Innate Immune Signalling Genetics of Pain, Cognitive Dysfunction and Sickness Symptoms in Cancer Pain Patients Treated with Transdermal Fentanyl

Daniel T. Barratt; Pål Klepstad; Ola Dale; Stein Kaasa; Andrew A. Somogyi

doi:10.1371/journal.pone.0137179

Abstract

Common adverse symptoms of cancer and chemotherapy are a major health burden; chief among these is pain, with opioids including transdermal fentanyl the mainstay of treatment. Innate immune activation has been implicated generally in pain, opioid analgesia, cognitive dysfunction, and sickness type symptoms reported by cancer patients. We aimed to determine if genetic polymorphisms in neuroimmune activation pathways alter the serum fentanyl concentration-response relationships for pain control, cognitive dysfunction, and other adverse symptoms, in cancer pain patients. Cancer pain patients (468) receiving transdermal fentanyl were genotyped for 31 single nucleotide polymorphisms in 19 genes: CASP1, BDNF, CRP, LY96, IL6, IL1B, TGFB1, TNF, IL10, IL2, TLR2, TLR4, MYD88, IL6R, OPRM1, ARRB2, COMT, STAT6 and ABCB1. Lasso and backward stepwise generalised linear regression were used to identify non-genetic and genetic predictors, respectively, of pain control (average Brief Pain Inventory < 4), cognitive dysfunction (Mini-Mental State Examination ≤ 23), sickness response and opioid adverse event complaint. Serum fentanyl concentrations did not predict between-patient variability in these outcomes, nor did genetic factors predict pain control, sickness response or opioid adverse event complaint. Carriers of the MYD88 rs6853 variant were half as likely to have cognitive dysfunction (11/111) than wild-type patients (69/325), with a relative risk of 0.45 (95% CI: 0.27 to 0.76) when accounting for major non-genetic predictors (age, Karnofsky functional score). This supports the involvement of innate immune signalling in cognitive dysfunction, and identifies MyD88 signalling pathways as a potential focus for predicting and reducing the burden of cognitive dysfunction in cancer pain patients.

Citation: Barratt DT, Klepstad P, Dale O, Kaasa S, Somogyi AA (2015) Innate Immune Signalling Genetics of Pain, Cognitive Dysfunction and Sickness Symptoms in Cancer Pain Patients Treated with Transdermal Fentanyl. PLoS ONE 10(9): e0137179. https://doi.org/10.1371/journal.pone.0137179

Editor: Chang-Qing Gao, Central South University, CHINA

Received: June 19, 2015; Accepted: August 14, 2015; Published: September 2, 2015

Copyright: © 2015 Barratt et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited

Data Availability: Due to identifying patient information, data are available upon request from the European Palliative Care Research Network which is responsible for the European Pharmacogenetic Opioid Study coordinated through St. Olav´s University Hospital, Trondheim University, Norway (contact: pal.klepstad@ntnu.no).

Funding: The Norwegian Research Council and the European Union’s 6th framework (Contract No. 037777; http://www.forskningsradet.no/en/Home_page/1177315753906; PK OD SK) and the National Health and Medical Research Council of Australia (Project Grant 1011521; https://www.nhmrc.gov.au/; AAS) financially supported the EPOS study.

Competing interests: The authors have declared that no competing interests exist.

Introduction

Common symptoms of cancer and its treatment are themselves a major health burden; chief among these is pain [1]. Opioids are the mainstay of cancer pain management, but many cancer patients experience inadequate pain relief and/or adverse opioid effects that reduce their quality of life [2,3].

Transdermal fentanyl is a strong opioid analgesic targeted to patients with stable opioid requirements [4]. It is commonly prescribed for the treatment of moderate to severe cancer pain [3], but has similar limitations to oral morphine with respect to inadequate pain control and adverse effects [5]. Clinical experience shows that individualisation is the key to successful opioid treatment. This is due to opioids’ narrow therapeutic index, as well as the large interindividual variability in their pharmacokinetics (PK) and pharmacodynamics (PD), and patients’ pain sensitivity, phenotype, perception and acceptance. Genetics can contribute to variability in these factors and thus variable opioid response [6].

The European Pharmacogenetic Opioid Study (EPOS) was a multinational collaborative effort to identify factors, particularly genetic, that determine opioid requirements for moderate-severe cancer pain [7]. In a sub-analysis of EPOS data, Klepstad et al. (2011) showed that transdermal fentanyl dose requirements for cancer pain are not associated with genetic variability in classical opioid and neuronal signalling genes, but with polymorphisms in STAT6; a cytokine- and growth factor-responsive transcription activator [7]. Those findings raise the question of whether genetic variability in non-classical opioid signalling pathways, in particular those involved in immune reactivity, might impact on fentanyl response in cancer pain patients.

There is growing evidence that opioids including fentanyl, as well as endogenous damage-associated molecular patterns (DAMPs) and peripheral inflammation associated with tissue damage from cancer or chemotherapy, can activate central innate immune cells (i.e. glia) via the toll-like receptor (TLR) family of pattern recognition receptors (e.g. TLR4) [8–11]. Activated glia release cytokines, chemokines, DAMPS and other inflammatory mediators that activate other glia, can facilitate peripheral immune cell infiltration into the CNS, and modify neuronal activity [12]. For example, immune mediators released by activated glia and infiltrating peripheral immune cells can enhance neuronal excitability, leading to central sensitisation and increased nociceptive signalling. This commonly manifests as hyperalgesia and/or allodynia that adds a pathological component to physiological pain associated with tissue (especially neuronal) damage, and is a major barrier to effective long-term opioid analgesia [9,10,12].

Other common symptoms experienced by cancer pain patients (e.g. nausea, fatigue, depression) are highly correlated and characteristic of cytokine-induced sickness responses, suggesting a shared neuroinflammatory mechanism [13]. Proinflammatory cytokine administration has also been shown to induce cognitive dysfunction in mice, and inflammatory cytokine immunotherapy is associated with significantly greater incidence of cognitive dysfunction in cancer patients when compared to primarily cytotoxic agents [14].

Therefore, we hypothesise that genetic variability in innate immune activation and inflammation pathways contributes to between-patient variability in pain control and adverse symptoms among cancer pain patients receiving transdermal fentanyl. The aim of this study was to investigate if common polymorphisms in genes involved in innate immune activation, inflammatory signalling, and consequent neuronal adaptation, alter the serum fentanyl concentration-response relationships for pain control, cognitive dysfunction, and adverse symptom complaint (sickness response and opioid side-effects), in EPOS cancer pain patients on transdermal fentanyl.

Materials and Methods

Subjects

EPOS is a multicentre (17 centres in 11 European countries) collaborative study of 2294 cancer patients with a malignant disease treated with an opioid for moderate to severe pain (step III at the WHO treatment ladder for cancer pain [15]). All EPOS patients were over 18 years of age, had a verified malignant disease, and were treated with regular opioids for their cancer pain for a minimum duration of 3 days. EPOS was conducted in accordance with the Declaration of Helsinki, the protocol was approved at each study centre’s local ethics committee (Regional Medical Research Ethics Committee, Central Norway Health Authority, Protocol reference number: 119–03, approved 27.09.03), and each participant provided written, informed consent [7].

Subjects selected for this study analysis were a subset of participants from EPOS maintained on transdermal fentanyl for pain control.

