Implications of OPRM1 and CYP2B6 variants on treatment outcomes in methadone-maintained patients in Ontario: Exploring sex differences.

Genetic variants in the OPRM1 and CYP2B6 genes, respectively coding for an opioid receptor and methadone metabolizers, have been linked to negative treatment outcomes in patients undergoing methadone maintenance treatment, with little consensus on their effect. This study aims to test the associations between pre-selected SNPs of OPRM1 and CYP2B6 and outcomes of continued opioid use, relapse, and methadone dose. It also aims to observe differences in associations within the sexes. 1,172 participants treated with methadone (nMale = 666, nFemale = 506) were included in this study. SNPs rs73568641 and rs7451325 from OPRM1 and all the tested CYP2B6 SNPs were detected to be in high linkage disequilibrium. Though no associations were found to be significant, noteworthy differences were observed in associations of OPRM1 rs73568641 and CYP2B6 rs3745274 with treatment outcomes between males and females. Further research is needed to determine if sex-specific differences are present.

Introduction

Background

Methadone maintenance treatment (MMT) targeted for patients with opioid use disorder (OUD) has been proven over time to decrease opioid cravings and use [1]. However, due to the chronic classification of OUD, MMT is not curative, but aims to maintain patients on a specific dose, controlling their opioid use and enabling them to regain stability [1-3]. Administered methadone binds to endogenous opioid receptors in the human brain, eliciting similar effects within the reward system as an opioid would, while suppressing withdrawal symptoms [4].

Though effective in reducing opioid use, MMT has been observed to have interindividual variability in methadone’s metabolism and methadone blood concentration for a given dose [5]. This can be potentially dangerous to patients, as prescribing physicians are unable to accurately predict the patient’s reaction to a methadone dose prior to administering it. If the methadone dose administered is too low, the patient can be at a high risk of relapse [6, 7]. Alternatively, if the dose is too high, the patient might be at a risk of overdosing, if supplementing with other opioids [8]. As such, a genetic predisposition for individual-based MMT outcomes has been the focus of much research [9-12].

The opioid receptor proteins, encoded by the mu opioid receptor 1 (OPRM1) gene, bind both endogenous and exogenous opioids, resulting in pain relief and feelings of euphoria [13]. Single nucleotide polymorphisms (SNPs) in OPRM1 have been associated with the number of opioid receptors present and their ability to function [14]. OPRM1 SNPs rs1799971 and rs1799972 have been previously implicated in opioid use disorder [15]. Interestingly, rs1799971, rs73568641, and rs10485058 have been associated with methadone plasma concentration, methadone dose, and opioid use changes [16].

The enzymes encoded by the cytochrome P450 family 2 subfamily B member 6 (CYP2B6) gene are involved in metabolizing 2 to 10% of clinically administered drugs, including methadone [17]. SNPs in this gene can lead loss or gain of function of the encoded proteins, possibly resulting in altered drug metabolism [18]. Many CYP2B6 SNPs have been implicated in altered methadone metabolism and plasma concentrations, most notably rs2279343 and rs10403955 [11, 19, 20]. Some studies have also found associations to adverse events in methadone patients, with rs8192719 and rs3745274 associated with overdose fatality [16, 21].

Disparities in opioid use patterns, health and social functioning, and polysubstance use in methadone patients have been observed between the sexes [22, 23]. Further, genetic differences between sexes have been detected in psychiatric disorders and traits, and studies have highlighted the presence of sex-dependent effects in models with common genetic variants [24, 25]. Though past studies have adjusted for sex in their analysis models, very few have been observed to assess the contribution of sex to the genetic predisposition to MMT outcomes using rigorous sex-based analyses, considering how findings might differ within males and within females.

Studying select OPRM1 and CYP2B6 SNPs in a European sample would allow us to not only confirm conclusions within the published literature but also test if the strength of these associations holds true to direct clinical MMT outcomes observable in patients, such as continued opioid use, relapse, and methadone dose. Additionally, having comparable male to female ratios within our sample enables us to robustly examine sex-based differences that have not been adequately highlighted in past studies.

Objectives

This study aims to report new genetic associations that have not been tested previously, as well as analyze associations with biological relevance from previous literature within a larger sample of European descent. The objectives of this study are to:

Test the association between pre-selected OPRM1 (rs73568641, rs7451325, rs10485058, rs1799971) and CYP2B6 (rs2279343, rs10403955, rs8192719, rs3745274) SNPs and continued opioid use, relapse, and methadone dose in MMT patients; and

Determine if there are differences in associations within and between the sexes through sex stratification and exploratory SNP x Sex interaction analyses.

Methods

This candidate gene study is reported according to Strengthening the Reporting of Genetic Association studies guideline, an extension of Strengthening the Reporting of Observational studies in Epidemiology statement [26]. An accompanying guideline checklist could be found in S1 File.

Study design and setting

This research reports data collected by the Genetics of Opioid Addiction (GENOA) study, which is an observational cohort study of 1,536 participants recruited from Canadian Addiction Treatment Centres across Ontario, Canada [27]. Data collected at the baseline (enrollment in the study) are the primary sources of information used. The data used include socio-demographic, opioid use-related, and treatment-related information, as well as information obtained from urine toxicology screen (UTS) results and blood samples. UTS results were also collected 3 months prior to study enrollment and up to a 12-month follow up period for measuring treatment outcomes. UTSs testing for opioid use were conducted regularly, on a weekly/biweekly basis, with results reported at 3-month intervals for the GENOA study. The GENOA study was approved by the Hamilton Integrated Research Ethics Board (#11056). All the participants enrolled in the study provided written informed consent.

Eligibility criteria

The participants selected for this study are those deemed eligible by the GENOA study eligibility criteria [27]. These required participants to be 18 years of age or older, have a Diagnostic and Statistical Manual of Mental Disorders [5th edition] OUD diagnosis, undergo an opioid substitution or antagonist therapy for OUD, and provide informed consent. Further inclusion criteria for all research questions addressed in this study include only participants who have provided a DNA sample and have received methadone as the primary opioid substitution or antagonist therapy.

For the measures of continued opioid use and relapse, participants must have had UTSs assessing the presence of opioids for a minimum duration of 3 months and 6 months, respectively. Participants taking prescription opioid medications were excluded due to the uncertainty of the opioid origin when reviewing the UTSs in these participants. These exclusion criteria did not apply to the methadone dose outcome measure as no UTSs were used for that set of analyses.

Outcomes and quantitative variables

Outcomes measured in this study include the following:

Continued opioid use while on MMT, defined as any opioid positive UTS (including opiates and oxycodone) observed over a duration of 3 to 15 months. It was measured as a binary variable.

Relapse while on MMT, defined as an event of opioid positive UTS following at least 3 months of opioid negative UTSs. It was measured as a binary variable.

Methadone dose while on MMT, defined as the amount of methadone a patient is administered at the time of study recruitment in milligrams. It was measured as a continuous variable.

Covariates for the measures of continued opioid use and relapse that were accounted for in the statistical models included: sex, age in years, methadone dose in milligrams, duration on MMT in months, and 5 principal components accounting for differences in ethnicity. Covariates accounted for in the measure of methadone dose were sex, age, duration on MMT, weight in kilograms, and the principal components. For the sex stratified analyses, the same variables as above were included in the additive models.

Genetic variants tested were identified from literature reviews, systematic reviews, candidate gene studies and genome-wide association studies as those related to OPRM1 or CYP2B6 and associated with altered methadone metabolism, methadone plasma concentrations, methadone dose, opioid use, or other treatment outcomes. Details about each selected SNP are shown in Table 1.

Table 1

Selected SNP details and genotype counts in European participants from the GENOA study.

Gene	Chr: Position (GRCh37)	SNP ID	Genotypes**	Genotype count	MAF*	HWE p-value*	Previously associated trait	Reference
OPRM1	6:154025139	rs73568641			0.152	0.151	Daily methadone dose	[28]
			CC	35
			CT	304
			TT	887
OPRM1	6:154016517	rs7451325			0.152	0.149	Daily methadone dose	[28]
			CC	35
			CT	303
			TT	888
OPRM1	6:154445215	rs10485058			0.132	0.901	Opioid positive urine screens of methadone patients	[16]
			GG	20
			GA	283
			AA	923
OPRM1	6:154360797	rs1799971			0.11	0.662	Opioid use disorder	[29]
			GG	16
			GA	237
			AA	973
CYP2B6	19:41515263	rs2279343			0.246	0.282	Higher S-Methadone plasma concentrations	[19]
			GG	67
			GA	470
			AA	689
CYP2B6	19:41509438	rs10403955			0.259	0.941	Higher S-Methadone plasma concentrations; lower apparent clearance of S-Methadone	[19]
			GG	83
			GT	470
			TT	673
CYP2B6	19:41518773	rs8192719			0.24	0.433	Increased frequency in methadone fatalities	[21]
			TT	65
			TC	458
			CC	703
CYP2B6	19:41512841	rs3745274			0.234	0.577	Increased frequency in methadone fatalities	[21]
			TT	63
			TG	447
			GG	716

MAF = minor allele frequency. HWE = Hardy-Weinberg equilibrium.

*Data from the GENOA study (N = 1226).

