Skip to main content
NCBI home page
Wiley Open Access Collection logoLink to Wiley Open Access Collection
. 2026 Aug 2;50(8):e70385. doi: 10.1111/acer.70385

The OPRM1 A118G G Allele Is Associated With Poorer Auditory‐Verbal Learning and Memory, Executive Functions, and Working Memory in Veterans Undergoing Residential Treatment for Alcohol Use Disorder

Roberto Andrade 1, M Windy McNerney 1,2, Eric P Kraybill 1, Claudia B Padula 1,2, Timothy C Durazzo 2,3,✉
PMCID: PMC13429951  PMID: 42543683

ABSTRACT

Background

The neurocognitive dysfunction observed in Alcohol Use Disorder (AUD) is related to a complex interplay between genetic, neurobiological, and environmental factors. The μ‐opioid receptor (MOR), encoded by the OPRM1 gene, modulates reward processing, stress responsivity, and excitatory–inhibitory activity within prefrontal–limbic and hippocampal circuits implicated in neurocognition. The common A118G (Asn40Asp, rs1799971) polymorphism alters MOR expression and signaling, but its contribution to domain‐specific cognitive functioning in AUD, particularly among Veterans, has received limited investigation.

Methods

Veterans (n = 127) in residential treatment for AUD completed a neurocognitive battery assessing the following domains of functioning: auditory‐verbal and visuospatial learning and memory, executive function, and working memory. Participants were genotyped for OPRM1 A118G, and A homozygotes and G‐allele carriers (AG/GG) were compared on the above measures, using bootstrapped generalized linear models, adjusted for age, education, biological sex, race, pre‐study alcohol consumption, and brain‐derived neurotrophic factor (BDNF) (rs6265) and catechol‐O‐methyltransferase (COMT) (rs4680) genotypes, which have been previously shown to be associated with neurocognition.

Results

G‐allele carriers demonstrated significantly worse performance than A homozygotes on measures of auditory‐verbal learning and memory, cognitive flexibility/set shifting, auditory working memory, and on a composite average of all measures. BDNF or COMT genotypes were not associated with any measure, and there were no significant interactions among OPRM1, BDNF, and COMT genotypes.

Conclusions

Findings suggest that altered MOR signaling in G‐allele carriers may be associated with differences in prefrontal network efficiency, potentially justifying future studies investigating the relationship between OPRM1 and glutamatergic plasticity. The results highlight a potential genetically influenced contributor to the neurocognitive heterogeneity in AUD and underscore the possible value of integrating MOR‐related biological markers into neurocognitive assessment of individuals with AUD.

OPRM1 G allele carriers (AG/GG; G‐carriers), compared to A allele homozygotes (AA), performed significantly worse on measures of auditory‐verbal learning and memory, executive skills, and working memory. G allele carriers also performed worse on a Test Composite (Z‐scores) that represented the average performance for each group across all tests.

graphic file with name ACER-50-0-g001.jpg

1. Introduction

Multiple premorbid and/or comorbid factors contribute to the heterogeneous pattern and magnitude of neurocognitive abnormalities experienced by those with alcohol use disorder (AUD) during the early phase of recovery (Durazzo et al. 2012; Stavro et al. 2013). The dysfunction observed in executive functions, learning and memory, and working memory is clinically relevant as it is linked to treatment engagement and risk of return to hazardous drinking (Bates et al. 2013; Tapert et al. 2004). Identification of the factors associated with neurocognitive dysfunction in AUD is necessary to better understand the mechanisms influencing recovery and development of more effective treatments.

Neurocognitive abnormalities in those with AUD are linked to frontostriatal and frontolimbic circuit dysfunction (Sullivan and Pfefferbaum 2005; Padula et al. 2023). At the molecular level, single nucleotide polymorphisms (SNPs) in brain‐derived neurotrophic factor (BDNF; Val66Met) and catechol‐O‐methyltransferase (COMT; Val158Met) are reported to regulate synaptic plasticity and cortical catecholamine activity, respectively, with downstream implications for aspects of learning, memory, and executive function (see Durazzo et al. 2012; Durazzo et al. 2023 and references therein). Our group has examined the association of these SNPs with neurocognition in veterans with AUD. We observed that COMT Met homozygotes were superior to Val homozygotes on measures of executive skills and showed trends for higher general intelligence and visuospatial skills, while COMT Val/Met heterozygotes showed significantly better general intelligence than Val homozygotes. COMT Val homozygotes performed better than heterozygotes on auditory‐verbal memory. BDNF genotype was not related to any neurocognitive domain (Durazzo et al. 2012). However, genes more proximal to the reinforcing properties of alcohol, particularly those regulating μ‐opioid receptor (MOR) signaling, have received comparatively limited attention for their potential domain‐specific neurocognitive associations.

The MOR, encoded by OPRM1 (rs1799971), is a key regulator of alcohol‐related reward, stress responsivity, and cue reactivity through opioid–dopamine interactions within mesocorticolimbic and prefrontal–hippocampal circuitry (Zhou et al. 2020; Ray et al. 2013). The common A118G (Asn40Asp) SNP alters receptor expression, binding affinity, and downstream signaling, with implications for reinforcement sensitivity and response to medications that involve MORs (Zhang et al. 2025). MOR signaling affects dopaminergic activity in the anterior frontal cortex primarily through modulation of GABAergic interneurons and mesocortical dopaminergic projections, which can indirectly impact executive function, learning, and working memory (Johnson and North 1992). Alterations of this opioid–dopamine coupling may adversely affect the aforementioned neurocognitive domains in those with AUD. Additionally, MORs are expressed at high density on GABAergic interneurons in the orbitofrontal cortex, anterior cingulate cortex (ACC), and hippocampus (Fields and Margolis 2015), which are regions that are well‐established to subserve aspects of executive functions, learning, and memory (Rolls and Grabenhorst 2008; Durazzo and Meyerhoff 2021; Eichenbaum 2004). Beyond GABAergic modulation, MORs indirectly impact glutamatergic and dopaminergic transmission, all of which influence long‐term potentiation (LTP) and synaptic efficiency (Fields and Margolis 2015). Dysregulation of these systems could compromise anterior frontal–hippocampal circuit functions, thereby disrupting aspects of executive functions, learning and memory, and working memory. The OPRM1 G allele (Asn40Asp) is also associated with altered protein conformation and opioid binding potential, which diminishes μ‐opioid modulation of inhibitory interneurons and alters excitatory–inhibitory balance (Zhang et al. 2025). This shift may disrupt the gating of glutamate‐and‐GABA‐mediated signaling across anterior frontal and hippocampal networks, impairing long‐term potentiation and the signal‐to‐noise ratio of task‐relevant neural activity during execution of measures assessing learning, memory, and executive functions. This is especially pertinent in AUD, where chronic, hazardous consumption levels can alter GABAergic, dopaminergic, and opioid activity in multiple functional circuits (Koob and Volkow 2016). A better understanding of the association of the OPRM1 gene with neurocognition in those with AUD could identify additional mechanisms related to neurocognitive dysfunction and recovery. Prior studies in AUD have primarily focused on the association of the OPRM1 SNP with clinical variables such as alcohol craving, subjective rewarding properties, and treatment response. Therefore, the relationship between OPRM1 A118G and neurocognitive performance in those with AUD is unknown.

