A European pharmacogenomic study of the response to opioids in advanced cancer patients identifies germline variants associated with nausea-vomiting side effect

doi:10.21203/rs.3.rs-4174606/v1

Download PDF

Research Article

A European pharmacogenomic study of the response to opioids in advanced cancer patients identifies germline variants associated with nausea-vomiting side effect

https://doi.org/10.21203/rs.3.rs-4174606/v1

This work is licensed under a CC BY 4.0 License

Version 1

posted

You are reading this latest preprint version

Opioids are the mainstay therapy for patients affected by moderate to severe cancer pain, yet in about 10–20% of the cases, patients do not benefit from the received analgesic treatment or experience side effects. Genetic variability might account for the variation in individual responses to opioids, both in terms of efficacy and toxicity. The aim of this genome-wide association study (GWAS) was to identify new genetic markers of opioid toxicity in terms of nausea-vomiting.

Cancer patients receiving morphine, oxycodone, buprenorphine, and fentanyl were recruited from different European cancer centers. Data about toxicity (nausea-vomiting score, NVS) and other relevant clinical information were collected. DNA samples were genotyped using Axiom Precision Medicine Research Arrays. Linear regression between genotypes of 2,059 patients and NVS was performed, using the REGENIE pipeline. Sex, age, study, and country were included in the model as covariates.

We found 68 variants associated with NVS (P-value < 1.0 x 10^− 6). Of note, 15 intronic variants on chromosome 2 were located in the NPAS2 gene, encoding a circadian transcription factor reported to play a role in another opioid side effect, the alteration of sleep. Some of these variants were previously identified as splicing quantitative trait loci of the NPAS2 gene.

This is the first GWAS, performed in more than two thousand individually genotyped patients treated with opioids for cancer pain, that investigated the genetic bases of opioid-induced nausea-vomiting. Although further studies are needed to confirm our findings and to characterize the functional role of the identified variants, our results emphasize the importance of performing large pharmacogenomic studies to identify germline variants associated with opioid response, with the ultimate goals of improving personalization of cancer pain therapies.

Advanced cancer patients experiencing moderate to severe pain are usually treated with opioids according to the third step of the World Health Organization (WHO) analgesic ladder (i.e., morphine, oxycodone, fentanyl, and buprenorphine) for relieving pain ¹. Not only is opioid therapy for cancer-related pain non-effective in some patients, but it also frequently causes side effects, including nausea and vomiting ^2,3. Indeed, nausea is experienced by half of cancer patients receiving opioids (range 3%-85%), whereas vomiting occurs less frequently (4%-50%) ⁴.

We and others previously hypothesized that genetic factors are involved in individual susceptibility to nausea and vomiting side effects. Indeed, a twin study demonstrated that 59% of phenotypic variability in alfentanil-induced nausea was explained by genetics ⁵. Some candidate-gene studies investigated polymorphisms in genes involved in opioid pathways or metabolism and reported associations with genetic variants and the interindividual variability in experiencing nausea and vomiting in cancer patients receiving opioids. For instance, a study on 16 candidate genes in 1,579 European patients reported significant associations among clinical characteristics, variants in the HTR3B, COMT and CHRM3 genes and nausea and vomiting in cancer patients treated with opioids ⁶. Additionally, a Japanese study in 32 cancer patients revealed an association between the low frequency of nausea and the UGT2B7*2 genotype ⁷. However, candidate gene studies often suffer from a lack of validation due to the small sample size or allele frequency discrepancies in patients from different genetic backgrounds.

We previously carried out an exome-wide association study ⁸ to overcome another limitation of the candidate-gene approach, which analyzes only a few genes, and does not take into account the genetic heterogeneity underlying a complex phenotype such as nausea and vomiting related to opioid therapy. However, our study was carried out on DNA pools and only one single nucleotide polymorphism (SNP) associated with nausea/vomiting in the discovery series was confirmed in the validation series, but with the opposite effect on the phenotype.

Herein, we individually genotyped by SNP-array more than two thousand European cancer patients treated with opioids for pain, and we carried out the largest genome-wide association study on individual susceptibility to nausea and vomiting to date.