**Alleles are on the + strand.

Data handling

DNA was extracted from blood samples and the genotyping was performed by the Genome Quebec Innovation Centre (Montreal, Canada) [30], using the Illumina Global Screening Array-24 v1.0. Standard genetic association study quality control checks were applied using PLINK v1.09 and the RStudio interface for R i386 3.5.1 [31-33]. Genotype imputation in participants of European ancestry (as confirmed by PCA, n = 1,226) was performed using the Haplotype Reference Consortium’s data as a reference panel via Michigan Imputation Server, using EAGLE2 and Minimac4 [34-36]. Post-imputation variant filtering was conducted, excluding SNPs with Rsq quality metrics of less than 0.3 and/or minor allele frequencies lower than 0.05.

SNPs reported in high linkage disequilibrium (r2>0.2) were pruned, keeping the SNP with the most reported clinical significance and published associations, as seen on NCBI’s SNP database [37]. As such, OPRM1 rs7451325, and CYP2B6 rs2279343, rs10403955, and rs8192719 were excluded. HaploView was used to visualize SNPs in linkage disequilibrium and calculate r-squared coefficients [38].

A detailed description as well as a flowchart outlining the different steps conducted to reach the final sample size are available in S2 File.

Bias

Measures were taken in this study to identify areas of bias and address them. However, there remained potential sources of bias that could not be avoided, and thus are reported here. Outcomes of continued opioid use and relapse were defined through UTSs as opposed to relying on patient self-reports to remain as objective and unbiased as possible. However, measures such as methadone dose and duration on MMT were self-reported, allowing for a potential of social desirability bias, where participants might provide false information in lieu of more accurate responses that might be viewed as less desirable. Social desirability bias could also have elicited differing responses within males and females as behaviours might seem more desirable in one sex but not the other [39]. In addition, the findings might be affected by volunteer bias, wherein the sample recruited could not have been representative of the entire OUD population receiving treatment. Furthermore, only participants of European ethnicities were included in the analyses conducted. This might result in data that are not generalizable or lack replicability in other ethnic populations. Lastly, since the nature of this study is observational, it is not possible to control for all variables present, and as such undetected biases could have contributed to the findings reported.

Statistical methods

Descriptive statistical analyses were conducted on the total samples and stratified by sex to describe the demographic and clinical characteristics of the sample. Continuous variables were expressed as means with standard deviations, while categorical variables were expressed as counts. Chi square tests were conducted for categorical variables and t-tests for continuous variables to measure differences between the sexes.

Separate regression analyses were performed to test the association between each set of gene SNPs and the outcomes of continued opioid use, relapse, and methadone dose. An additive genetic model was used for all variants and all tests. Logistic regressions were conducted to test the associations of continued opioid use and relapse, with the analyses testing for the association of having the minor allele and the outcomes as specified earlier. A linear regression model was used to test the association of having the minor allele and the outcome of methadone dose. All covariates were adjusted for, measuring their associations with the outcomes of interest. Furthermore, identical but separate regression analyses were conducted for male and female subsets, respectively. For analyzing sex differences, interaction analyses were performed with SNP x Sex as the interaction term in the regression models.

Samples with missing outcome values were excluded from the analysis. For the logistic regression analyses, missing values for the covariates of methadone dose and duration on MMT were imputed via mean substitution, from the averages of the values calculated per analysis. The same method was used to impute for missing weight and duration on MMT values for the linear regression.

Bonferroni corrected p-values of P<0.017 for OPRM1 SNPs and P<0.05 for CYP2B6 SNPs were used as thresholds for significance. All statistical analyses were performed on PLINK v1.09 and the RStudio interface for R i386 3.5.1 [31, 32].

Results

Participants

Samples from 1,226 participants and 5,563,682 SNPs passed quality control checks and filtering after imputation. After sample data cleanup and applying eligibility criteria for each outcome of interest, 1,129 samples were analyzed for continued opioid use, 944 samples for relapse, and 1,165 samples for methadone dose (S2 File).

Descriptive data

Participant demographics and clinical characteristics can be seen in Table 2. Of the 1,226 ethnically European participants, 57% were male and 43% were female. The majority of participants were never married, unemployed, on methadone, and not prescribed opioid medications. The mean duration on MMT, age of first opioid use, and total number of positive opioid urine screens, as well as continued opioid use and relapse outcomes, did not differ significantly between the sexes. The weight and dose of methadone administered were lower in females than males, as would have been expected, as individuals of lower weight tend to be prescribed lower doses of MMT. In addition, the ratio of employed to unemployed males (0.70) was significantly higher than that of females (0.37).

Table 2

Characteristics of participants of European ancestry with available genotype data in GENOA.

	Total	Male	Female	p-value
N (%)	1226	699 (57)	527 (43)
Age in years a , Mean (SD)	39 (11)	40 (11)	38 (11)	9.25E-03*
Weight in kg b , Mean (SD)	80 (21)	86 (20)	72 (19)	2.2E-16*
Marital status c , N (%)				3.51E-03*
Common law	236 (19)	118 (17)	118 (22)
Divorced	125 (10)	77 (11)	48 (9)
Currently married	144 (12)	95 (14)	48 (9)
Never married	555 (45)	328 (47)	227 (43)
Separated	134 (11)	64 (9)	70 (13)
Widowed	31 (3)	15 (2)	16 (3)
Employment d , N (%)				4.86E-07*
Employed	430 (35)	287 (41)	143 (27)
Unemployed	793 (65)	411 (59)	382 (73)
Methadone dose in mg e , Mean (SD), [Range]	75 (45), [1–400]	78 (47), [2–400]	71 (43), [1–280]	6.28E-03*
MAT f , N (%)				0.69
Methadone	1172 (96)	666 (96)	506 (96)
Suboxone	52 (4)	31 (4)	21 (4)
Duration on MMT in months g , Mean (SD)	45 (48)	45 (48)	44 (49)	0.74
Age of first opioid use h , Mean (SD)	25 (9)	25 (9)	25 (9)	0.93
Participant taking opioid prescription i , N (%)				0.83
Prescribed opioids	34 (3)	20 (3)	14 (3)
Not prescribed opioids	1192 (97)	679 (97)	513 (97)
Total number of opioid screens j † , Mean (SD)	74 (35)	74 (34)	75 (35)	0.57
Total number of positive opioid screens k , Mean (SD)	13 (21)	13 (20)	13 (22)	0.70
Continued opioid use outcome l , N (%)				0.32
Continued opioid use	893 (79)	513 (80)	380 (78)
No continued opioid use	236 (21)	127 (20)	109 (22)
Relapse outcome m , N (%)				0.30
Relapse	433 (46)	251 (47)	182 (44)
No relapse	511 (54)	279 (53)	232 (56)

260 of reported total included participants screened only for opiates.

*Significant difference between males and females.

All means were calculated excluding missing values.

aData available for nTotal = 1226, nMale = 699, nFemale = 527.

bData available for nTotal = 1216, nMale = 693, nFemale = 523.

cData available for nTotal = 1224, nMale = 697, nFemale = 527.

dData available for nTotal = 1223, nMale = 698, nFemale = 525.

eData available for nTotal = 1166, nMale = 664, nFemale = 502.

fData available for nTotal = 1224, nMale = 697, nFemale = 527.

gData available for nTotal = 1162, nMale = 661, nFemale = 501.

hData available for nTotal = 1197, nMale = 685, nFemale = 512.

iData available for nTotal = 1226, nMale = 699, nFemale = 527.

jData available for nTotal = 1226, nMale = 699, nFemale = 527.

kData available for nTotal = 1218, nMale = 692, nFemale = 526.

lData available for nTotal = 1129, nMale = 640, nFemale = 489.

mData available for nTotal = 944, nMale = 530, nFemale = 414.

Main results

Results of the sex-stratified association analyses between the OPRM1 SNPs (rs73568641, rs1799971, rs10485058) and continued opioid use, relapse, and methadone dose are shown in Table 3. No associations reached the Bonferroni adjusted significance threshold of P<0.017. However, some near-significant associations were observed within females but not within males, notably regarding rs73568641. Allele C expressed a potential of decreased odds of continued opioid use within females [OR = 0.71, 95%CI = 0.47,1.07, P = 0.098]. Its presence also signified a potentially more pronounced decrease in methadone dose in females [β = -7.99, SE = 3.73, P = 0.033] than in males [β = -2.36, SE = 3.33, P = 0.48].

Table 3

OPRM1 SNPs and associated outcomes.