Sample selection

Of EPOS participants, 728 were treated with transdermal fentanyl and shortlisted for the current study. Of these, patients were excluded from analysis because they received non-transdermal fentanyl for break-through pain during the study period (n = 15), were missing blood samples (n = 6), were missing data on serum fentanyl concentrations (n = 9), were missing samples for genetic evaluation (n = 8) and/or had an unclear fentanyl dose (n = 14). The PK pharmacogenetics of the remaining 676 patients has previously been published by the authors [16]. For the present study, patients were additionally excluded if they were scheduled other opioids in addition to fentanyl (n = 110) and/or were non-Caucasian (n = 10). Patients who received breakthrough opioids other than fentanyl, or weak opioids, in the previous 24 hours were not excluded. Of the remaining 556 patients, 468 had data required for analysis of at least one outcome measure (394 for all outcome measures) and were included in the final study analyses.

Data collection from EPOS

Relevant single time-point patient data used for the present study are outlined in Table 1, in addition to average pain assessed using the Brief Pain Inventory (BPI) [17]; and cognitive function assessed using the Mini-Mental State Examination (MMSE) [18,19]. Genetic clustering of EPOS patients into 4 main European ancestry subgroups was previously determined [20]. A complete list of patient data taken as part of the original EPOS study is in [7].

Download:

Table 1. Patient characteristics and investigated non-genetic variables (for n = 468 included in analyses).

https://doi.org/10.1371/journal.pone.0137179.t001

Genotyping

DNA was extracted from EDTA–treated whole blood [7]. Patients were genotyped for the following 20 SNPs in 14 genes using a custom Sequenom MassARRAY (iPLEX GOLD) multiplex panel designed and implemented at the Australian Genome Research Facility (Brisbane, Australia) [25]: Innate immune activation–TLR4 (rs4986790, rs4986791); TLR2 (rs3804100); MD2 (LY96) (rs11466004); MYD88 (rs6853). Inflammatory mediators–IL1B (rs1143627, rs1143634, rs16944); CASP1 (ICE) (rs554344, rs580253); IL6 (rs10499563); IL6R (rs8192284); IL10 (rs1800871, rs1800896); IL2 (rs2069762); CRP (rs2794521); TGFB1 (rs11466314, rs1800469); TNFA (rs1800629). Neuronal adaptation—BDNF (rs6265).

Reproducibility of the multiplex panel was confirmed by re-genotyping 100 randomly selected patients (99% for rs1143634 and 100% for all other SNPs).

COMT (rs4680), OPRM1 (rs1799971), ARRB2 (rs3786047, rs1045280, rs2271167, rs2036657), STAT6 (rs3024971, rs167769) and ABCB1 (rs1045642, rs2235013, rs1128503, rs4437575, rs2235033, rs1202170, rs7802773) SNPs had been genotyped previously [7]. For ABCB1, only rs1045642, rs2235013 and rs1128503 were included in the final analysis based on existing evidence of phenotype associations and near complete linkage disequilibrium (LD) (r² ≥ 0.9) with the other ABCB1 SNPs.

Details of SNP locations and nucleotide/amino acid changes are provided in S1 Table.

Data analysis

Data were analysed using the R statistical program [26] unless indicated otherwise. Chi-squared analysis was used to test for genotype deviations from Hardy-Weinberg Equilibrium.

Patient descriptive data were summarised as median ± standard deviation (SD) and range (minimum to maximum), or as counts, as appropriate.

Measures of fentanyl response.

Pathological, physiological and genetic variables were investigated for their association with four measures of fentanyl response: “pain control”; “cognitive dysfunction”; “sickness response”; and “opioid adverse event complaint”.

Patients with average pain of 3 or less measured on an 11-point NRS in the BPI were categorized as having “pain control”; higher scores were categorized as unacceptable pain [27]. Patients with a total MMSE of 23 or less were categorised as “cognitive dysfunction” [18,27]. Patients who reported two or more of the following were categorised as “sickness response”: nausea ≥ 50 (EORTC QLQ-C30 nausea and vomiting scale); tiredness ≥ 3 (EORTC QLQ-C30 item “Were you tired?”); and depression ≥ 3 (EORTC QLQ-C30 item “Did you feel depressed?”).

Based on previous EPOS studies, patients were categorised as “opioid adverse event complaint” if they reported nausea ≥ 50 (EORTC QLQ-C30 nausea and vomiting scale); constipation ≥ 50 (EORTC QLQ-C30 constipation scale); tiredness ≥ 3 (EORTC QLQ-C30 item “Were you tired?”); and/or had a total MMSE of 23 or less (“cognitive dysfunction”) [19,27,28].

The co-incidence of specific adverse events (nausea, tiredness, constipation, depression and cognitive dysfunction) was investigated using Fishers Exact Test.

Distributions of continuous variables were assessed using histograms and quantile-quantile (Q-Q) plots. If continuous variables were not normally distributed, optimal transformations to normalise the distributions were identified using the boxcox function in the MASS package [29]. Transformed data were then used in all subsequent analyses.

Identification of response predictors.

The absence of significant associations between responses and ancestral subgroup was first confirmed by chi-squared analysis (P > 0.05) before proceeding with further analyses.

Details of the subsequent statistical analysis pipeline are provided in S1 Protocol. Briefly, major non-genetic variables (listed in Table 1) to be controlled for in subsequent genotype analyses were identified by Lasso regression, including first order interactions with serum fentanyl concentration. A step-down regression model selection procedure based on cross-validation error was used to identify genetic factors associated with different responses, fixing major non-genetic predictors as the base model. Epistasis was also investigated by generalised multifactor dimensionality reduction (GMDR) analysis, incorporating major non-genetic predictors into the response score.

Given the number of genetic factors investigated, the likelihood of associations occurring by chance within the data for each outcome measure were investigated by comparing the performance of the final regression and GMDR models against control models using randomised permutations of paired response and non-genetic variable data.

To evaluate the data analysis approach of the current study against previously published EPOS findings [7], the analysis approach was also applied to identify predictors of fentanyl delivery rate (previously associated with STAT6 rs167769).

Results

Genetic variability

Allele and genotype frequencies for each SNP are given in S1 Table. No genotype frequencies significantly deviated from Hardy Weinberg Equilibrium (P > 0.1). Details of SNP linkage disequilibrium and haplotypes are given in S1 Text.

There was no significant association between ancestral subgroups and any response (point-wise P > 0.05).

Pain control

Of 430 patients with BPI scores, 210 were classified as having pain control. No major non-genetic predictors of pain control were identified by Lasso regression, including serum fentanyl concentrations (median ± SD in pain relieved = 5.4 ± 7.7 μM versus not pain relieved = 5.8 ± 13.6 μM). CRP rs2794521 variant genotype was associated with reduced pain control (optimal k = 2.4, CVE = 0.250 < base model = 0.251), however cross-validation performance of this model was no better than randomised controls (median (25–75^th percentile) CVE = 0.249 (0.247–0.251)). Reflecting this, the predictive performance of the model was very poor (area under the ROC curve = 0.53), and basic chi-squared analysis not significant (χ² = 3.6, point-wise P = 0.17).

No epistatic models for pain control performed better than randomised dataset controls.

STAT6 rs167769 was the sole predictor of fentanyl delivery rate, but fentanyl delivery rate was not a predictor of between-patient variability in pain control (median ± SD delivery rate in pain controlled = 50 ± 47 μg/hr versus unacceptable pain = 50 ± 54 μg/hr).

