The purpose of this investigation was to explore the combined effects of single nucleotide polymorphisms (SNPs) within LFA-1/ICAM-1/GSK-3β pathway and environmental hazards on susceptibility to Graves' opthalmopathy (GO) among a Chinese Han population. Altogether 305 GO patients and 283 Graves' disease (GD) subjects were recruited. Information relevant to the participants' age, gender, body mass index (BMI), regular physical activity, smoking history, alcohol intake, stressful work environment, stress at work, family history of thyroid disease and 131I treatment were summarized, and the participants’ related SNPs of LFA-1/ICAM-1/GSK-3β were also detected. Then the gene-gene and gene-environment interactions were evaluated by logistic regression model and multi-factor dimensionality reduction (MDR) modeling. The results exhibited that age, BMI, smoking history, stressful work, stress at home, family history of thyroid disease and 131I treatment appeared as potential indicators regulating GO risk, when either univariate or multivariate regression analysis was performed (all P<0.05). Moreover, rs12716977 (T>C) and rs2230433 (G>C) of LFA-1, rs1799969 (G>A) and rs5498 (A>G) of ICAM-1, as well as rs6438552 (T>C) and rs334558 (T>C) of GSK-3β were significantly associated with altered susceptibility to GO under the allelic models (all P<0.05). Also haplotype TGAATC acted as a protective factor against GO risk (P<0.05), whereas haplotype CGAACC largely elevated risk of GO (P<0.05). Besides, logistic regression analysis demonstrated that rs12716927, rs5498 and rs6438552 all would affect the influences exerted by age, BMI, smoking history, stressful work, stress at home, family history of thyroid disease or 131I treatment on GO susceptibility (all P<0.05). MDR modeling implied that the combined model of rs12716977, rs2230433 and rs1799969 was the supreme interactive model when BMI was co-assessed, and the interactive model of rs12716977, rs334558 and rs5491 was the most desirable among the smoking population. In conclusion, gene-gene and gene-environment interactions served as a crucial manner in affecting susceptibility to GO, providing solid evidences for screening effective GO-susceptible biomarkers and exploring potential GO treatment strategies.