Skip Navigation LinksHome > December 2013 - Volume 119 - Issue 6 > Effect of Catechol-o-methyltransferase-gene (COMT) Variants...
Anesthesiology:
doi: 10.1097/ALN.0000000000000013
Pain Medicine

Effect of Catechol-o-methyltransferase-gene (COMT) Variants on Experimental and Acute Postoperative Pain in 1,000 Women undergoing Surgery for Breast Cancer

Kambur, Oleg M.Sc.(Pharm.)*; Kaunisto, Mari A. Ph.D.; Tikkanen, Emmi M.Sc.; Leal, Suzanne M. Ph.D.§; Ripatti, Samuli Ph.D.; Kalso, Eija A. M.D., Ph.D.#

Free Access
Supplemental Author Material
Press Release
Article Outline
Collapse Box

Author Information

Collapse Box

Abstract

Background: Catechol-O-methyltransferase (COMT) metabolizes catecholamines in different tissues. Polymorphisms in COMT gene can attenuate COMT activity and increase sensitivity to pain. Human studies exploring the effect of COMT polymorphisms on pain sensitivity have mostly included small, heterogeneous samples and have ignored several important single nucleotide polymorphisms (SNPs). This study examines the effect of COMT polymorphisms on experimental and postoperative pain phenotypes in a large ethnically homogeneous female patient cohort.
Methods: Intensity of cold (+2-4°C) and heat (+48°C) pain and tolerance to cold pain were assessed in 1,000 patients scheduled for breast cancer surgery. Acute postoperative pain and oxycodone requirements were recorded. Twenty-two COMT SNPs were genotyped and their association with six pain phenotypes analyzed with linear regression.
Results: There was no association between any of the tested pain phenotypes and SNP rs4680. The strongest association signals were seen between rs165774 and heat pain intensity as well as rs887200 and cold pain intensity. In both cases, minor allele carriers reported less pain. Neither of these results remained significant after strict multiple testing corrections. When analyzed further, the effect of rs887200 was, however, shown to be significant and consistent throughout the cold pressure test. No evidence of association between the SNPs and postoperative oxycodone consumption was found.
Conclusions: SNPs rs887200 and rs165774 located in the untranslated regions of the gene had the strongest effects on pain sensitivity. Their effect on pain is described here for the first time. These results should be confirmed in further studies and the potential functional mechanisms of the variants studied.
Back to Top | Article Outline

What We Already Know about This Topic

* Variants of the gene for the catecholamine-metabolizing enzyme, COMT, are associated in some small studies with differences in sensitivity to pain
Back to Top | Article Outline

What This Article Tells Us That Is New

* In a more complete assessment of variants in the COMT gene in 1,000 women undergoing surgery for breast cancer, there were weak associations with sensitivity to heat or cold pain
* The lack of association of COMT variants with acute postoperative opioid requirement raises questions regarding the clinical relevance of this gene in acute postoperative pain
CATECHOL-O-METHYLTRANSFERASE (COMT) metabolizes catecholamines in glial cells and postsynaptic neurons in the central nervous system and other tissues.1 Metabolic activity of COMT is modulated by genetic variation which can lead to significant differences in COMT activity in humans.2,3 The role of COMT has previously been studied in pain sensitivity and opioid analgesia.4–9 However, most of the studies have focused on a few polymorphisms only.
The COMT gene is located in the long arm of chromosome 22. It consists of six exons covering more than 27 kb. Four of the exons are translated into protein, and promoters located in exon 3 regulate the synthesis of distinct types of messenger RNAs that code for the membrane-bound and soluble forms of the enzyme. Within the 27-kb genomic area, more than 900 genetic variants have been identified, the vast majority being located within the introns. The only common COMT single nucleotide polymorphism (SNP) that is nonsynonymous is rs4680 (Val108/158Met). The Met allele has been shown to reduce COMT activity three- to four-fold.10 In addition, several other polymorphisms have been suggested to modulate expression or activity of COMT. However, their effects on COMT activity in vivo and on pain phenotypes have not been characterized.11–16
COMT polymorphisms, SNP rs4680 in particular, have been studied in several painful conditions such as temporomandibular dysfunction, fibromyalgia, migraine, and headache. Low COMT activity has been suggested to increase the intensity of pain in these conditions (see study by Kambur and Männistö17 for references). In osteoarthritis, for example, the Val108/158Met variant has been associated with more intense pain with similar tissue destruction.18 Low COMT activity has also been related to increased pain sensitivity in experimental thermal and mechanical pain tests.19,20 The relevance of COMT polymorphism in acute postsurgery pain has been assessed in only two reports. One of them found an association between COMT SNP rs740603 and maximum postoperative pain ratings,6 and the other study reported associations between SNPs rs4818 and rs6269 and intensity of postoperative pain.21
Previous studies exploring the effect of COMT on pain have mostly been of small size and they have included both men and women of different ethnic origins. Most studies have analyzed only SNP rs4680 (Val108/158Met). Recent studies have suggested that other nearby SNPs (e.g., rs6269 and rs4818) and haplotypes containing these SNPs could be as important.19,20 There is much genetic variation within and around the COMT gene that has been overlooked in previous studies. Especially, synonymous and 3′- and 5′-untranslated region (UTR) SNPs remain largely uninvestigated even though they can also significantly affect the gene expression and thus have a phenotypic effect as well. In this study, our aim was to test the association between all common variation within and around the COMT gene and pain phenotypes in by far the largest homogenous dataset consisting of 1,000 Finnish female breast cancer patients. Detailed measures of experimental pain as well as acute postsurgery pain and oxycodone consumption were analyzed. Our study combines dense, high-quality genotype data, which allows a more comprehensive evaluation of COMT variants, and a large patient sample with rich phenotypic data.
Back to Top | Article Outline