Cognitive dysfunction

Of 438 patients with MMS data, 81 had cognitive dysfunction. Analysis of the co-incidence of specific adverse events showed cognitive dysfunction was unrelated to other adverse events, but there were significant positive associations between nausea, tiredness, depression and constipation (Table 2).

Download:

Table 2. Co-incidence of adverse events reported by cancer pain patients receiving transdermal fentanyl.

https://doi.org/10.1371/journal.pone.0137179.t002

Serum fentanyl concentrations were not associated with cognitive dysfunction (median ± SD in cognitive dysfunction = 7.1 ± 7.9 μM versus not cognitive dysfunction = 5.5 ± 11.3 μM). Older age and lower Karnofsky functional status were associated with increased cognitive dysfunction, with a modest predictive value (area under the ROC curve = 0.71). MYD88 rs6853 heterozygous and variant genotypes (combined) were associated with reduced cognitive dysfunction (Table 3 and Fig 1) (optimal k = 7.4). Addition of MYD88 rs6853 genotype (variant carrier versus non-carrier) slightly increased the predictive ability over the non-genetic model (area under the ROC curve = 0.73), and cross-validation performance of this model (CVE = 0.136) was better than randomised controls [median (25–75th percentile) CVE = 0.138 (0.138–0.138)]. The incidence of cognitive dysfunction in MYD88 rs6853 variant carriers was less than half that of wild-type patients [11/111 (10%) versus 69/325 (21%), respectively), with a relative risk of 0.45 when accounting for age and Karnofsky functional score (Table 3 and Fig 1).

Download:

Fig 1. Predictors of cognitive dysfunction in cancer patients receiving transdermal fentanyl.

Solid lines and filled circles are predicted frequency, and dotted lines and error bars are 95% confidence intervals, holding other variables to typical values. Hollow circles are unadjusted (raw) frequencies for each Karnofsky score, within 10-year bins from 30 years of age, or for each MYD88 rs6853 genotype group. Vertical bars above the x-axes of Karnofsky score and Age represent the distributions of patients’ Karnofsky scores and ages, respectively.

https://doi.org/10.1371/journal.pone.0137179.g001

Download:

Table 3. Variables associated with cognitive dysfunction in cancer pain patients receiving transdermal fentanyl.

https://doi.org/10.1371/journal.pone.0137179.t003

No epistatic models for cognitive dysfunction performed better than randomised dataset controls.

Sickness response

Of 418 patients with EORTC data, 148 were classified as “sickness response”. Serum fentanyl concentrations were not associated with sickness response (median ± SD in sickness response = 5.4 ± 8.0 μM versus not sickness response = 5.6 ± 12.6 μM). The use of breakthrough opioids was associated with increased, and male sex and an Italian treatment centre were associated with decreased, sickness response with modest predictive value (area under the ROC curve = 0.69).

The BDNF rs6265 variant was associated with increased sickness response (S2 Table) (optimal k = 3.9). However, adding this genetic factor only slightly improved predictive ability over the non-genetic model (area under the ROC curve = 0.70), and cross-validation performance of this model (CVE = 0.208) was not better than randomised controls [median (25–75th percentile) CVE = 0.210 (0.207–0.210)].

The incidence of sickness response was not significantly different between MYD88 rs6853 homozygous wild-type and variant carrier patients (67% versus 69%, respectively. OR (95% CI) = 1.1 (0.69 to 1.8)).

No epistatic models for sickness response performed better than randomised dataset controls.

Opioid adverse event complaint

Of 430 patients with adverse event data, 329 were classified as opioid adverse event complainers. Serum fentanyl concentrations were not associated with opioid adverse event complaint (median ± SD in complainers = 5.8 ± 9.2 μM versus non-complainers = 5.1 ± 16.1 μM). Depression and the use of breakthrough opioids were associated with increased complaint with modest predictive value (area under the ROC curve = 0.69). CASP1 rs554344 homozygous variant genotype and ARRB2 variant diplotype were associated with increased, and TGFB1 rs1800469 homozygous wildtype genotype associated with reduced, complaint (S3 Table) (optimal k = 2.1). Addition of these genetic factors increased the predictive ability over the non-genetic model (area under the ROC curve = 0.73), however cross-validation performance of this model (CVE = 0.157) was no better than randomised controls [median (25–75th percentile) CVE = 0.157 (0.156–0.159)].

No epistatic models for opioid adverse event complaint performed better than randomised dataset controls.

Discussion

This study set out to investigate if polymorphisms in genes implicated in neuroimmune activation alter the serum fentanyl concentration-response relationship in cancer pain patients. However, serum fentanyl concentrations did not predict between-patient variability in pain intensity or adverse events, reflecting similar previous findings with oral morphine [27]. This further emphasises the major hurdles to effective opioid use for cancer pain: large underlying heterogeneity in cancer pain phenotypes, a milieu of cancer/chemotherapy adverse effects, and unpredictable between-patient variability in opioid pharmacodynamics. Given the absence of clear serum fentanyl concentration-response relationships, the direct relationships between variables (genetic and non-genetic) and pain and adverse symptoms were investigated.

No major predictors of pain intensity were detected in this study, despite investigating variables encompassing both underlying pain and fentanyl pharmacodynamics. The data analysis approach of the current study was able to identify STAT6 rs167769 as a predictor of fentanyl dose as in a previous EPOS study [7], but dose itself (like serum concentrations) was not a predictor of between-patient variability in pain intensity.

Alternatively, a major genetic predictor of cognitive dysfunction in cancer pain patients receiving transdermal fentanyl was identified. Patients carrying the MYD88 rs6853 variant allele had half the risk of developing cognitive dysfunction: an effect seen with and without accounting for non-genetic variables.

MyD88 is an adapter protein for multiple TLRs (e.g. TLR4 and TLR2) and the interleukin 1 receptor (IL1R) [30,31]. Therefore, it plays a central role in both the initial innate immune activation by opioids and/or DAMPs, and the potentiation of immune activation via IL1B signalling. TLR4/MyD88 signalling is important for the development of post-operative- and post-traumatic brain injury-induced cognitive dysfunction in rats and mice [32–34], and glial modulators are currently being trialled for the potential to attenuate post-operative cognitive dysfunction [35]. Thus it is likely that MyD88 signalling is also key to cognitive dysfunction in cancer pain patients, but to our knowledge this has not been specifically investigated to date. Whilst the direct functional effects of the rs6853 SNP, located within the 3’ untranslated region of MYD88, are not currently known, it has been associated with decreased vaccine response and increased susceptibility to infection [36,37], as well as electroencephalogram spindle amplitude during anaesthesia [25]. Therefore, we hypothesise that cancer pain patients carrying the rs6853 SNP are at lower risk of developing cognitive dysfunction due to reduced MyD88 activity driving the neuroimmune signalling associated with cognitive impairment. Polymorphisms of more specific signalling systems (e.g. TLR4 and IL1B) were not significantly associated with cognitive dysfunction, indicating that the mechanistic importance of MyD88 across multiple cell types in parallel and sequential signalling systems may make it of specific importance to impairment of cognitive processing.

Our analysis also identified age and Karnofsky score as major predictors of cognitive dysfunction within the transdermal fentanyl cohort, reflecting previous EPOS findings when patients on any opioid were analysed in combination [28]. However, unlike Kurita and colleagues’ combined analysis [28], there was no major effect of daily opioid dose, lung cancer, time since diagnosis or breakthrough pain, on cognitive dysfunction within the subset of transdermal fentanyl patients included in the present study.

