in vivo exposure to fentanyl in healthy humans increases NK Duthey B (2011). Endogenous mu-opioid peptides modulate immune
response towards malignant melanoma. Exp Dermatol 20: 24–28.
cell numbers and cytotoxicity), buprenorphine does not
affect NK cell cytotoxicity and tramadol enhances it. This Boland J, Boland E, Brooks D (2013). Importance of the correct
change in NK cell cytotoxicity correlates with tumour growth diagnosis of opioid-induced respiratory depression in adult cancer
and metastasis in rodent models. Opioid dose and duration of patients and titration of naloxone. Clin Med 13: 149–151.
administration can influence outcome. Boland JW, Bennett MI (2016a). Opioids do not influence metastasis
Although clinical evidence is sparse, data suggest that in experimental animal cancer models. Pain 157: 1173.
perioperative opioid sparing may lead to better long-term
Boland JW, Foulds GA, Ahmedzai SH, Pockley AG (2014a). A
outcomes. However, high-quality perioperative and chronic
preliminary evaluation of the effects of opioids on innate and
cancer-related pain studies are needed. Given that current adaptive human in vitro immune function. BMJ Support Palliat Care
data from patients with cancer are inconclusive, definitive 4: 357–367.
recommendations about how adequate analgesia is best
achieved cannot be made and opioids for cancer-related pain Boland JW, McWilliams K, Ahmedzai SH, Pockley AG (2014b). Effects
of opioids on immunologic parameters that are relevant to anti-
will continue to be recommended.
tumour immune potential in patients with cancer: a systematic
literature review. Br J Cancer 111: 866–873.
Nomenclature of targets and ligands Boland JW, Pockley AG (2016b). Clinically relevant concentrations of
Key protein targets and ligands in this article are hyperlinked to opioids for in vitro studies. J Opioid Manag 12: 313–321.
corresponding entries in http://www.guidetopharmacology.
org, the common portal for data from the IUPHAR/BPS Guide Boland JW, Ziegler L, Boland EG, McDermid K, Bennett MI (2015). Is
regular systemic opioid analgesia associated with shorter survival in
to PHARMACOLOGY (Southan et al., 2016), and are perma-
adult patients with cancer? A systematic literature review. Pain 156:
nently archived in the Concise Guide to PHARMACOLOGY
2152–2163.
2015/16 (Alexander et al., 2015a,b).
Borman A, Ciepielewski Z, Wrona D, Stojek W, Glac W, Leszkowicz E
et al. (2009). Small doses of morphine can enhance NK cell
cytotoxicity in pigs. IntImmunopharmacol 9: 277–283.
Conflict of interest
Borner C, Kraus J, Bedini A, Schraven B, Hollt V (2008). T-cell
The authors declare no conflicts of interest. receptor/CD28-mediated activation of human T lymphocytes
induces expression of functional mu-opioid receptors. Mol
Pharmacol 74: 496–504.

Borner C, Lanciotti S, Koch T, Hollt V, Kraus J (2013). mu opioid

receptor agonist-selective regulation of interleukin-4 in T
References
lymphocytes. J Neuroimmunol 263: 35–42.
Afsharimani B, Baran J, Watanabe S, Lindner D, Cabot PJ, Parat MO Borner C, Stumm R, Hollt V, Kraus J (2007). Comparative analysis of
(2014). Morphine and breast tumor metastasis: the role of matrix- mu-opioid receptor expression in immune and neuronal cells. J
degrading enzymes. Clin Exp Metastasis 31: 149–158. Neuroimmunol 188: 56–63.
Afsharimani B, Cabot P, Parat MO (2011). Morphine and tumor Borner C, Warnick B, Smida M, Hartig R, Lindquist JA, Schraven B
growth and metastasis. Cancer Metastasis Rev 30: 225–238. et al. (2009). Mechanisms of opioid-mediated inhibition of human T
cell receptor signaling. J Immunol 183: 882–889.
Afsharimani B, Doornebal CW, Cabot PJ, Hollmann MW, Parat MO
(2015). Comparison and analysis of the animal models used to study Bortsov AV, Millikan RC, Belfer I, Boortz-Marx RL, Arora H, McLean
the effect of morphine on tumour growth and metastasis. Br J SA (2012). mu-Opioid receptor gene A118G polymorphism predicts
Pharmacol 172: 251–259. survival in patients with breast cancer. Anesthesiology 116: 896–902.