Materials and Methods

Ethics
Written informed consent was obtained from each subject participating in the study by either a research nurse or a physician. Surgery and all other treatments were part of the patients’ normal clinical care under the supervision of the physician in charge. The research protocol had been approved by the coordinating ethics committee (136/E0/2006) and the ethics committee of the Department of Surgery (Dnro 148/E6/05) of the Hospital District of Helsinki and Uusimaa.
Back to Top | Article Outline
Subjects and Initial Clinical Assessment
This study reports the data of 1,000 patients who were operated for breast cancer at the Unit for Breast Surgery at the Women’s Hospital, Helsinki University Central Hospital, Helsinki, Finland, between August 2006 and December 2010. The study cohort is described in detail in the accompanying article.22 In brief, either mastectomy or breast conserving surgery, with sentinel node biopsy, axillary clearance, or both, was performed. Before surgery, the patients filled in questionnaires on depression (Beck Depression Inventory)23 and anxiety (State and Trait Anxiety Inventory),24 and underwent experimental pain tests.
Back to Top | Article Outline
Experimental Pain
Contact heat pain and cold pain were assessed as described in the study by Kaunisto et al.22 For the heat pain, the volar surface of the forearm was first exposed to 43°C for 5 s and then 48°C also for 5 s. The patients assessed the intensity and unpleasantness of pain with a 0–10 numerical rating scale (NRS) at the end of each test; zero represented ‘‘no pain,” and 10 ‘‘worst imaginable pain.” All 1,000 women were evaluated for heat pain sensitivity.
Cold pain was measured by immersing the hand to a cold water (2-4°C) bath for the maximum time tolerated by the patients but not longer than 90 s. Time to withdrawal of the hand and pain intensity (NRS) and unpleasantness were measured every 15 s during the test and at withdrawal. The cold pain test was not performed on the first 100 patients studied (n = 900).
Back to Top | Article Outline
Anesthesia, Postsurgical Pain Assessment, and Administration of Oxycodone
The patients received standardized anesthesia during which a blood specimen was drawn for DNA isolation and banking.22 In the postanesthesia care unit, the patients were asked about the pain intensity at rest and during motion. Motion pain was assessed by asking the patient to raise the arm ipsilateral to the surgical intervention up to 90°. Patients were titrated with i.v. oxycodone by a research nurse who asked about the pain intensity every 5 min and administered oxycodone in doses of 1–3 mg until adequate analgesia (NRS < 4/10) was achieved. After this, pain intensity was recorded every 15 min until the patient needed the next dose of oxycodone. Time to the first dose of oxycodone and time needed to achieve satisfactory pain relief were documented. For the ward, the patients were provided with a patient- controlled analgesia device for up to 20 h except for 70 day-case surgery patients. The amount of oxycodone needed to achieve satisfactory pain relief, oxycodone consumption in the postanesthesia care unit, and the total amount of oxycodone consumed during 20 h after the operation were recorded.
Back to Top | Article Outline
SNP Selection and Genotyping
Table 1
Table 1
Image Tools
A total of 22 SNPs were genotyped from a 47 kb region (chr22: 18,299–18,346 kb, NCBI36/hg18 assembly) covering the COMT gene as well as 10 kb of 5′ and 3′ flanking regions (table 1). Information about the linkage disequilibrium (LD) within this area was used to select an optimal set of SNPs capturing most of the genetic information. The SNP selection was based on phase 3 data of the HapMap database and performed using the Tagger program included in the Haploview 4.1 software.**25 The HapMap database†† contained 30 SNPs having a minor allele frequency of greater than 10% in the European population (Utah residents with North-Western European ancestry from the Centre d’Etude du Polymorphisme Humain collection and Toscans in Italy). The aim was to capture all these SNPs by setting the limit for the pair-wise r2 to 0.9 or greater. Nineteen SNPs were chosen to be genotyped, and 17 of these were successfully included in the genotyping multiplexes. Thus, 28 of the 30 HapMap SNPs were captured with a mean r2 of 0.994. In addition to these 17 genotyped SNPs, two additional SNPs were selected based on their position in the exonic areas (rs4818 and rs16599) and three because they were known to be common based on HapMap phase 2 data but not available on phase 3 data (rs74063, rs165774, and rs2518824).
DNA was extracted from peripheral blood using the Autopure LS automated DNA purification instrument (Gentra Systems, Inc., Minneapolis, MN). The 22 SNPs were genotyped using the Sequenom MassARRAY system and the iPLEX Gold Single Base Extension chemistry (Sequenom, San Diego, CA) in a multiplex format.26 This method has excellent success (>95%) and accuracy (100%) rates.27 Both duplicate, Centre d’Etude du Polymorphisme Humain control and water control samples were included in each DNA plate to confirm the accuracy of the genotyping results. Genotyping was performed blind to phenotypic information.
Back to Top | Article Outline
Statistical Analysis
SNPs.
The entire sample was tested to determine whether the null hypothesis of Hardy–Weinberg equilibrium could be rejected by applying the chi-square method. The Haploview 4.1 program was used to determine the pair-wise LD between the SNPs (r2 value) and to identify haploblock structures using the CI algorithm.25
