In order to estimate the variance in depression explained by the genetic vulnerability, the stressors and their inter- actions, we tted linear mixed models controlling for relatedness for the whole sample as well as stratied by sex. The parameters of the model were estimated using GCTA 1.26.0 (student test to test the signicance of the xed effects) that accounts for twin relatedness using a Genetic Relatedness Matrix. Results: PRS for MDD signicantly predicted the depression score (maximum variance explained = 0.46%. p-value = 5.01e-08), which represents a substantial improvement com- pared to PRS predictions based on the previous PGC-MDD GWAS (PGC1, variance explained = 0.08%, p-value = 0.018). The depression score signicantly predicted lifetime DSM-IV MDD status (OR = 1.96, 95%CI 1.852.08, p-value = 3.0e-108, N = 8,607). The main effects of PSLE, NSLE, and lack of SS were also signicant, explaining respectively 12.9%, 0.3% and 3% of the depression score variance, with effects in the expected directions. We show a signicant interaction of the PRS with personal life events (0.12% of variance explained, pvalue = 0.0076) contributing positively to the risk of depression, predomi- nantly in women. The interaction was not signicant in men while explaining almost as much variance as the main effect in women. However, there was no signicant difference when comparing the size of the interaction across sexes (p- value = 0.21). Discussion: Our ndings point to an extra risk for indivi- duals with combined vulnerability and high number of reported personal life events beyond what would be expected from the additive contributions of these factors to the liability for depression, supporting the multiplicative diathesisstress model for this disease. Additionally, our results suggest possible differences in the aetiology of depression between women and men. Disclosure: Nothing to disclose. http://dx.doi.org/10.1016/j.euroneuro.2017.08.045 45. INTERACTION OF GENETIC RISK AND EARLY LIFE STRESS ON RISK FOR DEPRESSION INVESTIGATED IN A NATIONWIDE, DANISH CASE-COHORT STUDY Nis Suppli n ,1 , Esben Agerbo 2 , Klaus Kaae Andersen 3 , Veera Rajagopal 4 , Michael Benros 5 , Wesley Thompson 6 , Ole Mors 7 , Trine Munk-Olsen 8 , Merete Nordentoft 5 , Preben Bo Mortensen 4 , Katherine Musliner 8 1 Psychiatric Center Copenhagen 2 CIRRAU - Centre for Integrated Register-based Research, National Centre for Register-based Research, Aarhus University 3 Statistics and Pharmacoepidemiology, Danish Cancer Society Research Center 4 Aarhus University 5 Mental Health Centre Copenhagen, University of Copenhagen 6 Institute of Biological Psychiatry, Mental Health Centre Sct. Hans, Mental Health Services of the Capital Region of Denmark 7 Aarhus University Hospital, Risskov 8 National Center for Register-Based Research, Aarhus University Background: Risk for depression is attributable to both genes and life stress such as childhood maltreatment or death of a close relative. Candidate gene studies have indicated that life stress interacts with single genetic variants to inuence depression risk. Further, the genetic architecture of depression is highly polygenic, with genetic risk for depression primarily determined by small effects of multiple genetic variants which might also interact collec- tively with environmental factors. The aim of this study is to investigate gene-environment interaction in depression using three separate measures of genetic risk: a) candidate genetic variants, b) polygenic risk scores and c) a Genome- Wide Interaction Study (GWIS) based on 500,000 SNPs distributed across the genome. Methods: In the iPSYCH case-cohort study we identied 18,431 cases and 20,163 random subcohort members for whom GWAS analyses were available. All cases were diagnosed with depression at a Danish psychiatric hospital. We used nation- wide Danish registers to obtain information on the life stressors: childhood abuse, family disruption, disability pen- sion in a parent, severe somatic disease of the index person and severe somatic disease and death of rst-degree relatives. Information on life stress was operationalized as a cumulative time-dependent variable. Our analyses include 180 specic candidate genetic variants previously reported to take part in gene-environment interaction in depression or identied as directly associated with depression in GWASs. Further, we evaluate if a polygenetic risk score interacts with life stress to predict depression. Finally, we apply an agnostic approach investigating if any of 500,000 SNPs selected from across the genome interact with early life stress on risk for depression. We applied extended Cox models with Prentice weighting of cases for analyses. Sex, birth cohort and principal components were included as covariates. Results: At the time of abstract submission we have conducted preliminary analyses. Final analyses will be presented at the WCPG, 2017. Discussion: To our best knowledge, this study will be the largest original investigation to date on gene-environment interaction in depression. We expect that the results will provide new insights to the understanding of the combined effects of nature and nurture on risk for depression. Y. Ma, M. Li S806