The whole patient series recruited in the study comprised 2,193 patients. Some samples were excluded from the analysis because they failed quality check steps (Supplementary Fig. 1): in detail, 30 samples failed the Axiom Quality Controls pipeline, 23 samples were removed due to sex inconsistencies, six individuals were discarded due to a low call rate, and four duplicated or related patients were removed. PCA detected 30 individuals with a non-European ancestry who were excluded from the analysis. Finally, 41 patients did not have full NVS data; therefore, the GWAS was performed on 2,059 patients.

The personal and clinical information of the patients is shown in Table 1. Males and females were equally distributed. Approximately one third of patients received morphine, another third was treated with fentanyl, 26% received oxycodone and only 5% took buprenorphine. The most common types of cancer were gastro-enteric (∼ 18%), lung (∼ 18%) or breast (∼ 14%) cancer. More than two thirds of patients did not receive chemotherapy. Approximately 67% of patients were from the EPOS study, while 12% and 21% of subjects belonged to CERP and MOLO, respectively. Additionally, patients were mostly Italian, Norwegian, English and German. The median value for the NVS is 11.11 with an interquartile range of 33.33.

Table 1

Clinical characteristics of patients treated with opioids included in the GWAS for the nausea-vomiting phenotype.
Patient characteristic	NVS GWAS (n = 2059)
Age, years, median (range)	64 (18–91)
Sex, n (%)
Female	1046 (50.8)
Male	1013 (49.2)
Opioid, n (%)
Buprenorphine	103 (5.0)
Fentanyl	698 (33.9)
Morphine	713 (34.6)
Oxycodone	545 (26.5)
Tumor diagnosis, n (%)
Lung	360 (17.7)
Breast	296 (13.2)
Gastro-enteric	370 (18.2)
Pancreas	86 (4.2)
Prostate	209 (10.3)
Urinary traits	142 (7.0)
Head and neck	96 (4.7)
Gynecological	172 (8.5)
Other or unknown	328 (16.1)
Chemotherapy, n (%)
Yes	648 (31.5)
No	1411 (68.5)
Study
EPOS	1392 (67.6)
CERP	216 (10.5)
MOLO	451 (21.9)
Country of enrollment
Switzerland	83 (4.0)
Germany	132 (6.4)
Denmark	12 (0.6)
Finland	28 (1.4)
Great Britain	207 (10.0)
Greece	4 (0.2)
Island	122 (6.0)
Italy	945 (45.9)
Lithuania	48 (2.3)
Norway	383 (18.6)
Sweden	95 (4.6)
Average NVS, median (IQR)	11.11 (33.33)
IQR: interquartile range

We searched for clinical variables significantly associated with the NVS phenotype, by performing a multivariable linear regression with sex, age, administered opioid, chemotherapy, study, genotyping batch, country of enrollment and tumor type. We observed that females had higher NVS than male patients and that NVS slightly decreased with age. In addition, MOLO patients experienced less nausea-vomiting than patients in the CERP group. Finally, lower NVS was measured in patients enrolled in countries other than Italy (Table 2).

Table 2

Multivariable linear regression between clinical variables and NVS phenotype.
Characteristic		Beta	P-value
Sex	Male	1.0
	Female	5.1	8.3 x 10^− 5
Age		-0.10	0.023
Study	CERP	1.0
	MOLO	-4.6	0.035
	EPOS	0.13	0.96
Country of enrollment	Italy	1
	Other	-15	< 2.0 x 10^–16
Opioid	Morphine	1
	Buprenorphine	-3.1	0.26
	Fentanyl	0.84	0.60
	Oxycodone	-0.47	0.77
Tumor site	Lung	1
	Gastro-enteric	-0.30	0.86
	Breast	-0.31	0.88
	Prostate	-0.44	0.84
	Pancreas	-0.66	0.82
	Urinary tracts	0.51	0.83
	Head and neck	1.2	0.67
	Gynecologic	2.2	0.25
	Other	-2.9	0.11
Genotyping batch	I	1
	II	3.11	0.0956
	III	3.24	0.0655
	IV	2.35	0.1367

For each patient, we genotyped 888,799 variants; 799,417 of them passed the Axiom Quality Controls pipeline, 13,767 variants were removed due to missing genotype data, and 946 and 387,355 variants were discarded due to Hardy Weinberg disequilibrium and low allele frequency (MAF < 1%), respectively. Finally, we removed 13,759 non-autosomal variants and 13,765 variants in extended LD. This dataset comprising 369,825 variants was used in REGENIE Step 1. After imputation and filtering by MAF < 2% and imputation score R² < 0.3, the dataset for Step 2 included 7,669,761 polymorphisms.

