First, both men and women are affected by AD dementia. AD is not a disease unique to women, and declaring that AD is a “woman’s disease” is an over-simplification that impedes research and might affect care for both women and men. Why would it affect care? Because these differences can cause excess worry for women and potentially bias the diagnostic process, influencing general practitioners and leading to an over-diagnosis of AD among women and an under-diagnosis among men. This could result in suboptimal treatments and preventive interventions for both women and men.
Yes, it is true that a larger number of women are diagnosed with AD dementia than men. This is partly because many more women survive to older ages when dementia is most prevalent. However, in the United States, the incidence of AD dementia does not significantly differ between women and men, even after the age of 85. In contrast, multiple studies from Europe and some other regions of the world have reported a higher incidence among women at older ages, with the exception of the Cognitive Function and Aging Study from the United Kingdom which initially reported a higher incidence for men. Thus, sex and gender differences for the ‘risk’ or ‘incidence’ of AD may depend on the time period and geographic region where the study was conducted. The study of these differences could identify modifiable risk factors.
Second, in this era of precision medicine, even if the incidence of AD is the same, there is a critical need to determine whether the risk factors are the same for women and men (as we learned in cardiovascular disease, there can be significant differences). For women, sex-specific hormonal changes (e.g., early or premature menopause) and cardiometabolic risk factors, among others, have been recognized and are now being examined in more depth. However, there are very few risk factors that have been identified as being stronger for men. Given the number of men affected by AD, there is an urgent need to identify the most tractable areas for risk factor modification for them as well.
Third, the study of gender-related factors is important. Gender roles have changed over time in many parts of the world. A well-known example is that women historically have had significantly less access to education, but now they earn more bachelor’s degrees than men in the United States. Could this increase in education for women be related to observed reductions in the risk of AD? Other gender-related factors that could impact the risk and progression of AD include occupational attainment, socioeconomic differences, caregiver burden, and access to care (or willingness to seek care).
Fourth, it is important to sponsor, design, and conduct AD research that incorporates both female and male mice. Yes, it will cost more; and yes, studies of females have to consider and measure the estrous cycle, which is more burdensome. However, if researchers had not included female mice in their experiments, we still would not know that females have significantly more pathology among some strains of AD transgenic mice.
Similarly, in humans we need to move beyond just ‘adjusting’ for sex and examine whether a risk factor for AD is modified by sex. Some studies examining effect modification (i.e., interactions or with, or stratification by, sex) have reported important sex differences. For example mid-life cardiometabolic factors have been found to be stronger risk factors of AD for women and being single or widowed is a stronger risk factor for men. It is true that some studies will have spurious results due to smaller sample sizes. However, the point here is that we need to consistently look across all studies in order to move the field forward and distinguish real associations from spurious results. In addition, we acknowledge that some risk factors will not differ by sex. Reporting these ‘negative’ results is essential so that the field can move on to other foci. As evidenced in cardiology, the benefits far outweigh the risks.
WHERE DO WE GO FROM HERE?
It is always easy to call for more funding, in this case for sex and gender differences in AD, and to state that more research is needed. The hard part is identifying where to start, particularly because there has been less research on sex differences in AD, and the brain in general, compared to other diseases and organs. Herein, we briefly highlight potential areas of focus across the translational spectrum, which is by no means exhaustive. We hope that these ideas will stir discussion in the field and provide a foundation for the development of a more comprehensive plan to study sex and gender differences in AD in the coming years.
As discussed above, the incidence of AD dementia in the United States does not appear to differ by sex. However, emerging research suggests that risk factors and mechanisms may be different for men and women. Understanding these differences will lead to the development of better AD risk scores for each sex. In addition, when assessing sex differences in observational studies, there are four potential scenarios that should be considered: (1) Same frequency of the risk factor in women and men but a different effect (e.g., APOE genotype or other genetic variants); (2) Same effect of the risk factor but a different frequency (e.g., historically, women have had less access to education and men have a higher frequency of smoking); (3) Both different frequency and different effects of the risk factor (e.g., head trauma is more common in men compared to women, but women may be more susceptible to the adverse effects from the head injury); (4) Risk factors restricted to one sex (e.g., oophorectomy; prostate cancer and androgen deprivation therapy). Moving beyond ‘adjusting’ for sex, or observations of sex differences, will enhance our understanding of how to intervene and when to intervene.
Incorporating the assessment of sex differences at all stages of drug development, from non-human testing through Stage IV, is critically important but complex. Historically, some studies have utilized only animals of one sex in drug development. These animal studies should carefully evaluate if there is evidence for Pharmacokinetic/Pharmacodynamic (PK/PD) differences or other factors suggesting differences in metabolism or response by sex. Among clinical trials, Phase I studies have been traditionally conducted on male volunteers to avoid potential risks in women of childbearing age. Although Phase II and III studies include subjects of both sexes, and subgroup analysis by sex is required by regulatory agencies, the reporting of these results in the literature is not common practice and is not mandatory for failed trials. Therefore, we propose that sex differences, whether present or not, should be routinely reported. Lastly, the study of sex differences should not be limited to efficacy or metabolism, but should also focus on safety and adverse events, at all stages. Eight of the last ten prescription drugs pulled from the United States market were withdrawn because of greater health risks and increased adverse events among women.
Although current animal models are not fully representative of AD pathology and behavior in humans, the inclusion of both female and male animals in AD research is still important. These differences need to be examined across the lifespan, from developmental origins to adulthood and including aging processes. Notably, even if the pathology or behavior is the same and progresses at the same rate, the mechanisms by which a female versus a male develops the pathology could be completely different and warrants examination.
n short, consideration of sex differences is of significance in AD research at all levels, including basic science studies, clinical research, and clinical trials. Irrespective of the challenges ahead, addressing, instead of ignoring these differences, will bring us a step closer to developing and providing better care for both women and men.