Article
Determination of Genotype Combinations That Can
Predict the Outcome of the Treatment of Alcohol
Dependence Using the 5-HT3 Antagonist Ondansetron
Bankole A. Johnson, D.Sc., M.D.
Chamindi Seneviratne, M.D.
Xin-Qun Wang, M.S.
Nassima Ait-Daoud, M.D.
Ming D. Li, Ph.D.
Objective: The authors previously reported
that the 5:-HTTLPR-LL and rs1042173-TT
(SLC6A4-LL/TT) genotypes in the serotonin
transporter gene predicted a significant
reduction in the severity of alcohol consumption among alcoholics receiving the
5-HT3 antagonist ondansetron. In this
study, they explored additional markers
of ondansetron treatment response in
alcoholics by examining polymorphisms in
the HTR3A and HTR3B genes, which regulate directly the function and binding of
5-HT3 receptors to ondansetron.
Method: The authors genotyped one rare
and 18 common single-nucleotide polymorphisms in HTR3A and HTR3B in the
same sample that they genotyped for
SLC6A4-LL/TT in the previous randomized,
double-blind, 11-week clinical trial. Participants were 283 European Americans who
received oral ondansetron (4 mg/kg of body
weight twice daily) or placebo along
with weekly cognitive-behavioral therapy.
Associations of individual and combined
genotypes with treatment response on
drinking outcomes were analyzed.
Results: Individuals carrying one or
more of genotypes rs1150226-AG and
rs1176713-GG in HTR3A and rs17614942AC in HTR3B showed a significant overall
mean difference between ondansetron
and placebo in drinks per drinking day
(22.50; effect size=0.867), percentage of
heavy drinking days (220.58%; effect
size=0.780), and percentage of days abstinent (18.18%; effect size=0.683). Combining these HTR3A/HTR3B and SLC6A4-LL/TT
genotypes increased the target cohort
from approaching 20% (identified in the
previous study) to 34%.
Conclusions: The authors present initial
evidence suggesting that a combined fivemarker genotype panel can be used to
predict the outcome of treatment of alcohol dependence with ondansetron. Additional, larger pharmacogenetic studies
would help to validate these results.
(Am J Psychiatry 2013; 00:1–12)
A
lcoholism is a heterogeneous, complex disorder with
multiple subtypes or endophenotypes (1). Perhaps because of this heterogeneity, the therapeutic effect sizes of
the approved medications for the treatment of alcohol
dependence have been relatively small (2). A personalized
approach based on the patient’s genetic makeup is
increasingly being investigated for delivering optimum
treatment to suitable patients. Previously, in a randomized
double-blind clinical trial of alcohol-dependent individuals of European descent, we showed that treatment of
severe drinking with ondansetron, a specific serotonin-3
(5-HT3) antagonist, was significantly more efficacious in
a subpopulation carrying the serotonin-transporter-linked
polymorphic region LL (5:-HTTLPR-LL) and rs1042173TT (SLC6A4-LL/TT) genotype combination in the serotonin transporter (5-HTT) gene, SLC6A4 (3). Briefly, in that
study, we found that SLC6A4-LL/TT carriers treated with
ondansetron had 2.63 fewer standard drinks per drinking
day and a percentage of days abstinent 16.99% higher than
noncarriers of the SLC6A4-LL/TT genotypes. In contrast,
when patients were not subgrouped according to their
Am J Psychiatry 00:0, 2013
5:-HTTLPR and rs1042173 genotypes, there was no statistically significant difference between the ondansetron and
placebo groups.
Although the SLC6A4-LL/TT genotypes predicted ondansetron treatment response, ondansetron cannot bind
directly to 5-HTT molecules because it is an antagonist at
the 5-HT3A receptor subunit (4–7). Determination of the
5-HTT genotypes that we selected as being predictive of
ondansetron treatment response in our previous study (3)
was therefore indirect and based on previous in vitro and
in vivo experimental evidence indicating that SLC6A4-LL/
TT is associated with markedly reduced 5-HTT expression
levels in alcohol-dependent individuals (8–11). The postsynaptic 5-HT3 receptors are formed by 5-HT3A homopentamers or 5-HT3A and 5-HT3B heteropentamer
(5-HT3AB) complexes that evince even faster ion transport
and conduction. The 5-HT3B subunits do not bind with
serotonin or ondansetron but are important for stabilizing
the receptor complex at the cell surface (12, 13). The 5-HT3
A and B subunits are encoded by the HTR3A and HTR3B
genes located adjacently on chromosome 11, separated
ajp.psychiatryonline.org
1
PREDICTING THE OUTCOME OF ONDANSETRON TREATMENT OF ALCOHOL DEPENDENCE BY GENOTYPE
only by an intergenic region of 28 kb. Hodge et al. (14),
using a transgenic mouse model lacking the gene for the
5-HT3A subunit, showed that the reduction of alcohol
consumption produced by 5-HT3 antagonism is dependent on the presence of 5-HT3A-containing receptors.
Recently, a genetic association study conducted by Enoch
et al. (15) that examined African American men with a
lifetime diagnosis of alcohol, cocaine, or heroin dependence showed an association between these disorders and
the particular genotypes of HTR3A, HTR3B, and the 5:HTTLPR. These findings were supported by earlier work
from the same group that showed an association between
certain HTR3B genotypes and alcohol dependence with
antisocial behaviors (16).
Taken together, these risk association studies implicate
a collective effect of 5-HT3 and 5:-HTTLPR genotypes on
susceptibility for alcoholism. In the present study, we extended our previously tested hypothesis to include functional variants in the SLC6A4, HTR3A, and HTR3B genes
that would interact with each other to mediate ondansetron treatment response. This new extended hypothesis
was tested in the same sample that we examined in our
previously published phase 2 pharmacogenetic trial with
283 alcoholics of European descent (3). If such predictors
were found in the HTR3A and HTR3B genes, they would
help identify a larger population of alcoholics who would
respond to ondansetron.
Method
Participants
The study population analyzed here is identical to the sample
used in our previous study testing the genetic effects of the
SLC6A4 5:-HTTLPR-LL and rs1042173-TT genotypes on ondansetron’s efficacy (3). Briefly, all 283 participants were alcoholdependent individuals with no comorbid axis I diagnoses who
scored .8 on the Alcohol Use Disorders Identification Test (17)
and were enrolled in an 11-week randomized double-blind
clinical trial in which they received either oral ondansetron
(4 mg/kg of body weight twice daily) or placebo along with
weekly cognitive-behavioral therapy (18).
Study Design
We performed an a priori stratification based on participants’ 5:-HTTLPR genotype, with additional genotyping of
the single-nucleotide polymorphism (SNP) rs1042173 in the 3:untranslated region (3:-UTR) of SLC6A4. Candidate genotyping for polymorphisms in HTR3A and HTR3B were performed
retrospectively and were not used as stratification factors
(see Figures 1 and 2). We assessed the effect of genotype on
three different measures of alcohol consumption: drinks per
drinking day, percentage of heavy drinking days, and percentage of days abstinent. Information on daily alcohol consumption was collected for the 90 days prior to enrollment
and during the study period using the timeline follow-back
method (19). We employed a mixed-effects statistical model to
examine genetic associations with drinking patterns throughout the 3-month treatment period, rather than at a single time
point.
2
ajp.psychiatryonline.org
Genotyping
Genomic DNA was extracted from the blood of each subject
at baseline with a Gentra Puregene kit (QIAGEN, Inc., Valencia, Calif.). For SLC6A4 polymorphisms, genotyping data for 5:HTTLPR L/S and rs1042173 polymorphisms were obtained from
our previous pharmacogenetic trial (3, 20). For HTR3A and
HTR3B polymorphisms, we genotyped 10 SNPs in HTR3A and
nine SNPs in HTR3B using a commercially available TaqMan
premade genotyping assay (Applied Biosystems, Foster City,
Calif.) on an ABI 7900 platform. Average densities of their
distribution on the two genes were about one SNP every 2 kb
in HTR3A and every 5 kb in HTR3B. Candidate SNPs were
selected based on their location on the gene, having a minor
allele frequency .5%, and previously reported findings by
other researchers.
Because we did not detect a significant association of ondansetron treatment response with rs25531 genotypes alone or in
combination with 5:-HTTLPR genotypes in our previous study,
we did not examine 5:-HTTLPR triallelic effects in the present
study.
Genotyping procedures for ancestral informative marker SNPs
to assess population stratification using a 24-ancestral marker panel have been described previously (3, 20). Biological
and primer/probe sequence information for all of the abovementioned SLC6A4, HTR3A, and HTR3B polymorphisms is
presented in Table S1 in the data supplement that accompanies
the online edition of this article.
Statistical Analysis
Departure from Hardy-Weinberg equilibrium was assessed
using Haploview, version 4.0 (21). The quality of the clinical data
was assessed as described previously (3). Our primary outcome
variable was drinks per drinking day. Percentage of heavy
drinking days and percentage of days abstinent were secondary
outcome measures. A heavy drinking day is defined as consuming
$5 drinks for men or $4 drinks for women.
To study the effect of treatment and genotypes on our outcome
variables, we used mixed-effects linear regression models, which
can accommodate data missing at random. The models included
random intercept and slope (for temporal trend) and were adjusted for participants’ average drinking levels prior to the study,
age, gender, and proportions of genetic ancestry as covariates. Proportions of genetic ancestry were calculated using the Structure
program (http://pritch.bsd.uchicago.edu/software/structure2_2.
html) as described previously (3).