Outcome	SNP	N	Minor Allele	OR/BETA	95% CI/SE	P
Continued opioid use	rs73568641	1129	C	0.84	0.63, 1.10	0.21
	Male	640		0.99	0.67, 1.45	0.95
	Female	489		0.71	0.47, 1.07	0.098*
	rs1799971	1129	G	0.97	0.70, 1.36	0.88
	Male	640		1.11	0.72, 1.72	0.64
	Female	489		0.87	0.51, 1.48	0.61
	rs10485058	1129	G	0.96	0.71, 1.30	0.78
	Male	640		0.89	0.59, 1.36	0.60
	Female	489		1.00	0.64, 1.57	0.99
Relapse	rs73568641	944	C	0.98	0.76, 1.25	0.85
	Male	530		0.97	0.69, 1.34	0.82
	Female	414		1.04	0.70, 1.54	0.86
	rs1799971	944	G	0.82	0.61, 1.90	0.17
	Male	530		0.76	0.52, 1.09	0.14
	Female	414		0.94	0.58, 1.52	0.80
	rs10485058	944	G	1.10	0.83, 1.44	0.51
	Male	530		1.02	0.70, 1.49	0.91
	Female	414		1.15	0.77, 1.73	0.50
Methadone dose	rs73568641	1165	C	-4.24	2.49	0.089*
	Male	664		-2.36	3.33	0.48
	Female	501		-7.99	3.73	0.033**
	rs1799971	1165	G	0.20	2.90	0.95
	Male	664		2.59	3.76	0.49
	Female	501		-4.92	4.63	0.29
	rs10485058	1165	G	-0.45	2.72	0.87
	Male	664		-0.50	3.69	0.89
	Female	501		0.24	4.00	0.95

The minor alleles are also the reference and tested alleles. OR is odds ratio and BETA is the beta coefficient for the regression. 95% CI is the 95% confidence interval levels (lower, upper) and SE is the standard error. All results reported are odds ratios and 95% confidence intervals, except for the methadone dose outcomes, which are BETA coefficients and standard errors. P is the p-value for the t-statistic. The significance threshold is P<0.017.

*P<0.1.

**P<0.05.

Results of the sex-stratified association analyses between the CYP2B6 SNP rs3745274 and continued opioid use, relapse, and methadone dose are shown in Table 4. No associations were found to be significant (P<0.05). Nonetheless, a near-significant association between the T allele of rs3745274 and continued opioid use within males [OR = 0.73, 95%CI = 0.52, 1.014, P = 0.06] was observed.

Table 4

CYP2B6 SNPs and associated outcomes.

Outcome	SNP	N	Minor Allele	OR/BETA	95% CI/SE	P
Continued opioid use	rs3745274	1129	T	0.82	0.64, 1.05	0.11
	Male	640		0.73	0.52, 1.01	0.06*
	Female	489		0.95	0.66, 1.37	0.80
Relapse	rs3745274	944	T	0.91	0.73, 1.14	0.42
	Male	530		0.86	0.64, 1.16	0.32
	Female	414		1.07	0.76, 1.49	0.71
Methadone dose	rs3745274	1165	T	1.26	2.17	0.56
	Male	664		-1.17	2.99	0.70
	Female	501		4.19	3.18	0.19

The minor alleles are also the reference and tested alleles. OR is odds ratio and BETA is the beta coefficient for the regression. 95% CI is the 95% confidence interval levels (lower, upper) and SE is the standard error. All results reported are odds ratios and 95% confidence intervals, except for the methadone dose outcomes, which are BETA coefficients and standard errors. P is the p-value for the t-statistic. The significance threshold is P<0.05.

*P<0.1.

Exploratory analyses showcasing differences in associations between males and females were conducted. No significant results are reported. For detailed results see Tables G and H in S2 File.

Discussion

Key results

This study did not observe any associations that reached the significance threshold set. However, differences in the levels of significance within males and females were detected. Females with the C allele of OPRM1 rs73568641 showed higher significance levels and stronger protective properties towards continued opioid use than males, as well as a potentially decreased methadone dose. However, the T allele of CYP2B6 rs3745274 in males showed potential for being more protective and significant when it came to continued opioid use.

Interpretation

The possible involvement of the C allele of OPRM1’s rs73568641 in a decreased chance of opioid use and/or decreased methadone dose in females suggests the involvement of OPRM1 gene in not only opioid use disorder, but also treatment outcomes. The similar direction of association observed with respect to continued opioid use and methadone dose is interesting given that previous research has reported that higher methadone doses are more effective at decreasing opioid use while on MMT [40]. However, since the variable of methadone dose was accounted for in the analysis model of continued opioid use, the results of the associations can be viewed as independent. When compared to the literature, these associations conflict with the only other published findings. OPRM1 rs73568641 (allele C) seems to have an opposite effect in an African American population [28]. In a genome-wide association study subset (n = 383), it was found to slightly increase daily methadone dose [β = 0.681, P = 2.81E-08]. Unfortunately, no conclusions could be drawn due to the possibility that the differences observed between these findings could be a result of the ethnic contribution to the genetic makeup. This highlights the importance of ethnically diverse research and how interindividual differences of patients of different ethnic backgrounds could play a role in patient treatment outcomes.

While the role of the CYP2B6 rs3745274 SNP was not determined in this study with regards to an MMT outcome, other studies have reported evidence of association across different haplotypes of CYP2B6, especially those where this SNP is found, and plasma methadone concentrations. In a pharmacokinetics study, CYP2B6*6 carriers were observed to have higher S-methadone plasma concentrations than non-carriers [41]. It was also determined that CYP2B6 inhibition reduces methadone clearance and increases plasma methadone concentrations [41]. This was further supported by other studies where CYP2B6*6 was shown to have slower S-methadone intravenous clearance, slower R- and S-methadone oral clearance, higher plasma concentrations, and lower methadone dose requirements in carriers [20, 21, 42]. However, given methadone’s racemic mixture and findings supporting R-methadone’s heavier contribution to opioid effects, more evidence on the genetic effects on R-methadone metabolism is required [5]. When comparing CYP2B6 rs3745274 to literature findings on other treatment outcomes, the T allele seems to be associated with an increased frequency in methadone fatalities (P = 1.2E-03) in a sample of European ethnicity (n = 125) [21]. Though these fatality findings support the discussed literature, as a higher plasma concentration of methadone could also have negative effects and risks, such as death, the differing sample sizes of control and methadone-only groups (n = 255 and n = 125, respectively) could have contributed to such results.

This study was unique in stratifying analyses by sex and observing differential findings for each sex. The sex-based differences observed in the strengths of the associations could not be fully attributed to sample size, as seen in the strength of OPRM1 rs73568641’s associations in females despite having a smaller sample size than their male subset’s counterpart. This could be indicative of larger biology-based differences within the sexes, which could have influenced the results. Examples could be the differing CYP enzyme activities between the sexes that could affect drug metabolism, or neuroanatomical differences in the dopaminergic pathway that can influence the effects of a drug on the system [43, 44]. It is also possible that gender construct and its implications can affect the results, even if indirectly. Women are more likely to become dependent on prescribed opioids than males, experience faster dependence progression rates, and have higher relapse rates [23, 24]. Men, on the other hand, report higher prevalence cannabis use and are more likely to be employed and financially secure [22, 45]. These are only a few examples of how the behavioural and social functioning implications associated with gender can influence phenotypes measures, such as continued opioid use and relapse.

Limitations and generalizability

Aside from the sources of bias discussed earlier, some limitations in this study were faced and need to be addressed. Firstly, the findings are specific to a sample of European ethnic descent, making them not generalizable to samples of other ethnicities. Similarly, the sex-specific results may not be comparable to other study findings that do not conduct sex-stratified analyses. Another limitation is that there was a high degree of missingness within the data with respect to the measure of relapse, resulting in a smaller sample size for that set of analyses. Though a power analysis was conducted for the original GENOA project, it is not applicable due to the different SNPs analyzed in this specific study. Additionally, due to a lack of a reported and reliable effect size in the literature and the disputably misleading results of a post-hoc power analysis, an informative power calculation could not have been conducted [46]. Further data missingness was observed in the UTS results reported across the sample population. As the duration of UTS result collection ranged from 3 to 15 months, the outcomes of continued opioid use and relapse were not consistently measured. However, given that and the inevitable variability in how long participants had been on MMT, the duration on MMT was accounted for in all statistical models. An additional data-related limitation includes the inability to accurately use methadone dose as an indicator of treatment response in MMT patients. This is mostly due to the fact that patients on MMT could be at any of the induction, treatment, stabilization, or tapering stages, each of which characterized by a variable pattern of methadone dose administration. This participant variability also plays a role in the measurement of the relapse outcome, posing a challenge in accounting for all participants including those with some breakthrough opioid use while on treatment. Finally, since the exploratory between-sex analyses were insignificant, the interpretation of the sex-stratified results are made with caution. Though an insignificant interaction term could be interpreted as an absence of a difference between males and females, it could also be highly indicative of an under-powered study.

Conclusion

Given that the study had a larger sample size than most similar published research within this field, it was able to address a gap in the genetics of MMT research. Though none of the results were significant, this study identified a need for ethnically diverse research, and uncovered the important contribution sex measures have towards outcomes of continued opioid use and methadone dose in MMT patients. Future recommendations towards more powered studies including sex in the analysis models are made.

STREGA checklist.

(DOCX)

Click here for additional data file.

Appendix.

(DOCX)

Click here for additional data file.

14 May 2021

PONE-D-21-11145

Implications of OPRM1 and CYP2B6 variants on treatment outcomes in methadone-maintained patients in Ontario: Exploring sex differences

PLOS ONE

Dear Dr. Samaan,

Thank you for submitting your manuscript to PLOS ONE. After careful consideration, we feel that it has merit but does not fully meet PLOS ONE’s publication criteria as it currently stands. Therefore, we invite you to submit a revised version of the manuscript that addresses the points raised during the review process.

Please submit your revised manuscript by Jun 28 2021 11:59PM. If you will need more time than this to complete your revisions, please reply to this message or contact the journal office at plosone@plos.org. When you're ready to submit your revision, log on to https://www.editorialmanager.com/pone/ and select the 'Submissions Needing Revision' folder to locate your manuscript file.