The GWAS in 2,059 patients was performed with REGENIE software between genotypes and NVS phenotype, using age, sex, study, and country of origin as covariates. The results of this GWAS are shown in the Manhattan plot reported in Fig. 1. Although no associations reached the genome-wide statistical significance threshold (P-value < 5.0 x 10^− 8), 68 variants were associated with NVS with a nominal P-value < 1.0 x 10^− 6 (Supplementary Table 1) The most significant association was observed for rs6562126, which was negatively correlated with NVS (P-value = 1.2 x 10^− 7; beta = -4.17). This is an intronic variant of the long non-coding gene LINC00378, located on chromosome 13, approximately 296 kb downstream of the TDRD3 gene. Among the other top-significantly associated variants, about half of them mapped to three loci, two on chromosome 2 (with four variants at 6 Mb and 15 variants at 100 Mb, respectively), and one on chromosome 20, with 13 variants. Of note, the 15 variants on chromosome 2 mapped to the NPAS2 gene. They all negatively correlated with NVS, meaning that an increasing number of minor alleles of these variants in the patients’ genotypes was associated with less nausea/vomiting. The 13 variants on chromosome 20 mapped in an intergenic region, approximately 80 kb from the ZNF217 gene. All these polymorphisms, instead, had positive beta coefficients, and thus their minor alleles were associated with high NVS.

Focusing on variants mapping in the NPAS2 gene, we found, in the GTEx database, that six of them were NPAS2 splicing quantitative trait loci (sQTLs), in the esophagus mucosa. In particular, an increasing number of minor alleles of these SNPs were reported to be associated with a high level of transcript splicing alteration. Table 3 reports the results from GTEx.

Table 3

Six variants associated with NVS and mapping in the *NPAS2* gene were previously reported as *NPAS2* sQTLs, in GTEx.
Variant ID	Gene Symbol	Phenotype ID *	P-value	NES	Tissue
rs7558747	NPAS2	chr2:100977799:100979559:clu_45742:ENSG00000170485.16	2.0 x 10^− 8	-0.64	Esophagus - Mucosa
rs2289950	NPAS2	chr2:100977799:100979559:clu_45742:ENSG00000170485.16	1.9 x 10^− 6	-0.52	Esophagus - Mucosa
rs4851393	NPAS2	chr2:100977799:100979559:clu_45742:ENSG00000170485.16	2.3 x 10^− 6	-0.51	Esophagus - Mucosa
rs75107839	NPAS2	chr2:100977799:100979559:clu_45742:ENSG00000170485.16	2.3 x 10^− 6	-0.51	Esophagus - Mucosa
rs3768985	NPAS2	chr2:100977799:100979559:clu_45742:ENSG00000170485.16	2.4 x 10^− 6	-0.52	Esophagus - Mucosa
rs17025086	NPAS2	chr2:100977799:100979559:clu_45742:ENSG00000170485.16	2.7 x 10^− 6	-0.51	Esophagus - Mucosa
NES, normalized effect size.
* As reported in GTEx and defined as the intron (chr:start:end) and cluster of connected components (clu_) to which the intron belongs.

In addition, looking at the whole REGENIE output, we searched for variants previously reported for their possible association with nausea-vomiting in opioid-treated cancer patients ^6,8. Variants and their summary statistics, both from REGENIE analysis and the two published studies are shown in Supplementary Table 2. Two variants, rs41269255 and rs12305038, reported in ⁸, showed nominal P-values < 0.05 also in the present study and concordant beta coefficients.