Alexander SPH, Davenport AP, Kelly E, Marrion N, Peters JA, Benson Buggy DJ, Borgeat A, Cata J, Doherty DG, Doornebal CW, Forget P
HE et al. (2015a). The Concise Guide to Pharmacology 2015/16: G et al. (2015). Consensus statement from the BJA Workshop on Cancer
protein-coupled receptors. Br J Pharmacol 172: 5744–5869. and Anaesthesia. Br J Anaesth 114: 2–3.

British Journal of Pharmacology (2018) 175 2726–2736 2733

 J W Boland and A G Pockley

Byrne K, Levins KJ, Buggy DJ (2016). Can anesthetic-analgesic Gong L, Qin Q, Zhou L, Ouyang W, Li Y, Wu Yet al. (2014). Effects of
technique during primary cancer surgery affect recurrence or fentanyl anesthesia and sufentanil anesthesia on regulatory T cells
metastasis? Can J Anaesth 63: 184–192. frequencies. Int J Clin Exp Pathol 7: 7708–7716.

Cakmakkaya OS, Kolodzie K, Apfel CC, Pace NL (2014). Anaesthetic Grace PM, Maier SF, Watkins LR (2015). Opioid-induced central
techniques for risk of malignant tumour recurrence. Cochrane immune signaling: implications for opioid analgesia. Headache 55:
Database Syst Rev (11): CD008877. 475–489.
Campana G, Sarti D, Spampinato S, Raffaeli W (2010). Long-term Gupta K, Kshirsagar S, Chang L, Schwartz R, Law PY, Yee D et al.
intrathecal morphine and bupivacaine upregulate MOR gene (2002). Morphine stimulates angiogenesis by activating
expression in lymphocytes. Int Immunopharmacol 10: 1149–1152. proangiogenic and survival-promoting signaling and promotes
breast tumor growth. Cancer Res 62: 4491–4498.
Cieslinska A, Sienkiewicz-Szlapka E, Kostyra E, Fiedorowicz E, Snarska
J, Wronski K et al. (2015). mu-Opioid receptor gene (OPRM1) Hernandez MC, Flores LR, Bayer BM (1993). Immunosuppression by
polymorphism in patients with breast cancer. Tumour Biol 36: morphine is mediated by central pathways. JPharmacolExpTher 267:
4655–4660. 1336–1341.
Connolly C, Buggy DJ (2016). Opioids and tumour metastasis: does Hooijmans CR, Geessink FJ, Ritskes-Hoitinga M, Scheffer GJ (2015). A
the choice of the anesthetic-analgesic technique influence outcome systematic review and meta-analysis of the ability of analgesic drugs
after cancer surgery? Curr Opin Anaesthesiol 29: 468–474. to reduce metastasis in experimental cancer models. Pain 156:
Cronin-Fenton DP, Heide-Jorgensen U, Ahern TP, Lash TL, 1835–1844.
Christiansen PM, Ejlertsen B et al. (2015). Opioids and breast cancer Hou M, Zhou NB, Li H, Wang BS, Wang XQ, Wang XWet al. (2016).
recurrence: a Danish population-based cohort study. Cancer 121: Morphine and ketamine inhibit immune function of gastric cancer
3507–3514. patients by increasing percentage of CD4(+)CD25(+)Foxp3(+)
Desmond F, McCormack J, Mulligan N, Stokes M, Buggy DJ (2015). regulatory T cells in vitro. J Surg Res 203: 306–312.
Effect of anaesthetic technique on immune cell infiltration in breast Hutchinson MR, Zhang Y, Shridhar M, Evans JH, Buchanan MM,
cancer: a follow-up pilot analysis of a prospective, randomised, Zhao TX et al. (2010). Evidence that opioids may have toll-like
investigator-masked study. Anticancer Res 35: 1311–1319. receptor 4 and MD-2 effects. Brain Behav Immun 24: 83–95.
Doornebal CW, Vrijland K, Hau CS, Coffelt SB, Ciampricotti M,
Irwin M, Hauger RL, Brown M, Britton KT (1988). CRF activates
Jonkers J et al. (2015). Morphine does not facilitate breast cancer
autonomic nervous system and reduces natural killer cytotoxicity.
progression in two preclinical mouse models for human invasive
Am J Physiol 255 (5 Pt 2): R744–R747.