Please include the following items when submitting your revised manuscript:

A rebuttal letter that responds to each point raised by the academic editor and reviewer(s). You should upload this letter as a separate file labeled 'Response to Reviewers'.

A marked-up copy of your manuscript that highlights changes made to the original version. You should upload this as a separate file labeled 'Revised Manuscript with Track Changes'.

An unmarked version of your revised paper without tracked changes. You should upload this as a separate file labeled 'Manuscript'.

If you would like to make changes to your financial disclosure, please include your updated statement in your cover letter. Guidelines for resubmitting your figure files are available below the reviewer comments at the end of this letter.

If applicable, we recommend that you deposit your laboratory protocols in protocols.io to enhance the reproducibility of your results. Protocols.io assigns your protocol its own identifier (DOI) so that it can be cited independently in the future. For instructions see: http://journals.plos.org/plosone/s/submission-guidelines#loc-laboratory-protocols. Additionally, PLOS ONE offers an option for publishing peer-reviewed Lab Protocol articles, which describe protocols hosted on protocols.io. Read more information on sharing protocols at https://plos.org/protocols?utm_medium=editorial-email&utm_source=authorletters&utm_campaign=protocols.

We look forward to receiving your revised manuscript.

Kind regards,

Huiping Zhang

Academic Editor

PLOS ONE

Journal Requirements:

When submitting your revision, we need you to address these additional requirements.

1. Please ensure that your manuscript meets PLOS ONE's style requirements, including those for file naming. The PLOS ONE style templates can be found at

and

https://journals.plos.org/plosone/s/file?id=ba62/PLOSOne_formatting_sample_title_authors_affiliations.pdf

2.We note that you have indicated that data from this study are available upon request. PLOS only allows data to be available upon request if there are legal or ethical restrictions on sharing data publicly. For information on unacceptable data access restrictions, please see http://journals.plos.org/plosone/s/data-availability#loc-unacceptable-data-access-restrictions.

In your revised cover letter, please address the following prompts:

a) If there are ethical or legal restrictions on sharing a de-identified data set, please explain them in detail (e.g., data contain potentially identifying or sensitive patient information) and who has imposed them (e.g., an ethics committee). Please also provide contact information for a data access committee, ethics committee, or other institutional body to which data requests may be sent.

b) If there are no restrictions, please upload the minimal anonymized data set necessary to replicate your study findings as either Supporting Information files or to a stable, public repository and provide us with the relevant URLs, DOIs, or accession numbers. Please see http://www.bmj.com/content/340/bmj.c181.long for guidelines on how to de-identify and prepare clinical data for publication. For a list of acceptable repositories, please see http://journals.plos.org/plosone/s/data-availability#loc-recommended-repositories.

We will update your Data Availability statement on your behalf to reflect the information you provide.

[Note: HTML markup is below. Please do not edit.]

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #1: Yes

Reviewer #2: Partly

Reviewer #3: Partly

**********

2. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: Yes

Reviewer #2: I Don't Know

Reviewer #3: Yes

**********

3. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #1: Yes

Reviewer #2: Yes

Reviewer #3: No

**********

4. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #1: Yes

Reviewer #2: Yes

Reviewer #3: Yes

**********

5. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #1: The authors performed a genetic association study of methadone maintenance treatment (MMT) of opioid use disorder among ~1100 persons of European origin (~65% male) who had urine drug screen (UDS) data and methadone plasma levels available over a 3-15 month period, across a consortium of MMT clinics in Canada. They did array genotyping which yielded ~5 million SNPs after imputation. Despite this wealth of genetic data, they report on 3 OPRM1 and 2 CYP2B6 SNPs. There were 3 phenotypes: methadone plasma levels, any lapse to illicit opioid use (a single + UDS) and relapse (a single + UDS after at least 3 months of – UDS results). Logistic regression and linear regression analyses were implemented. Sex-specific results are reported. There were no statistically significant results. The following comments are relevant:

1. The authors must be clear about whether they analyzed (or plan to analyze) the methadone dose and other phenotypes using other SNPs, or whether they restricted analysis to only those 5 SNPs reported here. This has implications for multiple hypothesis testing.

2. The authors should consider analysis of relapse as a continuous variable, using % UDS +, instead of defining relapse as a single + UDS.

3. ~10-15% of the sample had missing data, but is it not possible to obtain the methadone dose or weight from the EHR?

4. The authors should report the numbers of patients in each of the categories for the two binary phenotypes in Table 2.

5. The authors might consider the use of generalized estimating equations in analysis of these data.

Reviewer #2: Comments to the Authors:

This manuscript employs a relatively large population of European MMT patients to examine the pharmacogenetic effects of variants in OPRM1 and CYP2B6 on MMT outcome measures. The authors further explore the effect of sex on these outcome measures (which the authors state has not been done in previous studies in the literature). While enthusiastic about the data set, the reviewer has reservations about the lack of information in the description of the study, the statistical methods and about the outcome measures used. Specifically, the outcome measures used by these authors are not consistent with those used in some previous literature, and, thus, cannot be used as a test of replication of previous findings (which is a stated objective of this manuscript). The reviewer believes that although the data analysis was likely done correctly, the manuscript could be improved before publication.

Major issues:

The authors state that sex-based analyses have not been done or have been overlooked by past studies. The reviewer finds the literature contains at least one study (Crist et al, 2018a) on OPRM1 genotypes and MMT efficacy to have used sex as a covariate in their statistical analyses. Although the Crist et al, 2018a study had fewer participants on MMT, sex was considered in their analysis, and the authors found no effect of sex on the measured outcome.

The authors state that the objective of the present study is to “replicate findings from the literature within a larger sample of European descent”. However, the defined outcome measures in the current study are not equivalent to those used in previous studies in the literature. The authors should reference those studies in the literature in which their outcome definitions have been used. For example, these authors employ binary data, rather than percent positive UTS over time by rs10485058 genotype (Crist et al, 2018a), as their outcome measure for “continued opioid use”. In addition, the authors use initial methadone dose at intake, rather than maximum methadone dose (Crist et al, 2018b; ref 28 - Smith et al, 2017), as their outcome measure (but, see question below about methadone dose at time of GENOA enrollment).

How long was each participant on MMT before enrollment in GENOA? The average time on MMT in the GENOA study was 3.7 years (Table 2), but the standard deviation is quite large. Did participants in the current study need to be on MMT for a minimum defined time to be included? If so, what was that minimum time requirement? The authors state that “primary sources of information used” were “data collected at baseline (enrollment in the study), 3 months prior to study enrollment, and up to 12 months follow up”. So, does the current manuscript encompass only data from a 15 months period that includes 3 months before enrollment into GENOA and 12 months of follow-up of participants after GENOA enrollment? Was UTS data available before enrollment of participants in the GENOA study, specifically at 3 months prior to enrollment? Were all participants enrolled in GENOA on a stable (maximal) dose of methadone at the time of enrollment, or were some participants enrolled before or during methadone titration up to a stable dose?

The authors should indicate how often UTS were done throughout the MMT period. Were they weekly or monthly? How were missed UTS appointments handled? Were they treated as missing or as positive? The definitions of “continued opioid use” and “relapse” do not seem rigorous enough because some level of continued opioid use is anticipated in real world settings of MMT, especially at the start of treatment. The reviewer suggests an analysis using percent positive UTS over time by genotype to determine MMT efficacy (see Crist et al, 2018a) especially at the start of MMT. Similarly, depending on the time on MMT of patients in the current study, the definition of relapse could be defined as more than 1 positive UTS after 3 months of clean UTS because some “breakthrough use” is to be expected as the clinician titrates the methadone dose to an effective level. The maximum dose of methadone is a better indication of therapeutic dose than the initial dose of methadone at intake (but, see question above about stable dose of methadone at time of GENOA enrollment). This is especially true for pharmacokinetic metabolic gene analysis, such as CYP2B6.

Regarding the discussion of SNP rs3745274 (*9 variant of CYP2B6), it is the R-enantiomer of methadone that is the therapeutic enantiomer and binds 50 times more strongly to the MOR (encoded by OPRM1), not the S-enantiomer. Because the authors did not determine the effect of rs3745274 on the R:S-enantiomer ratio, it is speculative to suggest that “[its] effect on continued opioid use in MMT patients could be explained as a decrease in the CYP2B6 gene activity, which could increase plasma methadone concentrations and subside the need for additional opioid intake.” The S-enantiomer is metabolized by CYP2B6, but the *9 variant is predicted to be benign by SIFT and PolyPhen, and was found not to significantly alter mean plasma methadone or the methadone:EDDP ratio (Figure 1 in reference 21 – Ahmad et al, 2017). Also, in Ahmad et al, the significant finding for rs3745274 is likely due to a skewing of the minor allele frequencies in different directions among the controls and methadone groups (overall predicted MAF=27%, controls=22%, methadone=31%) due to sampling error in small group sizes. (The reviewer also points out that the Ahmad et al study considered sex in their statistical analysis). Previous literature (ref 20 – Levran et al, 2013) indicates that it is the CYP2B6*6 haplotype (combined rs2279343 *4 with rs3745274 *9 genotypes) that led to differences in methadone dose requirements. The authors provide an image of the LD structure for CYP2B6 in their study population, but do not give the actual D’ and r2 values. From the image, it appears that although LD is high between *4 and *9 SNPs, it is not perfect. As such, to attempt to replicate previous findings in the literature, the authors should re-run the statistical analysis using the haplotype of these two SNPs, as was done in Levran et al, 2013.