In this pharmacogenomic study, on more than 2,000 advanced cancer patients treated with opioids to relieve pain, we looked for polymorphisms associated with nausea-vomiting side effect, one of these drugs’ most common adverse events. Our GWAS identified 68 variants associated with NVS at a nominal P-value < 1.0 x 10^− 6. The top-significant SNP, rs6562126, is an intronic variant of a long non-coding gene, LINC00378, and is near the Tudor domain containing 3 (TDRD3) gene. Unfortunately, no evidence for a role in nausea-vomiting has been reported thus far in the literature for both the variant and the genes where it maps. To our knowledge, the same is valid for most of the other identified variants.

Interestingly, 15 of the 68 variants mapped to the NPAS2 (Neuronal PAS Domain Protein 2) gene, a transcription factor regulating circadian rhythm genes ⁹. In particular, six of these variants are reported in the GTEx database as sQTLs of the NPAS2 gene in the esophageal mucosa. Still, no evidence is available about any possible role of these SNPs in affecting NPAS2 isoform function. Although any role of NPAS2 or circadian genes in the regulation of opioid-induced nausea-vomiting has not yet been reported, the influence of circadian rhythm on gastrointestinal toxicity due to other cancer therapies (e.g., chemotherapy) is under investigation (as reviewed in ¹⁰). Nonetheless, it has been reported that NPAS2 might affect another common opioid side effect, i.e., the alteration of the sleep-wake cycle. Indeed, it was observed that NPAS2 absence exacerbated the harmful effect of fentanyl on sleep in mice ¹¹. Overall, our finding of an association between germline variants in a circadian gene and nausea-vomiting in cancer patients treated with opioids for cancer pain is very interesting, considering the complex relationship existing between circadian rhythm, pain, and opioids, as reviewed in ¹²: indeed, both endogenous and exogenous opioids modify the circadian rhythm, pain sensitivity is altered by circadian rhythm dysregulation, and disrupted circadian rhythms can affect the efficacy of opioids. Further investigations are needed to understand the functional role of the NPAS2 gene (and its germline variants) in opioid-induced nausea-vomiting and in the feedback loop between opioids and circadian rhythms.

Comparing our findings with the previously identified associations between germline variants and nausea-vomiting in opioid-treated cancer patients, unfortunately, we did not validate any of the SNPs reported by Laugsand et al. ⁶, although more than 80% of the EPOS patients analyzed in that study were included in our dataset. However, in our study, EPOS patients were two thirds of the whole series, and possibly the remaining third (comprising CERP and MOLO patients) might be responsible for the lack of validation. Other reasons might reside in the differences between the analyses, in terms of phenotype considered (two ordinal variables for nausea and vomiting in ⁶ vs. composite NVS, herein), stratification (by country and use of anti-emetics in ⁶ vs. no stratification, in this study), and confounding factors included as covariates in the regression models.

On the other hand, two SNPs (rs41269255 and rs12305038), identified by Colombo et al. ⁸, were confirmed in our GWAS, with a nominal P-value < 0.05. In that study, a first discovery exome-wide study was carried out in the EPOS series (93% overlapping with EPOS patients in this study), and then, significant associations were tested in the CERP series (82% overlapping with CERP patients in this study) for validation. In detail, rs41269255 was not validated in that study in the CERP series, but herein, it had a concordant beta coefficient and P-value < 0.05 in the whole series. The rs12305038 variant was significant in both the discovery and validation series but with discordant beta values. Herein, we found that this SNP was associated with NVS at P-value = 0.025 and a beta coefficient concordant with that of the discovery study. Of note, several methodological differences between the two studies might have reduced the rate of result validation (e.g., DNA pooling strategy and exome sequencing vs. individual genome-wide genotyping by SNP-array; no adjustment for confounding factors in the exome data analysis) notwithstanding the overlap of patient series investigated in the two studies (> 90% of all the patients from the discovery and validation steps of Colombo’s study were included in the genotyped series herein, of which they constituted the 78%). It might be interesting to compare our results with those from other studies reporting associations between germline variants and nausea-vomiting due to different etiologies (i.e., chemotherapy-induced or post-operative nausea-vomiting ^13,14) to explore the possibility that shared genetic factors might control similar phenotypes.

