Prospective associations of adolescent obesity phenotypes with self-reported polycystic ovary syndrome diagnosis in young adulthood

Are empirically derived adolescent overweight/obesity phenotypes differentially associated with polycystic ovary syndrome (PCOS) in young adulthood? Self-reported PCOS diagnosis risk in young adulthood varied by empirically derived adolescent overweight/obesity phenotypes, with the highest risk observed among those in the 'mothers with obesity' and 'early puberty' phenotypes. Overweight and obesity during puberty are postulated to promote the development of PCOS. Much of the prior literature in this area is cross-sectional and defines weight status based solely on BMI, yet emerging research suggests that not all people with overweight/obesity have the same risk for chronic health conditions, including PCOS. Data came from 4838 female participants in the Growing Up Today Study (GUTS), an ongoing prospective cohort study in the USA that has followed children aged 9-14 into young adulthood (ages 31-37, with 16 waves of data collection between 1996 and 2019). We previously used latent class analysis to empirically derive obesity phenotypes among 2038 female participants aged 14-19 years with overweight/obesity in the sample, as determined by participants' self-reported height and weight status. Indicators in the latent class analysis were participants' maternal weight status, disordered eating behaviors, body image and weight concerns, depressive symptoms and pubertal timing. The derived obesity phenotypes included 'mothers with obesity', 'early puberty', 'high weight concerns', and 'mixed'. Among these participants and female participants without adolescent overweight/obesity, we used logistic regression with generalized estimating equations to examine associations of adolescent obesity phenotypes with self-reported PCOS diagnosis after age 19. Analyses were adjusted for potential confounders. Participants in all four obesity phenotypes were more likely than participants without overweight/obesity to report a PCOS diagnosis ('mothers with obesity' phenotype: odds ratio (OR) = 4.50, 95% CI = 2.61, 7.77; 'early puberty' phenotype: OR = 2.51, 95% CI = 1.59, 3.97; 'high weight concerns' phenotype: OR = 2.01, 95% CI = 1.24, 3.24; 'mixed' phenotype: OR = 1.94, 95% CI = 1.33, 2.82). Individuals in the 'mothers with obesity' phenotype had a significantly greater risk of PCOS diagnosis compared to those in the 'mixed' and 'high weight concerns' phenotypes (P < 0.05). Participants self-reported PCOS diagnosis, which may underestimate new-onset PCOS and limit our ability to establish a temporal order between overweight/obesity and PCOS development. Residual confounding may also explain some of the observed associations in our analysis. Despite the fact that participants were from all regions across the USA, the results may not be generalizable to non-White and socioeconomically diverse populations. Among females, the risk of PCOS in young adulthood varied by distinct adolescent obesity phenotypes. Those in the 'mothers with obesity' and 'early puberty' phenotypes had higher risks of PCOS, which suggests a potential underlying biological component. It may be beneficial to tailor PCOS surveillance according to these high-risk adolescent obesity phenotypes. This project was funded by research grants from the National Institutes of Health (R01 DK127585, U01 HL145386, and U01 CA176726). The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health. The authors have no competing interests to disclose. N/A.