Back to Top | Article Outline
Power Calculations.
The theoretical effect size needed for our study sample to have power to detect association with the previously pain-associated variant, rs4680, was calculated using R software environment.‡‡ The minor allele frequency was set to be the same as detected in our sample, 0.457. Additive inheritance model (each copy of the minor allele having an effect) was assumed. The results showed that in order to have a power greater than 0.9 to reach a P value of 0.05 (considered to be an appropriate level of evidence when replicating a previous association), the effect size should be 15% of the SD for the phenotype per allele (additive model). In our case, the minimum effect size of 0.36 NRS units on a scale 0–10 for heat pain and 0.39 NRS units for cold pain would produce a P value less than 0.05 with 90% certainty. Thus, our sample size provides enough power to detect meaningful effect sizes for rs4680.
Back to Top | Article Outline
Association Analyses.
Association between the COMT SNPs and pain phenotypes was tested using linear regression analysis. Additive, dominant, and recessive models were all considered. Analyses were performed with PLINK software.§§28
The experimental pain variables that were analyzed include intensity of thermal pain produced by heat stimulation (48°C, NRS) and two different cold pain–related variables: time to withdrawal and pain intensity (NRS) at 15 s. Those patients who withdrew their hand before the analyzed time point were coded as having the maximum NRS value (n = 118 at 15 s).
The postoperative pain-related variables that were studied include pain intensity during motion (NRS) when needing the first dose of oxycodone, the amount of oxycodone needed to achieve adequate analgesia (NRS <4/10), and total oxycodone consumption (mg·kg−1·20 h−1).
All association analyses were performed using variables adjusted for potential confounding factors, that is, on residual phenotypic scores after regressing out covariate effects. The covariates used for experimental pain phenotypes include age, body mass index, presence of any preoperative chronic pain condition (yes/no), and anxiety (State and Trait Anxiety Inventory score). For postoperative variables, the residual phenotypes were also adjusted for the effect of the type of surgery performed. When analyzing the association with total oxycodone consumption, the 70 day-case surgery patients who did not use the patient-controlled analgesia device were excluded.
Back to Top | Article Outline
Follow-up Analyses for SNP rs887200.
SNP rs887200 was explored further using parametric tests where homozygous carriers of risk allele were compared with other patients. Cold pain intensities across different time points were analyzed with two-way ANOVAs for repeated measurements using time and genotype as independent variables. Time to withdrawal was analyzed with Student unpaired t test.
Back to Top | Article Outline
Permutations and Correcting for Multiple Testing.
Due to the fact that the distributions of the studied quantitative traits were not normal, permutation procedures were used to obtain empirical P values. The associations were calculated using the max(T) permutation option of the PLINK software performing 10,000 permutations for each SNP. This option not only obtains empirical P values for each SNP but also produces P values adjusted for performing multiple tests in order to control for the family-wise error rate. Because this method takes into account the correlation due to LD between the studied SNPs, it provides a less stringent correction than using a Bonferroni correction to control for multiple testing. These P values were further multiplied by 18, to correct for the number of tested phenotypes (six) and statistical models (three). The level of significance was set at a corrected P value of 0.05.
Back to Top | Article Outline
Haplotype Analyses.
Fig. 1
Fig. 1
Image Tools
In addition to single-SNP analyses, haplotypes were also tested for association. The PLINK software was used for calculating the most likely haplotype combinations for each patient using the haploblock structure identified with the Haploview program as a guideline (fig. 1).25,28 Three different haplotype combinations, corresponding with the three identified haploblocks, were determined for each patient. Furthermore, we constructed the haplotypes previously associated with pain sensitivity phenotypes in a study by Diatchenko et al.19,20 who classified them as low pain sensitivity (LPS), average pain sensitivity, and high pain sensitivity. SNPs rs4633, rs4818, and rs4680 were used for this purpose. Haplotype-association analyses were performed using linear regression with an additive model, where all haplotypes having a frequency above 1% within a haploblock were tested one by one against all the others. Permutations were performed in a similar way as to the single-SNP analyses.
Back to Top | Article Outline
In Silico eQTL Database Analyses
The expression Quantitative Trait Loci-association patterns between COMT mRNA levels and genotypes of SNPs rs165744 and rs887200 were investigated using GENEVAR (GENe Expression VARiation) software‖‖ developed at Wellcome Trust Sanger Institute (Hinxton, Cambridge, United Kingdom). This interactive Java interface was used to perform searches in three available gene expression–profiling datasets containing data based on HapMap lymphoblastoid cell lines,29 three tissue types derived from female twins of the MuTHER study,30 and three cell types derived from umbilical cords of GenCord individuals.31
Back to Top | Article Outline