What statistics program did the authors use to impute the missing covariate data? How many imputations did the authors run? If missing data imputations were run in R, the authors should state the package they used – for instance, did they use ‘norm’? Could the authors have instead used the 'keep-pheno-on-missing-cov' option for covariates in PLINK 1.07? Did the authors test models other than the ‘Genotypic’ (ADD) model? Other genetic models were used in previous literature (see Crist et al, 2018a). For covariates that were not significant in the full model, the authors should re-run the statistics without them in the model to determine whether the significance of the outcome measures change.

Minor issues:

Some of the references are incorrect. For example, reference 32 in the manuscript is supposed to reference R (page 11, line 208) but is listed in the references as Cahn et al, 2011, J. Wildl Manage 75(8):1753-66. Was this supposed to be reference 33 for R as listed in the references section?

The authors should state which analyzed SNPs were directly genotyped (on the chip) and which were imputed.

The authors should clarify what the reference condition was for each statistical analysis. For instance, was the reference condition defined as an individual ‘having the minor allele and being positive for the particular outcome measure’, or was it ‘having the minor allele and not being positive for the outcome measure’?

References mentioned (but not referenced in the manuscript):

Crist RC, Doyle GA, Nelson EC, Degenhardt L, Martin NG, Montgomery GW, Saxon AJ, Ling W, Berrettini WH. A polymorphism in the OPRM1 3'-untranslated region is associated with methadone efficacy in treating opioid dependence. Pharmacogenomics J. 2018a Jan;18(1):173-179. doi: 10.1038/tpj.2016.89.

Crist RC, Li J, Doyle GA, Gilbert A, Dechairo BM, Berrettini WH. Pharmacogenetic analysis of opioid dependence treatment dose and dropout rate. Am J Drug Alcohol Abuse. 2018b ;44(4):431-440. doi: 10.1080/00952990.2017.1420795.

Reviewer #3: This paper aims to test associations between several OPRM1 and CYP2B6 SNPs on several methadone maintenance treatment outcomes in males and females separately.

The strength of the study is the large sample size as well as the high number of females, which is rare in MMT studies. Contradictory findings have been published on these SNPs, mostly in smaller studies, thus a large replicate study is interesting.

My main concerns of the manuscript are as follows:

- In the abstract and in the results sections (also in the tables), the results are always described in such a way that they first seem to be statistically significant, while they are not.

- Most included patients were taking methadone but it seems from table 2 that a few took suboxone. Why they were not excluded as the study is on MMT?

- For the outcomes on continued opioid use and relapse, they should be clarified. The frequency of UTS should be indicated to better describe these outcomes. Is the duration of 3 to 15 months for all patients?

- For the methadone dose outcome, more information on how long the dose was unchanged before inclusion should be added. Detailed descriptive statistics of the dose should be added (range, normal distribution in the population, …).

- The outcome of the methadone dose is questionable as methadone metabolism displays a high interindividual variability and as patients might be in different phase of their MMT (treatment introduction, stabilization or slow tapering of the dose for example). Also, in certain MMT prescribing center, there is a maximal dose not be exceeded but that could remain insufficient for certain patients who therefore continue to use opiates.

In general, when discussing the individual SNPs, it will help to include the gene name before the rs number (ex. OPRM1 rs73568641)

In the study design and setting section (p.5, line 107): it is not clear which data was collected 3 months prior to study enrollment.

In Table 2, the significantly different variables between male and female should be indicated (*). The opioid prescription variable should be clarified.

In the discussion (p.16, line 258): no trend was observed for rs73568641 on relapse in female.

**********

6. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: Yes: Wade Berrettini, MD, PhD

Reviewer #2: No

Reviewer #3: No

[NOTE: If reviewer comments were submitted as an attachment file, they will be attached to this email and accessible via the submission site. Please log into your account, locate the manuscript record, and check for the action link "View Attachments". If this link does not appear, there are no attachment files.]

While revising your submission, please upload your figure files to the Preflight Analysis and Conversion Engine (PACE) digital diagnostic tool, https://pacev2.apexcovantage.com/. PACE helps ensure that figures meet PLOS requirements. To use PACE, you must first register as a user. Registration is free. Then, login and navigate to the UPLOAD tab, where you will find detailed instructions on how to use the tool. If you encounter any issues or have any questions when using PACE, please email PLOS at figures@plos.org. Please note that Supporting Information files do not need this step.

10 Aug 2021

Reviewer #1: The authors performed a genetic association study of methadone maintenance treatment (MMT) of opioid use disorder among ~1100 persons of European origin (~65% male) who had urine drug screen (UDS) data and methadone plasma levels available over a 3-15 month period, across a consortium of MMT clinics in Canada. They did array genotyping which yielded ~5 million SNPs after imputation. Despite this wealth of genetic data, they report on 3 OPRM1 and 2 CYP2B6 SNPs. There were 3 phenotypes: methadone plasma levels, any lapse to illicit opioid use (a single + UDS) and relapse (a single + UDS after at least 3 months of – UDS results). Logistic regression and linear regression analyses were implemented. Sex-specific results are reported. There were no statistically significant results. The following comments are relevant:

1. The authors must be clear about whether they analyzed (or plan to analyze) the methadone dose and other phenotypes using other SNPs, or whether they restricted analysis to only those 5 SNPs reported here. This has implications for multiple hypothesis testing.

Response:

Thank you for your review and feedback. We would like to clarify that we do not report on methadone plasma level, we report methadone dose and opioid use. This study is a hypothesis-driven study, the SNPs presented in this candidate gene study have been selected a priori to be tested for the continued opioid use, relapse, and methadone dose phenotypes. Analysis was restricted to the outlined OPRM1 SNPs (rs73568641, rs7451325, rs10485058, rs1799971) and CYP2B6 SNPs (rs2279343, rs10403955, rs8192719, rs3745274) as highlighted in the ‘Objectives’ section as well as in Table 1. Future analysis that may include hypothesis-free testing will not impact the current study, and any future publications will cite and refer to this study to ensure readers are aware of any relevant publications from the same study sample.

2. The authors should consider analysis of relapse as a continuous variable, using % UDS +, instead of defining relapse as a single + UDS.

Response:

Though the reviewer’s suggestion is insightful, the measure of relapse as a %UDS positive would not be an accurate representation of the event. The use of %UDS positive has been outlined in the previous literature as a way to measure continued opioid use. It would not be accurate in measuring relapse events as it is not indicative of an individual returning to opioid use (testing positive) following a period of no opioid use (testing negative). To best represent this pattern, relapse was determined a priori to be measured as any opioid positive UDS following at least 3 months of opioid negative UDSs. Having it as a binary variable as opposed to a continuous one allows for all incidences of relapse to be viewed as such, in a population that is not normally distributed. In addition, %UDS will be challenging to interpret as how much change is clinically relevant.

3. ~10-15% of the sample had missing data, but is it not possible to obtain the methadone dose or weight from the EHR?

Response:

The methadone dose and weight measurements used in this study’s analyses were collected at the time of the participant interview as part of the greater GENOA study. The missing methadone dose data only contributed to less than 0.6% (7/1172) of the total analyzed population for the methadone dose phenotype. The missing weight data only contributed to 0.94% (11/1165) of the analyzed methadone dose sample, for which weight was considered a covariate. Since missingness in these variables was marginal and accounted for less than 10% of the sample population, any missing values were imputed using mean substitution when these variables were used as covariates, so as to not further affect the sample sizes.

For samples used in measures of continued opioid use (n=1129) and relapse (n=944), the smaller sample sizes were due to the exclusion of any participants who did not have UTSs assessing the presence of opioids for a minimum duration of 3 months and 6 months, respectively. Further exclusionary reasons included participants taking prescription opioid medications. These exclusionary reasons have been specified under the “Eligibility criteria” section of this manuscript and been discussed as limitations under “Limitations and generalizability”.

4. The authors should report the numbers of patients in each of the categories for the two binary phenotypes in Table 2.

Response:

Thank you for the suggestion. We have revised Table 2 to also include a summary breakdown of patient outcomes for the continued opioid use and relapse phenotypes measured.

5. The authors might consider the use of generalized estimating equations in analysis of these data.

Response:

The GEE model is best suited for longitudinal data and correlated observations; our data are binary and not looking at the individual in different time points for the primary study outcomes. Though analysis through generalized estimating equations might be helpful in determining the genetic variant-phenotype relationship in a longitudinal study such as the GENOA study, where outcomes of continued opioid use measured at multiple timepoints, the outcome is dichotomized, and it was not deemed the most appropriate method of analysis a priori. Generalized estimating equations are best used, for example, in family studies or multiple cohorts where multiple sources of correlation exist within the sample population.

Reviewer #2: Comments to the Authors:

This manuscript employs a relatively large population of European MMT patients to examine the pharmacogenetic effects of variants in OPRM1 and CYP2B6 on MMT outcome measures. The authors further explore the effect of sex on these outcome measures (which the authors state has not been done in previous studies in the literature). While enthusiastic about the data set, the reviewer has reservations about the lack of information in the description of the study, the statistical methods and about the outcome measures used. Specifically, the outcome measures used by these authors are not consistent with those used in some previous literature, and, thus, cannot be used as a test of replication of previous findings (which is a stated objective of this manuscript). The reviewer believes that although the data analysis was likely done correctly, the manuscript could be improved before publication.