To minimize type I error, we used two filtering steps. First, we
examined the drinking data for all 19 HTR3A and HTR3B SNPs
(see Table S1 in the online data supplement) and restricted
further analysis to those for which there was a difference of more
than 1.5 drinks per drinking day between the ondansetron and
placebo groups and a p level ,0.05 (Figure 1). The cutoff of 1.5
drinks per drinking day was chosen because in our primary
analysis, we detected a reduction of about 1.5 standard drinks
per drinking day in ondansetron-treated alcoholics who possessed the 5:-HTTLPR-LL genotype, which was the randomization factor (3). Thus, we set this cutoff with the empirical goal of
discovering other genetic markers that would be at least as
predictive of a response to ondansetron treatment. For all 19
HTR3A and HTR3B SNPs, we tested statistical models that included
two-way interaction terms between genotypes of a given SNP.
Genotype associations with drinking outcome measures for each
SNP were tested using dominant (homozygous major allele
genotype plus heterozygous genotype versus homozygous minor
allele genotype), recessive (homozygous minor allele genotype
plus heterozygous genotype versus homozygous major allele genotype), or heterozygous (heterozygous genotype versus homozygous
Am J Psychiatry 00:0, 2013
JOHNSON, SENEVIRATNE, WANG, ET AL.
FIGURE 1. SNP Selection Process and Statistical Analysis Workflowa
Selection of
genetic variants
Genes encoding the primary target of ondansetron were selected as
candidate genes. Polymorphisms within these genes were selected
based on prior known functional variants, variants with previously
known associations, and variants located in putative expression
regulatory regions.
Screening parameter Primary outcome measure across
treatment period assessed for each genotype group in all 19
polymorphisms
Screening
Selection criteria: reductions in drinks per drinking day by
ondansetron compared with placebo, within each genotype group;
1. >1.5 standard drinks and p<0.05, and
2. Sample size >5 individuals in each genotype-by-treatment group
Filtering parameter Secondary outcome measures percentage of heavy drinking days and percentage of days abstinent
across treatment period
Filtering
a
4 Genotypes
HTR3A: 3 SNPs
HTR3B: 1 SNP
3 Genotypes
HTR3A: 2 SNPs
HTR3B: 1 SNP
Selection criteria: reductions in percentage of heavy drinking days
and increases in percentage of days abstinent by ondansetron
compared with placebo, within each of the six genotype groups;
1. percentage of heavy drinking days >10% and p<0.05, and
2. percentage of days abstinent >10% and p<0.05
Combined effects were tested on drinks per drinking day, percentage
of heavy drinking days, and percentage of days abstinent using:
Testing for
combined effects
2 Candidate genes
HTR3A: 10 SNPs
HTR3B: 9 SNPs
4 Genotype
combinations
1. HTR3A and HTR3B genotype interactions
2. HTR3A and HTR3B genotypes and the previously reported two
SLC6A4 genotype interactions
SNP=single-nucleotide polymorphism. The primary outcome measure was drinks per drinking day; the secondary outcome measures were
percentage of heavy drinking days and percentage of days abstinent.
major allele genotype plus homozygous minor allele genotype)
genetic models. The differences in means between genotype-bytreatment combinations for all genotypes are presented in Table
S2 in the online data supplement. Second, in addition to the
above-mentioned screening process, to further minimize type I
error due to multiple testing, we calculated false discovery rate q
values (22) for all genotype-by-treatment association analyses
that survived the screening process, as shown in Table S2.
Subsequent to the screening process, the selected SNPs were
subjected to a filtering process based on their associations with
the two secondary outcome measures, percentage of heavy drinking days and percentage of days abstinent. Figure 1 summarizes the
SNP selection process used in the study. Only the SNPs that
survived filtering were analyzed for their interactive effects on
the outcome measures. All interactions were tested at two levels:
1) the effect of treatment on each of the three outcome measures
in those who possessed one, two, or all of the HTR3A and HTR3B
genotypes, regardless of their previously identified 5:-HTTLPR
and rs1042173 genotypes, and 2) the combined effect of the HTR3A
and HTR3B genotypes together with the previously identified
SLC6A4-LL/TT genotypes. The sample sizes included in the
combined genotypes-by-treatment groups are listed in Figure 2.
Statistical power analyses (23) were performed as described
previously (3).
Am J Psychiatry 00:0, 2013
Validation of detected genotype combinations using the
generalized multifactor dimensionality reduction (GMDR)
method. As described above and illustrated in Figure 1, our
statistical approach in identifying interactive effects of different
genotype combinations was based on pooling individually significant variants and analyzing their combined effects on the
drinking measures. This approach, however, could potentially
fail to identify individual variants that have significant epistatic
effects but have no significant main genetic effects on the
phenotypes of interest. To address this concern and validate our
results from the primary statistical analysis, we performed an
independent SNP-by-SNP interaction analysis using the GMDR
method (24), one of the commonly used approaches for detecting genetic epistatic effects (25, 26). The GMDR method
allowed us to test for the best interaction models consisting of
two to five variants by performing an exhaustive search using all
21 variants genotyped in this study. As the response variable for
GMDR analyses, we used predicted values for the three different
drinking outcome measures during the entire treatment period.
These predictive values were generated by using the same mixedeffects linear regression models that we described above for
detecting significant interactive effects of different genotype
combinations on the three outcome measures, which included
random intercept and slope (for temporal trend) adjusted for
ajp.psychiatryonline.org
3
PREDICTING THE OUTCOME OF ONDANSETRON TREATMENT OF ALCOHOL DEPENDENCE BY GENOTYPE
FIGURE 2. CONSORT Diagram of Alcohol-Dependent Participants in the Post Hoc Analysisa
Enrollment and
allocation
657 Individuals screened
283 Randomized based on
5′-HTTLPR alleles
Follow-up
44 LL and 99 Sx subjects randomized to receive placebo
90 Completed trial
50 Did not complete trial
2 Limiting adverse event; 15 Participant choice;
27 Lost to follow-up; 2 Lack of efficacy; 4 Other
101 Completed trial
42 Did not complete trial
4 Limiting adverse event; 10 Participant choice;
21 Lost to follow-up; 3 Lack of efficacy; 4 Other
Primary
analysis
49 LL and 91 Sx subjects randomized to receive ondansetron
374 Excluded
297 Did not meet inclusion criteria
62 Declined to participate
15 Randomization failures
49 LL and 91 Sx ondansetron subjects included
in intent-to-treat analysis
44 LL and 99 Sx placebo subjects included
in intent-to-treat analysis
140 Genotyped for SLC6A4, HTR3A, and HTR3B SNPs
143 Genotyped for SLC6A4, HTR3A, and HTR3B SNPs
Post hoc analysis
7 Genotyping failures
5 Genotyping failures
133 Included in screening and filtering for
significant statistical associations of genotypes
with primary and secondary outcome
measures of drinkingb
Genotypes associated with ondansetron vs. placebo response:
HTR3A: rs1150226-AG, rs1176713-GG
HTR3B: rs17614942-AC
SLC6A4: SLC6A4-LL, rs1042173-TT
41 Ondansetron
AG/GG/AC/LL/TT carriers
a
b
138 Included in screening and filtering for
significant statistical associations of genotypes
with primary and secondary outcome
measures of drinkingb
92 Ondansetron
noncarriers (all others)
51 Placebo
AG/GG/AC/LL/TT carriers
87 Placebo noncarriers
(all others)
5:-HTTLPR=serotonin-transporter-linked polymorphic region; SNP=single-nucleotide polymorphism.
The primary outcome measure was drinks per drinking day; the secondary outcome measures were percentage of heavy drinking days and
percentage of days abstinent.
age, gender, ethnicity defined by ancestral informative markers,
center, and corresponding baseline drinking outcome measures.
Results
Demographic Characteristics
The demographic characteristics of the sample are the
same as in our previous publication (3). Briefly, the present
study included all 283 participants who identified themselves as being of European descent. Within this European
American cohort, 122 Caucasian and 18 Hispanic individuals received ondansetron and 118 Caucasian and
25 Hispanic individuals received placebo. As discussed
previously (3), potential population stratification was
4
ajp.psychiatryonline.org
assessed with the 24-SNP ancestral informative marker
panel using the Structure program. Results from the
Structure program revealed no population admixture between those who identified themselves as Hispanic and
Caucasian or between participants in the ondansetron and
placebo groups. Therefore, all 283 participants were included
in the final analyses, and the individual ancestral scores
were used as covariates in place of self-reported ancestry.
Overall genotyping failure rates for all 21 polymorphisms examined were less than 3%. For genotyping
quality control, we performed Hardy-Weinberg equilibrium
tests on all 21 polymorphisms in HTR3A, HTR3B, and
SLC6A4, and none of the variants showed significant
deviations in either the ondansetron or the placebo group
Am J Psychiatry 00:0, 2013
JOHNSON, SENEVIRATNE, WANG, ET AL.
(see Table S1 in the data supplement). Thus, all 21 polymorphisms were included in the analyses detailed below.
Furthermore, there were no statistically significant differences in genotype frequencies between the ondansetron and placebo groups for all genotypes included in the
analyses (see Table S2 in the data supplement).
