During the menopausal transition and early postmenopause, participants in the Seattle Midlife Women’s Health Study (SMWHS) experienced one of the three symptom severity clusters identified through latent class analysis: severe hot flashes with moderate sleep, mood, cognitive, and pain symptoms (high-severity hot flash); low-severity hot flashes with moderate levels of all other symptom groups (moderate severity); and low levels of all symptom groups (low severity). In an effort to determine whether gene polymorphisms were associated with these symptom severity classes, we tested associations between gene polymorphisms in the estrogen synthesis pathways (cytochrome P450 19 [CYP 19] and 17 beta hydroxysteroid dehydrogenase [ 17HSDB1]) and the three symptom severity clusters. SMWHS participants ( N = 137) recorded symptoms monthly in diaries and provided buccal smears for genotyping. Multilevel latent class analysis with multinomial regression was used to determine associations between gene polymorphisms and symptom severity clusters. Only the 17HSDB1 polymorphisms ( rs615942 and rs592389) were associated significantly with the high-severity hot flash cluster versus the low-severity symptom cluster. None of the polymorphisms was associated with the moderate-severity cluster versus the low-severity symptom cluster. Findings of associations of the 17HSDB1 polymorphisms with the high-severity hot flash symptom cluster are consistent with those of an association between 17HSDB1 polymorphisms and hot flashes in the Study of Women and Health Across the Nation population and our previous findings of associations between these polymorphisms with greater estrone levels.