lobular and HER2(+) breast cancer. Pain 156: 1424–1432.
Jacobs R, Karst M, Scheinichen D, Bevilacqua C, Schneider U, Heine J
European Medicines Evaluation Agency (EMEA). (2000). Note for
et al. (1999). Effects of fentanyl on cellular immune functions in man.
guidance on repeated dose toxicity. CPMP/SWP/1042/99. London.
Int J Immunopharmacol 21: 445–454.
Available at: http://www.ema.europa.eu/docs/en_GB/
document_library/Scientific_guideline/2009/09/WC500003102.pdf Jaura AI, Flood G, Gallagher HC, Buggy DJ (2014). Differential effects
(cited 23.12.2016). of serum from patients administered distinct anaesthetic techniques
on apoptosis in breast cancer cells in vitro: a pilot study. Br J Anaesth
Fecho K, Maslonek KA, Dykstra LA, Lysle DT (1996). Evidence for
113 (Suppl 1): i63–i67.
sympathetic and adrenal involvement in the immunomodulatory
effects of acute morphine treatment in rats. J Pharmacol Exp Ther Juneja R (2014). Opioids and cancer recurrence. Curr Opin Support
277: 633–645. Palliat Care 8: 91–101.
Franchi S, Moretti S, Castelli M, Lattuada D, Scavullo C, Panerai AE Khabbazi S, Goumon Y, Parat MO (2015). Morphine modulates
et al. (2012). Mu opioid receptor activation modulates toll like interleukin-4- or breast cancer cell-induced pro-metastatic activation
receptor 4 in murine macrophages. Brain Behav Immun 26: 480–488. of macrophages. Sci Rep 5: 11389.
Franchi S, Panerai AE, Sacerdote P (2007). Buprenorphine ameliorates Khabbazi S, Nassar ZD, Goumon Y, Parat MO (2016a). Morphine
the effect of surgery on hypothalamus-pituitary-adrenal axis, natural decreases the pro-angiogenic interaction between breast cancer cells
killer cell activity and metastatic colonization in rats in comparison and macrophages in vitro. Sci Rep 6: 31572.
with morphine or fentanyl treatment. Brain BehavImmun 21:
767–774. Khabbazi S, Xie N, Pu W, Goumon Y, Parat MO (2016b). The TLR4-
active morphine metabolite morphine-3-glucuronide does not elicit
Gach K, Wyrebska A, Fichna J, Janecka A (2011). The role of morphine macrophage classical activation in vitro. Front Pharmacol 7: 441.
in regulation of cancer cell growth. Naunyn Schmiedebergs Arch
Pharmacol 384: 221–230. Koodie L, Yuan H, Pumper JA, Yu H, Charboneau R, Ramkrishnan S
et al. (2014). Morphine inhibits migration of tumor-infiltrating
Gaspani L, Bianchi M, Limiroli E, Panerai AE, Sacerdote P (2002). The leukocytes and suppresses angiogenesis associated with tumor
analgesic drug tramadol prevents the effect of surgery on natural growth in mice. Am J Pathol 184: 1073–1084.
killer cell activity and metastatic colonization in rats. J
Neuroimmunol 129: 18–24. Kraus J (2009). Regulation of mu-opioid receptors by cytokines. Front
Biosci(ScholEd) 1: 164–170.
Glattard E, Welters ID, Lavaux T, Muller AH, Laux A, Zhang D et al.
(2010). Endogenous morphine levels are increased in sepsis: a partial Langsdorf EF, Mao X, Chang SL (2011). A role for reactive oxygen
implication of neutrophils. PLoS One 5: e8791. species in endotoxin-induced elevation of MOR expression in the
nervous and immune systems. J Neuroimmunol 236: 57–64.
Gomez-Flores R, Weber RJ (2000). Differential effects of
buprenorphine and morphine on immune and neuroendocrine Lennon FE, Mirzapoiazova T, Mambetsariev B, Poroyko VA, Salgia R,
functions following acute administration in the rat mesencephalon Moss J et al. (2014). The mu opioid receptor promotes opioid and
periaqueductal gray. Immunopharmacology 48: 145–156. growth factor-induced proliferation, migration and epithelial

