Table 1 Demographic information, results and conclusions for GxE Studies of PTSD

Studies of Child Trauma on Child PTSD (2)

Amstadter et al. (2011)

103; PTSD-RI M= 24.09 (12.2)

14.63 (3.2)

40.8% EA; 45.6% AA; 13.6% O

Physical injury

NR^b

CRHR1

Yes

Yes

NR

rs12944712 was significantly related to higher acute PTSD Sxs and increasing trajectory of Sxs over time.

Drury et al. (2009)

88 (NR)

3-6(NR)

56% AA; 40% EA

Hurricane Katrina

NR

DAT1

NR

Yes

NR

The 9 allele increased risk of PTSD—both in the form of total and Criterion D Sxs.

Studies of Adults with History of Child and/or Adult Trauma on Adult PTSD (4)

Binder et al. (2008)`

900^c

40.8 (13.8)

95.2% AA;

Child abuse and non-child abuse

Yes

FKBP5

Yes

No

Yes

Significant interaction between FKBP5 polymorphisms and child abuse found for adult PTSD Sxs. The interaction for adult trauma was not significant.

2.2% EA;

0.6% L;

0.1% A;

0.9% Mixed;

1.0% Other

Nelson et al. (2009)

259 (17.8% LT)

NR

NR

Child abuse

NR

GABRA2

NR

No

Yes

Interactions b/w child trauma and SNP genotype provide consistent support for GxE interactions involving child trauma and SNP genotype. When separate variables were coded for the presence of one or two risk-associated alleles, significant Gx E interactions are only found for homozygous individuals.

Xie et al. (2010)

2427 (14.0% LT)

38.6 (10.8)

47.1% EA; 52.9% AA

Child adversity

Yes

FKBP5

Yes

No

Yes

In AAs, the interaction between child adversity and all 4 FKBP5 SNPs were associated with PTSD. SNP rs9470080 had strongest conditional effect; for AAs without child adversity, those homozygous for T allele had lowest risk of PTSD, while homogygotes with adversity had highest risk.

Xie et al. (2009)

1252 (18.3% LT)

38.9( 11)

46.5% EA; 53.5% AA

Both

Yes

SLC6A4

Yes

No

Yes

5-HTTLPR polymorphism alone did not predict PTSD; however it interacted with adult traumatic events and child adversity to increase the risk for PTSD, especially for those with high rates of both types of trauma exposure.

Studies of Adults which Assessed Adult Trauma (22)

Amstadter et al. (2009)

607 (3.6%)

22.6% ≤ 59

90% EA; 3.9% AA; 3.9% L; 1.7% Other; 0.5% Missing

2004 FL hurricanes

Yes

RGS2

No

No

Yes

GxE interaction such that rs4606 moderated risk of PTSD Sxs under high E stress and low social support.

Both LT and CT Sxs asses’d

77.4% ≥ 60

Kolassa et al. (2010a)

424 (80.2% LT; 48.8% CT)

34.8 (5.8)

100% Hutu or Tutsi

Rwandan genocide

Yes

COMT

Yes, LT PTSD

No, for LT and CT PTSD

Yes, LT PTSD

COMT genotype affected PTSD such that met/met homozygotes had higher risk for PTSD than those with Val allele independent of severity of traumatic load.

No, CT PTSD

No, CT PTSD

Kolassa et al. (2010b)

408 (81.1% LT)

34.68 (5.9)

100% Rwandan refugees

Rwandan genocide

Yes

SLC6A4

Yes , LT PTSD

Yes , LT PTSD

Yes , LT PTSD

Probability of developing PTSD was 100% for s homozygotes and there was no dose–response relationship between trauma and PTSD. However, when trauma approached extreme levels, genotype effect disappeared and PTSD approached 100%.

Kilpatrick et al. (2007)

589 (3.2% CT)

22.6%≤ 59

90% EA; 3.9% AA; 3.9% L; 1.7% Other; 0.5% Missing

2004 FL hurricanes

Yes

SLC6A4

No

No

Yes

5-HTTLPR increased risk of PTSD under low social support condition.

76.6% ≥ 60

Thakur et al. (2009)

41 (59% “acute” PTSD)

32 (NR)

95% EA; 5% Other

MVA^d

Yes

SLC6A4

NR

Yes

NR

Higher chronic PTSD was found in ll genotypes than sl and ss genotypes.

Dragan al. (2009)

107 (22.4% CT)

35.57 (12.89)

NR

Polish flood

Yes

DRD4

Yes

Yes

No

At least 1 copy of DRD4 long allele related to higher total PTSD and Avoidance/Numbing Sxs.

Comings et al. (1996)

56 (66%*)

43.6 (NR)

100% EA

Vietnam War

No

DRD2

NR

Yes

NR

59.5% of those with PTSD had D ₂ A1 allele; of the group that did not have PTSD, only 5.3% had D ₂ A1 allele.

