Menopause is a physiological progression event for women with multiple sclerosis: Commentary

Mid-life health is influenced by various factors, including both somatic and reproductive aging. ¹ These two constructs of biological aging are intertwined and difficult to disentangle, but this controversy can be informed by epidemiological facts and biological plausibility of effects.

Campagna and Bove outline potential mechanisms linking menopause to neurodegeneration in women with and without MS, noting that impacts may be gradual and challenging to detect. Jokubaitis agrees on the biological plausibility and the importance of perimenopause on symptoms but argues that associations with disability were not evident in a recent large registry study and may be overshadowed by somatic aging. All authors agree that puberty’s impact on MS ² makes the relevance of the other reproductive lifespan bookend compelling. However, the importance of ovarian aging is supported by more than symmetry alone. Sex dimorphism in MS diminishes substantially in midlife and this is unlikely to be driven only by somatic aging.^1,3 For all forms of MS presenting after age 50, the sex ratio approaches 1:1.^3,4 Furthermore, while women under 50 tend to have higher relapse rates and less disability accumulation than men, this dimorphism disappears around the time of menopause. ⁴

Differences in reported associations between self-reported menopause and disability may reflect study design. ⁵ Studies demonstrating significant associations included all MS subtypes, whereas the MSBase and Barcelona CIS analyses excluded women with primary progressive MS and early secondary progressive MS. These exclusions may represent left censoring and loss of informative data. Multivariable modeling also differed, particularly in handling body mass index (BMI) (a potential strong confounder or mediator), and data collection timeframes varied. Generational lifestyle changes may influence perimenopausal weight gain, and high efficacy therapies may alter interactions between inflammation and perimenopausal biology on disability. Finally, reliance on self-reported menopause may introduce responder bias that could vary by cohort.

What can future studies implement to clarify the contribution of ovarian aging to progression? First, including the full MS phenotype continuum would reduce loss of informative data. Second, investigators must define which aspects of the multi-year menopause transition are biologically relevant and avoid conflating the impact of pre-menopause subfertility, perimenopause hormonal fluctuations (including excessively high estrogen) and post-menopausal sex steroid deficiency. Furthermore, women differ in progression through perimenopause, and pace may matter. Longitudinal biological assessments of entire cohorts would better capture heterogeneity than single-time points such as last menstrual period. While no single hormone test fully captures these dynamics, anti-Müllerian hormone (AMH) has proven useful. In >400 women with MS, changes in AMH levels captured fast versus slow reproductive aging and demonstrated robust associations with clinical disability and gray matter volumes. ⁶ Approximately 200 women in the Silverman et al. article were in this prior AMH study. Careful distinction between confounders and mediators of perimenopause effects (e.g. metabolic dysfunction) is also essential. Sophisticated modeling can further disentangle the effects of reproductive aging from birthdate age, disease duration, and other aging factors, while monitoring for collinearity.

Finally, harmonizing terminology may be helpful. This writer recommends applying the term biological aging inclusively to encompass reproductive aging, recognizing it as a fundamental biological process deserving equal attention. Somatic aging may be reserved for processes occurring irrespective of biological sex.