Multiple Loci Modulate Opioid Therapy Response for Cancer Pain

Author Manuscript Published OnlineFirst on May 27, 2011; DOI: 10.1158/1078-0432.CCR-10-3028 Author manuscripts have been peer reviewed and accepted for publication but have not yet been edited.

1 Multiple loci modulate opioid therapy response for

2 cancer pain

4 Antonella Galvan 1, Frank Skorpen 2, Pål Klepstad 2,3, Anne Kari Knudsen 2, Torill 5 Fladvad 2, Felicia S. Falvella 1, Alessandra Pigni 1, Cinzia Brunelli 1, Augusto 6 Caraceni 1,2, Stein Kaasa 2,4, Tommaso A. Dragani 1 7 8 9 10 1 Fondazione IRCCS Istituto Nazionale Tumori, Milan, Italy; 2 European Palliative 11 Care Research Center, Medical Faculty, Norwegian University of Science and 12 Technology, Trondheim, Norway; 3 Department of Anaesthesiology and Emergency 13 Medicine and 4 Department of Oncology, St. Olavs University Hospital, Trondheim, 14 Norway 15 16 17 Conflict of interest statement 18 None declared. 19 20 Correspondence to: 21 Tommaso A. Dragani, Department of Predictive and Preventive Medicine, 22 Fondazione IRCCS Istituto Nazionale Tumori, Via G.A. Amadeo 42, 20133 23 Milan, Italy. Tel.: +39-0223902642, Fax: +39-0223902764. E-Mail: 24 [email protected] 25 Running Title: Genetic control of cancer pain response 26 Keywords: opioids, pharmacogenomics, pain, polygenic diseases, single-nucleotide 27 polymorphisms. 28

Downloaded from clincancerres.aacrjournals.org on September 24, 2021. © 2011 American Association for Cancer Research. Author Manuscript Published OnlineFirst on May 27, 2011; DOI: 10.1158/1078-0432.CCR-10-3028 Author manuscripts have been peer reviewed and accepted for publication but have not yet been edited.

1 Abstract

2 Purpose: Patients treated with opioid drugs for cancer pain experience different 3 relief responses, raising the possibility that genetic factors play a role in opioid 4 therapy outcome. In this study, we tested the hypothesis that genetic variations may 5 control individual response to opioid drugs in cancer patients. 6 Experimental design: We tested one million SNPs in European cancer patients 7 selected in a first series for extremely poor (pain relief ≤40%; n=145) or good (pain 8 relief ≥90%; n=293) responses to opioid therapy using a DNA pooling approach. 9 Candidate SNPs identified by SNP-array were genotyped in individual samples 10 constituting DNA pools as well as in a second series of 570 patients. 11 Results: Association analysis in 1008 cancer patients identified 8 SNPs significantly 12 associated with pain relief at a statistical threshold of P<1.0 x 10-3, with rs12948783, 13 upstream of the RHBDF2 gene, showing the best statistical association (P=8.1 x 10- 14 9). Functional annotation analysis of SNP-tagged genes suggested the involvement 15 of genes acting on neurological system processes. 16 Conclusion: Our results indicate that the identified SNP panel can modulate the 17 response of cancer patients to opioid therapy and may provide a new tool for 18 personalized therapy of cancer pain. 19

1 Translational relevance

2 Pain is a common complication of almost all types of advanced cancer, and 3 treatment with opioids produces variable responses of pain relief among individuals. 4 Indeed, a variable percentage of cancer patients experiences reduced or poor 5 benefit from opioid therapy, with impairment of their quality of life. Our association 6 study in 1008 European cancer patients identified 8 SNPs statistically associated 7 with pain relief, which together might modulate the response of cancer patients to 8 opioid therapy and might serve as targetable tools to design new drugs. Our 9 functional annotation analysis of SNP-tagged genes revealed for the first time the 10 involvement of genes acting on neurological system processes rather than on opioid 11 metabolism or pharmacodynamics. Thus, the identified SNP panel may provide a 12 new tool for personalized therapy of cancer pain. 13