Association of Drinking Outcome Measures With
HTR3A and HTR3B Polymorphisms at the Individual
Level
The drinks per drinking day estimates for all genotype
groups examined in this study are listed in Table S2 in
the data supplement. As shown in that table, genotypes
rs17614942-AC in HTR3B and rs1150226-AG, rs1176713GG, and rs1176719-AA in HTR3A were significantly associated with reductions of more than 1.5 drinks per
drinking day in response to ondansetron treatment
(p,0.05). Of the four HTR3A and HTR3B SNPs that survived the screening process, rs1176719-AA did not show
a statistically significant association with percentage of
heavy drinking days (reduction in percentage of heavy
drinking days, 15.35%; 95% confidence interval [CI]=
237.55, 6.86; p=0.175); furthermore, the association of
rs1176719-AA with percentage of days abstinent neither
was statistically significant nor reached the 10% threshold
(improvement in percentage of days abstinent, 3.48%; 95%
CI=22.98, 9.94; p=0.291). Hence, rs1176719 was excluded
(Figure 1) from further analyses performed to test the
interactive effects. The rs1176713 genotype GG showed
only a trend with percentage of heavy drinking days and
was not significantly associated with percentage of days
abstinent; yet, the magnitudes of percentage of heavy
drinking days and percentage of days abstinent (Table 1)
were similar to those of rs1150226-AG and rs17614942-AC.
We therefore included the following genotypes in the
subsequent multilocus interaction analyses: SLC6A4-LL/
TT, HTR3B-rs17614942-AC, and HTR3A-rs1150226-AG and
-rs1176713-GG. The false discovery rate q values for
associations of drinks per drinking day with the HTR3Brs17614942-AC and HTR3A-rs1150226-AG and -rs1176713GG genotypes ranged from 0.156 to 0.311 (see Table S2 in
the data supplement).
Combined Interactive Effects of Significantly
Associated HTR3A, HTR3B, and Previously Reported
SLC6A4-LL/TT Genotypes on Drinking
We identified three HTR3A/HTR3B (HTR3) genotypes
(HTR3B-rs176149-AC, HTR3A-rs1150226-AG, and HTR3Ars1176713-GG) in the above single-locus analyses in addition to the SLC6A4-LL/TT genotype combination in our
previous study; all of these polymorphisms were associated with lowering alcohol consumption from heavy to
nonheavy levels, reducing it by at least two standard drinks
on a given drinking day. Yet, if these five polymorphisms
are to be utilized as a marker panel for screening individual
DNA samples, it is imperative to know their predictability
when they are present alone or in combination with each
Am J Psychiatry 00:0, 2013
other. To examine combined interactive effects, we adopted
an analytical strategy similar to the collapsing method,
which is used widely in rare-variant analyses in genetic
association studies (27–31). The collapsing method investigates for the presence or absence of several rare alleles
in a region of interest and then treats them as a single
common variant; in our analysis, we investigated for the
presence or absence of the HTR3B-rs176149-AC, HTR3Ars1150226-AG, HTR3A-rs1176713-GG, and SLC6A4-LL/TT
genotypes and combined them in two ways. First, we assessed the combined effects of HTR3 genotypes identified
in this study, regardless of the presence or absence of
SLC6A4-LL/TT genotypes. Second, we assessed the combined effects of HTR3 and SLC6A4-LL/TT genotypes in
both the present study and the previous study.
1. Combined effects of HTR3A and HTR3B polymorphisms
on drinking outcome measures, regardless of presence or
absence of SLC6A4-LL/TT genotypes. Combination 1 indicates the effects of carrying only one of the three HTR3
genotypes on ondansetron treatment response. Included
subgroups are “rs1150226-AG and no rs176149-AC or
rs1176713-GG,” “rs1176713-GG and no rs1150226-AG or
rs176149-AC,” and “rs176149-AC and no rs1150226-AG
or rs1176713-GG.” All subgroups were present in less
than 5% of the total cohort, and overall, combination 1
represents 11% of the total study population.
Combination 2 indicates the effects of carrying any
two of the three HTR3 genotypes on ondansetron treatment response. Included subgroups are “rs1150226-AG
and rs176149-AC and no rs1176713-GG” and “rs176149AC and rs1176713-GG and no rs1150226-AG.” All but one
participant in combination 2 belonged to the subgroup
“rs1150226-AG and rs176149-AC and no rs1176713-GG.”
This can be partly explained by the low linkage disequilibrium (LD) between rs1176713 and the other two SNPs and
the high LD between rs17614942 and rs1150226. (The r2
values for linkage are presented in the LD plot in Figure 3.)
Combination 3 indicates the effects of possessing one or
any combination of two or all three HTR3 genotypes.
However, rs1176713-GG was found in only two of the
rs17614942-AC carriers and was not detected in any
rs1150226-AG carrier. This lack of individuals possessing
all three HTR3 genotypes in our sample resulted in
combination 3 being more like a pooled group of
combinations 1 and 2 (23% of the total study population)
rather than a new group consisting of individuals
possessing one, two, or all three HTR3 genotypes. The
included subgroups in combination 3 are “rs1150226-AG
and rs17614942-AC and no rs1176713-GG” (48%), “AG and
no AC or rs1176713-GG” (22%), “rs1176713-GG and no
rs1150226-AG or rs17614942-AC” (21%), “rs17614942-AC
and no rs1150226-AG or rs1176713-GG” (6%), and
“rs17614942-AC and rs1176713-GG and no rs1150226-AG”
(3%). As shown in Table 2 and Figure 4A,B,C, combination
3 was significantly associated with all three outcome
ajp.psychiatryonline.org
5
PREDICTING THE OUTCOME OF ONDANSETRON TREATMENT OF ALCOHOL DEPENDENCE BY GENOTYPE
TABLE 1. Significant Individual HTR3A and HTR3B SNP Associations With Three Drinking Measuresa
Genotype
Gene
HTR3B
HTR3A
a
Population Carrying the Genotype Combination
SNP
Reference
Associated
Ondansetron Group
Placebo Group
% in Total Cohort
rs17614942
rs1150226
rs1176713
AA/CC
AA/GG
AA/AG
AC
AG
GG
17/135
20/135
6/134
19/139
24/139
9/139
13
16
5
All comparisons are between the ondansetron and placebo groups. SNP=single-nucleotide polymorphism; EMD=estimated mean difference.
FIGURE 3. Linkage Disequilibrium Analysis of all HTR3A and HTR3B SNPs in the Total Cohort and Location of Significantly
Associated SNPsa
3′
5′
HTR3B
HTR3A
BF942819
NM_001161772.2
AM293589
NM_000869.5
2
11
3
10
20
4
16
7
13
5
15
25
5
2
56
13
11
9
6
10
0
3
0
0
0
3
14
5
2
14
1
5
2
0
2
0
2
rs1150220
rs10160548
rs1176719
rs2276302
rs1176720
rs1176722
0
1
0
2
64
67
1
18
42
51
65
23
0
4
3
6
0
5
7
0
1
13
27
6
17
43
27
7
5
4
0
3
0
rs33940208
0
0
8
5
43
13
3
16
41
41
10
0
2
10
0
0
6
37
56
8
15
39
2
14
4
0
2
10
rs1062613
6
0
1
0
14
98
40
1
0
3
41
1
0
13
0
1
0
6
3
1
rs1150226
rs17614942
0
0
2
0
12
1
0
1
3
6
14
0
0
11
29
5
0
5
2
2
8
2
10
56
0
0
4
5
0
0
8
9
3
16
2
0
rs1672717
25
14
5
8
11
72
44
9
7
32
49
6
rs2276307
6
9
36
35
rs1176744
rs1176746
rs12270070
rs4938056
rs11606194
rs3758987
1
rs1176713
NM_213621.3
NM_006028.3
19
75
56
81
33
50
48
11
13
3
0
2
2
2
6
0
1
1
1
a
SNPs=single-nucleotide polymorphisms. The numerals in the squares of the linkage disequilibrium plot are r2 values; the dotted lines point
toward the location of significantly associated SNPs in reference to various splice variants of HTR3A and HTR3B primary mRNA. The numbers
next to each splice variant indicate their accession numbers.
measures: drinks per drinking day, percentage of heavy
drinking days, and percentage of days abstinent. The effect
sizes for associations of these three measures with ondansetron treatment outcome in combination 3 were 0.867,
0.780, and 0.683, respectively.
6
ajp.psychiatryonline.org
2. Combined effects of HTR3A, HTR3B, and previously
reported SLC6A4-LL/TT on drinking outcome measures.
For our previously reported 5-HTT gene SLC6A4-LL/TT
genotype combination, ondansetron was associated with
a reduction of 2.08 drinks per drinking day (95% CI=23.75,
Am J Psychiatry 00:0, 2013
JOHNSON, SENEVIRATNE, WANG, ET AL.