Results

Genetic Association Studies
Twenty-two SNPs within or nearby the COMT gene were genotyped. The success rates, minor allele frequencies, and Hardy–Weinberg equilibrium values of the genotyped SNPs are presented in table 1. The average success rate of genotyping was 99.0%. None of the SNPs showed deviation from Hardy–Weinberg equilibrium (P < 0.05). Figure 1 presents the location of the studied COMT SNPs and the CI method–based LD block structures in the study subjects.
Table 2
Table 2
Image Tools
Table 3
Table 3
Image Tools
Fig. 2
Fig. 2
Image Tools
The clinical characteristics of the patients are shown in table 2. The results of the analyses testing association between COMT and experimental heat pain intensity are shown in table 3 and figure 2. Of the SNPs studied, rs165774 showed the strongest association with the intensity of heat pain at 48°C. The smallest uncorrected empirical P value of 0.003 was obtained when the dominant model was used. This P value remained less than 0.05 after correcting for the number of SNPs analyzed but not when the more stringent corrections were performed (P = 1). Regression coefficient (β) that represents the effect size and direction (using the major allele as the reference allele) was −0.46. The mean NRS of the patients homozygous for the major allele (G) of this SNP (56%) was 3.72 and the mean NRS of the heterozygous individuals (38%) was 3.20 whereas that of the individuals homozygous for the minor allele (5%) was 3.48.
Fig. 3
Fig. 3
Image Tools
Fig. 4
Fig. 4
Image Tools
Fig. 5
Fig. 5
Image Tools
Four to five SNPs, depending on the model used, showed nominal association with cold pain (table 3 and figs. 3 and 4). The strongest signal was seen between SNP rs887200 located within the 3′ prime region of the gene and the NRS score at 15 s using a recessive model. The uncorrected empirical P value was 0.004, but the association, however, did not sustain stringent multiple testing corrections (table 3). This SNP was relatively rare, only 25 of the 900 patients tested for the cold pain were homozygous for the minor allele (C). The regression coefficient for this SNP was −1.62 meaning that the homozygous individuals reported in average 1.6 points lower NRS values compared with the other groups. In a follow-up analysis, C/C-carriers reported lower pain intensity scores than other patients (two-way ANOVA for repeated measurements, effect of genotype: P < 0.0001), and the effect of rs887200 was consistent across all measured time points from 15 to 90 s as there was no time–genotype interaction (P > 0.05; fig. 4). Furthermore, C/C-carriers tolerated cold stimulation longer as compared with noncarriers (P = 0.005; fig. 5).
Of the other SNPs showing nominal evidence of association with cold pain, rs1544325 is of interest since using the dominant model it was associated with both of the cold pain–related phenotypes tested. The smallest uncorrected empirical P value of 0.007 (β = 6.11) was reached when association with cold pain tolerance (time to withdrawal) was tested (table 3).
Back to Top | Article Outline
Postoperative Pain and Oxycodone Consumption
Association between the COMT SNPs and several postoperative pain variables was studied. Three SNPs, rs4646312, rs2239393, and rs4818, all located within the same (middle) haploblock gave uncorrected empirical P values in the range of 0.008–0.01 for the phenotype “pain intensity during motion when the patient needed the first dose of oxycodone” using the dominant model (table 1, see Supplemental Digital Content 1, http://links.lww.com/ALN/B3). After multiple testing corrections, no evidence of association remained. In all cases, the minor allele (C, G, and G, respectively) carriers had higher postoperative pain scores than the patients being homozygous for the major allele (β = 0.25–0.28).
None of the studied SNPs showed any evidence of association (P < 0.05) with the other postoperative phenotypes, that is, the amount of oxycodone needed to achieve adequate analgesia and total oxycodone consumption. Results for the total oxycodone consumption are presented in table 2, Supplemental Digital Content 1, http://links.lww.com/ALN/B3.
Back to Top | Article Outline
Haplotype Analyses
In addition to single-SNP analysis, association between the pain phenotypes and the haplotypes constructed both based on the LD structure and previous literature was studied. Of the three haploblocks formed (fig. 1), blocks 1 and 2 consisted of seven SNPs (from rs6518591 to rs5993882 and from rs4646312 to 165774) and block 3 of five SNPs (from rs9332377 to rs887200). Within block 1, five haplotypes having a frequency above 1% could be constructed, whereas six haplotypes were identified within blocks 2 and 3 (table 2, see Supplemental Digital Content 1, http://links.lww.com/ALN/B3).
The strongest haplotype-association signal for experimental pain phenotypes was seen between heat pain intensity and haplotype TTACACA of block 2 (uncorrected P value 0.014). Carriers of this haplotype reported less heat pain (β = −0.31) compared with the carriers of all the other haplotypes. This haplotype has a frequency of 24% and was the only one of the six detected haplotypes containing the minor allele (A) of the SNP rs165774 associating with decreased heat pain sensitivity in single-SNP analyses as well. None of the haplotypes showed evidence of association with cold pain phenotypes. These results are presented in table 2, Supplemental Digital Content 1, http://links.lww.com/ALN/B3.
None of the haplotypes showed evidence of association with the phenotypes related to postoperative pain or oxycodone consumption.
In the studies by Diatchenko et al.,19,20 haplotypes consisting of three to four SNPs all located within our haploblock 2 have been suggested to have a significant effect on pain perception. To test this specific hypothesis, we constructed these haplotypes classified in the original study as LPS, average pain sensitivity, and high pain sensitivity haplotypes and tested association. Haplotype LPS (alleles CGG) was nominally associated with cold pain measured by NRS at 15 s (P = 0.031) and with postoperative pain (P = 0.045). However, the effect was opposite to what was expected based on the previous studies because the carriers of the LPS haplotype had a higher pain score than the others (β = 0.31 and 0.14, respectively), and after experimental-wise corrections, the P values were not significant. None of the LPS/high pain sensitivity/average pain sensitivity haplotypes showed any evidence of association to the other studied variables.
Back to Top | Article Outline
In Silico eQTL Analysis
The possible functional significance of the SNPs showing statistically significant evidence of association with pain phenotypes in this study (rs165744 and rs887200) was assessed analysing the association between the genotypes of these SNPs and the COMT-expression levels using the GENEVAR database. These SNPs did not show any significant effect on the expression levels of COMT in any of the three datasets available.
Back to Top | Article Outline