Breakthrough opioid use was more important than serum fentanyl concentrations for predicting adverse symptoms other than cognitive dysfunction. Transdermal fentanyl is typically associated with reduced incidence of nausea, sedation and constipation compared to oral opioids [38]. Therefore, the serum fentanyl concentration-adverse effect relationship may be less pronounced for transdermal fentanyl than for other opioids, with these adverse symptoms influenced more by acute effects of breakthrough opioid use, cancer and/or chemotherapy.

Innate immune activation can also be associated with a general sickness phenotype consisting of symptoms of nausea, tiredness and/or depression [13,14]. However cognitive dysfunction and sickness phenotype were unrelated in this study, with different predictors. No genetic predictors were identified for general sickness phenotype or opioid adverse event complaint, which share common symptoms of nausea and tiredness. Similarly, no genetic variants predicted nausea, tiredness, constipation or depression when analysing these symptoms separately (results not presented). This suggests innate immune activation may play a lesser role in symptoms of nausea, sedation, tiredness and depression compared to cognitive dysfunction, reflecting the differential clustering of these symptoms in cancer patients and thus likely mechanisms [14].

Patients’ pain and adverse effect phenotypes are the summation of the effects of cancer, chemotherapy and opioid treatment. Similarly, cancer, chemotherapy and opioids can all activate the neuroimmune activation pathway being investigated. The major limitation of this study is the cross-sectional design which could not distinguish between underlying (baseline) phenotype and response; that is, between the cancer, chemotherapy, and opioid effects. As discussed above, large variability in baseline (opioid-free) phenotype makes it difficult to establish a serum fentanyl concentration-response relationship across cancer pain patients: repeated measures designs would better demonstrate the concentration-response relationship and aid in separating underlying phenotype from fentanyl response, but are practically and ethically complex in the cancer pain population.

The MMSE was employed for assessing cognitive dysfunction due to its extensive use in research and clinical practice, its gold standard status for measuring cognitive dysfunction in patients with cancer, and a cut-off (a score of 23 or less that has high specificity for defining cognitive dysfunction) that has been applied and validated in cancer patient populations [19,39–43]. However, it is acknowledged that the MMSE has limited sensitivity to detect mild cognitive impairment [44–46], and future prospective studies would be improved by including multiple (validated) assessment tools that help better define mild cognitive impairment and assess specific cognitive domains.

Future directions

For cognitive dysfunction, the association with age and Karnofsky status, but not serum fentanyl concentrations or breakthrough opioid use, suggest underlying physiology and the impact of the cancer or chemotherapy may be the major contributors to this phenotype. It is probable that the protective effect of the MYD88 rs6853 variant is due to reduced innate immune activation in response to DAMPS associated with cancer and/or chemotherapy tissue damage, rather than to fentanyl. Therefore, the same association should be seen for other opioids in EPOS; this is currently under investigation in addition to MMSE sub-categories that may point to specific brain regions and processes especially impacted by MYD88 genetic variability.

The frequency of the rs6853 wild-type genotype (75%) relative to cognitive dysfunction (18%) prohibits a highly sensitive and specific prediction of risk for cognitive dysfunction based on this single SNP. Therefore additional protective or high risk MYD88 SNPs may remain to be identified. The characterisation of specific MyD88 pathways driving cancer- and chemotherapy-induced cognitive dysfunction will be important, and may point to additional genetic polymorphisms or biomarkers of these pathways useful for identifying patients at high risk of cognitive dysfunction. Further this may prompt the investigation of interventions targeting the MYD88 pathway to ameliorate cognitive dysfunction in cancer patients, similar to those being trialled for post-operative cognitive dysfunction [35].

Whilst candidate SNPs were identified for each response examined in this study, our randomised control analyses demonstrated a high probability that these were chance associations, with the exception of rs6853 for cognitive dysfunction. These candidate SNPs have been reported in supplementary material as they may be true positive associations and subsequently be confirmed in replicate studies.

Conclusions

Serum fentanyl concentrations were not a major determinant of between-patient variability in effective cancer pain control with transdermal fentanyl, further highlighting the importance and challenge of matching opioid and dose to individual patients from populations of highly heterogeneous pain phenotypes. For adverse symptoms other than cognitive dysfunction, breakthrough opioid use was more important than serum fentanyl concentrations, and they were not predicted by genetic polymorphisms relating to neuroimmune activation pathways. However, the MYD88 rs6853 variant was associated with significantly reduced risk of cognitive dysfunction, identifying MyD88 signalling pathways as a potential focus for predicting and reducing the burden of cognitive dysfunction in cancer patients.

Supporting Information

S1 Protocol. Statistical analysis pipeline.

https://doi.org/10.1371/journal.pone.0137179.s001

(DOC)

S1 Table. SNP allele and genotype frequencies in EPOS cancer pain patients receiving transdermal fentanyl.

https://doi.org/10.1371/journal.pone.0137179.s002

(DOCX)

S2 Table. Variables associated with sickness response (nausea, tiredness and/or depression complaint) in cancer pain patients receiving transdermal fentanyl.

https://doi.org/10.1371/journal.pone.0137179.s003

(DOCX)

S3 Table. Variables associated with opioid adverse event complaint (as defined by [27]) in cancer pain patients receiving transdermal fentanyl.

https://doi.org/10.1371/journal.pone.0137179.s004

(DOCX)

S1 Text. Supplementary results relating to SNP linkage disequilibrium and haplotypes.

https://doi.org/10.1371/journal.pone.0137179.s005

(DOC)

Author Contributions

Conceived and designed the experiments: DTB PK OD SK AAS. Performed the experiments: DTB. Analyzed the data: DTB. Contributed reagents/materials/analysis tools: PK OD SK AAS. Wrote the paper: DTB PK OD SK AAS.