Outcome Measure and Estimated Mean Difference
Drinks per Drinking Day
Percentage of Heavy Drinking Days
Percentage of Days Abstinent
EMD (%)
95% CI
p
EMD (%)
95% CI
p
EMD (%)
95% CI
p
–2.73
–1.81
–3.87
–4.59, –0.87
–3.51, –0.12
–6.95, –0.78
0.004
0.036
0.014
–20.45
–20.65
–23.98
–37.58, –3.32
–36.17, –5.12
–51.83, 3.87
0.019
0.009
0.091
17.95
19.75
18.20
0.95, 34.95
4.30, 35.19
–9.31, 45.70
0.039
0.012
0.195
20.41; p=0.015) relative to SLC6A4-LL/TT carriers who
received placebo regardless of their genotypes at the three
HTR3 SNPs; there were no statistically significant differences for percentage of heavy drinking days or percentage
of days abstinent between SLC6A4-LL/TT carriers in the
ondansetron and placebo groups. However, among
participants in the ondansetron group, SLC6A4-LL/TT
carriers, compared with noncarriers, were significantly
more likely to have more abstinent days (12.68%; 95%
CI=0.48, 24.89; p=0.042). Next, to study the combined
effects of the HTR3 and SLC6A4-LL/TT genotypes on the
three outcome variables, we pooled all subjects carrying
the above-mentioned combination 3 and/or SLC6A4-LL/
TT genotypes (combination 4) and compared them with
the noncarriers of combination 4.
Combination 4 indicates the effects of possessing the
following genotype subgroups: “LL/TT and no HTR3
genotypes” (32%), “any two HTR3 genotypes and no LL/
TT” (26%), “any one HTR3 genotype and no LL/TT” (25%),
“LL/TT combined with two HTR3 genotypes” (8.5%), and
“LL/TT combined with only one HTR3 genotype” (8.5%).
As shown in Table 2, the reductions in drinks per drinking
day in combination 4 were close to two standard drinks for
participants in the ondansetron group compared with
those in the placebo group (p=0.004) (Figure 4D). In
addition, carriers of combination 4 who were treated with
ondansetron also experienced significant improvements
in percentage of heavy drinking days (Figure 4E) and
percentage of days abstinent (Figure 4F). The effect sizes
for associations of drinks per drinking day, percentage of
heavy drinking days, and percentage of days abstinent
with ondansetron treatment outcome in combination 4
carriers were 0.593, 0.416, and 0.428, respectively.
Notably, in all of the above combinations, the ondansetron group not carrying the targeted genotype group
(i.e., groups designated as “all others” in Figure 4), compared with their placebo counterparts, showed a trend
toward worse outcomes on our three measures of alcohol
consumption.
The combination 4 group included four infrequent
(,5% of the total study population) subgroups carrying
SLC6A4-LL/TT and one or two of the HTR3 genotypes.
Associations of these infrequent genotype groups with the
primary outcome measure (drinks per drinking day) are
listed in Table S3 in the data supplement. As seen in that
table, reductions in drinks per drinking day in these groups
Am J Psychiatry 00:0, 2013
of individuals with markers for both HTR3 and SLC6A4
were around four standard drinks, which was greater than
the reductions detected for combinations 1–4 in Table 2.
The greatest reduction in drinks per drinking day was seen
in the combination of SLC6A4-LL/TT and rs1176713-GG
(combination 1; see Table S3).
Independent Determination of SNP-by-SNP
Interactive Effects Using the GMDR Method
To validate the significant interactive effects of various
genotype combinations detected by the statistical approach described earlier, we conducted an independent
SNP-by-SNP interaction analysis on all SNPs of the three
genes using the GMDR method. As shown in Table 3, the
detected significant five-variant genotype combination for
the three outcome measures, determined by the primary
statistical approach above, also appeared to be significant
based on the parameters of test accuracy, cross-validation
consistency, and permutated p value used commonly in
the evaluation of GMDR results (24, 32, 33). This provides
an independent determination of the best genotype combinations that can be used to predict treatment outcomes
for the three alcohol dependence measures.
Adverse Events
As reported previously (3), except for fatigue (p=0.019),
no other adverse events (including QT interval prolongations) occurred more frequently in the ondansetron group
than in the placebo group; nor did they occur more
frequently among carriers of the above-mentioned genotype combinations than among noncarriers (data not
shown). There were no life-threatening events, and the five
most commonly reported adverse events were insomnia
(20.5% and 22.3% in the ondansetron and placebo groups,
respectively), headache (20.9% and 19.4%), appetite disturbance (18.0% and 20.1%), fatigue (18.0% and 11.7%),
and diarrhea (13.1% and 15.2%).
Discussion
We identified three genotypes in the HTR3A and HTR3B
(collectively designated as HTR3) genes that were significantly associated with efficacy of ondansetron treatment for alcohol dependence in individuals with
European ancestry. We showed that possession of at least
one of the HTR3A-rs1150226-AG, HTR3A-rs1176713-GG,
and HTR3B-rs17614942-AC genotypes, along with the
ajp.psychiatryonline.org
7
PREDICTING THE OUTCOME OF ONDANSETRON TREATMENT OF ALCOHOL DEPENDENCE BY GENOTYPE
TABLE 2. Association of Frequent Genotype Combinations (>5%) with Primary and Secondary Outcome Measures of
Ondansetron Treatment Responsea
Population Carrying the Genotype Combination
Combination
1
Possessing only one of the three identified HTR3 genotypes
(HTR3A-rs1150226-AG alone or HTR3B-rs17614942-AC
alone or HTR3A-rs1176713-GG alone)
Possessing a combination of any two of the three
identified HTR3 genotypes (HTR3A-rs1150226-AG and
HTR3B-rs17614942-AC or HTR3A-rs1176713-GG
and HTR3B-rs17614942-AC)
Possessing any one, two, or all three of the identified HTR3 genotypes
(HTR3A-rs1150226-AG and/or HTR3A-rs1176713-GG and/or
HTR3B-rs17614942-AC)
Possessing any one, two, three, or all four of the
identified HTR3 and SLC6A4 genotypes (SLC6A4-LL/TT
and/or HTR3A-rs1150226-AG and/or HTR3Ars1176713-GG and/or HTR3B-rs17614942-AC)
2
3
4
a
Genotype Combination
Ondansetron
Group
Placebo
Group
% in Total
Cohort
15/133
16/138
11
14/133
18/138
12
29/133
34/138
23
41/133
51/138
34
All comparisons are between the ondansetron and placebo groups and are listed in descending order of magnitude of their reductions in
drinks per drinking day; mean differences for drinks per drinking day are in standard drinks; negative values indicate improvement from
baseline. EMD=estimated mean difference.
previously identified SLC6A4-LL/TT genotypes (i.e., combination 4), was predictive of ondansetron treatment
response; approximately 34% of the cohort carried these
genotypes.
A unique strength of our findings was that the fivegenotype combination 4 (Table 2) was associated not only
with a reduction in the amount of severe drinking (drinks
per drinking day) but also with two other measures of
treatment response—the frequency of heavy drinking
(percentage of heavy drinking days) and of abstinence
(percentage of days abstinent). Additional to these widely
used outcome measures, we also performed an exploratory
analysis for the genotype associations with an extreme
dichotomous responder versus nonresponder endpoint
efficacy variable, namely, percentage of subjects with no
heavy drinking days, which may be used by regulatory
agencies as an important outcome measure for larger
phase 3 trials (34). When only one heavy drinking day
during the final 4 weeks was allowed, we detected at least
a twofold increase in percentage of subjects with no
heavy drinking days among participants in the ondansetron group carrying one or more of the combination 4
genotypes compared with participants in the placebo
group or noncarriers in the ondansetron group. Notably,
however, the carriers of combination 3 who constituted
a subset of combination 4 (approximately 25% of the cohort) had more than a fivefold increase in percentage of
subjects with no heavy drinking days when the ondansetron and placebo groups were compared, and they might
define a group of “super-responders.”
Furthermore, the greatest reductions in drinks per
drinking day were seen in small subpopulations possessing both SLC6A4-LL/TT genotypes and any one or more of
the HTR3 genotypes. This finding supports our hypothesis
that the combined effects of genetic variants for the
8
ajp.psychiatryonline.org
presynaptic 5-HTT and postsynaptic 5-HT3 would predict
a greater response than genetic variations within each
individual gene, as they represent the state of both synaptic availability of serotonin and its postsynaptic receptor
levels. Another factor supporting this hypothesis is the
greater effect of SLC6A4 plus HTR3 than the combined
effect of the HTR3A and HTR3B genotypes. However,
results for the combined effect of SLC6A4 and HTR3
should be interpreted cautiously, as they were limited by
the small sample size within each placebo and ondansetron treatment-by-genotype group (see Table S3 in the
online data supplement), thereby reducing the statistical
power to draw definitive conclusions. Nevertheless, these
initial statistical trends generally supporting a molecular
basis for the use of ondansetron as a treatment for a
particular cohort of alcohol-dependent individuals warrant further corroboration using studies with larger sample sizes.
The second-best response for drinks per drinking day
was seen in the pooled group of individuals carrying any
one or a combination of the three HTR3 genotypes, regardless of LL/TT carrier status (combination 3 in Table 2).
From a biological perspective, this is quite explainable, since
5-HT3AB receptors are the primary target of ondansetron,
which in fact led us to select HTR3A and HTR3B as candidate genes for the present study. Even though ondansetron’s therapeutic effect on the selected SLC6A4 genotypes
may appear to be smaller than that on the target HTR3
genotypes, presumably because the biological effect of the
5-HTT on ondansetron’s pharmacodynamics is an indirect
one, their interaction is most intriguing in showing a more
robust response across different outcome measures of
alcohol consumption.
Notably, the significant associations that we detected
with SNPs rs1150226 and rs17614942 were through their
Am J Psychiatry 00:0, 2013
JOHNSON, SENEVIRATNE, WANG, ET AL.