Discussion

A few variants of the COMT gene have been much studied in relation to various aspects of pain in fairly small patient samples. The current study is unique in having analyzed 22 different SNPs in a large and homogenous patient cohort with rich phenotypic data. The main finding of our study is that there was no association between the most studied SNP rs4680 (Val108/158Met) and any of the studied pain phenotypes, whereas the strongest association signals were seen between experimental pain and COMT SNPs rs887200 and rs165774. These associations have not been described before.
Back to Top | Article Outline
Experimental Pain
SNP rs4680 (Val108/158Met), which modulates COMT activity, has been examined in most detail to date. No evidence of association between this SNP and the studied experimental pain phenotypes was seen in our study in agreement with the work by Kim et al.5 Our study was well powered to detect an association for this SNP had there been any. Any effect of rs4680 has been shown mainly in settings applying nociceptive stimuli of higher intensity and longer duration.9,19,32 It is possible that the effect is present only after a more intense stimulation than what was used in our study or in certain specific aspects of pain such as the rate of temporal summation of heat pain.19 It has also been suggested that the effect of rs4680 may depend on the genomic environment, for example, haplotype or cooccurring rare variants.19,20,33
Interesting evidence has also been provided for an association between pain sensitivity and certain COMT haplotypes (LPS/average pain sensitivity/high pain sensitivity) containing the rs4680.19,20 Our results, however, did not support these findings. There are several possible reasons for this lack of replication. The experimental pain phenotypes examined are rather different between the studies. In the original studies, a composite phenotype consisting of 16 parameters was analyzed.20 The strongest associations were seen for thermal pain, which was the only modality also assessed in our study.19 Its parameters, however, were different from those used in our study, and we did not assess heat pain threshold or tolerance. There are also other important differences as we studied breast cancer patients of whom approximately 25% had some previous chronic pain condition. Also, it should be noted that the sample sizes of the earlier studies are rather modest which could increase the risk of false-positive findings.
In our study, the strongest association was seen between heat pain intensity (48°C) and SNP rs165774 (uncorrected P = 0.003). Subjects carrying the A-allele showed lower pain intensity (mean NRS: G/A, 3.20; A/A, 3.48) than homozygous G-carriers (NRS 3.72). This result remained significant after correcting for the number of SNPs (P < 0.05) but did not sustain more stringent corrections. Rs165774 is located in the intronic region between exons 5 and 6 and is part of haploblock 2. Interestingly, a preliminary report showing association of rs165774 with temporomandibular dysfunction suggests that it might be important for the expression of an alternatively spliced, truncated COMT isoform showing differential expression pattern and higher efficiency in metabolizing norepinephrine.##
Regarding the cold pain phenotypes, the strongest association in our study was detected between SNP rs887200 and cold pain intensity. Homozygous minor allele (C/C) carriers were less sensitive to pain (mean NRS, 4.1) than the subjects having other genotypes (mean NRS, 5.8 and 5.7, respectively). The magnitude of the effect of rs887200 in our study was greater than that of other COMT SNPs, including several SNPs which have shown associations in earlier studies. This result did not remain significant after multiple testing corrections either. However, in the follow-up analysis taking into account individual nociceptive baselines as well as responses over the whole 90 s test, the effect of rs887200 was consistent across all measured time points (P < 0.0001) and C/C-carriers also tolerated cold stimulation longer as compared with noncarriers (P = 0.005; fig. 5).
The effect of rs887200 on pain sensitivity has not been studied before. It is located within the 3′ region of the gene, but it is also intragenic to a neighboring gene, ARVCF (armadillo repeat protein deleted in velocardiofacial syndrome). Other COMT 3′-UTR SNPs have been suggested to contribute to human neuropsychiatric phenotypes.34,35 Furthermore, in a recent animal study, eQTL analysis showed that in inbred mouse strains, COMT activity and related behavioral phenotypes are mainly regulated by a transposon located in 3′-UTR.36 Also in humans, one 3′-UTR SNP associated with neuropsychiatric phenotypes was connected to altered COMT expression.11 Regardless of the origin, the mechanisms of action of 3′-UTR variants are less evident because they do not alter the amino acid sequence of the protein. Possible mechanisms include attenuation of interaction of 3′-UTR area with negative modulators of expression such as microRNAs37 with a consequent increase in expression. The effect of rs887200 on COMT activity remains to be studied. The in silico eQTL searches did not provide evidence of this SNP affecting the COMT-expression levels significantly in the cell types available.
Back to Top | Article Outline
Postoperative Pain and Oxycodone Consumption
None of the SNPs showed significant association with the acute postoperative pain measures. Because of previous findings, SNP rs4818 was of special interest. This SNP as well as two other SNPs within the same haploblock, rs4646312 and rs2239393, showed nominal association (uncorrected P = 0.008–0.01). The carriers of the rarer alleles of these SNPs (rs4646312C, rs2239393G, and rs4818G) needed the first dose of oxycodone at higher postoperative pain scores during motion than the patients being homozygous for the major allele. In earlier studies, rs4818 has been associated with decreased experimental and postoperative pain intensity.19,21 Also, COMT SNP rs740603 in intron 1 has been previously suggested to be associated with maximum postoperative pain ratings.6 However, we were unable to replicate this finding.
Both the Val108/158Met variant and a COMT haplotype carrying the Met allele have been associated with variation in the efficacy of morphine in cancer pain treatment.8,38 However, no effect of Val108/158Met on opioid analgesia was seen in a larger but highly variable cohort of pain patients.39 In our study, no evidence of association with postoperative oxycodone consumption was seen even after adjusting the doses with age, body mass index, type of surgery, and other potential confounding factors.
Back to Top | Article Outline
Methodological Considerations
When multiple hypotheses are being tested, the risk of type I error and false-positive associations increases. This should be addressed by correcting for the multiple tests performed as was done in this study. We not only took into account the number of independent SNPs tested but also corrected for the number of phenotypes and statistical models used. Unfortunately, these very stringent correction methods also increase the probability of type II error (failure to reject the null hypothesis when it is false). We believe that it is possible and even likely that some of the SNPs showing border-line P values in this study have a true effect, but our study, even though it has 1,000 patients, is still underpowered to detect a statistically significant association. It becomes practically impossible to collect any larger, homogenous, and phenotypically well-characterized samples. With this sample size, the effect of the variant should be extremely strong for the results to remain statistically significant after multiple testing. Especially, regarding rare variants, the number of carriers can be small leaving effect sizes of these rare variants insufficient to reach the level of statistical significance.
It is commonly accepted that when replicating earlier findings, the criteria needed for evidence of association can be less stringent. However, in case of true replication also the direction of the effect (the risk allele) has to be in accordance with earlier findings. None of the replication efforts done in this study met these two criteria. In general, lack of replication of association signals that are based on candidate gene studies has resulted in the current wave of interest in genome-wide association studies where no prior hypothesis is needed.
In an ideal situation, the effects of several cooccurring SNPs can be simultaneously examined. From a statistical point of view, simultaneous assessment and testing of multiple SNPs (and phenotypes) are, however, difficult. Currently, evaluating COMT haplotypes appears to be the best way to assess the effect of COMT on pain and to consider multiple COMT SNPs at the same time. This method was used in our study as well, but the association signals seen were not more significant than the ones achieved by analyzing individual SNPs.
Our study population consists of patients having breast cancer. This can be regarded as a limitation because there are also some studies reporting associations between COMT variants and breast cancer.13,40,41 Theoretically, breast cancer risk variants could be overrepresented among our patients. Also, approximately 25% of the patients reported chronic pain and some high anxiety, which sensitize to acute pain. Nevertheless, this was taken into account by adjusting the data for these variables.
Back to Top | Article Outline