A major limitation of our study is the relatively small sample size, which did not allow for the identification of any locus associated with NVS at the genome-wide significance threshold. Nevertheless, this is the largest GWAS on NVS in individually genotyped, opioid-treated cancer patients performed thus far. Collecting such a wide and relatively homogeneous patient series was not easy and required several recruiting centers’ collaborative efforts. A broader and independent series is needed to confirm and strengthen our results and to better dissect the genetic complexity of opioid toxicity. Overall, our findings support the role of genetics in modulating nausea-vomiting due to opioid treatment. In our opinion, the hypothesis of a possible involvement of a circadian gene deserves further investigation. In conclusion, our study encourages the scientific community involved in opioid research to put their efforts (and patient series) together to increase the statistical power of pharmacogenomic studies for opioid therapy and their clinical applicability.

Patients Series, Data Collection, and Materials

This study included 2,193 European adult patients with locally advanced or metastatic tumors. These patients received step III WHO opioids to treat cancer pain. They were part of three studies: CERP, an Italian multicenter, randomized, and longitudinal phase IV clinical trial ¹⁵; EPOS, a European multicentric and cross-sectional study ¹⁶; and MOLO: an Italian, longitudinal study ¹⁷. The inclusion criteria for the three studies are listed in Supplementary Table 3.

The study protocol was approved by the Committees for Ethics of each recruiting hospital contributing to the EPOS and CERP studies and by the Ethics Committee of the Fondazione IRCCS Istituto Nazionale dei Tumori, Milan (Italy), for the MOLO study (INT 153/13). The research was conducted in accordance with the tenets of the Declaration of Helsinki. Patients signed a written informed consent form to agree to the use of their biological samples and data for the purposes of opioid pharmacogenomic research. Personal and clinical information was collected, such as age, sex, country of origin, cancer diagnosis, chemotherapy treatments at the time of recruitment, opioid taken, and the response to opioids and their quality of life, such as incidence and type of side effects (including nausea and vomiting). Specifically, we investigated one of the most common side effects of opioid treatment, i.e., nausea and vomiting. To this aim, we calculated a composite nausea vomiting score (NVS), as already described in ⁸. Briefly, nausea-vomiting was assessed using patient-reported outcome measures consisting of four-point rating scales that were converted in numerical values from 1 to 4. Then, we used the procedure described in the European Organization for Research and Treatment of Cancer’s Core Quality of Life Questionnaire (EORTC QLQ-C30) Scoring Manual ¹⁸ to calculate the mean of the QLQ-C30 scores for nausea and vomiting, and finally to linearly transform this raw score into the composite NVS, ranging from 0 to 100. For each retrospectively recruited patient from EPOS and CERP, genomic DNA samples were already available at Fondazione IRCCS Istituto Nazionale dei Tumori. Peripheral blood samples from prospectively enrolled patients were collected. Genomic DNA was extracted, using the DNeasy Blood & Tissue kit (Qiagen), and fluorometrically quantified using the Quant-iT PicoGreen dsDNA Assay (Thermo Fisher Scientific).

Multivariable linear regression with clinical variables

To understand which clinical variables were significantly associated with our phenotype of interest (NVS), a multivariable linear regression was performed using the glm() function in the R environment. The clinical variables in the model were sex, age, genotyping batch, the study in which each patient was recruited, the country of enrollment (defined as a binary value, with Italy as the reference category), the administered opioid, tumor diagnosis, and chemotherapy treatment. The clinical variables significantly associated with the NVS phenotype (P-value < 0.05) were then used as covariates in the GWAS.

Genotyping

Genome-wide genotyping was carried out using Axiom Precision Medicine Research Arrays on a GeneTitan multi-channel instrument (Thermo Fisher Scientific). We performed the genotype variant calling using the “Best Practice Workflow” and default quality check (QC) settings (except for the average call rate for passing samples ≥ 97%) with Axiom Analysis Suite v.5.01.38. After removing samples that failed Axiom QC, we extracted the genotypes of our patients and converted them into binary PLINK format. Per-sample and per-marker QC was carried out using the PLINK software v1.921 ¹⁹: in detail, we filtered out samples with discrepancies between the collected sex and that imputed from the genotypes, samples with missing call rate greater than 5%, and duplicated or related individuals. SNPs in high Hardy-Weinberg (HW) disequilibrium (P-value < 1.0 x 10^− 6), with missing genotype rate exceeding 1%, and minor allele frequency (MAF) < 1% were removed. We also filtered out SNPs mapping in regions with extended linkage disequilibrium (LD) ²⁰; finally, we retained only biallelic and autosomal variants.