Outcome Measure and Estimated Mean Difference
Drinks per Drinking Day
Percentage of Heavy Drinking Days
Percentage of Days Abstinent
EMD (%)
95% CI
p
EMD (%)
95% CI
p
EMD (%)
95% CI
p
–2.35
–4.41, –0.29
0.025
–19.05
–37.73, –0.36
0.046
16.12
–2.49, 34.74
0.090
–2.65
–4.65, –0.66
0.009
–22.43
–40.77, –4.09
0.017
20.23
2.06, 38.40
0.029
–2.50
–3.92, –1.08
0.0006
–20.58
–33.53, –7.62
0.002
18.18
5.26, 31.10
0.006
–1.71
–2.88, –0.54
0.004
–11.13
–21.94, –0.31
0.044
11.57
0.81, 22.33
0.035
heterozygous genotypes; that is, the heterozygotes showed
significantly greater response to ondansetron compared
with both homozygous genotype carriers. Although it is
argued that such a comparison utilizing the heterozygous
genetic model may not reflect underlying biological
mechanism(s), a dominant heterozygous effect over both
homozygotes, which is also referred to as molecular
heterosis, is an increasingly recognized phenomenon in
many human gene-disease association studies that include addictions (35–39). Another possible explanation for
detecting heterozygous associations could be the small
sample sizes of the homozygous minor allele genotype
groups, rather than the effects of molecular heterosis.
An additional caveat is that the preclinical studies of the
functionality of rs1150226 and rs1176713 of HTR3A and
rs17614942 of HTR3B have not been reported in the
literature, and therefore the exact molecular mechanism
for their interactive effects remains to be determined.
However, given their location within the genes, it is
possible that all three of these polymorphisms may alter
mRNA expression levels. The rs1150226 is located in the
promoter region of HTR3A within 1 kb of putative binding
sites for several transcription factors that have been shown
to be modulated by ethanol (40, 41). Furthermore, genetic
association analyses have shown an association of rs1150226
with alcoholism comorbid with antisocial personality disorder (16), altered temporal lobe activity (42), and clozapine
treatment outcome for schizophrenia (43). Both rs1176713,
which is a synonymous SNP located on exon 8 (Leu459Leu)
of HTR3A, and rs17614942, located in intron 8 of HTR3B,
could alter expression levels of functional 5-HT3AB subunits
through altered splicing of primary mRNA to form nonfunctional secondary mRNA. Interestingly, rs17614942 is
located in different regions in different mRNA isoforms
(Figure 3). In primary mRNA isoforms AM293589 and
NM.006028.3 (http://www.ncbi.nlm.nih.gov/IEB/Research/
Acembly/av.cgi?c=geneid&org=9606&l=9177), the two longer
HTR3B mRNA transcripts that are found to be present in the
Am J Psychiatry 00:0, 2013
brain, rs17614942 is located in intron 8. In rodents, intron 8
consists of a splice acceptor site that results in the long
isoform (44); even though this site is not known to be present
in humans, it underlines the importance of the intron 8
region in the formation of various mRNA splice variants. In
the shorter primary mRNA isoform BF942819 (http://
www.ncbi.nlm.nih.gov/IEB/Research/Acembly/av.cgi?
c=geneid&org=9606&l=9177), which has not yet been reported to be present in the brain, rs17614942 is located in
the 3:-UTR, which can also lead to changes in functional 5HT3B subunit expression level through different molecular
mechanisms, such as altered mRNA polyadenylation,
miRNA binding, and decay. In vivo experiments would
be required to characterize accurately the functional effects of the variants in combination 4, both individually
and collectively. We therefore propose our theorization
as a framework on which to launch further molecular
investigations (3).
Another caveat regarding our findings is that the
sample size was relatively small for performing statistical comparisons for all different genotype combinations
constituted by the five variants included in combination 4. It is therefore not possible to dissect fully the
relative contribution of each individual genotype with
treatment response. However, an analysis that tested
for all 16 different combinations of the five variants was
well beyond the scope of this phase 2 clinical trial and
would require a multicenter clinical trial that enrolled
several thousand treatment-seeking alcohol-dependent
individuals.
Another consideration was that we used a descriptive
filtering process to control for type I error, by minimizing
the number of inferential contrasts, to understand the
relationship between ondansetron treatment response
and genotype. Thus, an additional replication study is
needed in a much larger sample to validate our results. Furthermore, because the frequency of the identified alleles may vary in different ethnic populations,
ajp.psychiatryonline.org
9
PREDICTING THE OUTCOME OF ONDANSETRON TREATMENT OF ALCOHOL DEPENDENCE BY GENOTYPE
FIGURE 4. Comparison of Means for the Three Drinking Outcome Measuresa
A
D
8
p=0.0009
7
Difference from baseline
Drinks per Drinking Day
8
6
6
5
5
4
4
3
3
2
2
1
1
0
Ondansetron Placebo
Genotype
combination 3
60
Ondansetron Placebo
All others
60
p=0.0371
50
50
40
40
30
30
20
20
10
10
0
Mean difference from baseline
Ondansetron Placebo
Genotype
combination 4
p=0.0019
Difference from baseline
Percentage of Heavy Drinking Days
p=0.0041
E
p=0.0033
Ondansetron Placebo
Genotype
combination 3
0
Ondansetron Placebo
All others
C
Percentage of Days Abstinent
0
Ondansetron Placebo
All others
B
p=0.0437
Ondansetron Placebo
Genotype
combination 4
Ondansetron Placebo
All others
F
p=0.0044
50
50
p=0.0208
p=0.0058
40
40
30
30
20
20
10
10
0
p=0.0351
0
Ondansetron
Placebo
Genotype
combination 3
a
p=0.0001
7
p=0.0006
Ondansetron
Placebo
All others
Ondansetron
Placebo
Genotype
combination 4
Ondansetron
Placebo
All others
Graphs in the left-hand column are for participants in the ondansetron or placebo group carrying any one or a combination of the HTR3Ars1150226-AG, HTR3A-rs1176713-GG, and/or HTR3B-rs17614942-AC genotypes (genotype combination 3) who showed the best treatment
response and for those not carrying these genotypes (“all others”). Graphs in the right-hand column are for participants in the ondansetron or
placebo group carrying any one or a combination of the SLC6A4-LL/TT, HTR3A-rs1150226-AG, HTR3A-rs1176713-GG, and/or HTR3Brs17614942-AC genotypes (genotype combination 4) and for those not possessing any of these genotypes (“all others”).
additional studies are needed to test the applicability of
our findings to other ethnic populations. Nevertheless,
our findings were bolstered by the fact that the combination 4 variants also were included in the detected
significant genetic epistatic models by GMDR independent
10
ajp.psychiatryonline.org
epistatic analyses, which were performed to validate the
findings from the primary analysis.
Interestingly, a previous association study conducted by
Enoch et al. (15) suggested that rs1176744 polymorphisms
in HTR3B and SLC6A4 (LALA) were associated with the
Am J Psychiatry 00:0, 2013
JOHNSON, SENEVIRATNE, WANG, ET AL.
TABLE 3. SNP-by-SNP Interaction Models Detected to Be Significant Using the GMDR Method, With the Closest Resemblance
to Combinations 1–4 From the Primary Analysisa
Outcome Measure and Interaction Model
Drinks per drinking day
HTR3A: rs1176713–SLC6A4: 5:-HTTLPR
HTR3A: rs1176713–SLC6A4: 5:-HTTLPR, rs1042173
HTR3A: rs1150226,rs1176713–HTR3B:rs17614942–SLC6A4:
5:-HTTLPR, rs1042173
Percentage of heavy drinking days
HTR3A: rs1176713–SLC6A4: 5:-HTTLPR
HTR3A: rs1176713–SLC6A4: 5:-HTTLPR, rs1042173
HTR3A: rs1150226,rs1176713–HTR3B:rs17614942–SLC6A4:
5:-HTTLPR, rs1042173
Percentage of days abstinent
HTR3A: rs1176713–SLC6A4: 5:-HTTLPR
HTR3A: rs1176713–SLC6A4: 5:-HTTLPR, rs1042173
HTR3A: rs1150226-rs1176713–HTR3B:rs17614942–SLC6A4:
5:-HTTLPR, rs1042173
a
b
Test Accuracy
Cross-Validation
Consistency
Permutated pb
0.674
0.595
0.483
10/10
10/10
10/10
0.0001
0.039
0.576
0.612
0.594
0.565
7/10
10/10
10/10
0.003
0.031
0.094
0.593
0.628
0.598
8/10
10/10
10/10
0.033
0.003
0.029
SNP=single-nucleotide polymorphism; GMDR=generalized multifactor dimensionality reduction.
Generated with 106 permutations.
development of alcohol dependence and, speculatively,
ondansetron treatment response. Confirming these findings, an independent association analysis conducted by
our team (45) also showed a highly significant association
between alcohol dependence and a genotype combination comprising HTR3A-rs10160548, HTR3B-rs1176744 and
-rs3782025, SLC6A4-rs1042173, and 5:-HTTLPR polymorphisms.
In conclusion, we have shown that polymorphisms in
the SLC6A4, HTR3A, and HTR3B genes are predictors of
reduced drinking in response to ondansetron. Based on
these findings, we hypothesize that the combination 4
genotypes can be used to screen for individuals who
would respond to ondansetron. Furthermore, the effect
sizes of combination 4 across different measures of
alcohol consumption appear to be medium to high.