1 Introduction

2 Pain is the most dreaded symptom in patients with malignant disease. Indeed, 3 each year several million men and women of all ages suffer from some form of 4 invasive cancer associated with pain requiring chronic opioid treatment. At present, 5 opioids are virtually the only analgesics able to control moderate and severe cancer 6 pain. 7 While opioids are widely used to treat cancer pain, clinical studies indicate 8 great variability among individuals in the efficacy of these drugs and their side effects 9 (1, 2). Many studies have investigated the potential genetic basis of individual 10 variability in opioid drug response, but almost all have focused on a few candidate 11 genes involved in opioid pharmacokinetics or pharmacodynamics, and results are not 12 consistent among these often small studies (3, 4). For example, the catecholamines 13 dopamine, epinephrine and norepinephrine and their hydroxylated metabolites are 14 substrates for the enzyme catechol-O-methyltransferase encoded by the COMT 15 gene, in which genetic variations have been associated with variations in pain 16 perception and in doses of opioids required for pain treatment; however, 17 discrepancies in the results from different association studies on COMT 18 polymorphisms and pain perception have been reported, possibly due to the small 19 size of the studies (reviewed in (5)). 20 The human mu 1 opioid receptor (OPRM1) gene, which is the main 21 pharmacologic target of opioids, contains more than 100 polymorphisms, and these 22 polymorphisms might affect opioid binding affinity to the receptor and, consequently, 23 opioid-induced analgesia, tolerance, and dependence. However, a recent meta- 24 analysis casts doubt on the clinical relevance of the OPRM1 118A>G polymorphism 25 suggested by both preclinical and clinical studies as a candidate in personalized 26 cancer pain therapy (6). 27 A complex genetics underlying inter-individual differences in opioid response 28 is expected in light of the biochemical and biological complexity of the drug response 29 and of pain control. Indeed, opioid effects are governed by drug-related absorption, 30 distribution, metabolism, intrinsic efficacy of the receptors involved, and by signal 31 pathways downstream of the receptors. Since each of these steps in pain control is 32 modulated by several genes, each of which can exist in allelic forms in the general

1 population, a genome-wide comprehensive analysis would be required to identify all 2 genetic factors involved. 3 Herein, we conducted the first genome-wide association study (GWAS) in 4 cancer patients treated with opioids using a 1-million single-nucleotide polymorphism 5 (SNP)-array to identify new genetic variations modulating individual pain relief.

6 Patients and methods

7 Study design 8 The European Pharmacogenetic Opioid Study (EPOS) represents a 9 multicenter effort to examine symptoms, pharmacology, and pharmacogenomics 10 related to the use of opioids. Recruitment criteria of patients included a regularly 11 scheduled opioid treatment stable for at least 3 days. Opioids administered to >5% of 12 patients included morphine, oxycodone, and fentanyl. The study includes 2294 13 cancer patients treated with opioids for moderate or severe pain and recruited from 14 17 centers of 11 European countries. The participation of the Research Steering 15 Committee from the European Association for Palliative Care (EAPC) facilitated 16 international cooperation in the study. 17 The study collected demographic and personal data, as well as malignant 18 diagnosis and opioid treatment from cancer patients. Opioid doses were converted to 19 equivalent total daily oral morphine doses (7). Pain was assessed by completing the 20 Brief Pain Inventory (BPI) (8). Functional status was measured by the Karnofsky 21 performance status score. Further details about the EPOS series, including opioid 22 doses and recruitments, have been reported (7). The study collected whole blood for 23 genetic analysis and has established a biobank.

24 Pain relief phenotype and DNA pool construction 25 The pain relief phenotype under study is semi-quantitative and determined 26 based on the BPI, a robust and psychometric validated method to assess the 27 subjective severity of pain. Pain relief is measured using an 11-point numerical rating 28 scale, from 0%, representing “no pain relief,” to 100% or “complete pain relief", i.e., 29 0%, 10%, 20%, etc. (9). 30 For DNA pool preparation, we first defined the cancer patients as “good” or

1 “poor” responders to opioid therapy, based on their pain relief phenotype score 2 ≥90% or ≤40%, respectively (Fig. 1). Selection was stratified by gender and country 3 of origin, with a 1:2 matching of “poor” and “good” responders, to obtain a balanced 4 representation of patients in the two groups. DNA pools of “good” (n=293, mean pain 5 relief, 94.1%) and “poor” (n=145, mean pain relief, 27.0%) responders were created 6 using equal amounts of fluorimetrically measured DNA from each sample.