A principal component analysis (PCA) was performed with PLINK v.2 ²¹ to correct for population stratification. We projected the first four principal components (PCs) of our patients together with those of 1,563 individuals from five different populations, selected from 1000 Genomes Project ²²: Africans, Americans, South-East Asians, East Asians, and Europeans. The scatterplot of the PC1 vs. PC2 and PC3 vs. PC4 was then plotted by grouping and coloring dots at a continental level. Patients who did not cluster with Europeans were removed from the dataset.

Genotype imputation to the whole genome sequence was performed using the TopMED Imputation Server, setting GRCh38/hg38 as build, TopMED as reference Panel, and phasing the data with Eagle v2.4 ^23–26. The imputed genotypes were then filtered to exclude rare variants (MAF < 2%) and SNPs with a low-quality imputation (R² info score ≤ 0.3) ²⁷.

Genome-wide association study

We performed a GWAS using the REGENIE software ²⁸, using default settings of the pipeline (https://rgcgithub.github.io/regenie/options/), that consisted of two different steps. In step 1, ridge regression was performed, using the non-imputed dataset, to define genetic predictors, calculated with the Leave One Chromosome Out method, to be used in the second step. Then, Step 2 of REGENIE analysis was tested on 7,669,761 imputed germline variants. The regression model in the REGENIE pipeline was performed with the following covariates: age, sex, the country of enrollment, and the study in which each patient was recruited (coded as a dummy variable). To graphically visualize the associations with NVS, a Manhattan plot was drawn using the qqman library ²⁹, and the function manhattan() in R environment. Intronic variants of the NPAS2 gene were searched in the GTEx portal (accessed on 08/24/2023) for a possible role as splicing QTL.

Acknowledgments and Funding

The research leading to these results has received funding from AIRC under MFAG 2019 - ID. 22950 - P.I. Colombo Francesca. The funding organization had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript.

Author contributions

Conceptualization, F.C. and F.M; formal analysis, F.M., and F.C.; resources, A.T.C., M.SH., A.P., E.Z., F.S., P.K, S.K., O.C., M.C.M., M.C.P.; investigation – performing experiments, S.N. and F.C.; investigation – data collection, M.SH., C.B., A.P., E.Z., F.S., P.K, S.K., O.C., M.C.M., M.C.P.; data curation, C.B., M.SH., S.N., F.M, and F.C; writing—original draft preparation, F.M. and F.C.; writing—review and editing F.M., M.SH., A.P., P.K., and F.C.; funding acquisition, F.C. All authors have read and agreed to the published version of the manuscript.

Data availability statement

The data that support the findings of this study are available on request from the corresponding author. The data are not publicly available due to privacy or ethical restrictions.

Competing Interests Statement

The authors declare no competing interests.