Clinical trials randomized on the basis of combination 4
genotypes are thus needed to test the replicability of our
findings.
Received Sept. 4, 2012; revision received Feb. 7, 2013; accepted
April 8, 2013 (doi: 10.1176/appi.ajp.2013.12091163). From the
Department of Psychiatry and Neurobehavioral Sciences and the
Department of Public Health Sciences, University of Virginia, Charlottesville. Address correspondence to Dr. Johnson (bankolejohnson@virginia.
edu).
Dr. Johnson has served as a consultant for ADial Pharmaceuticals
(for which he also serves as Chairman), D&A Pharma, Eli Lilly,
Organon, and Psychological Education Publishing Company. Dr. Li
has served as a consultant and board member for ADial Pharmaceuticals. The other authors report no financial relationships with
commercial interests.
ClinicalTrials.gov identifier: NCT00382642.
The authors thank the National Institute on Alcohol Abuse and
Alcoholism for its support through grants 5 R01 AA010522-17 and 1
R01 AA021163-01 (to Dr. Johnson) and grant 5 R01 AA019720-02 (to
Dr. Ait-Daoud); the National Institute on Drug Abuse for its support of
Dr. Li through grant 5 R01 DA012844-12; and Robert H. Cormier, Jr.,
B.A., for his assistance with manuscript preparation.
Am J Psychiatry 00:0, 2013
References
1. Enoch M-A, Schuckit MA, Johnson BA, Goldman D: Genetics of
alcoholism using intermediate phenotypes. Alcohol Clin Exp Res
2003; 27:169–176
2. Johnson BA: Update on neuropharmacological treatments for
alcoholism: scientific basis and clinical findings. Biochem
Pharmacol 2008; 75:34–56
3. Johnson BA, Ait-Daoud N, Seneviratne C, Roache JD, Javors MA,
Wang X-Q, Liu L, Penberthy JK, DiClemente CC, Li MD: Pharmacogenetic approach at the serotonin transporter gene as
a method of reducing the severity of alcohol drinking. Am J
Psychiatry 2011; 168:265–275
4. McBride WJ, Lovinger DM, Machu T, Thielen RJ, Rodd ZA, Murphy
JM, Roache JD, Johnson BA: Serotonin-3 receptors in the actions
of alcohol, alcohol reinforcement, and alcoholism. Alcohol Clin
Exp Res 2004; 28:257–267
5. Niesler B, Walstab J, Combrink S, Möller D, Kapeller J, Rietdorf J,
Bönisch H, Göthert M, Rappold G, Brüss M: Characterization of
the novel human serotonin receptor subunits 5-HT3C, 5-HT3D,
and 5-HT3E. Mol Pharmacol 2007; 72:8–17
6. Papke RL, Porter Papke JK, Rose GM: Activity of alpha7-selective
agonists at nicotinic and serotonin 5HT3 receptors expressed in
Xenopus oocytes. Bioorg Med Chem Lett 2004; 14:1849–1853
7. Boyd GW, Low P, Dunlop JI, Robertson LA, Vardy A, Lambert JJ,
Peters JA, Connolly CN: Assembly and cell surface expression of
homomeric and heteromeric 5-HT3 receptors: the role of oligomerization and chaperone proteins. Mol Cell Neurosci 2002;
21:38–50
8. Johnson BA, Javors MA, Roache JD, Seneviratne C, Bergeson SE,
Ait-Daoud N, Dawes MA, Ma JZ: Can serotonin transporter genotype predict serotonergic function, chronicity, and severity of
drinking? Prog Neuropsychopharmacol Biol Psychiatry 2008;
32:209–216
9. Barr CS, Newman TK, Becker ML, Champoux M, Lesch KP, Suomi
SJ, Goldman D, Higley JD: Serotonin transporter gene variation
is associated with alcohol sensitivity in rhesus macaques exposed to early-life stress. Alcohol Clin Exp Res 2003; 27:812–817
10. Preuss UW, Soyka M, Bahlmann M, Wenzel K, Behrens S, de
Jonge S, Krüger M, Bondy B: Serotonin transporter gene regulatory region polymorphism (5-HTTLPR), [3H]paroxetine binding in healthy control subjects and alcohol-dependent patients
ajp.psychiatryonline.org
11
PREDICTING THE OUTCOME OF ONDANSETRON TREATMENT OF ALCOHOL DEPENDENCE BY GENOTYPE
11.
12.
13.
14.
15.
16.
17.
18.
19.
20.
21.
22.
23.
24.
25.
26.
27.
12
and their relationships to impulsivity. Psychiatry Res 2000; 96:
51–61
Heinz A, Jones DW, Mazzanti C, Goldman D, Ragan P, Hommer
D, Linnoila M, Weinberger DR: A relationship between serotonin transporter genotype and in vivo protein expression and
alcohol neurotoxicity. Biol Psychiatry 2000; 47:643–649
Dubin AE, Huvar R, D’Andrea MR, Pyati J, Zhu JY, Joy KC, Wilson
SJ, Galindo JE, Glass CA, Luo L, Jackson MR, Lovenberg TW,
Erlander MG: The pharmacological and functional characteristics of the serotonin 5-HT(3A) receptor are specifically modified by a 5-HT(3B) receptor subunit. J Biol Chem 1999; 274:
30799–30810
Krzywkowski K, Davies PA, Feinberg-Zadek PL, Bräuner-Osborne
H, Jensen AA: High-frequency HTR3B variant associated with
major depression dramatically augments the signaling of the human 5-HT3AB receptor. Proc Natl Acad Sci USA 2008; 105:722–727
Hodge CW, Kelley SP, Bratt AM, Iller K, Schroeder JP, Besheer J:
5-HT(3A) receptor subunit is required for 5-HT3 antagonist-induced
reductions in alcohol drinking. Neuropsychopharmacology 2004;
29:1807–1813
Enoch MA, Gorodetsky E, Hodgkinson C, Roy A, Goldman D:
Functional genetic variants that increase synaptic serotonin and
5-HT3 receptor sensitivity predict alcohol and drug dependence. Mol Psychiatry 2011; 16:1139–1146
Ducci F, Enoch MA, Yuan Q, Shen PH, White KV, Hodgkinson C,
Albaugh B, Virkkunen M, Goldman D: HTR3B is associated with
alcoholism with antisocial behavior and alpha EEG power: an
intermediate phenotype for alcoholism and co-morbid behaviors. Alcohol 2009; 43:73–84
Bohn MJ, Babor TF, Kranzler HR: The Alcohol Use Disorders
Identification Test (AUDIT): validation of a screening instrument
for use in medical settings. J Stud Alcohol 1995; 56:423–432
Kadden R, Carroll K, Donovan D, Cooney N, Monti P, Abrams D,
Litt M, Hester R: Cognitive-Behavioral Coping Skills Therapy
Manual: A Clinical Research Guide for Therapists Treating Individuals With Alcohol Abuse and Dependence (NIH Publication
No 92-1895). Washington, DC, US Department of Health and
Human Services, 1992
Sobell LC, Sobell MB: Timeline follow-back: a technique for
assessing self-reported alcohol consumption, in Measuring Alcohol Consumption: Psychosocial and Biochemical Methods.
Edited by Litten RZ, Allen JP. Totowa, NJ, Humana Press, 1992,
pp 41–72
Seneviratne C, Huang W, Ait-Daoud N, Li MD, Johnson BA:
Characterization of a functional polymorphism in the 3: UTR of
SLC6A4 and its association with drinking intensity. Alcohol Clin
Exp Res 2009; 33:332–339
Barrett JC, Fry B, Maller J, Daly MJ: Haploview: analysis and
visualization of LD and haplotype maps. Bioinformatics 2005;
21:263–265
Benjamini Y, Drai D, Elmer G, Kafkafi N, Golani I: Controlling the
false discovery rate in behavior genetics research. Behav Brain
Res 2001; 125:279–284
Cohen J: Statistical Power Analysis for the Behavioral Sciences,
2nd ed. Hillsdale, NJ, Lawrence Erlbaum Associates, 1988
Lou XY, Chen GB, Yan L, Ma JZ, Zhu J, Elston RC, Li MD: A generalized combinatorial approach for detecting gene-by-gene
and gene-by-environment interactions with application to nicotine dependence. Am J Hum Genet 2007; 80:1125–1137
Cordell HJ: Detecting gene-gene interactions that underlie human diseases. Nat Rev Genet 2009; 10:392–404
Moore JH, Williams SM: Epistasis and its implications for personal genetics. Am J Hum Genet 2009; 85:309–320
Dering C, Hemmelmann C, Pugh E, Ziegler A: Statistical analysis
of rare sequence variants: an overview of collapsing methods.