The goal of this study was to investigate the association of OPRM1 A118G genotype with neurocognition in Veterans in treatment for AUD. We predicted that G‐allele carriers (GG/AG), after adjusting for BDNF (Val66Met) and COMT (Val158Met) genotypes, and salient demographics, exhibit poorer performance than A homozygotes on measures of auditory‐verbal and visuospatial learning and memory, executive functions, and working memory.

2. Methods

2.1. Participants

Veterans with AUD (n = 127) were recruited from the VA Palo Alto Health Care System (VAPAHCS) and were in residential treatment at the time of study. Treatment program duration ranged from 28 to 90 days. Participants in this study were from research examining neurobiological predictors of resumption of alcohol consumption following AUD treatment, and clinical trials investigating the efficacy of repetitive transcranial magnetic stimulation (rTMS) as an adjunct treatment for AUD (see Padula et al. 2024, https://www.clinicaltrials.gov/search?id=NCT03291431; Durazzo et al. 2025, NCT03191266, https://www.clinicaltrials.gov/search?id=NCT03191266). Participants provided written informed consent prior to the initiation of study procedures. All study procedures were approved by Stanford University and VAPAHCS, institutional review boards, and observed the ethical standards of the Declaration of Helsinki. All neurocognitive data from participants in the rTMS clinical trials were obtained before initiation of the rTMS intervention. See Table 1 for participant demographic and clinical characteristics.

TABLE 1.

Demographic and clinical variables.

Measure OPRM1 A homozygotes (n = 93) OPRM1 G‐carriers (n = 34) Group comparisons (p‐value)
Age 48.6 (13.5) 49.6 (16.8) 0.75
Education (years) 14.2 (2.1) 13.5 (1.6) 0.09
Male (%) 79 (85) 30 (88) 0.78
Race (%) a All > 0.20
Asian 2 (2) 0 (0)
Black 12 (13) 2 (6)
Native American 4 (4) 3 (9)
Native Hawaiian/Pacific Islander 2 (2) 0 (0)
White 70 (75) 26 (77)
More than one race 3 (3) 3 (9)
Ethnicity (%)
Latino 23 (25) 10 (29)
Days abstinent from alcohol at study entry 39 (54) 40 (44) 0.88
Lifetime major depressive disorder (%) 52 (56) 22 (65) 0.42
PTSD, past month (%) 51 (55) 14 (42) 0.23
Stimulant use disorder, past month (%) 13 (14) 2 (6) 0.21
Cannabis use disorder, past month (%) 16 (17) 5 (15) 0.74
Panic disorder, past month (%) 11 (12) 2 (6) 0.51
Obsessive‐compulsive disorder, past month (%) 5 (5) 1 (3) 0.99
Beck Depression Inventory‐II 19 (11) 18 (12) 0.85
Beck Anxiety Inventory 15 (12) 13 (10) 0.30
PTSD checklist for DSM‐5 57 (2) 54 (3) 0.52
Smoking status (%) All > 0.21
Never 25 (26) 8 (24)
Former 34 (37) 8 (24)
Current 34 (37) 18 (52)
Number alcohol drinking days 3 months prior to study 49 (28) 44 (32) 0.35
Number alcohol‐containing drinks 3 months prior to study 772 (669) 534 (520) 0.06
Alcohol‐containing drinks per drinking day 3 months prior to study 15 (11) 10 (8) 0.03
Number of DSM‐5 alcohol use disorder criteria met (median)

10

Min = 1

Max = 11

10

Min = 4

Max = 11

 0.71
Alcohol use disorder identification test (median)

29

Min = 3

Max = 40

29

Min = 7

Max = 40

 0.67
Cannabis use disorder identification test

9

Min = 4

Max = 28

9

Min = 4

Max = 22

 0.66
Open in a new tab

Note: Mean (standard deviation), unless otherwise specified.

Abbreviation: PTSD, post‐traumatic stress disorder.

a

A separate category was formed for individuals endorsing more than one race.

Participants met Diagnostic and Statistical Manual of Mental Disorders‐5 (DSM‐5) criteria for AUD and 95% of the sample were in the severe range of AUD symptomatology. Principal inclusion criteria were adults aged 18 and older, fluent and fully literate in English, and in residential treatment. Exclusion criteria were: (1) active suicidal ideation representing imminent risk for suicide. (2) biomedical diseases/conditions known to adversely affect brain neurobiology or neurocognition (e.g., space‐occupying cerebral lesion(s), cerebrovascular accident, multiple sclerosis, Parkinson's disease). (3) traumatic brain injury resulting in a loss of consciousness > 10 min. (4) clinically documented severe impairment of visual and/or auditory acuity or motor skills. (5) history of schizophrenia spectrum, bipolar, or other psychotic disorders. The following comorbidities were permitted due to their high prevalence in alcohol use disorders, particularly in Veterans: hepatitis C, type‐2 diabetes, hypertension, unipolar mood disorders (major depression, substance‐induced mood disorder), anxiety disorders (generalized anxiety disorder, panic disorder) and post‐traumatic stress disorder (PTSD) (Nguyen et al. 2020; Durazzo et al. 2020; Gilman and Abraham 2001; Stinson et al. 2005). Participants who met DSM‐5 criteria for current substance use disorder were included, given the high prevalence of comorbid substance abuse in AUD (Stinson et al. 2005; Mannes et al. 2021; Dawson et al. 2005); the most common substance use disorders were stimulant and cannabis use disorders. Participants were urine‐tested for illicit substances and breathalyzed for recent alcohol consumption before assessment. No participant tested positive for illicit or non‐prescribed substances or had a detectable blood alcohol level at the time of assessment.