Caraceni A et al (2012) Use of opioid analgesics in the treatment of cancer pain: evidence-based recommendations from the EAPC. Lancet Oncol 13:e58–68
Teunissen SCCM et al (2007) Symptom prevalence in patients with incurable cancer: a systematic review. J Pain Symptom Manage 34:94–104
Klepstad P et al (2005) Pain and pain treatments in European palliative care units. A cross sectional survey from the European Association for Palliative Care Research Network. Palliat Med 19:477–484
Oosten AW, Oldenmenger WH, Mathijssen RHJ, Van Der Rijt CC (2015) D. A Systematic Review of Prospective Studies Reporting Adverse Events of Commonly Used Opioids for Cancer-Related Pain: A Call for the Use of Standardized Outcome Measures. J Pain 16:935–946
Angst MS et al (2012) Aversive and reinforcing opioid effects: a pharmacogenomic twin study. Anesthesiology 117:22–37
Laugsand EA et al (2011) Clinical and genetic factors associated with nausea and vomiting in cancer patients receiving opioids. Eur J Cancer 47:1682–1691
Fujita KI et al (2010) Association of UGT2B7 and ABCB1 genotypes with morphine-induced adverse drug reactions in Japanese patients with cancer. Cancer Chemother Pharmacol 65:251–258
Colombo F et al (2020) Identification of genetic polymorphisms modulating nausea and vomiting in two series of opioid-treated cancer patients. Sci Rep 10
Peng L, u, Bai G, Pang Y (2021) Roles of NPAS2 in circadian rhythm and disease. Acta Biochim Biophys Sin (Shanghai) 53:1257–1265
Hofmeister EN, Fisher S, Palesh O, Innominato PF (2020) Does circadian rhythm influence gastrointestinal toxicity? Curr Opin Support Palliat Care 14:120–126
Gamble MC et al (2022) A role for the circadian transcription factor NPAS2 in the progressive loss of non-rapid eye movement sleep and increased arousal during fentanyl withdrawal in male mice. Psychopharmacology 239:3185–3200
Bumgarner JR, McCray EW, Nelson RJ (2023) The disruptive relationship among circadian rhythms, pain, and opioids. Front Neurosci 17
Jin Y et al (2021) An Update in Our Understanding of the Relationships Between Gene Polymorphisms and Chemotherapy-Induced Nausea and Vomiting. Int J Gen Med Volume 14:5879–5892
Klenke S, Frey UH (2020) Genetic variability in postoperative nausea and vomiting. Eur J Anaesthesiol 37:959–968
Corli O et al (2016) Are strong opioids equally effective and safe in the treatment of chronic cancer pain? A multicenter randomized phase IV ‘real life’ trial on the variability of response to opioids. Ann Oncol 27:1107–1115
Kurita GP et al (2011) Prevalence and Predictors of Cognitive Dysfunction in Opioid-Treated Patients With Cancer: A Multinational Study. J Clin Oncol 29:1297–1303
Shkodra M et al (2022) Cancer pain: Results of a prospective study on prognostic indicators of pain intensity including pain syndromes assessment. Palliat Med 36:1396–1407
Fayers P et al (2001) EORTC QLQ-C30 Scoring Manual. European Organisation for Research and Treatment of Cancer
Purcell S et al (2007) PLINK: a tool set for whole-genome association and population-based linkage analyses. Am J Hum Genet 81:559–575
Price AL et al (2008) Long-Range LD Can Confound Genome Scans in Admixed Populations. Am J Hum Genet 83:132–135
Chang CC et al (2015) Second-generation PLINK: Rising to the challenge of larger and richer datasets. Gigascience 4:7
Delaneau O et al (2014) Integrating sequence and array data to create an improved 1000 Genomes Project haplotype reference panel. Nat Commun 5:3934
Fuchsberger C, Abecasis GR, Hinds D (2015) A. minimac2: faster genotype imputation. Bioinformatics 31:782–784
Loh PR et al (2016) Reference-based phasing using the Haplotype Reference Consortium panel. Nature Genetics 2016 48:11 48, 1443–1448
Das S et al (2016) Next-generation genotype imputation service and methods. Nature Genetics 2016 48:10 48, 1284–1287
Taliun D et al (2021) Sequencing of 53,831 diverse genomes from the NHLBI TOPMed Program. Nature 2021 590:7845 590, 290–299
Verlouw JAM et al (2021) A comparison of genotyping arrays. European Journal of Human Genetics 2021 29:11 29, 1611–1624
Mbatchou J et al (2021) Computationally efficient whole-genome regression for quantitative and binary traits. Nat Genet 53:1097–1103
Turner SD (2014) qqman: an R package for visualizing GWAS results using Q-Q and manhattan plots. bioRxiv 005165. 10.1101/005165

Supplementary Tables 2 and 3 are not available with this version

The authors declare no competing interests.

Download PDF

Version 1

posted

You are reading this latest preprint version

A European pharmacogenomic study of the response to opioids in advanced cancer patients identifies germline variants associated with nausea-vomiting side effect

Status:

Version 1

Abstract

Figures

Introduction

Results

Discussion

Materials and Methods

Patients Series, Data Collection, and Materials

Multivariable linear regression with clinical variables

Genotyping

Genome-wide association study

Declarations

References

Supplementary Tables

Additional Declarations

Supplementary Files

Status:

Version 1