7 SNP genotyping 8 SNP-array analysis of each of the two DNA pools was carried out using the 9 HumanOmni1-Quad BeadChip (Illumina, San Diego, CA), which allows analysis of 10 1,140,419 genetic markers, including more than 20,000 markers in more than 300 11 important gene regions related to drug absorption, distribution, metabolism, and 12 excretion; several of the SNPs map to opioid receptor or metabolism genes 1. We 13 performed 12 replicas to verify genotype reproducibility and to estimate technical 14 variability. Data were obtained as intensity signals, which were used to determine the 15 allele frequencies of each genetic marker and to reconstruct the number of 16 chromosomes carrying each of the two possible alleles. Individual samples were 17 genotyped using MassARRAY (Sequenom, Inc., San Diego, CA), as described (10).

18 Gene analysis 19 To explore the role of the associated variants in the biological context, we 20 mapped the best statistically associated SNPs in the DNA pools (n=486, P<1.0 x 10- 21 3) using Ensembl database to identify genes possibly involved in pain relief 22 response, and analyzed the gene list for enrichment in ontology annotation terms 23 using the DAVID Functional Annotation Tool and Clustering (11) 2.

24 Statistical analyses 25 Chromosome counts of the “poor” and “good” pain relief groups, from which 26 DNA pools were prepared, were reconstructed based on allelic frequencies 27 assessed in SNP-array hybridization, and the resulting 2x2 contingency tables were 28 tested by chi-square analysis. Association analyses between SNPs and the pain 29 relief phenotype (categorical variable “poor” or “good” responders, or quantitative 30 variable) were carried out using PLINK software (12), which included analysis of the

1 http://www.illumina.com/products/humanomni1_quad_beadchip_kits.ilmn

1 Hardy-Weinberg equilibrium (HWE), linkage disequilibrium between SNPs, and 2 population-based associations between phenotypes and genotype/allelotype 3 (Supplementary Fig. 1).

4 Results

5 The study population for which pain relief values were available included 1907 6 patients from 4 ethnic groups: (i) Caucasian, including Hispanic (n=1859), (ii) African 7 (n=13), (iii) Oriental (n=4), and (iv) Other (n=29), and 11 European countries of 8 origin. Patients included in the DNA pools constituted the first series (n=438), 9 whereas the second series of 570 EPOS patients was selected for pain relief 10 phenotype score ≥90% or ≤50% to extend the results of the GWAS (Fig. 1). Table 1 11 lists the phenotypic characteristics of patients analyzed in the present study. The 12 pain relief mean value was slightly higher in the second series (76.5 ± 1.1 versus 13 71.9 ± 1.6; P=0.014, ANOVA), whereas no statistically significant differences were 14 observed in the mean total oral morphine equivalent opioid dose (Table 1). 15 Preliminary analysis for factors that might confound the assessment of the 16 pain relief phenotype revealed a weak effect of gender (P=0.019) and a strong effect 17 of country of origin (P<2.2 x 10-16, ANOVA; Supplementary Table 1). No statistically 18 significant association between the pain relief phenotype and age of patients, 19 ethnicity, or tumor type was found. Thus, we normalized the pain relief phenotype by 20 the country mean value and adjusted analyses for gender as covariate.

21 Genes involved in neurological system processes may modulate pain relief 22 In our search for genetic variations associated with individual pain relief, we 23 analyzed the EPOS subjects showing the extreme phenotype values of pain relief in 24 order to reduce the bias in the subjective pain relief assessment of the borderline 25 categories and to allow testing genetic effects on pain therapy response on the best 26 candidate subjects (Fig. 1). 27 GWAS using a DNA pooling strategy and 1-million SNP-array detected 28 173,851 informative SNPs, with overall minor allele frequency >0.1 and coefficient of 29 variation of allelic frequency (standard deviation/mean) >0.08. Statistical 30 associations of these SNPs are visualized by Manhattan plot (Fig. 2). Analysis of