2.2. Medical, Psychiatric, Substance, and Alcohol Consumption History

Medical history for participants was obtained via self‐report and confirmed via available electronic medical records. Psychiatric symptomatology and diagnoses were assessed by the Mini‐International Neuropsychiatric Interview for DSM‐5 (MINI). Participants also completed the Clinical Interview for DSM‐5 Alcohol Use Disorder and in‐house questionnaires assessing demographics, history of biomedical and other illicit and prescribed substance use. The Timeline Follow‐back (TLFB) was used to obtain alcohol consumption (total number of drinking days, average drinks/day, total number of drinks) over the 3 months prior to study participation. PTSD symptomatology was evaluated with the PTSD Checklist for DSM‐5 (PCL‐5). Major depressive symptomatology was assessed with the Beck Depression Inventory‐II (BDI‐II), and anxiety symptomatology with the Beck Anxiety Inventory (BAI). Nicotine dependence level was evaluated via the Fagerstrom Test for Nicotine Dependence (FTND). See Nguyen et al. (2020) for citations for the above measures.

2.3. Neurocognitive Measures

The domains evaluated and the constituent measures are as follows: Auditory‐verbal and visuospatial learning and memory: [California Verbal Learning Test‐II (CVLT‐II, n = 127), Brief Visuospatial Memory Test‐Revised (BVMT‐R, n = 124)]. CVLT‐II measures used were Total Recall from trials 1–5 (CVLT‐II Total Recall), and Short and Long Delay Free and Cued Recall. For the BVMT‐R, total recall from trials 1–3 (BVMT‐R Total Recall) and Delayed Free Recall were used. Executive functions: Delis‐Kaplan Executive Function System, Color‐Word Interference Test [(DKEFS CWI, n = 122), Trail Making Test (TMT, n = 121), Wisconsin Card Sorting Test‐64 (WCST‐64, n = 124)]. Measures used from the DKEFS CWI were Interference and Interference/Switching and TMT B were used from the TMT. Perseverative responses and non‐perseverative errors were used from the WCST‐64; however, these variables and other WCST‐64 dependent measures were highly skewed, and no transformation method produced distributions that were acceptably symmetrical. Consequently, the WCST‐64 perseverative responses and non‐perseverative errors were excluded from analyses (see Data S1 for details). Working memory: [WAIS‐IV Digit Span (DS, n = 124)]. DS Total Score was used. CWI Interference and Interference/Switching and TMT B were square root transformed due to their skewed distributions and showed acceptably symmetrical distributions following transformation. See Padula and Durazzo (2022) for citations for the above measures. A Test Composite measure was formed by converting all measures to Z‐scores and calculating the arithmetic average of the 11 tests.

2.4. Genotyping

Genomic DNA was extracted from saliva using the Oragene collection kit (DNAGenotek). OPRM1 rs1799971 (A118G; Asn40Asp) was genotyped using a modified protocol from Troisi et al. (2011). After DNA extraction, samples were checked for quantity and quality via nanodrop and reisolated with low yield or poor quality in a similar manner to our previous reports (Durazzo, McNerney, et al. 2023; Padula et al. 2021). The PCR reaction was carried out in master mix (Qiagen) containing forward and reverse primers at a volume of 25 μL. PCR was initiated at 95°C for 5 min and then 30 cycles at: 94°C for 30 s, 63°C for 30 s, and 72°C for 30 s. Final extension was held at 72°C for 10 min and the product was digested in DPN II (New England Biolabs) at 37°C for 3 h and subsequently visualized on an 8% acrylamide gel. Due to the low frequency of OPRM1 G allele homozygotes (n = 4), AG genotype (n = 30), and G homozygotes were grouped as G‐carriers (n = 34) and were compared to A homozygotes for primary analyses. The OPRM1 allelic frequency was within Hardy–Weinberg equilibrium (χ 2 = 0.21, p = 0.89). Both BDNF Val66Met (rs6265) (χ 2 = 0.63, p = 0.43) and COMT Val158Met (rs4680) (χ 2 = 3.32, p = 0.07) were within Hardy–Weinberg equilibrium. For additional information on BDNF genotyping (see Durazzo, McNerney, et al. 2023; Padula et al. 2021) for COMT genotyping.

2.5. Data Analyses

Comparisons of OPRM1 genotype on clinical and demographic variables were conducted with univariate analysis of variance, chi‐square, Fisher's Exact test, or Mann–Whitney test, where appropriate (see Table 1). Group differences on these variables were considered statistically significant at p < 0.05. All statistical analyses were completed with SPSS v30 and R (4.5.3).