Conclusions

Of the 22 COMT SNPs studied, 3′-UTR SNP rs887200 and intronic rs165774, whose effect on pain have not been described before, showed strongest evidence of association to pain sensitivity. Their association with human pain phenotypes should be replicated in other cohorts. Our results provide basis for future studies on the mechanism of action and effects of COMT activity.
This study was initiated in collaboration with the late K. Steven LaForge, Ph.D., Senior Research Fellow, Institute of Molecular Medicine, University of Helsinki, Helsinki, Finland; the authors thank his enthusiasm and expert knowledge. The authors are indebted to Eija R. Ruoppa (R.N.) and Minna Kaiponen (R.N.), Department of Anesthesia, Helsinki University Central Hospital, Helsinki, Finland, for excellent assistance throughout the project.
** Haploview. Available at: http://www.broadinstitute.org/scientific-community/science/programs/medical-and-population-genetics/haploview/haploview. Accessed June 14, 2012. Cited Here...
†† HapMap database. Available at: http://www.hapmap.org/. Accessed February 25, 2010. Cited Here...
‡‡ Available at: http://www.r-project.org/. Accessed March 25, 2013. Cited Here...
§§ Available at: http://pngu.mgh.harvard.edu/~purcell/plink/. Accessed May 17, 2013. Cited Here...
‖‖ Available at: http://www.sanger.ac.uk/resources/software/genevar/. Accessed May 24, 2012. Cited Here...
## 90th General Session and Exhibition of the International Association of Dental Research, Iguaçu Falls, Brazil. Meloto C, Segall SK, Gauthier J, Rizzatti-Barbosa CM, Diatchenko L: Characterizing a novel COMT isoform: Potential implication on TMD. J Dent Res 2012; 91(Spec Iss B):abstract 968. Available at: https://iadr.confex.com/iadr/2012rio/webprogram/Paper166239.html, www.dentalresearch.org. Accessed October 29, 2013. Cited Here...
Back to Top | Article Outline

References

1. Männistö PT, Kaakkola S. Catechol-O-methyltransferase (COMT): Biochemistry, molecular biology, pharmacology, and clinical efficacy of the new selective COMT inhibitors. Pharmacol Rev. 1999;51:593–628

2. Boudíková B, Szumlanski C, Maidak B, Weinshilboum R. Human liver catechol-O-methyltransferase pharmacogenetics. Clin Pharmacol Ther. 1990;48:381–9

3. Weinshilboum RM, Raymond FA. Inheritance of low erythrocyte catechol-o-methyltransferase activity in man. Am J Hum Genet. 1977;29:125–35

4. Akil M, Kolachana BS, Rothmond DA, Hyde TM, Weinberger DR, Kleinman JE. Catechol-O-methyltransferase genotype and dopamine regulation in the human brain. J Neurosci. 2003;23:2008–13

5. Kim H, Mittal DP, Iadarola MJ, Dionne RA. Genetic predictors for acute experimental cold and heat pain sensitivity in humans. J Med Genet. 2006;43:e40

6. Kim H, Lee H, Rowan J, Brahim J, Dionne RA. Genetic polymorphisms in monoamine neurotransmitter systems show only weak association with acute post-surgical pain in humans. Mol Pain. 2006;2:24

7. Oswald LM, McCaul M, Choi L, Yang X, Wand GS. Catechol-O-methyltransferase polymorphism alters hypothalamic-pituitary-adrenal axis responses to naloxone: A preliminary report. Biol Psychiatry. 2004;55:102–5