2734 British Journal of Pharmacology (2018) 175 2726–2736

 Opioids and immune function in patients with cancer pain

mesenchymal transition (EMT) in human lung cancer. PLoS One 9: Ramaswamy S, Langford R (2017). Antinociceptive and
e91577. immunosuppressive effect of opioids in an acute postoperative
setting: an evidence-based review. BJA Education 17: 105–110.
Lennon FE, Mirzapoiazova T, Mambetsariev B, Salgia R, Moss J,
Singleton PA (2012). Overexpression of the mu-opioid receptor in Sacerdote P, Bianchi M, Gaspani L, Manfredi B, Maucione A, Terno G
human non-small cell lung cancer promotes Akt and mTOR et al. (2000). The effects of tramadol and morphine on immune
activation, tumor growth, and metastasis. Anesthesiology 116: responses and pain after surgery in cancer patients. Anesth Analg 90:
857–867. 1411–1414.

Liang X, Liu R, Chen C, Ji F, Li T (2016). Opioid system modulates the Sacerdote P, Bianchi M, Gaspani L, Panerai AE (1999). Effects of
immune function: a review. Translational perioperative and pain tramadol and its enantiomers on concanavalin-A induced-
medicine 1: 5–13. proliferation and NK activity of mouse splenocytes: involvement of
serotonin. Int J Immunopharmacol 21: 727–734.
Martucci C, Panerai AE, Sacerdote P (2004). Chronic fentanyl or
buprenorphine infusion in the mouse: similar analgesic profile but Saurer TB, Ijames SG, Lysle DT (2009). Evidence for the nucleus
different effects on immune responses. Pain 110: 385–392. accumbens as a neural substrate of heroin-induced immune
alterations. J Pharmacol ExpTher 329: 1040–1047.
Mathew B, Lennon FE, Siegler J, Mirzapoiazova T, Mambetsariev N,
Sammani S et al. (2011). The novel role of the mu opioid receptor in Saurer TB, Ijames SG, Lysle DT (2006). Neuropeptide Y Y1 receptors
lung cancer progression: a laboratory investigation. Anesth Analg mediate morphine-induced reductions of natural killer cell activity. J
112: 558–567. Neuroimmunol 177: 18–26.

Mellon RD, Bayer BM (1998). Role of central opioid receptor subtypes Shao YJ, Liu WS, Guan BQ, Hao JL, Ji K, Cheng XJ et al. (2017).
in morphine-induced alterations in peripheral lymphocyte activity. Contribution of opiate analgesics to the development of infections in
Brain Res 789: 56–67. advanced cancer patients. Clin J Pain 33: 295–299.

Mossner R, Lesch KP (1998). Role of serotonin in the immune Shavit Y, Ben-Eliyahu S, Zeidel A, Beilin B (2004). Effects of fentanyl
system and in neuroimmune interactions. Brain Behav Immun 12: on natural killer cell activity and on resistance to tumor metastasis in
249–271. rats. Dose and timing study. Neuroimmunomodulation 11: 255–260.

Myles PS, Peyton P, Silbert B, Hunt J, Rigg JR, Sessler DI (2011). Shen JC, Sun HL, Zhang MQ, Liu XY, Wang Z, Yang JJ (2014).
Perioperative epidural analgesia for major abdominal surgery for Flurbiprofen improves dysfunction of T-lymphocyte subsets and
cancer and recurrence-free survival: randomised trial. BMJ 342: natural killer cells in cancer patients receiving post-operative
d1491. morphine analgesia. Int J Clin Pharmacol Ther 52: 669–675.

Narahara H, Kadoi Y, Hinohara H, Kunimoto F, Saito S (2013). Shou J, Zhang Z, Lai Y, Chen Z, Huang J (2016). Worse outcome in
Comparative effects of flurbiprofen and fentanyl on natural killer cell breast cancer with higher tumor-infiltrating FOXP3+ Tregs: a
cytotoxicity, lymphocyte subsets and cytokine concentrations in systematic review and meta-analysis. BMC Cancer 16: 687.
post-surgical intensive care unit patients: prospective, randomized Singleton PA, Mirzapoiazova T, Hasina R, Salgia R, Moss J (2014).
study. J Anesth 27: 676–683. Increased mu-opioid receptor expression in metastatic lung cancer. Br
Nguyen J, Luk K, Vang D, Soto W, Vincent L, Robiner S et al. (2014). J Anaesth 113 (Suppl 1): i103–i108.
Morphine stimulates cancer progression and mast cell activation and Southan C, Sharman JL, Benson HE, Faccenda E, Pawson AJ,
impairs survival in transgenic mice with breast cancer. Br J Anaesth Alexander SP et al. (2016). The IUPHAR/BPS guide to
113 (Suppl 1): i4–13. PHARMACOLOGY in 2016: towards curated quantitative
interactions between 1300 protein targets and 6000 ligands. Nucl
Page GG (2003). The immune-suppressive effects of pain. Adv Exp
Acids Res 44: D1054–D1068.
Med Biol 521: 117–125.
Sun X, Yang C, Li K, Ding S (2015). The impact of anesthetic
Page GG (2005). Immunologic effects of opioids in the presence or
techniques on survival for patients with colorectal cancer:
absence of pain. JPain SymptomManage 29 (5 Suppl): S25–S31.
evidence based on six studies. Hepatogastroenterology 62:
Page GG, Blakely WP, Ben-Eliyahu S (2001). Evidence that 299–302.
postoperative pain is a mediator of the tumor-promoting effects of
Suzuki M, Sakurada T, Gotoh K, Watanabe S, Satoh N (2013).
surgery in rats. Pain 90: 191–199.
Correlation between the administration of morphine or oxycodone
Pei L, Tan G, Wang L, Guo W, Xiao B, Gao X et al. (2014). and the development of infections in patients with cancer pain. Am J
Comparison of combined general-epidural anesthesia with general Hosp Palliat Care 30: 712–716.
anesthesia effects on survival and cancer recurrence: a meta-
Tabellini G, Borsani E, Benassi M, Patrizi O, Ricotta D, Caimi L et al.
analysis of retrospective and prospective studies. PLoS One 9:
(2014). Effects of opioid therapy on human natural killer cells. Int
e114667.
Immunopharmacol 18: 169–174.
Pergolizzi J, Boger RH, Budd K, Dahan A, Erdine S, Hans G et al.
Takeuchi Y, Nishikawa H (2016). Roles of regulatory T cells in cancer
(2008). Opioids and the management of chronic severe pain in the
immunity. Int Immunol 28: 401–409.
elderly: consensus statement of an International Expert Panel with
focus on the six clinically most often used World Health Vogelaar FJ, Lips DJ, van Dorsten FRC, Lemmens VE, Bosscha K
Organization step III opioids (buprenorphine, fentanyl, (2016). Impact of anaesthetic technique on survival in colon cancer:
hydromorphone, methadone, morphine, oxycodone). Pain Pract 8: a review of the literature. Gastroenterology Report 4: 30–34.
287–313.
Wang S, Li Y, Liu XD, Zhao CX, Yang KQ (2013). Polymorphism of
Portenoy RK, Sibirceva U, Smout R, Horn S, Connor S, Blum RH et al. A118G in mu-opioid receptor gene is associated with risk of
(2006). Opioid use and survival at the end of life: a survey of a hospice esophageal squamous cell carcinoma in a Chinese population. Int J
population. J Pain Symptom Manage 32: 532–540. Clin Oncol 18: 666–669.