References

1. Cleeland CS. Cancer-related symptoms. Semin Radiat Oncol. 2000; 10:175–90. pmid:11034629
- View Article
- PubMed/NCBI
- Google Scholar
2. van den Beuken-van Everdingen MH, de Rijke JM, Kessels AG, Schouten HC, van Kleef M, Patijn J. Prevalence of pain in patients with cancer: a systematic review of the past 40 years. Ann Oncol. 2007; 18:1437–49. pmid:17355955
- View Article
- PubMed/NCBI
- Google Scholar
3. Klepstad P, Kaasa S, Cherny N, Hanks G, de Conno F. Pain and pain treatments in European palliative care units. A cross sectional survey from the European Association for Palliative Care Research Network. Palliat Med. 2005; 19:477–84. pmid:16218160
- View Article
- PubMed/NCBI
- Google Scholar
4. Ripamonti CI. Pain management. Ann Oncol. 2012; 23 Suppl 10:x294–301. pmid:22987980
- View Article
- PubMed/NCBI
- Google Scholar
5. Tassinari D, Drudi F, Rosati M, Maltoni M. Transdermal opioids as front line treatment of moderate to severe cancer pain: a systemic review. Palliat Med. 2011; 25:478–87. pmid:21708854
- View Article
- PubMed/NCBI
- Google Scholar
6. Somogyi AA, Barratt DT, Coller JK. Pharmacogenetics of opioids. Clin Pharmacol Ther. 2007; 81:429–44. pmid:17339873
- View Article
- PubMed/NCBI
- Google Scholar
7. Klepstad P, Fladvad T, Skorpen F, Bjordal K, Caraceni A, Dale O, et al. Influence from genetic variability on opioid use for cancer pain: a European genetic association study of 2294 cancer pain patients. Pain. 2011; 152:1139–45. pmid:21398039
- View Article
- PubMed/NCBI
- Google Scholar
8. Stevens CW, Aravind S, Das S, Davis RL. Pharmacological characterization of LPS and opioid interactions at the toll-like receptor 4. Br J Pharmacol. 2013; 168:1421–9. pmid:23083095
- View Article
- PubMed/NCBI
- Google Scholar
9. Hutchinson MR, Shavit Y, Grace PM, Rice KC, Maier SF, Watkins LR. Exploring the neuroimmunopharmacology of opioids: an integrative review of mechanisms of central immune signaling and their implications for opioid analgesia. Pharmacol Rev. 2011; 63:772–810. pmid:21752874
- View Article
- PubMed/NCBI
- Google Scholar
10. Grace PM, Hutchinson MR, Maier SF, Watkins LR. Pathological pain and the neuroimmune interface. Nat Rev Immunol. 2014; 14:217–31. pmid:24577438
- View Article
- PubMed/NCBI
- Google Scholar
11. Hutchinson MR, Zhang Y, Shridhar M, Evans JH, Buchanan MM, Zhao TX, et al. Evidence that opioids may have toll-like receptor 4 and MD-2 effects. Brain Behav Immun. 2010; 24:83–95. pmid:19679181
- View Article
- PubMed/NCBI
- Google Scholar
12. Ji RR, Xu ZZ, Gao YJ. Emerging targets in neuroinflammation-driven chronic pain. Nat Rev Drug Discov. 2014; 13:533–48. pmid:24948120
- View Article
- PubMed/NCBI
- Google Scholar
13. Dantzer R, O'Connor JC, Freund GG, Johnson RW, Kelley KW. From inflammation to sickness and depression: when the immune system subjugates the brain. Nat Rev Neurosci. 2008; 9:46–56. pmid:18073775
- View Article
- PubMed/NCBI
- Google Scholar
14. Lee BN, Dantzer R, Langley KE, Bennett GJ, Dougherty PM, Dunn AJ, et al. A cytokine-based neuroimmunologic mechanism of cancer-related symptoms. Neuroimmunomodulation. 2004; 11:279–92. pmid:15316238
- View Article
- PubMed/NCBI
- Google Scholar
15. WHO. WHO's pain ladder for adults. WHO; 2013 [cited 2013 August]. Available: http://www.who.int/cancer/palliative/painladder/en/.
16. Barratt DT, Bandak B, Klepstad P, Dale O, Kaasa S, Christrup LL, et al. Genetic, pathological and physiological determinants of transdermal fentanyl pharmacokinetics in 620 cancer patients of the EPOS study. Pharmacogenet Genomics. 2014; 24:185–94. pmid:24469018
- View Article
- PubMed/NCBI
- Google Scholar
17. Cleeland CS, Ryan KM. Pain assessment: global use of the Brief Pain Inventory. Ann Acad Med Singapore. 1994; 23:129–38. pmid:8080219
- View Article
- PubMed/NCBI
- Google Scholar
18. Folstein MF, Folstein SE, McHugh PR. "Mini-mental state". A practical method for grading the cognitive state of patients for the clinician. J Psychiatr Res. 1975; 12:189–98. pmid:1202204
- View Article
- PubMed/NCBI
- Google Scholar
19. Tombaugh TN, McIntyre NJ. The mini-mental state examination: a comprehensive review. J Am Geriatr Soc. 1992; 40:922–35. pmid:1512391
- View Article
- PubMed/NCBI
- Google Scholar
20. Galvan A, Fladvad T, Skorpen F, Gao X, Klepstad P, Kaasa S, et al. Genetic clustering of European cancer patients indicates that opioid-mediated pain relief is independent of ancestry. Pharmacogenomics J. 2012; 12:412–6. pmid:21747411
- View Article
- PubMed/NCBI
- Google Scholar
21. Cockcroft DW, Gault MH. Prediction of creatinine clearance from serum creatinine. Nephron. 1976; 16:31–41. pmid:1244564
- View Article
- PubMed/NCBI
- Google Scholar
22. Moksnes K, Fredheim OM, Klepstad P, Kaasa S, Angelsen A, Nilsen T, et al. Early pharmacokinetics of nasal fentanyl: is there a significant arterio-venous difference? Eur J Clin Pharmacol. 2008; 64:497–502. pmid:18180911
- View Article
- PubMed/NCBI
- Google Scholar
23. Karnofsky D, Abelmann W, Craver L, Burchenal J. The use of nitrogen mustards in the palliative treatment of carcinoma. Cancer. 1948; 1:634–56.
- View Article
- Google Scholar
24. Aaronson NK, Ahmedzai S, Bergman B, Bullinger M, Cull A, Duez NJ, et al. The European Organization for Research and Treatment of Cancer QLQ-C30: a quality-of-life instrument for use in international clinical trials in oncology. J Natl Cancer Inst. 1993; 85:365–76. pmid:8433390
- View Article
- PubMed/NCBI
- Google Scholar
25. Mulholland CV, Somogyi AA, Barratt DT, Coller JK, Hutchinson MR, Jacobson GM, et al. Association of innate immune single-nucleotide polymorphisms with the electroencephalogram during desflurane general anaesthesia. J Mol Neurosci. 2014; 52:497–506. pmid:24352713
- View Article
- PubMed/NCBI
- Google Scholar
26. R Core Team. R: A Language and Environment for Statistical Computing. Vienna, Austria: R Foundation for Statistical Computing; 2013.
27. Klepstad P, Borchgrevink PC, Dale O, Zahlsen K, Aamo T, Fayers P, et al. Routine drug monitoring of serum concentrations of morphine, morphine-3-glucuronide and morphine-6-glucuronide do not predict clinical observations in cancer patients. Palliat Med. 2003; 17:679–87. pmid:14694919
- View Article
- PubMed/NCBI
- Google Scholar
28. Kurita GP, Sjogren P, Ekholm O, Kaasa S, Loge JH, Poviloniene I, et al. Prevalence and predictors of cognitive dysfunction in opioid-treated patients with cancer: a multinational study. J Clin Oncol. 2011; 29:1297–303. pmid:21357785
- View Article
- PubMed/NCBI
- Google Scholar
29. Venables W, Ripley B. Modern Applied Statistics with S. 4th ed. New York: Springer; 2002.
30. Deguine J, Barton GM. MyD88: a central player in innate immune signaling. F1000Prime Rep. 2014; 6:97. pmid:25580251
- View Article
- PubMed/NCBI
- Google Scholar
31. Janssens S, Beyaert R. A universal role for MyD88 in TLR/IL-1R-mediated signaling. Trends Biochem Sci. 2002; 27:474–82. pmid:12217523
- View Article
- PubMed/NCBI
- Google Scholar
32. Lu SM, Yu CJ, Liu YH, Dong HQ, Zhang X, Zhang SS, et al. S100A8 contributes to postoperative cognitive dysfunction in mice undergoing tibial fracture surgery by activating the TLR4/MyD88 pathway. Brain Behav Immun. 2015; 44:221–34. pmid:25449673
- View Article
- PubMed/NCBI
- Google Scholar
33. Wang Y, He H, Li D, Zhu W, Duan K, Le Y, et al. The role of the TLR4 signaling pathway in cognitive deficits following surgery in aged rats. Mol Med Rep. 2013; 7:1137–42. pmid:23426570
- View Article
- PubMed/NCBI
- Google Scholar
34. Mao SS, Hua R, Zhao XP, Qin X, Sun ZQ, Zhang Y, et al. Exogenous administration of PACAP alleviates traumatic brain injury in rats through a mechanism involving the TLR4/MyD88/NF-kappaB pathway. J Neurotrauma. 2012; 29:1941–59. pmid:22583372
- View Article
- PubMed/NCBI
- Google Scholar
35. Gunma University Hospital, Department of Anesthesiology. Preventive effect of minocycline on postoperative cognitive dysfunction in Japanese elderly patients: A randomized, double-blind, placebo-controlled trial. Unique trial Number: UMIN000014620. Database: UMIN-CTR Clinical Trial Database, University hospital Medical Information Network (UMIN) Center (Japan) [internet]. Available: https://upload.umin.ac.jp/cgi-open-bin/ctr/ctr.cgi?function=brows&action=brows&recptno=R000016796&type=summary&language=E.
36. Capparelli R, De Chiara F, Di Matteo A, Medaglia C, Iannelli D. The MyD88 rs6853 and TIRAP rs8177374 polymorphic sites are associated with resistance to human pulmonary tuberculosis. Genes Immun. 2013; 14:504–11. pmid:24067789
- View Article
- PubMed/NCBI
- Google Scholar
37. Dhiman N, Ovsyannikova IG, Vierkant RA, Ryan JE, Pankratz VS, Jacobson RM, et al. Associations between SNPs in toll-like receptors and related intracellular signaling molecules and immune responses to measles vaccine: preliminary results. Vaccine. 2008; 26:1731–6. pmid:18325643
- View Article
- PubMed/NCBI
- Google Scholar
38. Skaer TL. Dosing considerations with transdermal formulations of fentanyl and buprenorphine for the treatment of cancer pain. J Pain Res. 2014; 7:495–503. pmid:25170278
- View Article
- PubMed/NCBI
- Google Scholar
39. Fayers PM, Hjermstad MJ, Ranhoff AH, Kaasa S, Skogstad L, Klepstad P, et al. Which mini-mental state exam items can be used to screen for delirium and cognitive impairment? J Pain Symptom Manage. 2005; 30:41–50. pmid:16043006
- View Article
- PubMed/NCBI
- Google Scholar
40. Mystakidou K, Tsilika E, Parpa E, Galanos A, Vlahos L. Brief cognitive assessment of cancer patients: evaluation of the Mini-Mental State Examination (MMSE) psychometric properties. Psychooncology. 2007; 16:352–7. pmid:16991106
- View Article
- PubMed/NCBI
- Google Scholar
41. Iracleous P, Nie JX, Tracy CS, Moineddin R, Ismail Z, Shulman KI, et al. Primary care physicians' attitudes towards cognitive screening: findings from a national postal survey. Int J Geriatr Psychiatry. 2010; 25:23–9. pmid:19513985
- View Article
- PubMed/NCBI
- Google Scholar
42. Milne A, Culverwell A, Guss R, Tuppen J, Whelton R. Screening for dementia in primary care: a review of the use, efficacy and quality of measures. Int Psychogeriatr. 2008; 20:911–26. pmid:18533066
- View Article
- PubMed/NCBI
- Google Scholar
43. Shulman KI, Herrmann N, Brodaty H, Chiu H, Lawlor B, Ritchie K, et al. IPA survey of brief cognitive screening instruments. Int Psychogeriatr. 2006; 18:281–94. pmid:16466586
- View Article
- PubMed/NCBI
- Google Scholar
44. Ismail Z, Rajji TK, Shulman KI. Brief cognitive screening instruments: an update. Int J Geriatr Psychiatry. 2010; 25:111–20. pmid:19582756
- View Article
- PubMed/NCBI
- Google Scholar
45. Lonie JA, Tierney KM, Ebmeier KP. Screening for mild cognitive impairment: a systematic review. Int J Geriatr Psychiatry. 2009; 24:902–15. pmid:19226524
- View Article
- PubMed/NCBI
- Google Scholar
46. Mitchell AJ. A meta-analysis of the accuracy of the mini-mental state examination in the detection of dementia and mild cognitive impairment. J Psychiatr Res. 2009; 43:411–31. pmid:18579155
- View Article
- PubMed/NCBI
- Google Scholar