We tested the a priori hypothesis that OPRM1 G‐carriers demonstrate poorer performance on the cognitive measures acquired in this study. A two‐step approach was employed for comparisons of OPRM1 genotypes (A homozygotes vs. G‐carriers) on neurocognitive measures. First, we used the glmnet package (4.1‐10) elastic net procedure in R (4.5.3) to determine the salient predictors for the 11 individual neurocognitive tests and the Test Composite. The following 15 predictors were included in the elastic net models for each neurocognitive measure, given their potential to serve as suppressor or confounder variables of the associations of OPRM1 genotype with the dependent measures acquired in this study: OPRM1 genotype (G‐carriers, A homozygotes), BDNF genotype (Met‐carriers, Val homozygotes), COMT genotype (Val homozygotes, Val/Met, Met homozygotes), current anxiety disorder diagnosis (anxiety disorder; yes, no), lifetime history of Major Depressive Disorder diagnosis (MDD; no history, past, current), current PTSD diagnosis (PTSD; yes, no), current cannabis use disorder diagnosis (CUD; yes, no), current stimulant use disorder (SUD; yes, no), race (see Table 1 of main manuscript for race categories), biological sex (male, female), smoking status (non‐smoker, former smoker, current smoker), age, years of education, number of days abstinent from alcohol at time of study (days abstinent), average drinks/drinking day 3 months prior to study (average drinks). Although OPRM1, BDNF, and COMT genotypes were equivalent on all of the above measures (except average drinks), we included anxiety disorders, MDD diagnosis (Marazziti et al. 2010; Pan et al. 2019), PTSD (Scott et al. 2015), and smoking status (Padula and Durazzo 2022) because of their associations with the neurocognitive measures obtained in this study (see Data S1 for detailed description and results from the elastic net models). Second, variables selected as predictors from the elastic net models for the individual neurocognitive measures and Test Composite were entered as predictors in SPSS generalized linear model with bootstrapping (GENLIN; distribution = normal; link = identity; 1000 bootstrapped samples). Although BDNF and COMT genotypes were not selected as statistically significant predictors of any neurocognitive test or Test Composite in elastic net models (see supplemental statistical analyses in DataS1 for details), they were still included as predictors in GENLIN models to test for interactions among OPRM1, COMT, and BDNF genotypes. Additionally, irrespective of selection or non‐selection in the elastic net models, biological sex, age, education were included as covariates in GENLIN models, given their well‐established parametric associations with the cognitive domains assessed in this study (Heaton et al. 2004). Finally, average drinks was considered as a covariate because of the significant differences on this measure between OPRM1 genotypes. p < 0.05 was considered statistically significant for effects for all variables and interactions among OPRM1, COMT, and BDNF genotypes. Significant effects for OPRM1, COMT, and BDNF genotypes were followed with two‐tailed t‐tests. To control for multiplicity of pairwise t‐tests, we employed the Dubey and Armitage‐Parmar (D/AP) procedure [see (Sankoh et al. 1997)], which adjusted t‐test p‐values for the number of neurocognitive tests (11) and the average Spearman intercorrelation among the 11 neurocognitive measures (r = 0.51) across participants; this procedure resulted in an adjusted alpha level of p < 0.016 for pairwise two‐tailed t‐tests for all genotypes. The D/AP procedure was used instead of the standard Bonferroni correction for multiple comparisons because the standard Bonferroni correction (and Bonferroni‐variants) assumes orthogonality among dependent measures, which was not apparent with the neurocognitive test acquired in this study. P < 0.016 was considered statistically significant for genotype two‐tailed pairwise comparisons for the Test Composite. Cohen's d effect sizes (ES) were calculated for mean comparisons of all genotypes and interpreted as follows: small d = 0.20–0.49; moderate d = 0.50–0.79; and large d ≥ 0.80 (Cohen 1988).

3. Results

3.1. Participant Demographics and Clinical Variables

See Table 1 for group comparisons on demographic, clinical, and alcohol consumption variables. OPRM1 A homozygotes had significantly greater average drinks per drinking day 3 months prior to study than OPRM1 G‐carriers. No other differences between OPRM1 genotypes were observed on demographic or clinical variables.

3.2. Predictor Selection for Each Neurocognitive Measure From Elastic Net Models

As can be seen in Table 2, BDNF, COMT, MDD, PTSD, CUD, SUD, smoking status, race, days abstinent prior to study, and average drinks per drinking day 3 months prior to study were not selected as salient predictors of any individual neurocognitive measure or Test Composite.

TABLE 2.

Variables selected as predictors of neurocognitive tests from elastic net models.

Measure Variables selected
BVMT‐R total recall Age
BVMT‐R delayed recall Age
CVLT‐II total recall OPRM1, sex, age
CVLT‐II short delay free recall OPRM1, sex, age
CVLT‐II short delay cued recall Sex, age
CVLT‐II long delay free recall OPRM1, sex, age
CVLT‐II long delay cued recall Sex, age
DKEFS CWI interference OPRM1, age
DKEFS CWI interference/switching Age
Trail making test B OPRM1, age
WAIS‐IV digit span total OPRM1, age, education
Test composite (Z‐score) OPRM1, sex, age
Open in a new tab

Abbreviations: BVMT‐R, Brief Visuospatial Memory Test‐Revised; CVLT‐II, California Verbal Learning Test‐II; DKEFS CWI, Delis‐Kaplan Executive Function System, Color‐Word Interference; WAIS‐IV, Wechsler Adult Intelligence Scale‐Fourth Edition.

3.3. Comparison of OPRM1 Genotypes on Neurocognitive Tests With GENLIN

Significant effects for OPRM1 genotype were observed for the following CVLT‐II measures: Total recall [B = 5.51, standard error of the estimate (SE) = 1.94, 95% confidence interval (CI) = 1.54 to 8.92, p = 0.009], short delay free recall (B = 1.56, SE = 0.61, CI = 0.24 to 2.72, p = 0.015) and long delay free (B = 1.58, SE = 0.57, CI = 0.40 to 2.61, p = 0.007). Significant effects for OPRM1 genotype were observed for TMT B (B = −1.18, SE = 0.37, CI = −1.91 to −0.45, p = 0.003), DS total (B = 2.74, SE = 0.86, CI = 0.89 to 4.25, p = 0.003), and test composite (B = 0.72, SE = 0.18, CI = 0.38 to 1.08, p < 0.001). For all measures with significant effects for OPRM1 genotype, G‐carriers were inferior to A homozygotes. No significant differences were observed between OPRM1 G‐carriers and A homozygotes on BVMT‐R total recall (B = 1.82, SE = 1.38, CI = 0.71 to 4.37, p = 0.16), BVMT‐R delayed free recall (B = 0.67, SE = 0.47, CI = 0.23 to 1.63, p = 0.16), CVLT‐II short delay cued recall (B = 1.42, SE = 0.62, CI = 0.26 to 2.76, p = 0.026) and CVLT‐II long delayed cued recall (B = 1.40, SE = 0.61, CI = 0.22 to 2.65, p = 0.029), CWI Interference (B = −0.29, SE = 0.13, CI = −0.53 to −0.29, p = 0.017), and CWI Interference/Switching (B = −0.24, SE = 0.15, CI = −0.54 to −0.05, p = 0.12). See Figure 1A–L for OPRM1 genotype pairwise comparisons and effect sizes for all measures. Greater age was related to poorer performance on all individual measures and the Test Composite (all p < 0.001). Higher education was associated with better performance on CWI interference, DS total and TMT B (all p < 0.030). Females were superior to males on all CVLT‐II measures (all p < 0.01; see Table S3). Higher average drinks per drinking day over 3 months prior to study were related to worse performance on TMT B (p = 0.043) and Test Composite (p = 0.026); no other significant associations between alcohol consumption variables and neurocognitive measures were observed. BDNF (all p > 0.16) and COMT (all p > 0.023) genotypes were not significant predictors of any individual neurocognitive test or the Test Composite and there were no significant interactions among the three genotypes (all p > 0.15). See Tables S1 and S2 for BDNF and COMT genotype pairwise comparisons across neurocognitive measures.