Response:

Thank you for your review and feedback. Please see below for detailed responses.

Major issues:

The authors state that sex-based analyses have not been done or have been overlooked by past studies. The reviewer finds the literature contains at least one study (Crist et al, 2018a) on OPRM1 genotypes and MMT efficacy to have used sex as a covariate in their statistical analyses. Although the Crist et al, 2018a study had fewer participants on MMT, sex was considered in their analysis, and the authors found no effect of sex on the measured outcome.

Response:

Though many studies adjust for sex in their analysis models, as was done in the referenced Crist et al study, the analysis is not designed a priori to test for differences within or between sex groups. To date and to the knowledge of the authors, very few have sought to specifically test the genetic contribution to MMT outcomes using sex-stratified analyses, considering findings separately within females and within males. The language used in the manuscript has been adjusted to reflect that.

The authors state that the objective of the present study is to “replicate findings from the literature within a larger sample of European descent”. However, the defined outcome measures in the current study are not equivalent to those used in previous studies in the literature. The authors should reference those studies in the literature in which their outcome definitions have been used. For example, these authors employ binary data, rather than percent positive UTS over time by rs10485058 genotype (Crist et al, 2018a), as their outcome measure for “continued opioid use”. In addition, the authors use initial methadone dose at intake, rather than maximum methadone dose (Crist et al, 2018b; ref 28 - Smith et al, 2017), as their outcome measure (but, see question below about methadone dose at time of GENOA enrollment).

Response:

Though many studies have previously measured opioid positive urine tests as a continuous response outcome (i.e. the above referenced Crist et al, 2018a paper), our current study and previous literature have shown that opioid use is not normally distributed within this sample population, with a mean of about 20-50% of opioid urine screens testing positive (Bawor et al, 2015; Kamal et al, 2007; Hser et al, 2014). Please see a comment below elaborating on study data distribution and normality regarding positive UTS results. Numerous studies have also opted for measuring treatment response as a binary variable, as shown in Table 4 of Crist et al’s published review (Crist et al, 2018).

With regards to the methadone dose measure, we have chosen to analyze this phenotype as the dose reported at enrollment since it was collected at the time of the participant interview as part of the greater GENOA study. We acknowledge that this measure of methadone dose might not be the most accurate indicator of MMT response as patients could be at different stages within their treatment. Though we have tried to account for this variability by adjusting for participant duration on MMT, we have discussed this limitation under the “Limitations and generalizability” section of the revised manuscript.

As the main objective of this study was to test the association between the select SNPs outlined (as proven by the literature to be biologically relevant) and the phenotypes of continued opioid use, relapse, and methadone dose, a focus was placed on examining and reporting these associations within our larger sample population as opposed to replicating the specific findings of previous research, matching their outcome measures and sample population. We have modified the language within our objectives to clarify this intention.

References mentioned in response:

Bawor, M., Dennis, B. B., Tan, C., Pare, G., Varenbut, M., Daiter, J., ... & Samaan, Z. (2015). Contribution of BDNF and DRD2 genetic polymorphisms to continued opioid use in patients receiving methadone treatment for opioid use disorder: an observational study. Addiction science & clinical practice, 10(1), 1-9.

Kamal, F., Flavin, S., Campbell, F., Behan, C., Fagan, J., & Smyth, R. (2007). Factors affecting the outcome of methadone maintenance treatment in opiate dependence. Irish medical journal, 100(3), 393-397.

Hser, Y. I., Saxon, A. J., Huang, D., Hasson, A., Thomas, C., Hillhouse, M., ... & Ling, W. (2014). Treatment retention among patients randomized to buprenorphine/naloxone compared to methadone in a multi‐site trial. Addiction, 109(1), 79-87.

Crist, R. C., Clarke, T. K., & Berrettini, W. H. (2018). Pharmacogenetics of opioid use disorder treatment. CNS drugs, 32(4), 305-320.

How long was each participant on MMT before enrollment in GENOA? The average time on MMT in the GENOA study was 3.7 years (Table 2), but the standard deviation is quite large. Did participants in the current study need to be on MMT for a minimum defined time to be included? If so, what was that minimum time requirement?

Response:

For the GENOA study, there was no minimum duration on MMT required for participants to be included. There was also no minimum amount of time required to be on MMT for the analysis conducted in this current study, so as to not limit the sample size. However, for the outcomes of continued opioid use and relapse, participants had to have a minimum of 3 months’ worth of urine screens while on MMT to be able to quantify these outcomes.

To account for the variability in the treatment duration across participants, all statistical models adjusted for duration on MMT in months as a covariate.

The authors state that “primary sources of information used” were “data collected at baseline (enrollment in the study), 3 months prior to study enrollment, and up to 12 months follow up”. So, does the current manuscript encompass only data from a 15 months period that includes 3 months before enrollment into GENOA and 12 months of follow-up of participants after GENOA enrollment?

Response:

Yes, that is correct. The current manuscript only includes data from urine toxicology screens (UTSs) collected from the GENOA study at 3-month intervals. Since data availability were inconsistent across patients, a duration range of 3 to 15 months was used to prevent the exclusion of patients who did not have more than 3 months’ worth of UTS results recorded. The 15 months period includes data collected 3 months prior to enrollment into GENOA, at baseline, and up to a 12 month follow up after GENOA enrollment, if available. All other data included in this current study were collected at baseline (time of enrollment in the GENOA study). We have revised the language used under the “Study design and setting” section of this manuscript to more clearly explain that.

Was UTS data available before enrollment of participants in the GENOA study, specifically at 3 months prior to enrollment?

Response:

These data were available in the medical records for various durations depending on when the participant started the treatment program, however not accessible to the research team until after the participant was recruited and consented for access to medical records. When available, based on if the participants were enrolled in the MMT program for 3 months prior to study enrollment (baseline), UTS data for 3 months prior to enrollment were used to measure outcomes of continued opioid use and relapse for the respective participants.

Were all participants enrolled in GENOA on a stable (maximal) dose of methadone at the time of enrollment, or were some participants enrolled before or during methadone titration up to a stable dose?

Response:

As outlined earlier, not all participants enrolled in the GENOA study were on a stable (maximal) dose at the time of enrollment. Participants were included at any stage of treatment while on MMT, including induction, stabilization, treatment, and tapering. On average, participants have been in the treatment program for 3.7 years. Though this was a limitation due to data unavailability, the duration on MMT and methadone dose variables were accounted for in the statistical models where appropriate. This limitation has been further discussed in the “Limitations and generalizability” section of the revised manuscript.

The authors should indicate how often UTS were done throughout the MMT period. Were they weekly or monthly?

Response:

UTSs for opioids were performed regularly by clinics on a weekly or sometimes biweekly basis, and then recorded in the GENOA study at 3-month intervals. Please find the UTS frequency information included under the section “Study design and setting” of the revised manuscript.

How were missed UTS appointments handled? Were they treated as missing or as positive?

Response:

As we did not have data on missed UTS appointments, any UTS data that were not recorded and represented in the minimum of 3 months amalgamated UTS results in the GENOA study were handled as missing and not included in the analysis. Since data availability were inconsistent across participants, a duration range of 3 to 15 months was used to prevent the exclusion of participants who did not have more than 3 months’ worth of UTS results recorded.

The definitions of “continued opioid use” and “relapse” do not seem rigorous enough because some level of continued opioid use is anticipated in real world settings of MMT, especially at the start of treatment. The reviewer suggests an analysis using percent positive UTS over time by genotype to determine MMT efficacy (see Crist et al, 2018a) especially at the start of MMT. Similarly, depending on the time on MMT of patients in the current study, the definition of relapse could be defined as more than 1 positive UTS after 3 months of clean UTS because some “breakthrough use” is to be expected as the clinician titrates the methadone dose to an effective level.

Response:

The feedback is really appreciated. Though the reviewer’s suggestion is not incorrect, defining continued opioid use as percent positive UTS over time would not be the most ideal way of measuring that variable as continued opioid use is not normally distributed within the MMT patient population, as explained in the comments above. For the reviewers and not included as part of the manuscript, please see below the Q-Q plot and data distribution histogram of continued opioid use measured as a continuous variable of percent positive opioid UTS out of total opioid UTS. As displayed by the histogram, the data are skewed towards low % positive opioid UTS. When tested for normality by the Shapiro-Wilk test and as can be observed by the Q-Q plot, the variable was not normally distributed. The data also could not be transformed into a normal distribution as was evident by log and inverse transformations. Thus, the authors have opted to measure continued opioid use as binary incidence outcomes and analyze the association via a logistic regression.

In terms of the definition of relapse, we have opted to measure incidences of relapse as any positive UTS results following 3 months of negative ones. In doing so, we have opted for a more conservative approach in identifying those who have relapsed. We do acknowledge that there is variability in opioid use trends across patients on MMT and as a result cannot account for all scenarios within our model while accurately representing the population. We have included that as a limitation discussed in our manuscript.

The maximum dose of methadone is a better indication of therapeutic dose than the initial dose of methadone at intake (but, see question above about stable dose of methadone at time of GENOA enrollment). This is especially true for pharmacokinetic metabolic gene analysis, such as CYP2B6.

