Table 2 Prediction of pleiotropy using Polygenic Risk Score

Goes et al. [54]

2196 cases (BP) and 8148 controls

The PGC SCZ results

The PRS analysis showed a genetic overlap between BP with mood-incongruent psychotic features and SCZ.

Demirkan et al. [55]

1738 cases (MDD) and 1802 controls

2286 cases (MDD and anxiety) and 1205 controls

MDD-PRS explained up to 0.7% of the variance in depression in the study sample. The MDD-PRS was associated with anxiety and explained up 2.1% of the anxiety variance in the study population.

Peyrot et al. [56]

7544 cases (MDD) and 7754 controls

1645 cases (MDD) and 340 controls

Persons with both high MDD-PRS and history of childhood trauma are at risk for developing MDD in adolescence.

Neuropsychiatric GWAS Consortium Bipolar Disorder Working Group [57]

7481 cases (BP) and 9250 controls

Replication study: 4493 cases (BP) and 42,542 controls

675 cases (BP) and 1297 controls

SCZ-PRS contributes to the risk of bipolar disorder.

Ruderfer et al. [58]

9369 cases (SCZ) and 8723 controls

10,410 cases (BP) and 10,700 controls

There is a significant correlation between a BP-PRS and the clinical dimension of mania in SCZ patients. BP-PRS was associated with only the manic factors in SCZ patients, the association between BP-PRS and mania was largest at the high end of the mania distribution. BP-PRS explained 2% of the variance.

Ikeda et al. [59]

236 cases (METH-dependent), 864 controls

560 cases (SCZ), 548 controls

There was a shared genetic risk between METH-induced psychosis and SCZ.

Solovieff et al. [60]

3322 cases (SCZ), 3587 controls [9]

845 cases (PTSD), 1693 controls

There was an association between BP-PRS and PTSD severity.

Byrne et al. [61]

6324 cases (MDD), 6678 controls

4 different MDD and PDD target samples were used: 2104 PPD cases, 3149 MDD cases, 9447 PPD screened controls, 3468 MDD screened controls

BP-PRS explained 0.1% of the post-partum-depression variance.

Ferentinos et al. [62]

Cohort from the PGC MDD and BP mega-analysis [57, 63]

1966 cases from the RADIANT studies (MDD)

MDD-PRS predicted depression episodicity, and episodicity was better predicted with MDD-PRS than with BP-PRS.

Wiste et al. [64]

Meta-analysis from PGC 2011 (BP) [57]

1274 cases (MDD)

Further replication: 992 cases (MDD), 585 cases (MDD)

BP polygenic genetic load was associated with bipolar-like presentation in MDD. The results were, however, inconclusive since they were not replicatable. BP-PRS explained 0.8%–1.1% of the variance in depression traits.

Musliner et al. [65]

Results of the combined GWAS of MDD by the PGC [63]

HRS target dataset, 8761 participants.

Stressful life events did not mediate or confound the association between MDD-PRS and depressive symptoms, however; MDD-PRS and stressful life events were independent, significant predictors of depressive symptoms and MDD-PRS explained less than 1% of the variance in depressive symptoms.

Derks et al. [66]

8690 cases (SCZ), 11,831 controls

314 cases (SCZ), 148 controls

No significant correlation between SCZ-PRS and quantitative domains of SCZ symptoms in SCZ cases and controls.

Mullins et al. [67]

7 discovery datasets (MDD, BP)

4 validation/target sets (3 sets for suicide attempt, 1 from suicide ideation).

MDD-PRS predicted suicidal ideation. There was no polygenic association between suicide attempt and suicidal ideation, suggesting that suicide attempts and suicidal ideation are not part of the same spectrum, thus the tendency to act on suicidal thoughts may have another proponent than suicidal ideation.