FIGURE 1.

FIGURE 1

Open in a new tab

For all group comparisons, p < 0.016 was considered statistically significant. Cohen's d effect size is provided for mean group differences. Error bars are standard error of the mean. (A) Brief Visuospatial Memory Test‐Revised (BVMT‐R) Total Recall Raw Score comparison of OPRM1 A homozygotes (AA) and G allele carriers (G‐carriers). (B) Brief Visuospatial Memory Test‐Revised (BVMT‐R) Delayed Recall Raw Score comparison of OPRM1 A homozygotes (AA) and G allele carriers (G‐carriers). (C) California Verbal Learning Test‐II (CVLT‐II) Total Recall (sum of trials 1–5) Raw Score comparison of OPRM1 A homozygotes (AA) and G allele carriers (G‐carriers). (D) California Verbal Learning Test‐II (CVLT‐II) Short Delay Free Recall Raw Score comparison of OPRM1 A homozygotes (AA) and G allele carriers (G‐carriers). (E) California Verbal Learning Test‐II (CVLT‐II) Short Delay Cued Recall Raw Score comparison of OPRM1 A homozygotes (AA) and G allele carriers (G‐carriers). (F) California Verbal Learning Test‐II (CVLT‐II) Long Delay Free Recall Raw Score comparison of OPRM1 A homozygotes (AA) and G allele carriers (G‐carriers). (G) California Verbal Learning Test‐II (CVLT‐II) Long Delay Cued Recall Raw Score comparison of OPRM1 A homozygotes (AA) and G allele carriers (G‐carriers). (H) Wechsler Adult Intelligence Scale‐Fourth Edition, Digit Span Total Raw Score comparison of OPRM1 A homozygotes (AA) and G allele carriers (G‐carriers). (I) Delis‐Kaplan Executive Function System, Color‐Word Interference Test, Interference Raw Score (square root transformed) comparison of OPRM1 A homozygotes (AA) and G allele carriers (G‐carriers). Higher score indicates poorer performance. (J) Delis‐Kaplan Executive Function System, Color‐Word Interference Test, Interference/Switching Raw Score (square root transformed) comparison of OPRM1 A homozygotes (AA) and G allele carriers (G‐carriers). Higher score indicates poorer performance. (K) Trail Making Test B Raw Score (square root transformed) comparison of OPRM1 A homozygotes (AA) and G allele carriers (G‐carriers). Higher score indicates poorer performance. (L) Test Composite Score comparison of OPRM1 A homozygotes (AA) and G allele carriers (G‐carriers).

4. Discussion

Findings from the current study indicated OPRM1 G allele carriers (AG/GG) were inferior to A homozygotes on measures of auditory‐verbal learning and memory free recall, executive functions, and auditory working memory, as well as on a composite index of all measures administered. The poorer performance of OPRM1 G allele carriers was not moderated by BDNF or COMT genotypes, and BDNF or COMT genotypes were not significant predictors of any individual neurocognitive measure or the Test Composite. Psychiatric and substance use disorders, and cigarette smoking status were not significant predictors of the neurocognitive measures obtained in this study.

Prior OPRM1 research in AUD has predominantly focused on allelic associations with craving, reward sensitivity, and treatment response, rather than neurocognition (Fields and Margolis 2015). The association between G‐allele carriers and poorer neurocognitive performance in this study aligns with prior evidence from non‐AUD samples linking this variant to reduced μ‐opioid receptor binding and atypical neural activation across prefrontal and limbic circuits (Zhou et al. 2020). These findings are consistent with emerging research implicating the MOR system in cognitive control processes relevant to neurocognitive functioning. Finally, our results parallel studies in non‐AUD populations showing that OPRM1 G‐allele carriers demonstrate worse working memory than A homozygotes (Hajj et al. 2022). This consistency suggests that OPRM1 variation may contribute to individual differences in neurocognitive functioning observed in those with AUD. Collectively, these results suggest that the influence of the OPRM1 SNP may extend beyond reward processes to affect aspects of learning and memory, executive functions and working memory.

A substantial body of literature demonstrates that the OPRM1 A118G substitution is related to stress reactivity, reward sensitivity, and increased vulnerability to developing substance use disorders (Lovallo et al. 2015). The G allele has been associated with altered protein conformation and altered glycosylation, resulting in diminished receptor signaling. Neuroimaging studies showed that G‐allele carriers display atypical activation within anterior frontal and striatal regions during reward anticipation and cognitive control tasks (Zhou et al. 2020). However, rather than attributing these findings solely to dopaminergic mechanisms, growing evidence points to the broader role of MORs as a neuromodulator of both inhibitory and downstream excitatory transmission (Fields and Margolis 2015). For example, MORs modulate GABAergic inhibition and glutamatergic synaptic plasticity, including long‐term potentiation within hippocampal and prefrontal circuits involved in learning and memory (Schmidt‐Wilcke et al. 2018; Shi et al. 2020; Chartoff and Connery 2014; Drake et al. 2007), which may be associated with differences in synaptic plasticity and functional connectivity. In people with AUD, such effects may be further amplified by alcohol‐related alterations to regional GABA and glutamate concentrations, and macrostructural and microstructural abnormalities (Meyerhoff et al. 2013; Buhler and Mann 2011; Demirakca et al. 2011). These proposed neurobiological mechanisms must be considered speculative in the context of the cross‐sectional design of the present study, as functional neuroimaging, receptor binding, and molecular measures were not obtained, but suggest future assessments of these pathways/circuits are warranted.