2 http://david.abcc.ncifcrf.gov/home.jsp

1 chromosome counts obtained from allele frequencies in DNA pools of “good” and 2 “poor” responders identified 486 SNPs at statistical threshold of P<1.0 x 10-3 (chi- 3 square test; Fig. 2; Supplementary Fig. 1). 4 To explore the role of the associated variants in the biological context, we 5 identified the genes tagged by these 486 SNPs. A list of 226 known genes, either 6 near or containing a SNP, was obtained and analyzed using the Functional 7 Annotation Tool 3, which identified as the best statistically significant gene ontology 8 annotation terms those linked to the nervous system, such as the "neurological 9 system process", "transmission of nerve impulse" and "synaptic transmission". 10 Notably, the best term was the "neurological system process" which included 31 11 genes (P=3.5 x 10-5, Table 2). Interestingly, 10 of these 31 genes were members of 12 the olfactory receptor family. Use of Functional Annotation Clustering gave similar 13 results, with the most highly significant cluster including the annotation terms 14 "transmission of nerve impulse", "synaptic transmission" and "cell-cell signaling" (not 15 shown).

16 A genetic profile controls pain relief 17 Among the 486 candidate SNPs identified by statistically significant 18 differences (P<1.0 x 10-3, chi-square test) between “poor” and “good” responders in 19 chromosome counts, we selected 72 SNPs associated with the pain relief phenotype 20 (at P<1.0 x 10-5 threshold value) for genotyping in individual samples constituting the 21 DNA pools, and in the second series of 570 EPOS patients by MassARRAY (Table 22 1, Fig. 1). Six SNPs failed PCR or MassEXTEND primer design, and another seven 23 failed genotyping. Of the 59 genotyped SNPs, two showed significant deviation from 24 HWE (P<1.0 x 10-3), reducing the number of markers to 57 SNPs. 25 Association analysis in the first series indicated that 54 of the 57 SNPs were 26 significantly associated with the poor and good responder categories at P<0.05 27 (logistic analysis), confirming the robustness of our DNA pooling approach, which 28 produces reliable results and is time- and cost-effective, in agreement with other 29 reports (13, 14). SNP rs12948783, which maps to chromosome 17 upstream of the 30 RHBDF2 gene, showed the strongest association (P=1.1 x 10-7). 31 Analysis of the total sample size of 1008 cancer patients for the quantitative 32 pain relief phenotype, allowing an increased statistical power of the study, pointed to

1 8 SNPs associated at statistical threshold of P<1.0 x 10-3, with the strongest allelic 2 association again for SNP rs12948783 (RHBDF2) (P=1.1 x 10-8; linear model, Table 3 3; Supplementary Fig. 1). 4 Genotype-phenotype analysis for rs12948783 SNP in 936 patients (72 5 genotypes were missing) showed that patients homozygous for the rare allele, i.e., 6 the AA genotype, experienced a low normalized pain relief value (80% ± 9%, mean ± 7 S.E., n=23) that did not differ statistically from that of patients with the GA genotype 8 (88.5% ± 3%, n=243). Comparison of the subjects carrying the rare allele (AA or GA 9 genotypes, n=266) with patients carrying the common GG genotype (104% ± 1%, 10 n=670) showed a highly statistically significant association (P=8.1 x 10-9), with 11 patients carrying the GG genotype and with almost complete pain relief (Fig. 3).

12 Discussion

13 Opioid efficacy for the treatment of pain in cancer patients varies greatly 14 among individuals, suggesting a role for genetic factors in pain relief outcome. Unlike 15 previous pharmacogenetic studies that have tested genetic variations of a few 16 candidate genes in a small number of subjects, our present study represents the first 17 use of genome-wide scanning in a large number of opioid-treated cancer patients to 18 identify genetic variations that may explain differences in pain relief. 19 Pain relief, an inherently subjective parameter, was indeed the phenotype 20 quantified through the BPI method by cancer patients included in the EPOS study; 21 surprisingly, we found a significant correlation between pain relief and country of 22 origin, but not with ethnicity. The association with country of origin might rest in 23 subtle variations in phenotype definitions related to each country or in organizational 24 differences between health care services in various European countries that 25 introduce variability in selection of patients at EPOS recruitment, although the BPI 26 questionnaire is officially recognized and recommended by EAPC and has been 27 validated across cultures and languages. The lack of association between pain relief 28 and ethnicity may be genuine, although the non-homogeneous distribution of 29 patients into 4 ethnic groups (97.6% of patients were ethnically Caucasian) may 30 undermine the statistical power to detect effects linked to ethnicity. 31 Analysis of candidate SNPs defined a set of 8 genetic markers associated