8. Rakvåg TT, Klepstad P, Baar C, Kvam TM, Dale O, Kaasa S, Krokan HE, Skorpen F. The Val158Met polymorphism of the human catechol-O-methyltransferase (COMT) gene may influence morphine requirements in cancer pain patients. Pain. 2005;116:73–8

9. Zubieta JK, Heitzeg MM, Smith YR, Bueller JA, Xu K, Xu Y, Koeppe RA, Stohler CS, Goldman D. COMT val158met genotype affects mu-opioid neurotransmitter responses to a pain stressor. Science. 2003;299:1240–3

10. Lotta T, Vidgren J, Tilgmann C, Ulmanen I, Melén K, Julkunen I, Taskinen J. Kinetics of human soluble and membrane-bound catechol-O-methyltransferase: A revised mechanism and description of the thermolabile variant of the enzyme. Biochemistry. 1995;34:4202–10

11. Bray NJ, Buckland PR, Williams NM, Williams HJ, Norton N, Owen MJ, O’Donovan MC. A haplotype implicated in schizophrenia susceptibility is associated with reduced COMT expression in human brain. Am J Hum Genet. 2003;73:152–61

12. Dempster EL, Mill J, Craig IW, Collier DA. The quantification of COMT mRNA in post mortem cerebellum tissue: Diagnosis, genotype, methylation and expression. BMC Med Genet. 2006;7:10

13. Ji Y, Olson J, Zhang J, Hildebrandt M, Wang L, Ingle J, Fredericksen Z, Sellers T, Miller W, Dixon JM, Brauch H, Eichelbaum M, Justenhoven C, Hamann U, Ko Y, Brüning T, Chang-Claude J, Wang-Gohrke S, Schaid D, Weinshilboum R. Breast cancer risk reduction and membrane-bound catechol-O-methyltransferase genetic polymorphisms. Cancer Res. 2008;68:5997–6005

14. Chen J, Lipska BK, Halim N, Ma QD, Matsumoto M, Melhem S, Kolachana BS, Hyde TM, Herman MM, Apud J, Egan MF, Kleinman JE, Weinberger DR. Functional analysis of genetic variation in catechol-O-methyltransferase (COMT): Effects on mRNA, protein, and enzyme activity in postmortem human brain. Am J Hum Genet. 2004;75:807–21

15. Tsao D, Shabalina SA, Gauthier J, Dokholyan NV, Diatchenko L. Disruptive mRNA folding increases translational efficiency of catechol-O-methyltransferase variant. Nucleic Acids Res. 2011;39:6201–12

16. Lee SG, Joo Y, Kim B, Chung S, Kim HL, Lee I, Choi B, Kim C, Song K. Association of Ala72Ser polymorphism with COMT enzyme activity and the risk of schizophrenia in Koreans. Hum Genet. 2005;116:319–28

17. Kambur O, Männistö PT. Catechol-O-methyltransferase and pain. Int Rev Neurobiol. 2010;95:227–79

18. van Meurs JB, Uitterlinden AG, Stolk L, Kerkhof HJ, Hofman A, Pols HA, Bierma-Zeinstra SM. A functional polymorphism in the catechol-O-methyltransferase gene is associated with osteoarthritis-related pain. Arthritis Rheum. 2009;60:628–9

19. Diatchenko L, Nackley AG, Slade GD, Bhalang K, Belfer I, Max MB, Goldman D, Maixner W. Catechol-O-methyltransferase gene polymorphisms are associated with multiple pain-evoking stimuli. Pain. 2006;125:216–24

20. Diatchenko L, Slade GD, Nackley AG, Bhalang K, Sigurdsson A, Belfer I, Goldman D, Xu K, Shabalina SA, Shagin D, Max MB, Makarov SS, Maixner W. Genetic basis for individual variations in pain perception and the development of a chronic pain condition. Hum Mol Genet. 2005;14:135–43

21. Lee PJ, Delaney P, Keogh J, Sleeman D, Shorten GD. Catecholamine-o-methyltransferase polymorphisms are associated with postoperative pain intensity. Clin J Pain. 2011;27:93–101

22. Kaunisto MA, Jokela R, Tallgren M, Kambur O, Tikkanen E, Tasmuth T, Sipilä R, Palotie A, Estlander A-M, Leidenius M, Ripatti S, Kalso EA. Pain in 1,000 women treated for breast cancer: A prospective study of pain sensitivity and postoperative pain. ANESTHESIOLOGY. 2013;119:1410–21

23. Beck AT, Ward CH, Mendelson M, Mock J, Erbaugh J. An inventory for measuring depression. Arch Gen Psychiatry. 1961;4:561–71

24. Spielberger CD, Gorsuch RL, Lushane PR, Vagg PR, Jacobs AG Manual for the State-Trait Anxiety Inventory (Form Y). 1983 Palo Alto Consulting Psychologists Press, Inc.