We observed that OPRM1 A homozygotes consumed a higher number of alcohol‐containing drinks per drinking days than G‐carriers, which is consistent with prior findings in non‐treatment‐seeking “problem drinkers” (Ray et al. 2013). However, the neurocognitive differences between OPRM1 G‐carriers and A homozygotes observed remained significant on those measures where average drinks per drinking day was a statistically significant predictor. Of note, females were superior to males on all CVLT‐II measures (the number of females across OPRM1 genotypes was equivalent). The few cross‐sectional studies that compared males and females with AUD reported females performed more poorly than males on measures of visual working memory, spatial planning and problem solving, and cognitive flexibility (Acker 1986; Fama et al. 2020; Flannery et al. 2007). The significantly better performance of females than males across measures of auditory‐verbal learning and memory must be considered preliminary, despite the overall moderate effect sizes and p‐values that survived correction for multiple comparisons, given the small number of females (n = 18) in this study.

This study has limitations that may affect the generalizability of the findings. The study sample was composed primarily of males; therefore, the interactions of biological sex with the SNPs genotyped in this sample could not be meaningfully interrogated. While we did specifically include race in our model and found no association, we emphasize the importance of race and ancestry in genomic SNP analysis. It is well‐documented that allele frequencies and functional endpoints vary across ancestry groups, so future studies with sufficiently large samples should investigate this association. We did not obtain a normal control reference group to determine if the findings for OPRM1 genotypes were apparent in a healthy, non‐AUD sample. While there is some potential neurotransmission convergence between COMT, BDNF, and OPRM1, these genes are located on different chromosomes and show overlapping enrichment pathways specific to G‐coupled protein activation (see snpXplorer). Comorbid factors (e.g., diet/nutrition, exercise, and subclinical hepatic, cardiac, or cerebrovascular dysfunction) that were not assessed in this study may have influenced the reported findings. Finally, the exclusively Veteran sample, though demographically similar to many studies with non‐Veterans, limits generalizability.

In conclusion, the inferior performance of OPRM1 G allele carriers on measures of auditory‐verbal learning and memory, executive skills, and working memory is novel in those with AUD. Results support studies indicating that the pattern and magnitude of neurocognitive abnormalities demonstrated by those with AUD are not solely attributable to hazardous alcohol consumption (Staudt et al. 2023; Durazzo et al. 2012). Inclusion of a larger sample of females is necessary to determine if the associations of the OPRM1 G allele with neurocognition are mediated or moderated by biological sex. Overall, these findings contribute to the growing recognition that the MOR system may contribute to neurocognitive variability, not only in reward and affective processes but also in aspects of executive functions, learning and memory, and working memory in those with AUD. Future research incorporating functional neuroimaging (fMRI, PET) and molecular assays may elucidate the association of OPRM1 with brain functional connectivity, receptor binding, and neurotransmitter dynamics in those with AUD. Determination of whether OPRM1‐related effects are domain‐specific or related to global neurocognitive dysfunction will require further study incorporating a more comprehensive assessment battery.

Funding

This study was supported by a Department of Veteran Affairs Rehabilitation Research and Development Merit Review Award (T.C.D., RX002303; NCT03191266), Stanford University NeuroChoice Initiative (T.C.D. & C.B.P., clinical trial NCT03191266) and Department of Veteran Affairs Clinical Science Research and Development Career Development Award (C.B.P., IK2 CX001356) and by use of resources and facilities of the Palo Alto Veterans Administration Health Care System. The study sponsors had no role in the execution of this project or the preparation of this manuscript.

Conflicts of Interest

The authors declare no conflicts of interest.

Supporting information

Table S1: Pairwise comparisons of brain‐derived neurotrophic factor (BDNF) genotypes on neurocognitive measures.

Table S2: Pairwise comparisons of catechol‐o‐methyltransferase (COMT) genotypes on neurocognitive measures.

Table S3: Pairwise comparisons of females and males on CVLT‐II measures.

ACER-50-0-s001.docx (71.4KB, docx)

Acknowledgments

We extend our gratitude to the U.S. Armed Services Veteran study participants who made this research possible.

Data Availability Statement

The data that support the findings of this study are available from the corresponding author upon reasonable request.