3 http://david.abcc.ncifcrf.gov/home.jsp

1 with individual response to opioid therapy for cancer pain (Table 3). Of these 8 2 SNPs, the best statistically associated SNP maps at the 5’-proximal region, 1892 bp 3 from the transcriptional start site of the rhomboid 5 homolog 2 (Drosophila) 4 (RHBDF2) gene, of unknown function (Fig. 3). rs7104613 is located in the intron 5 region of the spondin 1 extracellular matrix protein (SPON1) gene, which encodes a 6 cell adhesion protein that promotes the attachment of spinal cord and sensory 7 neuron cells and the outgrowth of neurites in vitro. rs10413396 maps at -101 bp from 8 the 5’ region of zinc finger protein 235 (ZNF235) gene, which is located in a cluster 9 of genes encoding members of regulatory proteins characterized by nucleic acid- 10 binding zinc finger domains. 11 None of the genetic variations found in the present study mapped in the opioid 12 receptor genes suggested by previous studies as candidates for individualized opioid 13 treatment (reviewed in (4)). Instead, our results point to the involvement of genes 14 related to transmission of nerve impulse, not opioid pharmacology, in mediating the 15 main modulatory effect of genetic variability on the perception of pain. Nevertheless, 16 it remains possible that opioid therapy outcome is the result of interaction of complex 17 biological systems involved in both pathways, especially considering that the 18 biological function of some genes that we identified is still unknown or uncertain. 19 Our findings could represent the starting point to allow estimation of the 20 probability of being a poor responder to opioid treatment, thus leading to 21 personalized cancer pain therapy in which, for example, individuals at genetic risk of 22 poor response would receive higher starting doses of opioids. Moreover, genes 23 tagged by these polymorphisms may provide a target for new therapeutic strategies 24 to control cancer pain perception.

25 Acknowledgements

26 The authors thank all centers and principal investigators of each center 27 contributing to the EPOS study: Trondheim NORWAY (Pål Klepstad, Stein Kaasa), 28 Oslo, Ullevål NORWAY (Jon Håvard Loge), Oslo, Radiumhospitalet NORWAY 29 (Kristin Bjordal), Copenhagen DENMARK (Per Sjøgren), Turku FINLAND (Eeva 30 Salminen), Essen GERMANY (Marianne Kloke), Aachen GERMANY (Lucas 31 Radbruch), Dresden GERMANY (Rainer Sabatowski), Athens GREECE (Eriphili 32 Argyra), Reykjavik ICELAND (Valgerdur Sigurdardottir, Sigridur Gunnarsdottir), Forli

1 ITALY (Marco Maltoni), Pavia ITALY (Danilo Miotti), Milan ITALY (Alessandra Pigni, 2 Augusto Caraceni), Vilnius LITHUANIA (Irena Poviloniene), Stockholm SWEDEN 3 (Staffan Lundström), St. Gallen SWITZERLAND (Florian Strasser), and 4 Sutton/Chelsea UNITED KINGDOM (Andrew Davies). 5 We thank Harvard-Partners Center for Genetics and Genomics Genotyping 6 Facility, Cambridge, MA, for custom genotyping by MassARRAY.

7 Disclosure of Potential Conflicts of Interest

8 The authors declare no competing financial interests.

9 Grant support

10 This work was supported by the Associazione and Fondazione Italiana 11 Ricerca Cancro (AIRC and FIRC), the Fondazione Floriani, Milan, Italy, the 12 Norwegian Research Council and the EU’s 6th framework (contract no. 037777). 13