[ref1] 1. Cleeland CS. Cancer-related symptoms. Semin Radiat Oncol. 2000; 10:175–90. pmid:11034629
View Article
PubMed/NCBI
Google Scholar

[2] View Article

[3] PubMed/NCBI

[4] Google Scholar

[ref2] 2. van den Beuken-van Everdingen MH, de Rijke JM, Kessels AG, Schouten HC, van Kleef M, Patijn J. Prevalence of pain in patients with cancer: a systematic review of the past 40 years. Ann Oncol. 2007; 18:1437–49. pmid:17355955
View Article
PubMed/NCBI
Google Scholar

[6] View Article

[7] PubMed/NCBI

[8] Google Scholar

[ref3] 3. Klepstad P, Kaasa S, Cherny N, Hanks G, de Conno F. Pain and pain treatments in European palliative care units. A cross sectional survey from the European Association for Palliative Care Research Network. Palliat Med. 2005; 19:477–84. pmid:16218160
View Article
PubMed/NCBI
Google Scholar

[10] View Article

[11] PubMed/NCBI

[12] Google Scholar

[ref4] 4. Ripamonti CI. Pain management. Ann Oncol. 2012; 23 Suppl 10:x294–301. pmid:22987980
View Article
PubMed/NCBI
Google Scholar

[14] View Article

[15] PubMed/NCBI

[16] Google Scholar

[ref5] 5. Tassinari D, Drudi F, Rosati M, Maltoni M. Transdermal opioids as front line treatment of moderate to severe cancer pain: a systemic review. Palliat Med. 2011; 25:478–87. pmid:21708854
View Article
PubMed/NCBI
Google Scholar

[18] View Article

[19] PubMed/NCBI

[20] Google Scholar

[ref6] 6. Somogyi AA, Barratt DT, Coller JK. Pharmacogenetics of opioids. Clin Pharmacol Ther. 2007; 81:429–44. pmid:17339873
View Article
PubMed/NCBI
Google Scholar

[22] View Article

[23] PubMed/NCBI

[24] Google Scholar

[ref7] 7. Klepstad P, Fladvad T, Skorpen F, Bjordal K, Caraceni A, Dale O, et al. Influence from genetic variability on opioid use for cancer pain: a European genetic association study of 2294 cancer pain patients. Pain. 2011; 152:1139–45. pmid:21398039
View Article
PubMed/NCBI
Google Scholar

[26] View Article

[27] PubMed/NCBI

[28] Google Scholar

[ref8] 8. Stevens CW, Aravind S, Das S, Davis RL. Pharmacological characterization of LPS and opioid interactions at the toll-like receptor 4. Br J Pharmacol. 2013; 168:1421–9. pmid:23083095
View Article
PubMed/NCBI
Google Scholar

[30] View Article

[31] PubMed/NCBI

[32] Google Scholar

[ref9] 9. Hutchinson MR, Shavit Y, Grace PM, Rice KC, Maier SF, Watkins LR. Exploring the neuroimmunopharmacology of opioids: an integrative review of mechanisms of central immune signaling and their implications for opioid analgesia. Pharmacol Rev. 2011; 63:772–810. pmid:21752874
View Article
PubMed/NCBI
Google Scholar