References

  1. Acker, C. 1986. “Neuropsychological Deficits in Alcoholics: The Relative Contributions of Gender and Drinking History.” British Journal of Addiction 81: 395–403. [DOI] [PubMed] [Google Scholar]
  2. Bates, M. E. , Buckman J. F., and Nguyen T. T.. 2013. “A Role for Cognitive Rehabilitation in Increasing the Effectiveness of Treatment for Alcohol Use Disorders.” Neuropsychology Review 23: 27–47. [DOI] [PMC free article] [PubMed] [Google Scholar]
  3. Buhler, M. , and Mann K.. 2011. “Alcohol and the Human Brain: A Systematic Review of Different Neuroimaging Methods.” Alcoholism, Clinical and Experimental Research 35: 1771–1793. [DOI] [PubMed] [Google Scholar]
  4. Chartoff, E. H. , and Connery H. S.. 2014. “It's MORe Exciting Than Mu: Crosstalk Between Mu Opioid Receptors and Glutamatergic Transmission in the Mesolimbic Dopamine System.” Frontiers in Pharmacology 5: 116. [DOI] [PMC free article] [PubMed] [Google Scholar]
  5. Cohen, J. 1988. Statistical Power Analysis for the Behavioral Sciences. Lawrence Erlbaum Associates. [Google Scholar]
  6. Dawson, D. A. , Grant B. F., Stinson F. S., Chou P. S., Huang B., and Ruan W. J.. 2005. “Recovery From DSM‐IV Alcohol Dependence: United States, 2001‐2002.” Addiction 100: 281–292. [DOI] [PubMed] [Google Scholar]
  7. Demirakca, T. , Ende G., Kammerer N., et al. 2011. “Effects of Alcoholism and Continued Abstinence on Brain Volumes in Both Genders.” Alcoholism, Clinical and Experimental Research 35: 1678–1685. [DOI] [PubMed] [Google Scholar]
  8. Drake, C. T. , Chavkin C., and Milner T. A.. 2007. “Opioid Systems in the Dentate Gyrus.” Progress in Brain Research 163: 245–263. [DOI] [PubMed] [Google Scholar]
  9. Durazzo, T. , Hutchison K., Fryer S., Mon A., and Meyerhoff D.. 2012. “Associations of Cigarette Smoking and Polymorphisms in Brain‐Derived Neurotrophic Factor and Catechol‐O‐Methyltransferase With Neurocognition in Alcohol Dependent Individuals During Early Abstinence.” Frontiers in Pharmacology 3: 178. [DOI] [PMC free article] [PubMed] [Google Scholar]
  10. Durazzo, T. C. , Kraybill E. P., Stephens L. H., et al. 2025. “Intermittent Theta Burst to the Left Dorsolateral Prefrontal Cortex Promoted Decreased Alcohol Consumption and Improved Outcomes in Those With Alcohol Use Disorder: A Randomized, Double‐Blind, Placebo‐Controlled Clinical Trial.” Drug and Alcohol Dependence 270: 112641. [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. Durazzo, T. C. , McNerney M. W., Hansen A. M., Gu M., Sacchet M. D., and Padula C. B.. 2023. “BDNF rs6265 Met Carriers With Alcohol Use Disorder Show Greater Age‐Related Decline of N‐Acetylaspartate in Left Dorsolateral Prefrontal Cortex.” Drug and Alcohol Dependence 248: 109901. [DOI] [PubMed] [Google Scholar]
  12. Durazzo, T. C. , and Meyerhoff D. J.. 2021. “GABA Concentrations in the Anterior Cingulate and Dorsolateral Prefrontal Cortices: Associations With Chronic Cigarette Smoking, Neurocognition, and Decision Making.” Addiction Biology 26: e12948. [DOI] [PMC free article] [PubMed] [Google Scholar]
  13. Durazzo, T. C. , Nguyen L. C., and Meyerhoff D. J.. 2020. “Medical Conditions Linked to Atherosclerosis Are Associated With Magnified Cortical Thinning in Individuals With Alcohol Use Disorders.” Alcohol and Alcoholism 55: 382–390. [DOI] [PMC free article] [PubMed] [Google Scholar]
  14. Eichenbaum, H. 2004. “Hippocampus: Cognitive Processes and Neural Representations That Underlie Declarative Memory.” Neuron 44: 109–120. [DOI] [PubMed] [Google Scholar]
  15. Fama, R. , Le Berre A. P., and Sullivan E. V.. 2020. “Alcohol's Unique Effects on Cognition in Women: A 2020 (Re)view to Envision Future Research and Treatment.” Alcohol Research: Current Reviews 40: 3. [DOI] [PMC free article] [PubMed] [Google Scholar]
  16. Fields, H. L. , and Margolis E. B.. 2015. “Understanding Opioid Reward.” Trends in Neurosciences 38: 217–225. [DOI] [PMC free article] [PubMed] [Google Scholar]
  17. Flannery, B. , Fishbein D., Krupitsky E., et al. 2007. “Gender Differences in Neurocognitive Functioning Among Alcohol‐Dependent Russian Patients.” Alcoholism, Clinical and Experimental Research 31: 745–754. [DOI] [PubMed] [Google Scholar]
  18. Gilman, S. E. , and Abraham H. D.. 2001. “A Longitudinal Study of the Order of Onset of Alcohol Dependence and Major Depression.” Drug and Alcohol Dependence 63: 277–286. [DOI] [PubMed] [Google Scholar]
  19. Hajj, A. , Khoury R., Hachem R., et al. 2022. “Clinical and Genetic Factors Associated With Self‐Reported Cognitive Deficits in Women With Breast Cancer: The “CAGE‐Cog” Study.” BMC Cancer 22: 996. [DOI] [PMC free article] [PubMed] [Google Scholar]
  20. Heaton, R. K. , Miller S. W., Taylor M. J., and Grant I.. 2004. Revised Comprehensive Norms for an Expanded Halstead‐Reitan Battery, In Series Revised Comprehensive Norms for an Expanded Halstead‐Reitan Battery. Psychological Assessment Resources, Inc. [Google Scholar]
  21. Johnson, S. W. , and North R. A.. 1992. “Opioids Excite Dopamine Neurons by Hyperpolarization of Local Interneurons.” Journal of Neuroscience 12: 483–488. [DOI] [PMC free article] [PubMed] [Google Scholar]
  22. Koob, G. F. , and Volkow N. D.. 2016. “Neurobiology of Addiction: A Neurocircuitry Analysis.” Lancet. Psychiatry 3: 760–773. [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. Lovallo, W. R. , Enoch M. A., Acheson A., et al. 2015. “Cortisol Stress Response in Men and Women Modulated Differentially by the Mu‐Opioid Receptor Gene Polymorphism OPRM1 A118G.” Neuropsychopharmacology 40: 2546–2554. [DOI] [PMC free article] [PubMed] [Google Scholar]
  24. Mannes, Z. L. , Shmulewitz D., Livne O., Stohl M., and Hasin D. S.. 2021. “Correlates of Mild, Moderate, and Severe Alcohol Use Disorder Among Adults With Problem Substance Use: Validity Implications for DSM‐5.” Alcoholism, Clinical and Experimental Research 45: 2118–2129. [DOI] [PMC free article] [PubMed] [Google Scholar]
  25. Marazziti, D. , Consoli G., Picchetti M., Carlini M., and Faravelli L.. 2010. “Cognitive Impairment in Major Depression.” European Journal of Pharmacology 626: 83–86. [DOI] [PubMed] [Google Scholar]