Genet Epidemiol 2011; 35(suppl 1):S12–S17
ajp.psychiatryonline.org
28. Li B, Leal SM: Methods for detecting associations with rare
variants for common diseases: application to analysis of sequence data. Am J Hum Genet 2008; 83:311–321
29. Asimit J, Zeggini E: Rare variant association analysis methods for
complex traits. Annu Rev Genet 2010; 44:293–308
30. Bansal V, Libiger O, Torkamani A, Schork NJ: Statistical analysis
strategies for association studies involving rare variants. Nat Rev
Genet 2010; 11:773–785
31. Madsen BE, Browning SR: A groupwise association test for rare
mutations using a weighted sum statistic. PLoS Genet 2009; 5:
e1000384
32. Li MD, Lou XY, Chen G, Ma JZ, Elston RC: Gene-gene interactions
among CHRNA4, CHRNB2, BDNF, and NTRK2 in nicotine dependence. Biol Psychiatry 2008; 64:951–957
33. Li MD, Mangold JE, Seneviratne C, Chen GB, Ma JZ, Lou XY,
Payne TJ: Association and interaction analyses of GABBR1 and
GABBR2 with nicotine dependence in European- and AfricanAmerican populations. PLoS ONE 2009; 4:e7055
34. Falk D, Wang XQ, Liu L, Fertig J, Mattson M, Ryan M, Johnson B,
Stout R, Litten RZ: Percentage of subjects with no heavy drinking days: evaluation as an efficacy endpoint for alcohol clinical
trials. Alcohol Clin Exp Res 2010; 34:2022–2034
35. Gratacòs M, González JR, Mercader JM, de Cid R, Urretavizcaya
M, Estivill X: Brain-derived neurotrophic factor Val66Met and
psychiatric disorders: meta-analysis of case-control studies
confirm association to substance-related disorders, eating disorders, and schizophrenia. Biol Psychiatry 2007; 61:911–922
36. Pooley EC, Fairburn CG, Cooper Z, Sodhi MS, Cowen PJ, Harrison
PJ: A 5-HT2C receptor promoter polymorphism (HTR2C-759C/T)
is associated with obesity in women, and with resistance to
weight loss in heterozygotes. Am J Med Genet B Neuropsychiatr
Genet 2004; 126B:124–127
37. Lee HS, Kim SH, Lee HJ, Kim L, Lee SK, Jang DW, Lee MS, Son BG,
Suh KY, Kim S: Gender-specific molecular heterosis of dopamine D2 receptor gene (DRD2) for smoking in schizophrenia.
Am J Med Genet 2002; 114:593–597
38. Retz W, Rösler M, Supprian T, Retz-Junginger P, Thome J: Dopamine D3 receptor gene polymorphism and violent behavior:
relation to impulsiveness and ADHD-related psychopathology. J
Neural Transm 2003; 110:561–572
39. Gosso MF, de Geus EJ, Polderman TJ, Boomsma DI, Heutink P,
Posthuma D: Catechol O-methyl transferase and dopamine D2
receptor gene polymorphisms: evidence of positive heterosis
and gene-gene interaction on working memory functioning. Eur
J Hum Genet 2008; 16:1075–1082
40. Szabo G, Mandrekar P: Ethanol-mediated regulation of transcription factors in immunocompetent cells. Front Biosci 2002;
7:a80–a89
41. Wang H, Zhou H, Chervenak R, Moscatello KM, Brunson LE,
Chervenak DC, Wolcott RM: Ethanol exhibits specificity in its
effects on differentiation of hematopoietic progenitors. Cell
Immunol 2009; 255:1–7
42. Windemuth A, Calhoun VD, Pearlson GD, Kocherla M, Jagannathan
K, Ruaño G: Physiogenomic analysis of localized fMRI brain activity
in schizophrenia. Ann Biomed Eng 2008; 36:877–888
43. Souza RP, de Luca V, Meltzer HY, Lieberman JA, Kennedy JL:
Influence of serotonin 3A and 3B receptor genes on clozapine
treatment response in schizophrenia. Pharmacogenet Genomics 2010; 20:274–276
44. Brüss M, Eucker T, Göthert M, Bönisch H: Exon-intron organization of the human 5-HT3A receptor gene. Neuropharmacology 2000; 39:308–315
45. Seneviratne C, Franklin J, Beckett K, Ma JZ, Ait-Daoud N, Payne TJ,
Johnson BA, Li MD: Association, interaction, and replication analysis of genes encoding serotonin transporter and 5-HT3 receptor
subunits A and B in alcohol dependence. Hum Genet (in press)
Am J Psychiatry 00:0, 2013
Data Supplement for Johnson et al., Determination of Genotype Combinations That Can Predict the Outcome of the Treatment of Alcohol Dependence Using
the 5-HT3 Antagonist Ondansetron. Am J Psychiatry (doi: 10.1176/appi.ajp.2013.12091163)
HTR3B
HTR3A
SLC6A4
SNP No.
Gene
TABLE S1. Biological Information on SLC6A4, HTR3B, and HTR3A Polymorphisms Examined in This
Studya
dbSNP ID
Minor Allele
Frequency
(Minor
Allele)b
p values for Deviation from
Hardy-Weinberg Equilibriumc
Location
Within the
Gene
Chromosomal
Location
25,588,443 to
25,588,485d
L/S
0.450 (S)
1.000
0.779
Alleles
OND Group
Placebo Group
Primers and Probe Sequences / Context Sequence
ID of ABI Primers and Probes
1
5-HTTLPR
promoter
2
rs1042173
25,549,137
T/G
0.425 (G)
0.790
0.930
3
rs1150226
113,350,751
G/A
0.058 (A)
0.545
0.675
C___7488474_10
4
rs1062613
3-UTR
5-near
gene
5-UTR
Forward:
TCCTCCGCTTTGGCGCCTCTTCC
Reverse:
TGGGGGTTGCAGGGGAGATCCTG
C_7473190_10
113,351,216
C/T
0.989
0.258
C___7488465_10
5
rs33940208
Exon
1(L168L)
113,351,287
C/T
1.000
1.000
C__22275584_10
6
7
8
rs1176722
rs1176720
rs2276302
Intron 1
Intron 2
Intron 3
113,353,684
113,354,963
113,355,350
A/G
C/G
A/G
0.793
0.678
1.000
0.645
1.000
0.061
C___7488431_10
C___1646939_10
C___1646938_1_
9
rs1176719
Intron 4
113,357,397
A/G
0.365
0.878
C___1372135_1_
10
rs10160548
Intron 5
113,361,891
G/T
0.356
0.255
C___1372126_10
11
rs1150220
Intron 7
113,363,096
C/T
0.214 (T)
not
reported
(T=0.017e)
0.159 (A)
0.186 (G)
0.319 (G)
not
reported
(A=0.258e)
0.341 (G)
not
reported
(T=0.233e)
1.000
0.525
C___7488415_1_
12
rs1176713
113,365,635
G/A
0.257 (G)
1.000
0.835
C___1372122_10
13
rs3758987
113,280,485
T/C
0.252 (C)
0.746
0.452
C__27512451_10
14
15
16
17
rs11606194
rs4938056
rs12270070
rs1176746
113,286,191
113,291,749
113,304,257
113,307,811
T/C
C/T
G/C
G/A
0.097 (C)
0.376 (T)
0.226 (C)
0.425 (A)
0.766
0.615
0.325
0.154
1.000
0.240
0.587
0.354
C__26324626_10
C___3053567_10
C___3053563_10
C___7488605_10
Exon 8
(L459L)
5-near
gene
Intron 2
Intron 2
Intron 2
Intron 4
Page 1 of 6
Exon 5
113,308,238
A/C
0.292 (C)
0.972
0.437
C___7488596_1_
(Y129S)
19
rs2276307
Intron 6
113,309,097
A/G
0.226 (G)
0.821
0.870
C___1646949_1_
20
rs1672717
Intron 6
113,317,943
A/G
0.425 (G)
1.000
1.000
C___3053558_10
21 rs17614942
Intron 8
113,321,587
C/A
0.049 (A)
0.987
1.000
C__32983576_10
a
All chromosomal locations are in reference to: NC_000017.9 for SLC6A4, NM_006028.3 for HTR3B, and NM_213621.1 for HTR3A.
SNP=single-nucleotide polymorphism; ABI=Applied Biosystems (Foster City, CA); OND=ondansetron.
b
From HapMap CEU (European ancestry) sample.
c
Data from this study.
d
Reference citation: Garcia LF, Aluja A, Fibla J, Cuevas L, Garcia O: Incremental effect for antisocial personality disorder genetic risk
combining 5-HTTLPR and 5-HTTVNTR polymorphisms. Psychiatry Res 2010; 177:161–166.
e
pilot_1_CEU_low_coverage_panel.