[34] View Article

[35] PubMed/NCBI

[36] Google Scholar

[ref10] 10. Grace PM, Hutchinson MR, Maier SF, Watkins LR. Pathological pain and the neuroimmune interface. Nat Rev Immunol. 2014; 14:217–31. pmid:24577438
View Article
PubMed/NCBI
Google Scholar

[38] View Article

[39] PubMed/NCBI

[40] Google Scholar

[ref11] 11. Hutchinson MR, Zhang Y, Shridhar M, Evans JH, Buchanan MM, Zhao TX, et al. Evidence that opioids may have toll-like receptor 4 and MD-2 effects. Brain Behav Immun. 2010; 24:83–95. pmid:19679181
View Article
PubMed/NCBI
Google Scholar

[42] View Article

[43] PubMed/NCBI

[44] Google Scholar

[ref12] 12. Ji RR, Xu ZZ, Gao YJ. Emerging targets in neuroinflammation-driven chronic pain. Nat Rev Drug Discov. 2014; 13:533–48. pmid:24948120
View Article
PubMed/NCBI
Google Scholar

[46] View Article

[47] PubMed/NCBI

[48] Google Scholar

[ref13] 13. Dantzer R, O'Connor JC, Freund GG, Johnson RW, Kelley KW. From inflammation to sickness and depression: when the immune system subjugates the brain. Nat Rev Neurosci. 2008; 9:46–56. pmid:18073775
View Article
PubMed/NCBI
Google Scholar

[50] View Article

[51] PubMed/NCBI

[52] Google Scholar

[ref14] 14. Lee BN, Dantzer R, Langley KE, Bennett GJ, Dougherty PM, Dunn AJ, et al. A cytokine-based neuroimmunologic mechanism of cancer-related symptoms. Neuroimmunomodulation. 2004; 11:279–92. pmid:15316238
View Article
PubMed/NCBI
Google Scholar

[54] View Article

[55] PubMed/NCBI

[56] Google Scholar

[ref15] 15. WHO. WHO's pain ladder for adults. WHO; 2013 [cited 2013 August]. Available: http://www.who.int/cancer/palliative/painladder/en/.

[ref16] 16. Barratt DT, Bandak B, Klepstad P, Dale O, Kaasa S, Christrup LL, et al. Genetic, pathological and physiological determinants of transdermal fentanyl pharmacokinetics in 620 cancer patients of the EPOS study. Pharmacogenet Genomics. 2014; 24:185–94. pmid:24469018
View Article
PubMed/NCBI
Google Scholar

[59] View Article

[60] PubMed/NCBI

[61] Google Scholar

[ref17] 17. Cleeland CS, Ryan KM. Pain assessment: global use of the Brief Pain Inventory. Ann Acad Med Singapore. 1994; 23:129–38. pmid:8080219
View Article
PubMed/NCBI
Google Scholar

[63] View Article

[64] PubMed/NCBI

[65] Google Scholar

[ref18] 18. Folstein MF, Folstein SE, McHugh PR. "Mini-mental state". A practical method for grading the cognitive state of patients for the clinician. J Psychiatr Res. 1975; 12:189–98. pmid:1202204
View Article
PubMed/NCBI
Google Scholar

[67] View Article

[68] PubMed/NCBI

[69] Google Scholar

[ref19] 19. Tombaugh TN, McIntyre NJ. The mini-mental state examination: a comprehensive review. J Am Geriatr Soc. 1992; 40:922–35. pmid:1512391
View Article
PubMed/NCBI
Google Scholar

[71] View Article

[72] PubMed/NCBI

[73] Google Scholar

[ref20] 20. Galvan A, Fladvad T, Skorpen F, Gao X, Klepstad P, Kaasa S, et al. Genetic clustering of European cancer patients indicates that opioid-mediated pain relief is independent of ancestry. Pharmacogenomics J. 2012; 12:412–6. pmid:21747411
View Article
PubMed/NCBI
Google Scholar

[75] View Article

[76] PubMed/NCBI

[77] Google Scholar

[ref21] 21. Cockcroft DW, Gault MH. Prediction of creatinine clearance from serum creatinine. Nephron. 1976; 16:31–41. pmid:1244564
View Article
PubMed/NCBI
Google Scholar

[79] View Article

[80] PubMed/NCBI

[81] Google Scholar

[ref22] 22. Moksnes K, Fredheim OM, Klepstad P, Kaasa S, Angelsen A, Nilsen T, et al. Early pharmacokinetics of nasal fentanyl: is there a significant arterio-venous difference? Eur J Clin Pharmacol. 2008; 64:497–502. pmid:18180911
View Article
PubMed/NCBI
Google Scholar

[83] View Article

[84] PubMed/NCBI

[85] Google Scholar

[ref23] 23. Karnofsky D, Abelmann W, Craver L, Burchenal J. The use of nitrogen mustards in the palliative treatment of carcinoma. Cancer. 1948; 1:634–56.
View Article
Google Scholar

[87] View Article

[88] Google Scholar

[ref24] 24. Aaronson NK, Ahmedzai S, Bergman B, Bullinger M, Cull A, Duez NJ, et al. The European Organization for Research and Treatment of Cancer QLQ-C30: a quality-of-life instrument for use in international clinical trials in oncology. J Natl Cancer Inst. 1993; 85:365–76. pmid:8433390
View Article
PubMed/NCBI
Google Scholar

[90] View Article

[91] PubMed/NCBI

[92] Google Scholar

[ref25] 25. Mulholland CV, Somogyi AA, Barratt DT, Coller JK, Hutchinson MR, Jacobson GM, et al. Association of innate immune single-nucleotide polymorphisms with the electroencephalogram during desflurane general anaesthesia. J Mol Neurosci. 2014; 52:497–506. pmid:24352713
View Article
PubMed/NCBI
Google Scholar

[94] View Article

[95] PubMed/NCBI

[96] Google Scholar

[ref26] 26. R Core Team. R: A Language and Environment for Statistical Computing. Vienna, Austria: R Foundation for Statistical Computing; 2013.

[ref27] 27. Klepstad P, Borchgrevink PC, Dale O, Zahlsen K, Aamo T, Fayers P, et al. Routine drug monitoring of serum concentrations of morphine, morphine-3-glucuronide and morphine-6-glucuronide do not predict clinical observations in cancer patients. Palliat Med. 2003; 17:679–87. pmid:14694919
View Article
PubMed/NCBI
Google Scholar

[99] View Article

[100] PubMed/NCBI

[101] Google Scholar

[ref28] 28. Kurita GP, Sjogren P, Ekholm O, Kaasa S, Loge JH, Poviloniene I, et al. Prevalence and predictors of cognitive dysfunction in opioid-treated patients with cancer: a multinational study. J Clin Oncol. 2011; 29:1297–303. pmid:21357785
View Article
PubMed/NCBI
Google Scholar

[103] View Article

[104] PubMed/NCBI

[105] Google Scholar

[ref29] 29. Venables W, Ripley B. Modern Applied Statistics with S. 4th ed. New York: Springer; 2002.