25. Barrett JC, Fry B, Maller J, Daly MJ. Haploview: Analysis and visualization of LD and haplotype maps. Bioinformatics. 2005;21:263–5

26. Jurinke C, van den Boom D, Cantor CR, Köster H. Automated genotyping using the DNA MassArray technology. Methods Mol Biol. 2002;187:179–92

27. Lahermo P, Liljedahl U, Alnaes G, Axelsson T, Brookes AJ, Ellonen P, Groop PH, Halldén C, Holmberg D, Holmberg K, Keinänen M, Kepp K, Kere J, Kiviluoma P, Kristensen V, Lindgren C, Odeberg J, Osterman P, Parkkonen M, Saarela J, Sterner M, Strömqvist L, Talas U, Wessman M, Palotie A, Syvänen AC. A quality assessment survey of SNP genotyping laboratories. Hum Mutat. 2006;27:711–4

28. Purcell S, Neale B, Todd-Brown K, Thomas L, Ferreira MA, Bender D, Maller J, Sklar P, de Bakker PI, Daly MJ, Sham PC. PLINK: A tool set for whole-genome association and population-based linkage analyses. Am J Hum Genet. 2007;81:559–75

29. Stranger BE, Montgomery SB, Dimas AS, Parts L, Stegle O, Ingle CE, Sekowska M, Smith GD, Evans D, Gutierrez-Arcelus M, Price A, Raj T, Nisbett J, Nica AC, Beazley C, Durbin R, Deloukas P, Dermitzakis ET. Patterns of cis regulatory variation in diverse human populations. PLoS Genet. 2012;8:e1002639

30. Nica AC, Parts L, Glass D, Nisbet J, Barrett A, Sekowska M, Travers M, Potter S, Grundberg E, Small K, Hedman AK, Bataille V, Tzenova Bell J, Surdulescu G, Dimas AS, Ingle C, Nestle FO, di Meglio P, Min JL, Wilk A, Hammond CJ, Hassanali N, Yang TP, Montgomery SB, O’Rahilly S, Lindgren CM, Zondervan KT, Soranzo N, Barroso I, Durbin R, Ahmadi K, Deloukas P, McCarthy MI, Dermitzakis ET, Spector TDMuTHER Consortium. . The architecture of gene regulatory variation across multiple human tissues: The MuTHER study. PLoS Genet. 2011;7:e1002003

31. Dimas AS, Deutsch S, Stranger BE, Montgomery SB, Borel C, Attar-Cohen H, Ingle C, Beazley C, Gutierrez Arcelus M, Sekowska M, Gagnebin M, Nisbett J, Deloukas P, Dermitzakis ET, Antonarakis SE. Common regulatory variation impacts gene expression in a cell type-dependent manner. Science. 2009;325:1246–50

32. Schmahl C, Ludäscher P, Greffrath W, Kraus A, Valerius G, Schulze TG, Treutlein J, Rietschel M, Smolka MN, Bohus M. COMT val158met polymorphism and neural pain processing. PLoS One. 2012;7:e23658

33. Kim H, Neubert JK, San Miguel A, Xu K, Krishnaraju RK, Iadarola MJ, Goldman D, Dionne RA. Genetic influence on variability in human acute experimental pain sensitivity associated with gender, ethnicity and psychological temperament. Pain. 2004;109:488–96

34. Stein MB, Fallin MD, Schork NJ, Gelernter J. COMT polymorphisms and anxiety-related personality traits. Neuropsychopharmacology. 2005;30:2092–102

35. Shifman S, Bronstein M, Sternfeld M, Pisanté-Shalom A, Lev-Lehman E, Weizman A, Reznik I, Spivak B, Grisaru N, Karp L, Schiffer R, Kotler M, Strous RD, Swartz-Vanetik M, Knobler HY, Shinar E, Beckmann JS, Yakir B, Risch N, Zak NB, Darvasi A. A highly significant association between a COMT haplotype and schizophrenia. Am J Hum Genet. 2002;71:1296–302

36. Segall SK, Nackley AG, Diatchenko L, Lariviere WR, Lu X, Marron JS, Grabowski-Boase L, Walker JR, Slade G, Gauthier J, Bailey JS, Steffy BM, Maynard TM, Tarantino LM, Wiltshire T. Comt1 genotype and expression predicts anxiety and nociceptive sensitivity in inbred strains of mice. Genes Brain Behav. 2010;9:933–46

37. Lewis BP, Shih IH, Jones-Rhoades MW, Bartel DP, Burge CB. Prediction of mammalian microRNA targets. Cell. 2003;115:787–98

38. Rakvåg TT, Ross JR, Sato H, Skorpen F, Kaasa S, Klepstad P. Genetic variation in the catechol-O-methyltransferase (COMT) gene and morphine requirements in cancer patients with pain. Mol Pain. 2008;4:64

39. Lötsch J, von Hentig N, Freynhagen R, Griessinger N, Zimmermann M, Doehring A, Rohrbacher M, Sittl R, Geisslinger G. Cross-sectional analysis of the influence of currently known pharmacogenetic modulators on opioid therapy in outpatient pain centers. Pharmacogenet Genomics. 2009;19:429–36

40. Gaudet MM, Chanock S, Lissowska J, Berndt SI, Peplonska B, Brinton LA, Welch R, Yeager M, Bardin-Mikolajczak A, Garcia-Closas M. Comprehensive assessment of genetic variation of catechol-O-methyltransferase and breast cancer risk. Cancer Res. 2006;66:9781–5

41. Peterson NB, Trentham-Dietz A, Garcia-Closas M, Newcomb PA, Titus-Ernstoff L, Huang Y, Chanock SJ, Haines JL, Egan KM. Association of COMT haplotypes and breast cancer risk in caucasian women. Anticancer Res. 2010;30:217–20

Supplemental Digital Content

Back to Top | Article Outline

© 2013 American Society of Anesthesiologists, Inc.

Publication of an advertisement in Anesthesiology Online does not constitute endorsement by the American Society of Anesthesiologists, Inc. or Lippincott Williams & Wilkins, Inc. of the product or service being advertised.
Login

Article Tools

Images

Share