Abstract
OBJECTIVE: Sex-specific differences impact vulnerability to alcohol effects, dependence (AD) and related phenotypes. Previous studies demonstrated that sex-related hormone and protein levels may contribute to those differences in sex-specific manner. However, previous studies were limited by sample size, the depths of phenotyping assessments and/or the number of hormones and proteins measured at the same time. Here, we investigated the association of the sex-related hormone (testosterone, estrogens, and progesterone) and protein (FSH, LH, SHBG, and albumin) levels with AD in males and females treated in community-based programs compared to age and sex-matched controls. We also investigated the associations of AD with the combined plasma sex-related hormone/protein “signatures” in the same groups of study subjects. METHODS: Sex-related hormone and protein levels were measured in plasma samples collected from 401 male and female subjects meeting DSM-IV-TR criteria for AD and 401 age and sex matched controls. Psychiatric Research Interview for Substance and Mental Disorder (PRISM) was used to capture DSM-IV diagnoses of substance use and other psychiatric disorders. All analyses were stratified by sex. Associations between each plasma sex-related hormone/protein level and AD was assessed with multivariable linear regression adjusted for age and BMI for males and age, BMI, and menopause status for females. Progesterone was defined as a continuous variable and a binary variable based on a lower laboratory detection limit of 0.1 ng/mL. We used principal component analysis to define hormone/protein signatures. Hormones and proteins with factor loadings =/> |0.4| were considered important drivers for the corresponding principal components (PCs). Resulting PCs were entered into a sex-stratified logistic regression model with adjustment of age and BMI to determine which PCs were associated with AD. RESULTS: Male AD subjects had higher estradiol and albumin levels (p <0.00625) as well as progesterone and LH levels (p <0.05) compared to controls. Progesterone was detected in a larger proportion of AD subjects than in the controls (62.6% vs. 47.7%; p = 0.017). In females, albumin levels were significantly higher (p <0.00385) while estrone and progesterone levels were lower (p <0.05) in AD subjects than the controls. Total testosterone (TT) levels were relatively higher in AD subjects but differences between groups were not significant (p >0.05). In males, the first PC (PC1) that accounted for the majority of data variance was significantly associated with AD (p=0.016) and was mostly driven by TT, estrone, estradiol and SHBG. PC3 (driven by albumin, progesterone and estradiol) and PC4 (driven by albumin and SHBG) were also associated with AD (p=0.007 and p<0.001, respectively). In females, PC1 (driven by estrone, estradiol, SHBG) was not significantly associated with AD. However, PC3 (driven by Albumin and progesterone), PC4 (driven by TT and Albumin) and PC5 (driven by progesterone and TT), were significantly associated with AD (p=0.009, p=0.001, p=0.004, respectively). CONCLUSION: Our findings indicate that individual hormone/protein levels as well as hormone/protein combination signatures associated with AD are different in male and female alcoholics. Future studies should investigate physiological mechanisms reflected in hormone/protein signatures captured by PC analyses.