Response:

We agree that having the maximum dose of methadone would have been a helpful measure in the association with SNPs in the CYP2B6. However, dose at intake for the study does not mean this was the initiation dose. 823 participants have been in the treatment program for 1 year or longer. It is expected that the dose may change during the treatment program, and therefore it is challenging to identify the maximum dose. Though we have tried to account for variability in methadone dose across participants at different stages of MMT by adjusting for participant duration on MMT, we have discussed this limitation under the “Limitations and generalizability” section of the revised manuscript.

Regarding the discussion of SNP rs3745274 (*9 variant of CYP2B6), it is the R-enantiomer of methadone that is the therapeutic enantiomer and binds 50 times more strongly to the MOR (encoded by OPRM1), not the S-enantiomer. Because the authors did not determine the effect of rs3745274 on the R:S-enantiomer ratio, it is speculative to suggest that “[its] effect on continued opioid use in MMT patients could be explained as a decrease in the CYP2B6 gene activity, which could increase plasma methadone concentrations and subside the need for additional opioid intake.” The S-enantiomer is metabolized by CYP2B6, but the *9 variant is predicted to be benign by SIFT and PolyPhen, and was found not to significantly alter mean plasma methadone or the methadone:EDDP ratio (Figure 1 in reference 21 – Ahmad et al, 2017). Also, in Ahmad et al, the significant finding for rs3745274 is likely due to a skewing of the minor allele frequencies in different directions among the controls and methadone groups (overall predicted MAF=27%, controls=22%, methadone=31%) due to sampling error in small group sizes. (The reviewer also points out that the Ahmad et al study considered sex in their statistical analysis). Previous literature (ref 20 – Levran et al, 2013) indicates that it is the CYP2B6*6 haplotype (combined rs2279343 *4 with rs3745274 *9 genotypes) that led to differences in methadone dose requirements.

Response:

Thank you for this feedback. We have reworded the language in the discussion section to be more accurate of the findings in the literature and not be speculative. We have also provided a response for the inclusion of sex as a covariate in the literature in a previous comment.

The authors provide an image of the LD structure for CYP2B6 in their study population, but do not give the actual D’ and r2 values. From the image, it appears that although LD is high between *4 and *9 SNPs, it is not perfect. As such, to attempt to replicate previous findings in the literature, the authors should re-run the statistical analysis using the haplotype of these two SNPs, as was done in Levran et al, 2013.

Response:

The LD structure for CYP2B6 along with the r-squared values can be found in Figure B of S2 File. The r-squared values are those highlighted within the plot squares. We have revised the figure caption to clarify that. Due to a high r-squared coefficient between rs2279343 (*4) and rs3745274 (*9) of 81, they were deemed to be in high LD and were treated as such. Though replicating findings of CYP2B6 haplotype analyses would have been interesting, conducting a haplotype analysis for this study would not provide additional information on variant-phenotype associations as SNPs rs2279343 and rs3745274 (which was tested) are in high LD.

What statistics program did the authors use to impute the missing covariate data? How many imputations did the authors run? If missing data imputations were run in R, the authors should state the package they used – for instance, did they use ‘norm’? Could the authors have instead used the 'keep-pheno-on-missing-cov' option for covariates in PLINK 1.07?

Response:

No statistical package was used for the missing phenotypic data imputations. Averages of the duration on MMT and methadone dose values were used to substitute for any missing values in excel. The term ‘mean substitution’ has been used to clarify the meaning behind imputed variables within this manuscript.

Did the authors test models other than the ‘Genotypic’ (ADD) model? Other genetic models were used in previous literature (see Crist et al, 2018a).

Response:

Within the scope of this manuscript, only an additive model was used to test statistical associations. As opioid use and MMT response are complex traits with multiple genes and multiple variants of a gene contributing to their outcomes, an additive model was seen as best fitting.

The referenced Crist et al study has also reported using an additive genetic model to test the association between the rs10485058 genotype with their outcome of self-reported relapse.

For covariates that were not significant in the full model, the authors should re-run the statistics without them in the model to determine whether the significance of the outcome measures change.

Response:

Thank you for the suggestion, however, we retained these covariates for their clinical relevance and not statistical significance. If we are to exclude these variables, the results may be confounded.

Minor issues:

Some of the references are incorrect. For example, reference 32 in the manuscript is supposed to reference R (page 11, line 208) but is listed in the references as Cahn et al, 2011, J. Wildl Manage 75(8):1753-66. Was this supposed to be reference 33 for R as listed in the references section?

Response:

Thank you for highlighting this. The references list and in-text citations have been fixed to match the references. Please note that due to this change being implemented through an imbedded reference manager, the change in reference numbers will not be viewed as tracked.

The authors should state which analyzed SNPs were directly genotyped (on the chip) and which were imputed.

Response:

We have included data on which SNPs from those of interest were imputed in the S2 File.

SNPs that were directly genotyped include: rs10485058, rs1799971, rs8192719, rs3745274.

SNPs that were imputed include: rs73568641, rs7451325, rs2279343, rs10403955.

The authors should clarify what the reference condition was for each statistical analysis. For instance, was the reference condition defined as an individual ‘having the minor allele and being positive for the particular outcome measure’, or was it ‘having the minor allele and not being positive for the outcome measure’?

Response:

All results reported in this manuscript, as outlined in Table 3 and Table 4, test for the minor allele. The table legends specify that the minor allele is the reference and tested allele. In terms of outcomes, the statistical analyses test for having the minor allele and being positive for a phenotype (i.e. cases) in the logistic regressions. The language under the “Statistical methods” section has been revised to reflect this.

References mentioned (but not referenced in the manuscript):

Crist RC, Doyle GA, Nelson EC, Degenhardt L, Martin NG, Montgomery GW, Saxon AJ, Ling W, Berrettini WH. A polymorphism in the OPRM1 3'-untranslated region is associated with methadone efficacy in treating opioid dependence. Pharmacogenomics J. 2018a Jan;18(1):173-179. doi: 10.1038/tpj.2016.89.

Crist RC, Li J, Doyle GA, Gilbert A, Dechairo BM, Berrettini WH. Pharmacogenetic analysis of opioid dependence treatment dose and dropout rate. Am J Drug Alcohol Abuse. 2018b ;44(4):431-440. doi: 10.1080/00952990.2017.1420795.

Reviewer #3: This paper aims to test associations between several OPRM1 and CYP2B6 SNPs on several methadone maintenance treatment outcomes in males and females separately.

The strength of the study is the large sample size as well as the high number of females, which is rare in MMT studies. Contradictory findings have been published on these SNPs, mostly in smaller studies, thus a large replicate study is interesting.

My main concerns of the manuscript are as follows:

- In the abstract and in the results sections (also in the tables), the results are always described in such a way that they first seem to be statistically significant, while they are not.

Response:

Thank you for highlighting this. We have revised the language in the abstract and results sections to be more reflective of the actual results of the study.

- Most included patients were taking methadone but it seems from table 2 that a few took suboxone. Why they were not excluded as the study is on MMT?

Response:

Table 2 showcases the overall study sample that was collected as part of the GENOA study and had genotyped data available (please refer to Figure A in S2 File for a more detailed flow diagram of the sample size). Participants included in this study’s analyses, however, were only those on methadone maintenance treatment (n=1172) as the sample size of those on suboxone was too small for analysis (n=52). Anyone who was administered suboxone was excluded, as mentioned under the section “Eligibility criteria”. For the purpose of showcasing the data availability for this sample, we have displayed numbers of participants from the GENOA study who were administered methadone versus suboxone in Table 2. We have corrected the abstract to clarify that 1172 participants treated with methadone were included in this study.

- For the outcomes on continued opioid use and relapse, they should be clarified. The frequency of UTS should be indicated to better describe these outcomes. Is the duration of 3 to 15 months for all patients?

Response:

Urine toxicology screens (UTSs) for opioids were performed regularly on a weekly or sometimes biweekly basis, and then recorded in the GENOA study at 3-month intervals. Since data availability were inconsistent across patients, a duration range of 3 to 15 months was used to prevent the exclusion of patients who did not have more than 3 months’ worth of UTS results recorded. The mean total number of opioid UTSs is 74.34 (+/- 34.64), as reported in Table 2. Incomplete data could have been a result of patient incarcerations, relocations, hospitalizations, mortality, or other unaccounted outcomes.

To best account for the differences in the duration of opioid UTSs, duration on MMT was included as a covariate in all analysis models.

Please find the UTS frequency information also included under the section “Study design and setting”.

- For the methadone dose outcome, more information on how long the dose was unchanged before inclusion should be added. Detailed descriptive statistics of the dose should be added (range, normal distribution in the population, …).

Response:

Though reporting on how long the dose was unchanged prior to participant inclusion in the study would have been informative, that information is unfortunately unavailable. At the time of enrollment in the study, participants were asked if their dose had remained stable for the past 3 months or if it had changed. Because the yes/no responses were self-reported, we cannot ascertain that they were accurate or be sure of the direction of that change (increased or decreased dose). As a result, this measure was not found to be informative and was not included in the manuscript. Also, methadone doses have the potential to change on a weekly basis, as clinically indicated. We acknowledge that the stability of the methadone dose is a limitation of the study and have included that in the limitations’ discussion of the manuscript.

We have also revised Table 2 to include more detailed descriptive statistics on the methadone dose, including the range of methadone doses reported.

