Just like the onset of puberty before it, the timing of menopause is quite variable in the human population. But when exactly menopause begins can have a significant effect on health.
Through genetic studies on reproductive longevity, researchers have identified a link between menopause age and disease, including bone health, type 2 diabetes, and cancer (1). There are several common gene variants that influence age at menopause and disease outcome, but in a new study, a team of researchers identified rare mutations that exert an even stronger effect on menopause timing (2). These mutations also affect the number of de novo mutations passed down to that person’s children, influencing cancer risk for generations to follow.
“This is an area of women’s health that has not been researched extensively,” said study coauthor Anna Murray, a reproductive genomics researcher at the University of Exeter. “Three years ago, we published the largest genome-wide association study for the timing of menopause. In that analysis, it was very obvious that genes that predisposed to cancer were enriched in associations with menopause timing.”
However, genome-wide association studies look at regions of the genome rather than at single genes. “We needed to look at individual genes and the consequences of genetic variation,” Murray said. “That’s where rare variants really help because they often have a much bigger effect on protein expression.”
To identify these variants, Murray and her colleagues analyzed whole exome sequencing data from over 100,000 postmenopausal women in the UK Biobank. By categorizing genes according to their predicted functions and linking them to the women’s age at menopause, the researchers found rare mutations in nine genes, five of which (ETAA1, PALB2, PNPLA8, SAMHD1, and ZNF518A) were not previously associated with menopause. In people with these mutations, menopause may begin up to six years earlier than in people without — a significant increase over common variants, which shift the onset of menopause by an average of about six weeks.
Some genes that influence menopause timing also carry implications for cancer risk. These include BRCA2, CHEK2, and PALB2, which are associated with a heightened overall risk of cancer, but have a disproportionately strong effect on hormone-sensitive tumors such as breast and ovarian cancers (3). In Murray’s new study, SAMHD1 also associated with increased cancer risk, especially of mesothelioma, breast, and prostate cancers.
“Ovarian cancer is rare, so it’s helpful to understand the mechanism of disease,” said Sharyn Lewin, a gynecologic oncologist at Holy Name Medical Center who was not involved in the study. “Studying and characterizing rare mutations may ultimately help us understand familial cancer syndromes, which may inform genetic counseling and screening recommendations for at-risk individuals.”
Studying and characterizing rare mutations may ultimately help us understand familial cancer syndromes, which may inform genetic counseling and screening recommendations for at-risk individuals.
- Sharyn Lewin, Holy Name Medical Center
Murray and her colleagues suspected that some variants that affect proteins involved in detecting and repairing DNA breakages might not only raise the risk of cancer but also prevent the repair or removal of oocytes with damaged DNA. When this happens, the damaged DNA remains in the oocyte pool and can be passed down to children, resulting in de novo mutations that may influence their health or cancer risk. Comparing genomic sequences from “trios” (two parents and one child) participating in the UK’s 100,000 Genome Project supported their hypothesis, revealing that mothers who carried more variants associated with early menopause tended to have children with more maternally-derived de novo mutations.
One of the study’s limitations is the scarcity of the data needed to interrogate and replicate its findings. Few databases contain enough individuals with fully sequenced genomes or exomes; trios are even more difficult to find. When Murray and her colleagues attempted to replicate their trio findings using data from the Icelandic deCODE genetics database, they were unable to do so (4).
Murray pointed to the smaller size of that database as a likely cause. “When we meta-analyzed the two, the effect we found was still significant,” she said. “But these associations are hard to find, so we need much bigger sample sizes to be robust.”
Differences in database populations or in the way samples were stored, processed, or analyzed could also affect the findings’ replicability. “Replication is the cornerstone of scientific research,” said Lewin. “Discrepancies can highlight the need for further investigations to fully understand the underlying reasons.”
The lack of diversity in genomic data may also limit the findings’ applicability. “It’s very European-focused,” Murray said. “We looked at other ethnicities in the UK Biobank data, but the numbers are small. The lack of diversity will affect the usability of these types of genetic scores for predicting reproductive lifespan or cancer risk, so we’d like to do more in that area.”
References
- Ruth, K.S. et al. Genetic insights into biological mechanisms governing human ovarian ageing. Nature 596, 393–397 (2021).
- Stankovic, S. et al. Genetic links between ovarian ageing, cancer risk and de novo mutation rates. Nature 633, 608–614 (2024).
- National Cancer Institute. Genetics of Breast and Gynecologic Cancers (PDQ®)–Health Professional Version. (2024).
- Kong, A. et al. Detection of sharing by descent, long-range phasing and haplotype imputation. Nat Genet 40, 1068–1075 (2008).