1 References

2 1. Riley J, Ross JR, Rutter D, Wells AU, Goller K, du Bois R, et al. No pain relief 3 from morphine? Individual variation in sensitivity to morphine and the need to 4 switch to an alternative opioid in cancer patients. Support Care Cancer. 5 2006;14:56-64. 6 2. Klepstad P, Kaasa S, Cherny N, Hanks G, de Conno F, Research Steering 7 Committee of the EAPC. Pain and pain treatments in european palliative care 8 units. A cross sectional survey from the european association for palliative 9 care research network. Palliat Med. 2005;19:477-84. 10 3. Klepstad P. Pharmacogenetic considerations in the treatment of cancer pain. In: 11 Cancer Pain: Assessment and management. 2nd ed. New York: Cambridge 12 University Press; 2010. p. 180-94. 13 4. Skorpen F, Laugsand EA, Klepstad P, Kaasa S. Variable response to opioid 14 treatment: Any genetic predictors within sight? Palliat Med. 2008;22:310-27. 15 5. Andersen S, Skorpen F. Variation in the COMT gene: Implications for pain 16 perception and pain treatment. Pharmacogenomics. 2009;10:669-84. 17 6. Walter C, Lotsch J. Meta-analysis of the relevance of the OPRM1 118A>G genetic 18 variant for pain treatment. Pain. 2009;146:270-5. 19 7. Knudsen AK, Brunelli C, Kaasa S, Apolone G, Corli O, Montanari M, et al. Which 20 variables are associated with pain intensity and treatment response in 21 advanced cancer patients? - implications for a future classification system for 22 cancer pain. Eur J Pain. 2010. 23 8. Daut RL, Cleeland CS, Flanery RC. Development of the wisconsin brief pain 24 questionnaire to assess pain in cancer and other diseases. Pain. 25 1983;17:197-210. 26 9. Holen JC, Lydersen S, Klepstad P, Loge JH, Kaasa S. The brief pain inventory: 27 Pain's interference with functions is different in cancer pain compared with 28 noncancer chronic pain. Clin J Pain. 2008;24:219-25. 29 10. Galvan A, Falvella FS, Frullanti E, Spinola M, Incarbone M, Nosotti M, et al. 30 Genome-wide association study in discordant sibships identifies multiple 31 inherited susceptibility alleles linked to lung cancer. Carcinogenesis. 32 2010;31:462-5.

1 11. Huang da W, Sherman BT, Lempicki RA. Systematic and integrative analysis of 2 large gene lists using DAVID bioinformatics resources. Nat Protoc. 2009;4:44- 3 57. 4 12. Purcell S, Neale B, Todd-Brown K, Thomas L, Ferreira MA, Bender D, et al. 5 PLINK: A tool set for whole-genome association and population-based linkage 6 analyses. Am J Hum Genet. 2007;81:559-75. 7 13. Macgregor S, Zhao ZZ, Henders A, Nicholas MG, Montgomery GW, Visscher 8 PM. Highly cost-efficient genome-wide association studies using DNA pools 9 and dense SNP arrays. Nucleic Acids Res. 2008;36:e35. 10 14. Bosse Y, Bacot F, Montpetit A, Rung J, Qu HQ, Engert JC, et al. Identification of 11 susceptibility genes for complex diseases using pooling-based genome-wide 12 association scans. Hum Genet. 2009;125:305-18. 13

1 Figure legends

2 Figure 1. Distribution of the pain relief phenotype (0 to 100%) in cancer patients of 3 the EPOS series. Bar indicates the number of cancer patients genotyped in 4 the first (red) and second series (green) or not analyzed (yellow). 5 Figure 2. Manhattan plot of the GWAS in DNA pools of poor or good responders to 6 opioid therapy for cancer pain. Points represent the statistical association of 7 each of >173,000 SNPs (chromosome X shown as chromosome 23) with pain 8 relief category (chi-square test on reconstructed chromosome counts). 486 9 SNPs showed statistical association at P<1.0 x 10-3. Red points indicate 10 SNPs and their relevant genes as confirmed by genotyping of 1008 11 individuals (see Table 3). 12 Figure 3. RHBDF2 on chromosome 17 is a major pain relief modifier locus in the 13 EPOS series of opioid-treated cancer patients. Cancer patients carrying the 14 rare allele (either the AA or GA genotype; n=266) at SNP rs12948783 located 15 in the 5’-proximal region of the RHBDF2 gene showed a statistically 16 significant lower pain relief as compared to patients with the common GG 17 genotype (n=670), who showed an almost complete response to opioid 18 therapy (P=8.1 x 10-9, ANOVA). Dots and bars represent mean ± S.E. values 19 of the normalized pain relief phenotype.