- The outcome of the methadone dose is questionable as methadone metabolism displays a high interindividual variability and as patients might be in different phase of their MMT (treatment introduction, stabilization or slow tapering of the dose for example). Also, in certain MMT prescribing center, there is a maximal dose not be exceeded but that could remain insufficient for certain patients who therefore continue to use opiates

Response:

That is a great point. We agree that there is interindividual variability in the level of methadone metabolism across patients and did anticipate that to be represented in the association between the CYP2B6 gene SNPs and the methadone dose outcome.

With regards to patients being in different phases of their MMT, we unfortunately were not able to precisely account for that, although on average patients were in treatment for 3.7 years, and have thus reported it as a limitation of the study. Though as mentioned earlier, we did account for the patients’ duration on MMT in the statistical models.

Regarding the maximal dose of methadone provided by treatment centers, we do acknowledge that there are certain maximal doses that cannot be exceeded and that the methadone dose affects treatment outcomes of patients, and have as such accounted for methadone dose as a covariate in the measure of continued opioid use.

In general, when discussing the individual SNPs, it will help to include the gene name before the rs number (ex. OPRM1 rs73568641)

Response:

We have revised the manuscript text where appropriate to include the gene name before outlining the rs-ID of the SNP of interest.

In the study design and setting section (p.5, line 107): it is not clear which data was collected 3 months prior to study enrollment.

Response:

Only UTS results were collected 3 months prior to study enrollment. We have revised the language under the “Study design and setting section” to clarify that.

In Table 2, the significantly different variables between male and female should be indicated (*). The opioid prescription variable should be clarified.

Response:

Differences between the male and female variables shown in Table 2 were calculated, p-values displayed, and significant differences outlined with an asterisk. We have further revised the “opioid prescription” variable to say “Participant taking opioid prescription” for clarity.

In the discussion (p.16, line 258): no trend was observed for rs73568641 on relapse in female.

Response:

Thank you for highlighting that. We have revised the text in the discussion to reflect the results of the study more accurately.

Submitted filename: Response to Reviewers.docx

Click here for additional data file.

12 Oct 2021

PONE-D-21-11145R1Implications of OPRM1 and CYP2B6 variants on treatment outcomes in methadone-maintained patients in Ontario: Exploring sex differencesPLOS ONE

Dear Dr. Samaan,

Thank you for submitting your manuscript to PLOS ONE. After careful consideration, we feel that it has merit but does not fully meet PLOS ONE’s publication criteria as it currently stands. Therefore, we invite you to submit a revised version of the manuscript that addresses the points raised during the review process.

Please address a couple of very minor issues from Reviewer 3.

Please ensure that your decision is justified on PLOS ONE’s publication criteria and not, for example, on novelty or perceived impact.

Please submit your revised manuscript by Nov 26 2021 11:59PM. If you will need more time than this to complete your revisions, please reply to this message or contact the journal office at plosone@plos.org. When you're ready to submit your revision, log on to https://www.editorialmanager.com/pone/ and select the 'Submissions Needing Revision' folder to locate your manuscript file.

Please include the following items when submitting your revised manuscript:If you would like to make changes to your financial disclosure, please include your updated statement in your cover letter. Guidelines for resubmitting your figure files are available below the reviewer comments at the end of this letter.

A rebuttal letter that responds to each point raised by the academic editor and reviewer(s). You should upload this letter as a separate file labeled 'Response to Reviewers'.

A marked-up copy of your manuscript that highlights changes made to the original version. You should upload this as a separate file labeled 'Revised Manuscript with Track Changes'.

An unmarked version of your revised paper without tracked changes. You should upload this as a separate file labeled 'Manuscript'.

If applicable, we recommend that you deposit your laboratory protocols in protocols.io to enhance the reproducibility of your results. Protocols.io assigns your protocol its own identifier (DOI) so that it can be cited independently in the future. For instructions see: https://journals.plos.org/plosone/s/submission-guidelines#loc-laboratory-protocols. Additionally, PLOS ONE offers an option for publishing peer-reviewed Lab Protocol articles, which describe protocols hosted on protocols.io. Read more information on sharing protocols at https://plos.org/protocols?utm_medium=editorial-email&utm_source=authorletters&utm_campaign=protocols.

We look forward to receiving your revised manuscript.

Kind regards,

Huiping Zhang

Academic Editor

PLOS ONE

Journal Requirements:

Please review your reference list to ensure that it is complete and correct. If you have cited papers that have been retracted, please include the rationale for doing so in the manuscript text, or remove these references and replace them with relevant current references. Any changes to the reference list should be mentioned in the rebuttal letter that accompanies your revised manuscript. If you need to cite a retracted article, indicate the article’s retracted status in the References list and also include a citation and full reference for the retraction notice.

[Note: HTML markup is below. Please do not edit.]

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. If the authors have adequately addressed your comments raised in a previous round of review and you feel that this manuscript is now acceptable for publication, you may indicate that here to bypass the “Comments to the Author” section, enter your conflict of interest statement in the “Confidential to Editor” section, and submit your "Accept" recommendation.

Reviewer #1: All comments have been addressed

Reviewer #3: (No Response)

**********

2. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #1: Partly

Reviewer #3: Yes

**********

3. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: Yes

Reviewer #3: Yes

**********

4. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #1: Yes

Reviewer #3: Yes

**********

5. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #1: Yes

Reviewer #3: Yes

**********

6. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #1: (No Response)

Reviewer #3: Thank you for your response.

My comments have been addressed except for the addition of the range (min-max or interquartile) of methadone dose in Table 2. The number of significant digits (for methadone dose but not only) used in table 2 should also be reviewed.

And there is an inversion in table 2 in the Continued opioid use frequency.

**********

7. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: Yes: Wade Berrettini

Reviewer #3: No

[NOTE: If reviewer comments were submitted as an attachment file, they will be attached to this email and accessible via the submission site. Please log into your account, locate the manuscript record, and check for the action link "View Attachments". If this link does not appear, there are no attachment files.]

While revising your submission, please upload your figure files to the Preflight Analysis and Conversion Engine (PACE) digital diagnostic tool, https://pacev2.apexcovantage.com/. PACE helps ensure that figures meet PLOS requirements. To use PACE, you must first register as a user. Registration is free. Then, login and navigate to the UPLOAD tab, where you will find detailed instructions on how to use the tool. If you encounter any issues or have any questions when using PACE, please email PLOS at figures@plos.org. Please note that Supporting Information files do not need this step.

25 Nov 2021

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #1: (No Response)

Reviewer #3: Thank you for your response.

My comments have been addressed except for the addition of the range (min-max or interquartile) of methadone dose in Table 2. The number of significant digits (for methadone dose but not only) used in table 2 should also be reviewed.

And there is an inversion in table 2 in the Continued opioid use frequency.

Response: Thank you for your comments and review. In response to Reviewer #3’s comment, the range that was included in the previous revision as the difference between the highest and lowest methadone doses in Table 2 has been modified to show the “minimum-maximum” values. Table 2 values (of methadone dose and other descriptors) have been edited where applicable to show significant digits that are proportional to the magnitude of the SD values presented. Finally, the inversion in the “Continued opioid use” frequency in Table 2 has been corrected; thank you for noting that.

Submitted filename: Response to Reviewers 2.docx

Click here for additional data file.

29 Nov 2021

Implications of OPRM1 and CYP2B6 variants on treatment outcomes in methadone-maintained patients in Ontario: Exploring sex differences

PONE-D-21-11145R2

Dear Dr. Samaan,

We’re pleased to inform you that your manuscript has been judged scientifically suitable for publication and will be formally accepted for publication once it meets all outstanding technical requirements.

Within one week, you’ll receive an e-mail detailing the required amendments. When these have been addressed, you’ll receive a formal acceptance letter and your manuscript will be scheduled for publication.

An invoice for payment will follow shortly after the formal acceptance. To ensure an efficient process, please log into Editorial Manager at http://www.editorialmanager.com/pone/, click the 'Update My Information' link at the top of the page, and double check that your user information is up-to-date. If you have any billing related questions, please contact our Author Billing department directly at authorbilling@plos.org.

If your institution or institutions have a press office, please notify them about your upcoming paper to help maximize its impact. If they’ll be preparing press materials, please inform our press team as soon as possible -- no later than 48 hours after receiving the formal acceptance. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information, please contact onepress@plos.org.

Kind regards,

Huiping Zhang

Academic Editor

PLOS ONE

Additional Editor Comments (optional):

Reviewers' comments:

6 Dec 2021

PONE-D-21-11145R2

Implications of OPRM1 and CYP2B6 variants on treatment outcomes in methadone-maintained patients in Ontario: Exploring sex differences

Dear Dr. Samaan:

I'm pleased to inform you that your manuscript has been deemed suitable for publication in PLOS ONE. Congratulations! Your manuscript is now with our production department.

If your institution or institutions have a press office, please let them know about your upcoming paper now to help maximize its impact. If they'll be preparing press materials, please inform our press team within the next 48 hours. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information please contact onepress@plos.org.

If we can help with anything else, please email us at plosone@plos.org.

Thank you for submitting your work to PLOS ONE and supporting open access.

Kind regards,

PLOS ONE Editorial Office Staff

on behalf of

Dr. Huiping Zhang

Academic Editor

PLOS ONE