[ref30] 30. Deguine J, Barton GM. MyD88: a central player in innate immune signaling. F1000Prime Rep. 2014; 6:97. pmid:25580251
View Article
PubMed/NCBI
Google Scholar

[108] View Article

[109] PubMed/NCBI

[110] Google Scholar

[ref31] 31. Janssens S, Beyaert R. A universal role for MyD88 in TLR/IL-1R-mediated signaling. Trends Biochem Sci. 2002; 27:474–82. pmid:12217523
View Article
PubMed/NCBI
Google Scholar

[112] View Article

[113] PubMed/NCBI

[114] Google Scholar

[ref32] 32. Lu SM, Yu CJ, Liu YH, Dong HQ, Zhang X, Zhang SS, et al. S100A8 contributes to postoperative cognitive dysfunction in mice undergoing tibial fracture surgery by activating the TLR4/MyD88 pathway. Brain Behav Immun. 2015; 44:221–34. pmid:25449673
View Article
PubMed/NCBI
Google Scholar

[116] View Article

[117] PubMed/NCBI

[118] Google Scholar

[ref33] 33. Wang Y, He H, Li D, Zhu W, Duan K, Le Y, et al. The role of the TLR4 signaling pathway in cognitive deficits following surgery in aged rats. Mol Med Rep. 2013; 7:1137–42. pmid:23426570
View Article
PubMed/NCBI
Google Scholar

[120] View Article

[121] PubMed/NCBI

[122] Google Scholar

[ref34] 34. Mao SS, Hua R, Zhao XP, Qin X, Sun ZQ, Zhang Y, et al. Exogenous administration of PACAP alleviates traumatic brain injury in rats through a mechanism involving the TLR4/MyD88/NF-kappaB pathway. J Neurotrauma. 2012; 29:1941–59. pmid:22583372
View Article
PubMed/NCBI
Google Scholar

[124] View Article

[125] PubMed/NCBI

[126] Google Scholar

[ref35] 35. Gunma University Hospital, Department of Anesthesiology. Preventive effect of minocycline on postoperative cognitive dysfunction in Japanese elderly patients: A randomized, double-blind, placebo-controlled trial. Unique trial Number: UMIN000014620. Database: UMIN-CTR Clinical Trial Database, University hospital Medical Information Network (UMIN) Center (Japan) [internet]. Available: https://upload.umin.ac.jp/cgi-open-bin/ctr/ctr.cgi?function=brows&action=brows&recptno=R000016796&type=summary&language=E.

[ref36] 36. Capparelli R, De Chiara F, Di Matteo A, Medaglia C, Iannelli D. The MyD88 rs6853 and TIRAP rs8177374 polymorphic sites are associated with resistance to human pulmonary tuberculosis. Genes Immun. 2013; 14:504–11. pmid:24067789
View Article
PubMed/NCBI
Google Scholar

[129] View Article

[130] PubMed/NCBI

[131] Google Scholar

[ref37] 37. Dhiman N, Ovsyannikova IG, Vierkant RA, Ryan JE, Pankratz VS, Jacobson RM, et al. Associations between SNPs in toll-like receptors and related intracellular signaling molecules and immune responses to measles vaccine: preliminary results. Vaccine. 2008; 26:1731–6. pmid:18325643
View Article
PubMed/NCBI
Google Scholar

[133] View Article

[134] PubMed/NCBI

[135] Google Scholar

[ref38] 38. Skaer TL. Dosing considerations with transdermal formulations of fentanyl and buprenorphine for the treatment of cancer pain. J Pain Res. 2014; 7:495–503. pmid:25170278
View Article
PubMed/NCBI
Google Scholar

[137] View Article

[138] PubMed/NCBI

[139] Google Scholar

[ref39] 39. Fayers PM, Hjermstad MJ, Ranhoff AH, Kaasa S, Skogstad L, Klepstad P, et al. Which mini-mental state exam items can be used to screen for delirium and cognitive impairment? J Pain Symptom Manage. 2005; 30:41–50. pmid:16043006
View Article
PubMed/NCBI
Google Scholar

[141] View Article

[142] PubMed/NCBI

[143] Google Scholar

[ref40] 40. Mystakidou K, Tsilika E, Parpa E, Galanos A, Vlahos L. Brief cognitive assessment of cancer patients: evaluation of the Mini-Mental State Examination (MMSE) psychometric properties. Psychooncology. 2007; 16:352–7. pmid:16991106
View Article
PubMed/NCBI
Google Scholar

[145] View Article

[146] PubMed/NCBI

[147] Google Scholar

[ref41] 41. Iracleous P, Nie JX, Tracy CS, Moineddin R, Ismail Z, Shulman KI, et al. Primary care physicians' attitudes towards cognitive screening: findings from a national postal survey. Int J Geriatr Psychiatry. 2010; 25:23–9. pmid:19513985
View Article
PubMed/NCBI
Google Scholar

[149] View Article

[150] PubMed/NCBI

[151] Google Scholar

[ref42] 42. Milne A, Culverwell A, Guss R, Tuppen J, Whelton R. Screening for dementia in primary care: a review of the use, efficacy and quality of measures. Int Psychogeriatr. 2008; 20:911–26. pmid:18533066
View Article
PubMed/NCBI
Google Scholar

[153] View Article

[154] PubMed/NCBI

[155] Google Scholar

[ref43] 43. Shulman KI, Herrmann N, Brodaty H, Chiu H, Lawlor B, Ritchie K, et al. IPA survey of brief cognitive screening instruments. Int Psychogeriatr. 2006; 18:281–94. pmid:16466586
View Article
PubMed/NCBI
Google Scholar

[157] View Article

[158] PubMed/NCBI

[159] Google Scholar

[ref44] 44. Ismail Z, Rajji TK, Shulman KI. Brief cognitive screening instruments: an update. Int J Geriatr Psychiatry. 2010; 25:111–20. pmid:19582756
View Article
PubMed/NCBI
Google Scholar

[161] View Article

[162] PubMed/NCBI

[163] Google Scholar

[ref45] 45. Lonie JA, Tierney KM, Ebmeier KP. Screening for mild cognitive impairment: a systematic review. Int J Geriatr Psychiatry. 2009; 24:902–15. pmid:19226524
View Article
PubMed/NCBI
Google Scholar

[165] View Article

[166] PubMed/NCBI

[167] Google Scholar

[ref46] 46. Mitchell AJ. A meta-analysis of the accuracy of the mini-mental state examination in the detection of dementia and mild cognitive impairment. J Psychiatr Res. 2009; 43:411–31. pmid:18579155
View Article
PubMed/NCBI
Google Scholar

[169] View Article

[170] PubMed/NCBI

[171] Google Scholar

Figures

Abstract

Introduction

Materials and Methods

Subjects

Sample selection

Data collection from EPOS

Genotyping

Data analysis

Measures of fentanyl response.

Identification of response predictors.

Results

Genetic variability

Pain control

Cognitive dysfunction

Sickness response

Opioid adverse event complaint

Discussion

Future directions

Conclusions

Supporting Information

S1 Protocol. Statistical analysis pipeline.

S1 Table. SNP allele and genotype frequencies in EPOS cancer pain patients receiving transdermal fentanyl.

S2 Table. Variables associated with sickness response (nausea, tiredness and/or depression complaint) in cancer pain patients receiving transdermal fentanyl.

S3 Table. Variables associated with opioid adverse event complaint (as defined by [27]) in cancer pain patients receiving transdermal fentanyl.

S1 Text. Supplementary results relating to SNP linkage disequilibrium and haplotypes.

Author Contributions

References