British Journal of Pharmacology (2018) 175 2726–2736 2735

 J W Boland and A G Pockley

Wang Y, Sun J, Zheng R, Shao Q, Gao W, Song B et al. (2016). Yeager MP, Colacchio TA, Yu CT, Hildebrandt L, Howell AL, Weiss J
Regulatory T cells are an important prognostic factor in breast cancer: et al. (1995). Morphine inhibits spontaneous and cytokine-enhanced
a systematic review and meta-analysis. Neoplasma 63: 789–798. natural killer cell cytotoxicity in volunteers. Anesthesiology 83:
500–508.
Weber RJ, Pert A (1989). The periaqueductal gray matter mediates
opiate-induced immunosuppression. Science 245: 188–190. Yeager MP, Procopio MA, DeLeo JA, Arruda JL, Hildebrandt L, Howell
Weng M, Chen W, Hou W, Li L, Ding M, Miao C (2016). The effect AL (2002). Intravenous fentanyl increases natural killer cell
cytotoxicity and circulating CD16(+) lymphocytes in humans.
of neuraxial anesthesia on cancer recurrence and survival after
cancer surgery: an updated meta-analysis. Oncotarget 7: Anesth Analg 94: 94–99.
15262–15273. Zhang EY, Xiong J, Parker BL, Chen AY, Fields PE, Ma X et al. (2011).
Wigmore T, Farquhar-Smith P (2016). Opioids and cancer: friend or Depletion and recovery of lymphoid subsets following morphine
foe? Curr Opin Support Palliat Care 10: 109–118. administration. Br J Pharmacol 164: 1829–1844.

Williams JP, Thompson JP, McDonald J, Barnes TA, Cote T, Zhang YF, Xu QX, Liao LD, Xu XE, Wu JY, Wu ZYet al. (2015).
Rowbotham DJ et al. (2007). Human peripheral blood mononuclear Association of mu-opioid receptor expression with lymph node
cells express nociceptin/orphanin FQ, but not mu, delta, or kappa metastasis in esophageal squamous cell carcinoma. Dis Esophagus 28:
opioid receptors. Anesth Analg 105: 998–1005 table. 196–203.

Xie N, Gomes FP, Deora V, Gregory K, Vithanage T, Nassar ZD et al. Zylla D, Gourley BL, Vang D, Jackson S, Boatman S, Lindgren B et al.
(2017). Activation of mu-opioid receptor and toll-like receptor 4 by (2013). Opioid requirement, opioid receptor expression, and clinical
plasma from morphine-treated mice. Brain Behav Immun 61: outcomes in patients with advanced prostate cancer. Cancer 119:
244–258. 4103–4110.

2736 British Journal of Pharmacology (2018) 175 2726–2736