18
rs1176744
Page 2 of 6
TABLE S2. Mean Drinks per Drinking Day Estimates in All Examined Genotype Groupsa
DDD in Ondansetron Group
Gene
HTR3A
SNP
rs1150226
rs1062613
Genotype
DDD in Placebo Group
95% CI
Estimate
Lower
Upper
Ondansetron vs. Placebo Comparison
95% CI
N
Estimate
95% CI
N
Estimate
-1.82
Lower
Upper
7.16
24
Lower
Upper
p-Value for
DDD
Difference
-3.53
-0.12
0.036
FDRb
p-Value for
Genotype
Frequency
Difference
0.311
0.968
AG
4.18
2.91
5.44
20
6.00
4.84
GG
6.11
5.53
6.69
113
6.04
5.49
6.6
115
0.07
-0.68
0.81
0.859
0.970
CC
3.70
1.37
6.04
6
4.47
1.62
7.32
4
-0.76
-4.44
2.91
0.682
0.474
TC
5.21
4.36
6.05
48
6.29
5.5
7.07
54
-1.08
-2.21
0.05
0.061
0.416
TT
6.27
5.61
6.92
81
5.9
5.25
6.56
82
0.36
-0.51
1.24
0.417
0.675
rs33940208
CC
5.76
5.19
6.34
123
6.01
5.47
6.55
126
-0.24
-0.97
0.47
0.498
0.754
CT
6.06
4.09
8.04
12
6.12
4.34
7.9
14
-0.06
-2.30
2.19
0.962
0.754
rs1176722
AA
2.25
-2
6.51
2
4.69
1.78
7.6
4
-2.43
-7.59
2.72
0.355
0.408
rs1176720
rs2276302
rs1176719
rs10160548
rs1150220
AG
5.93
4.93
6.93
36
6.51
5.49
7.54
32
-0.58
-1.98
0.82
0.415
0.329
GG
5.88
5.25
6.5
94
6.01
5.4
6.62
99
-0.13
-0.95
0.68
0.749
0.618
CC
2.21
-2.02
6.45
2
7.55
4.2
10.91
3
-5.34
-10.76
0.08
0.053
0.663
CG
5.85
4.88
6.82
38
6.08
5.12
7.04
36
-0.23
-1.57
1.09
0.730
0.578
GG
5.83
5.22
6.44
95
5.94
5.34
6.54
101
-0.11
-0.92
0.7
0.788
0.861
AA
6.13
5.42
6.85
67
6.13
5.35
6.9
57
0.01
-1.01
1.02
0.989
0.423
AG
5.77
4.97
6.57
57
5.97
5.28
6.66
73
-0.20
-1.20
0.8
0.695
0.203
GG
3.82
2.05
5.59
11
5.94
4.12
7.76
10
-2.12
-4.65
0.4
0.099
0.761
AA
4.33
2.50
6.16
11
7.65
5.92
9.38
11
-3.32
-5.78
-0.86
0.008
AG
5.94
5.08
6.81
46
5.22
4.45
5.98
60
0.73
-0.38
1.83
0.197
0.278
GG
5.88
5.21
6.54
78
6.46
5.76
7.15
69
-0.58
-1.49
0.34
0.215
0.373
0.172
0.872
GG
4.69
3.11
6.26
14
5.73
4.52
6.93
24
-1.04
-2.97
0.89
0.289
0.119
GT
5.82
5.08
6.55
68
6.25
5.48
7.02
59
-0.44
-1.45
0.58
0.400
0.088
TT
6.01
5.21
6.81
53
5.88
5.1
6.66
57
0.13
-0.95
1.21
0.811
0.641
AA
2.69
-0.46
5.84
4
6.79
4.25
9.33
5
-4.10
-8.15
-0.06
0.047
0.833
AG
5.60
4.66
6.53
40
5.91
5.1
6.71
52
-0.31
-1.5
0.87
0.610
0.615
Page 3 of 6
TABLE S2. Continued
DDD in Ondansetron Group
Gene
SNP
rs1176713
HTR3B
rs3758987
rs11606194
rs4938056
rs12270070
rs1176746
rs1176744
rs2276307
Genotype
DDD in Placebo Group
95% CI
Estimate
Lower
Upper
Ondansetron vs. Placebo Comparison
95% CI
N
Estimate
Lower
Upper
95% CI
N
Estimate
Lower
Upper
p-Value for
DDD
Difference
p-Value for
Genotype
Frequency
Difference
FDRb
GG
5.96
5.34
6.58
91
5.95
5.3
6.6
82
0.01
-0.85
0.86
0.983
0.365
GG
4.43
1.94
6.93
6
8.34
6.45
10.23
9
-3.91
-6.99
-0.82
0.013
AG
5.45
4.58
6.32
46
5.58
4.79
6.36
57
-0.13
-1.25
0.99
0.824
0.837
AA
6.05
5.4
6.7
82
6.06
5.38
6.74
73
-0.01
-0.91
0.89
0.982
0.189
CC
6.52
4.81
8.22
11
4.92
3.00
6.83
9
1.60
-0.96
4.16
0.220
0.533
CT
5.76
4.9
6.62
51
6.32
5.57
7.08
62
-0.57
-1.65
0.52
0.306
0.322
TT
5.73
5.02
6.43
70
5.85
5.14
6.56
69
-0.13
-1.1
0.85
0.801
0.969
CC
0.08
-5.83
5.99
1
2.14
-3.45
7.74
1
-2.07
-10.23
6.09
0.620
0.846
CT
6.66
5.13
8.18
14
5.66
4.50
6.82
25
0.995
-0.90
2.89
0.302
0.450
TT
5.71
5.14
6.29
117
6.14
5.58
6.7
114
-0.43
-1.17
0.32
0.261
0.669
CC
6.56
5.63
7.48
40
5.86
5
6.71
47
0.70
-0.51
1.91
0.259
0.566
CT
5.36
4.61
6.1
62
6.23
5.48
6.97
61
-0.88
-1.9
0.15
0.095
0.645
TT
5.86
4.78
6.94
30
5.98
4.94
7.02
31
-0.12
-1.58
1.34
0.870
0.892
CC
3.3
-0.05
6.65
3
6.18
4.07
8.3
7
-2.88
-6.83
1.07
0.153
0.213
0.172
0.511
CG
6.2
5.34
7.06
48
5.83
4.93
6.74
42
0.37
-0.83
1.57
0.548
0.147
GG
5.64
4.99
6.29
84
6.09
5.47
6.71
91
-0.45
-1.31
0.41
0.302
0.480
AA
5.44
4.22
6.67
22
6.16
5.01
7.31
25
-0.72
-2.38
0.94
0.394
0.729
AG
6.08
5.27
6.9
54
5.87
5.11
6.62
61
0.22
-0.85
1.28
0.696
0.690
GG
5.65
4.87
6.44
59
6.12
5.31
6.92
54
-0.47
-1.54
0.61
0.398
0.405
AA
6.14
5.4
6.87
65
6.08
5.37
6.78
70
0.05
-0.92
1.04
0.909
0.939
AC
5.37
4.56
6.18
56
6.01
5.20
6.82
54
-0.64
-1.72
0.45
0.249
0.608
CC
5.38
3.77
6.98
13
5.36
3.88
6.84
15
0.02
-2.15
2.19
0.986
0.757
0.684
AA
5.65
4.98
6.32
80
6.09
5.39
6.78
77
-0.44
-1.34
0.47
0.343
AG
6.32
5.44
7.19
46
5.9
5.12
6.67
55
0.42
-0.72
1.56
0.473
0.549
GG
4.35
2.25
6.46
8
6.64
4.65
8.64
8
-2.29
-5.19
0.62
0.122
0.892
Page 4 of 6
TABLE S2. Continued
DDD in Ondansetron Group
Gene
SNP
rs1672717
rs17614942
Genotype
DDD in Placebo Group
95% CI
Estimate
Lower
Upper
6.33
AA
5.54
4.75
Ondansetron vs. Placebo Comparison
95% CI
N
Estimate
57
95% CI
Lower
Upper
-0.65
-1.73
0.43
0.236
0.414
Estimate
Lower
Upper
6.19
5.39
7
53
p-Value for
Genotype
Frequency
Difference
p-Value for
DDD
Difference
N
FDRb
AG
5.54
4.78
6.29
62
5.76
5.05
6.48
66
-0.23
-1.22
0.77
0.655
0.813
GG
7.4
6.03
8.76
17
6.37
5.15
7.60
21
1.03
-0.79
2.84
0.269
0.640
AC
3.86
2.49
5.22
17
6.59
5.31
7.88
19
-2.73
-4.6
-0.87
0.004
AA
5.58
0.03
11.13
1
4.04
-1.65
9.73
1
1.54
-6.38
9.47
0.703
0.964
CC
6.09
5.52
6.66
117
5.97
5.42
6.53
119
0.12
-0.61
0.84
0.757
0.958
0.156
0.896
a
Genetic polymorphisms that were associated with more than 1.5 drinks/drinking day difference between the ondansetron and
placebo groups, and that had more than five subjects within each treatment group, are enclosed within boxes. SNP=singlenucleotide polymorphism; DDD=drinks per drinking day; CI=confidence interval; FDR=false discovery rate.
b
As mentioned in the Method section, FDR values were assessed only for the variants that survived the “screening process”.
Page 5 of 6
TABLE S3. Combinations of Genotypes Significantly Associated with the Primary
Outcome Measure, Drinks per Drinking Daya
Combination
Genotype Combination
Population
Carrying the
Genotype
Combination
Estimated Mean
Difference for DDD
(95% CI) and pValue
1
SLC6A4-LL/TT + HTR3A-rs1176713-GG
O=3
P=3
(2%)
–7.64
(–12.65 to –2.63)
0.003
2
SLC6A4-LL/TT + HTR3B-rs17614942-AC
O=6
P=4
(4%)
–4.25
(–7.87 to –0.63)
0.021
3
SLC6A4-LL/TT + HTR3A-rs1150226-AG + HTR3Brs17614942-AC
O=4
P=3
(3%)
–4.05
(–8.41 to 0.31)
0.069
4
SLC6A4-LL/TT + HTR3A-rs1150226-AG
O=5
P=3
(3%)
–3.92
(–8.09 to 0.26)
0.066
a
All comparisons are between the ondansetron and placebo groups and are listed in descending order of magnitude
of their drinks per drinking day reductions; mean differences for DDD are in standard drinks; negative values
indicate improvement from baseline. DDD=drinks per drinking day; CI=confidence interval; O=ondansetron;
P=placebo.
Page 6 of 6