Bachmann et al. (2005)

160 (73.8%*)

55.7 (4.2)

NR

Vietnam War

No

GCCR

NR

No

NR

N363S and Bcll GR polymorphisms not more frequent in PTSD patients than controls.

Gelernter et al. (1999)

139 (37.4%*)

With PTSD: 44.6 (3.6)

100% EA

Vietnam War

No

DRD2

NR

No

NR

No allelic association between DRD2 TaqI “A” system alleles and PTSD.

Without PTSD: NR

Grabe et al. (2009)

1,663 (4.03% LT)

With ≥ 1 traumatic experience: 57.6 (15.6); without traumatic experience: 50.0 (13.3)

100% EA

Community based sample; variety of events

Yes

SLC6A4

Yes

Yes

Yes

GxE interaction found between high expression of L _A allele and frequent trauma.

Koenen et al. (2009)

590 (3.2% CT)

<60 = 22.7%

90.7% EA; 9.5% Other

2004 FL hurricanes

NR

SLC6A4

Yes

No

Yes

County-level crime and employment rate modified association between genotype and PTSD risk. The s’ allele associated with decreased risk in low-risk environments and increased risk in high-risk environments.

Mellman et al. (2009)

118 (47% LT)

39.9(16.3)

NR

Various

NR

SLC6A4

NR

Yes

NR

5HT2A G allele significantly associated with PTSD.

Mustapic et al. (2007)

167 (85% CT and LT)

With PTSD: 40.3 (7.2); Without PTSD 38.12 (4.2)

100% Croatian Caucasian

Combat-related trauma

No

DBH

NR

Yes

NR

PTSD associated with significantly lower plasma DBH activity in those carrying CC genotype.

Sayin et al. (2010)

77 (23.3% CT and 50.0% LT)

NR

NR

Mild physical trauma

NR

SLC6A4

Yes

No

No

Having L allele for 5-HTT gene- linked polymorphic region may cause milder hyperarousal symptoms in those patients who have developed PTSD.

Segman et al. (2002)

206 (50.5% CT)

With PTSD: 39.7 (11.7); Without PTSD: 33.9 (10.2)

100% Jewish of definite Ashkenazi or non-Ashkenazi origin

Various (e.g., road accidents, terrorism)

NR

DAT1

NR

Yes

NR

The nine repeat allele at the DAT1 locus associated with increased risk for PTSD.

Bailey et al. (2010)

200 (36.5%*^e)

NR

100% Armenian

1988 Armenian Earthquake

NR

DRD2, DAT1

NR

No

NR

Neither DRD2 nor DAT1 associated with PTSD.

Sarapas et al. (2011)

40 (50% total; both CT and LT assessed)

With PTSD: 57.30 (13.2); Without PTSD: 51.20 (15.9)

100% EA

9/11 attacks

Yes

FKBP5

Yes, child trauma

No

NR

Comparison of LT versus CT PTSD identified overlapping genes with altered expression suggesting enduring markers, while some markers present only in CT PTSD may reflect state measures. As a follow-up, direct comparisons of expression in CT PTSD, LT-only PTSD, and control groups identified FKBP5 and MHC Class II as state markers, and also identified several trait markers. An analysis of indirect effects revealed that homozygosity for any of 4 PTSD risk-related polymorphisms at FKBP5 predicted FKBP5 expression, which mediated indirect effects of genotype on plasma cortisol and PTSD severity.

Yes, other trauma

Valente et al. (2011a)

99 (66.5%*)

With PTSD: 37.9 (8.7); Without PTSD: 44 (13.8)

NR

Urban violence

Yes

COMT

No, child trauma

Yes

NR

Found significant association (between met allele and PTSD in victims of violence

Valente et al. (2011b)

99 (66.5%*)

With PTSD: 37.9 (8.7); Without PTSD: 44 (13.8)

NR

Being victim of an urban violence that could be characterized as criterion A

Yes

BDNF, DAT1, SLC6A4

No

Yes, DAT1

NR

Only the nine repeat allele of the DAT1 was associated with an increased risk of PTSD after being exposed to urban violence.

No, SLC6A4

Hauer et al. (2011)

126 (11.9% CT)

Homozygotes: 67.1(10.8); Heterozygotes: 65.8(9.3)

NR

Cardiac surgery

NR

GCCR

No

Yes

NR

Homozygous Bcll *G carriers at an increased risk for PTSD stress.

Ressler et al. (2011)

NR

NR

Majority AA

NR

NR

PACAP, PAC1

NR

Yes, women only

NR

Alterations in the PACAP–PAC1 pathway involved in abnormal and sex-specific stress responses underlying PTSD. These sex-specific effects may occur via oestrogen regulation of ADCYAP1R1.

Tang et al. (2010)

227 (30.4% CT)

43.9 (12.8)

100% AA

NR

NR

DBH

ME for adult; No ME for child

No

NR

No relationship between sDβH and PTSD (i.e., Sx or Dx)