Table 1. Phenotypic characteristics of EPOS cancer patients examined for response to opioid therapy Subject characteristics First series Second series Good responders Poor responders No. of subjects 293 145 570 Gender Male 150 73 308 Female 143 72 262 Median age (range) a 61 (32 - 88) 63 (18 - 86) 62 (27 - 91) Country of origin CH 28 14 22 DE 63 34 62 DK 0 0 17 FI 0 0 18 GB 33 15 109 IS 0 0 82 IT 22 11 141 LT 0 0 29 NO 121 58 61 SE 26 13 29 Pain relief (percentage score; mean ± 94.1 ± 0.3 27.0 ± 1.1 76.5 ± 1.1 S.E.) Total oral morphine equivalent opioid 287 ± 28 402 ± 56 293 ± 17 dose (mg; mean ± S.E.) a Age in years. b CH, Switzerland, DE, Germany, DK, Denmark, FI, Finland, GB, Great Britain, IS, Iceland, IT, Italy, LT, Lithuania, NO, Norway, SE, Sweden.

Author manuscriptshavebeenpeerreviewedandacceptedforpublicationbutnotyetedited. Author ManuscriptPublishedOnlineFirstonMay27,2011;DOI:10.1158/1078-0432.CCR-10-3028

Table 2. Gene ontology analysis by DAVID of the 226 genes defined by the 486 SNPs most significantly (P < 0.001) associated with allelic imbalance in the SNP-array analysis of DNA pools from “poor” or “good” responders to opioid therapy for cancer pain clincancerres.aacrjournals.org Term Gene P ‡ Genes count

GO:0050877~neurological system process 31 3.5 x 10-5 ABLIM1, OR8U9, AGTPBP1, SYT5, KCNA1, COL2A1, OR4C6, RAX2, OR4S2, ELOVL4, OR8K3, DMD, OR4C11, OR5AS1, KCNE1, OR5AP2, COL4A4, COL18A1, OR5R1, OR11L1, PCDH15, AFF2, NRXN1, OR5T3, ACCN1, SLC17A6, SBF2, ABAT, TBL1X, NCAN, PTENP1

GO:0007155~cell adhesion 17 6.1 x 10-3 COL18A1, PLXNC1, NRP1, C21ORF29, NELL1, ACTN1, COL2A1, PCDH15, NRXN1, BTBD9, ITGBL1, CDH12, CD96, SORBS1, NCAN, JAM2, SPON1 on September 24, 2021. © 2011American Association for Cancer Research. GO:0022610~biological adhesion 17 6.2 x 10-3 COL18A1, PLXNC1, NRP1, C21ORF29, NELL1, ACTN1, COL2A1, PCDH15, NRXN1, BTBD9, ITGBL1, CDH12, CD96, SORBS1, NCAN, JAM2, SPON1

GO:0019226~transmission of nerve impulse 11 6.8 x 10-3 COL4A4, ACCN1, SLC17A6, SYT5, AGTPBP1, SBF2, DMD, KCNA1, ABAT, NRXN1, NCAN

GO:0007268~synaptic transmission 10 7.1 x 10-3 COL4A4, ACCN1, SLC17A6, SYT5, AGTPBP1, DMD, KCNA1, ABAT, NRXN1, NCAN

GO:0050890~cognition 20 7.4 x 10-3 COL18A1, ABLIM1, OR5R1, OR8U9, OR11L1, PCDH15, AFF2, COL2A1, OR4C6, OR5T3, RAX2, OR4S2, ELOVL4, OR8K3, OR4C11, OR5AS1, OR5AP2, KCNE1, TBL1X, PTENP1

‡ Calculated by the DAVID functional annotation tool, using a modified Fisher exact test.

Author Manuscript Published OnlineFirst on May 27, 2011; DOI: 10.1158/1078-0432.CCR-10-3028 Author manuscripts have been peer reviewed and accepted for publication but have not yet been edited.