  26. Meyerhoff, D. J. , Durazzo T. C., and Ende G.. 2013. “Chronic Alcohol Consumption, Abstinence and Relapse: Brain Proton Magnetic Resonance Spectroscopy Studies in Animals and Humans.” Current Topics in Behavioral Neurosciences 13: 511–540. [DOI] [PubMed] [Google Scholar]
  27. Nguyen, L. C. , Durazzo T. C., Dwyer C. L., et al. 2020. “Predicting Relapse After Alcohol Use Disorder Treatment in a High‐Risk Cohort: The Roles of Anhedonia and Smoking.” Journal of Psychiatric Research 126: 1–7. [DOI] [PMC free article] [PubMed] [Google Scholar]
  28. Padula, C. B. , and Durazzo T. C.. 2022. “Active Cigarette Smoking Is Associated With Increased Age‐Related Decline on Measures of Visuospatial Learning and Memory and Executive Function in Alcohol Use Disorder.” Alcohol and Alcoholism (Oxford, Oxfordshire) 57: 656–663. [DOI] [PubMed] [Google Scholar]
  29. Padula, C. B. , Durazzo T. C., and Meyerhoff D. J.. 2023. “Alcohol‐Related Disorders.” In The SAGE Handbook of Clinical Neuropsychology, edited by Boyle G. J., Stern Y., Stein D. J., and Sahakian B. J.. SAGE Publications Ltd. [Google Scholar]
  30. Padula, C. B. , Hansen A., Hughes R. L., and McNerney M. W.. 2021. “Dimensions of Craving Interact With COMT Genotype to Predict Relapse in Individuals With Alcohol Use Disorder Six Months After Treatment.” Brain Sciences 11: 62. [DOI] [PMC free article] [PubMed] [Google Scholar]
  31. Padula, C. B. , McCalley D. M., Tenekedjieva L. T., et al. 2024. “A Pilot, Randomized Clinical Trial: Left Dorsolateral Prefrontal Cortex Intermittent Theta Burst Stimulation Improves Treatment Outcomes in Veterans With Alcohol Use Disorder.” Alcohol: Clinical and Experimental Research 48: 164–177. [DOI] [PubMed] [Google Scholar]
  32. Pan, Z. , Park C., Brietzke E., et al. 2019. “Cognitive Impairment in Major Depressive Disorder.” CNS Spectrums 24: 22–29. [DOI] [PubMed] [Google Scholar]
  33. Ray, L. A. , Bujarski S., MacKillop J., Courtney K. E., Monti P. M., and Miotto K.. 2013. “Subjective Response to Alcohol Among Alcohol‐Dependent Individuals: Effects of the μ‐Opioid Receptor (OPRM1) Gene and Alcoholism Severity.” Alcoholism, Clinical and Experimental Research 37, no. Suppl 1: E116–E124. [DOI] [PMC free article] [PubMed] [Google Scholar]
  34. Rolls, E. T. , and Grabenhorst F.. 2008. “The Orbitofrontal Cortex and Beyond: From Affect to Decision‐Making.” Progress in Neurobiology 86: 216–244. [DOI] [PubMed] [Google Scholar]
  35. Sankoh, A. J. , Huque M. F., and Dubey S. D.. 1997. “Some Comments on Frequently Used Multiple Endpoint Adjustment Methods in Clinical Trials.” Statistics in Medicine 16: 2529–2542. [DOI] [PubMed] [Google Scholar]
  36. Schmidt‐Wilcke, T. , Fuchs E., Funke K., et al. 2018. “GABA‐From Inhibition to Cognition: Emerging Concepts.” Neuroscientist 24: 501–515. [DOI] [PubMed] [Google Scholar]
  37. Scott, J. C. , Matt G. E., Wrocklage K. M., et al. 2015. “A Quantitative Meta‐Analysis of Neurocognitive Functioning in Posttraumatic Stress Disorder.” Psychological Bulletin 141: 105–140. [DOI] [PMC free article] [PubMed] [Google Scholar]
  38. Shi, M.‐M. , Fan K.‐M., Qiao Y.‐N., et al. 2020. “Hippocampal μ‐Opioid Receptors on GABAergic Neurons Mediate Stress‐Induced Impairment of Memory Retrieval.” Molecular Psychiatry 25: 977–992. [DOI] [PMC free article] [PubMed] [Google Scholar]
  39. Staudt, J. , Kok T., de Haan H. A., Walvoort S. J. W., and Egger J. I. M.. 2023. “Neurocognitive Recovery in Abstinent Patients With Alcohol Use Disorder: A Scoping Review for Associated Factors.” Neuropsychiatric Disease and Treatment 19: 2039–2054. [DOI] [PMC free article] [PubMed] [Google Scholar]
  40. Stavro, K. , Pelletier J., and Potvin S.. 2013. “Widespread and Sustained Cognitive Deficits in Alcoholism: A Meta‐Analysis.” Addiction Biology 18: 203–213. [DOI] [PubMed] [Google Scholar]
  41. Stinson, F. S. , Grant B. F., Dawson D. A., Ruan W. J., Huang B., and Saha T.. 2005. “Comorbidity Between DSM‐IV Alcohol and Specific Drug Use Disorders in the United States: Results From the National Epidemiologic Survey on Alcohol and Related Conditions.” Drug and Alcohol Dependence 80: 105–116. [DOI] [PubMed] [Google Scholar]
  42. Sullivan, E. V. , and Pfefferbaum A.. 2005. “Neurocircuitry in Alcoholism: A Substrate of Disruption and Repair.” Psychopharmacology 180: 583–594. [DOI] [PubMed] [Google Scholar]
  43. Tapert, S. F. , Ozyurt S. S., Myers M. G., and Brown S. A.. 2004. “Neurocognitive Ability in Adults Coping With Alcohol and Drug Relapse Temptations.” American Journal of Drug and Alcohol Abuse 30: 445–460. [DOI] [PubMed] [Google Scholar]
  44. Troisi, A. , Frazzetto G., Carola V., et al. 2011. “Social Hedonic Capacity Is Associated With the A118G Polymorphism of the Mu‐Opioid Receptor Gene (OPRM1) in Adult Healthy Volunteers and Psychiatric Patients.” Social Neuroscience 6: 88–97. [DOI] [PubMed] [Google Scholar]
  45. Zhang, Z. , Hu S., Geng S., et al. 2025. “A Multiscale Atlas of Oprm1‐Expressing Neurons in the Central Nervous System: Brain‐Wide Distribution, Circuit Functions, and Translational Therapeutic Implications.” Neurochemistry International 188: 105998. [DOI] [PubMed] [Google Scholar]
  46. Zhou, H. , Rentsch C. T., Cheng Z., et al. 2020. “Association of OPRM1 Functional Coding Variant With Opioid Use Disorder: A Genome‐Wide Association Study.” JAMA Psychiatry 77: 1072–1080. [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Table S1: Pairwise comparisons of brain‐derived neurotrophic factor (BDNF) genotypes on neurocognitive measures.

Table S2: Pairwise comparisons of catechol‐o‐methyltransferase (COMT) genotypes on neurocognitive measures.

Table S3: Pairwise comparisons of females and males on CVLT‐II measures.

ACER-50-0-s001.docx (71.4KB, docx)

Data Availability Statement

The data that support the findings of this study are available from the corresponding author upon reasonable request.

Articles from Alcohol, Clinical & Experimental Research are provided here courtesy of Wiley

RESOURCES