Table 3. List of 8 SNPs associated with the quantitative pain relief phenotype in 1008 European cancer patients treated with opioids a. SNP Chromoso Position Gene Minor Common First and second combined series First series Second me (Mb) allele allele series Frequency BETA b P b P , DNA P , Individual P b of the pools c genotyping b minor allele rs13421094 2 157.8 G A 0.14 0.09789 8.0 x 10-5 6.5 x 10-6 5.8 x 10-4 0.367 rs12211463 6 93.5 G T 0.19 0.07732 6.6 x 10-4 3.9 x 10-6 5.8 x 10-4 0.073 rs7757130 6 113.4 A C 0.07 -0.1157 5.4 x 10-4 2.9 x 10-7 1.5 x 10-6 0.757 rs2473967 6 113.6 C A 0.07 -0.1434 6.2 x 10-5 1.3 x 10-10 8.1 x 10-6 0.212 rs2884129 10 17.6 G T 0.2 -0.09343 7.5 x 10-5 1.9 x 10-6 2.6 x 10-4 0.03 rs7104613 11 14 SPON1 T C 0.07 -0.1172 7.3 x 10-4 2.2 x 10-8 6.8 x 10-5 0.759 rs12948783 17 72 RHBDF2 A G 0.15 -0.1415 1.1 x 10-8 5.8 x 10-7 2.1 x 10-8 0.041 rs10413396 19 49.5 ZNF235 T G 0.05 -0.1447 4.1 x 10-4 3.8 x 10-7 2.2 x 10-4 0.99 a SNPs genotyped in individual cancer patients by MassARRAY and showing statistically significant association (P <1.0 x 10-3) with pain relief; SNPs with statistically significant deviation from the Hardy-Weinberg equilibrium (P <1.0 x 10-3) were excluded. b P and BETA regression coefficient values were obtained by allelic association of SNPs with country-normalized pain relief phenotype (linear model adjusted by gender). c P values obtained by chi-square analysis on reconstructed chromosome numbers in DNA pools.

Downloaded from clincancerres.aacrjournals.org on September 24, 2021. © 2011 American Association for Cancer Research. Downloaded from Figure 1 Author manuscriptshavebeenpeerreviewedandacceptedforpublicationbutnotyetedited. Author ManuscriptPublishedOnlineFirstonMay27,2011;DOI:10.1158/1078-0432.CCR-10-3028 clincancerres.aacrjournals.org on September 24, 2021. © 2011American Association for Cancer Research. Figure 2 Downloaded from Author manuscriptshavebeenpeerreviewedandacceptedforpublicationbutnotyetedited. Author ManuscriptPublishedOnlineFirstonMay27,2011;DOI:10.1158/1078-0432.CCR-10-3028 clincancerres.aacrjournals.org on September 24, 2021. © 2011American Association for Cancer Research. Downloaded from Figure 3 Author manuscriptshavebeenpeerreviewedandacceptedforpublicationbutnotyetedited. Author ManuscriptPublishedOnlineFirstonMay27,2011;DOI:10.1158/1078-0432.CCR-10-3028 clincancerres.aacrjournals.org on September 24, 2021. © 2011American Association for Cancer Research. Author Manuscript Published OnlineFirst on May 27, 2011; DOI: 10.1158/1078-0432.CCR-10-3028 Author manuscripts have been peer reviewed and accepted for publication but have not yet been edited.

Multiple loci modulate opioid therapy response for cancer pain

Antonella Galvan, Frank Skorpen, Pål Klepstad, et al.

Clin Cancer Res Published OnlineFirst May 27, 2011.

Updated version Access the most recent version of this article at: doi:10.1158/1078-0432.CCR-10-3028

Supplementary Access the most recent supplemental material at: Material http://clincancerres.aacrjournals.org/content/suppl/2021/03/17/1078-0432.CCR-10-3028.DC1

Author Author manuscripts have been peer reviewed and accepted for publication but have not yet been Manuscript edited.

E-mail alerts Sign up to receive free email-alerts related to this article or journal.

Reprints and To order reprints of this article or to subscribe to the journal, contact the AACR Publications Subscriptions Department at [email protected].

Permissions To request permission to re-use all or part of this article, use this link http://clincancerres.aacrjournals.org/content/early/2011/05/27/1078-0432.CCR-10-3028. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC) Rightslink site.