Tool may predict premature menopause among childhood cancer survivors
Click Here to Manage Email Alerts
A multi-institutional team of researchers created a statistical model capable of predicting the risk for premature menopause among women who survived childhood cancer.
Seven percent of childhood cancer survivors develop primary ovarian insufficiency — also known as premature menopause — within 5 years of their cancer diagnosis. Another 12% develop premature menopause more than 5 years after diagnosis.
In comparison, only 1% to 2% of women in the general population develop premature menopause.
Investigators evaluated data from women who had been diagnosed with cancer before age 21 years and underwent treatment at North American hospitals from 1970 to 1999. The study population included women with self-reported menstrual history information and no malignant neoplasms within 5 years of initial cancer diagnosis.
Investigators included 7,891 female childhood cancer survivors — 922 of whom had primary ovarian insufficiency — from the Childhood Cancer Survivor Study in the development of the risk prediction model. They included 1,349 female survivors — 101 of whom had primary ovarian insufficiency — from the St. Jude Lifetime Cohort in a validation study.
The risk model showed strong discriminatory and positive predictive performance, suggesting that they are appropriate for use in clinical practice.
“Because women with premature menopause are not producing estrogen and other hormones, it can affect bone health, cardiovascular health and various other areas of health,” research team leader Yan Yuan, PhD, professor at University of Alberta School of Public Health, told Healio. “For the physicians who care for these survivors, being aware of these risks can enable them to provide a better long-term care plan for those patients.”
Healio spoke with Yuan about how they developed this risk prediction tool, the efficacy it demonstrated so far and next steps in research.
Healio: How did you develop this model and what need does it address?
Yuan: In the general population, we know the rate of premature menopause is about 1% to 2%. However, in this group of survivors, we probably have about 18% to 20%. That indicates that many of these survivors will have a shortened fertility window. Before this study, we didn’t know who was at higher risk and at what age, so the problem becomes knowing what to recommend as far as counseling survivors about fertility preservation and other health issues. We used treatment information to give a more accurate risk of premature menopause between ages 21 and 40.
Healio: What factors does your model consider when predicting premature menopause?
Yuan: It’s mostly about the cancer treatment — what kind of radiation they had and the dosage. We can provide the most accurate prediction if we know the dosage of radiation specific to the ovaries. If organ-specific dosage information isn’t available, we can approximate it to use the pelvic and abdominal radiation dose. Another factor is chemotherapy drugs. We included 20 commonly used chemotherapy drugs and their dosages in our prediction algorithm. We also consider bone marrow transplant, which is another important risk factor. Other considerations are age at diagnosis and race.
Healio: How has this prediction model performed so far?
Yuan: Our model has performed well. We validated it on an external dataset, and it has shown very good ability to discriminate those women who will have premature menopause versus those who don’t. It has also given us very high prediction accuracy, as well.
Healio: What still needs to be validated?
Yuan: To validate it, it’s important to use additional data. There is a cancer survivor cohort in the U.K., as well as one in the Netherlands and one in Switzerland. If we can validate this tool in the international cohort, it will ensure the generalizability of our model.
Healio: What are your next steps in this research?
Yuan: We want to understand how chemotherapy drugs affect premature menopause. There is a lot of causality information that is still not very clear. What is available now in the literature doesn’t seem to agree with what we found during the development of these tools. There is room to improve the scientific information on how these chemotherapy drugs may affect fertility.
Healio: Is there anything else you’d like to mention?
Luan: Before we published our study, another group published a study on radiation and female fertility. Their result was based purely on mathematical models. Their result was very different from ours in terms of their predictions. They predicted much higher risk than what we showed in our population. Our model is more accurate. We’re not exaggerating risk, but we’re not under-counting the risk. I think this is very important for any scientific work.
References:
- Clark RA, et al. Lancet Oncol. 2024;doi:10.1016/S1470-2045(19)30818-6.
- Im C, et al. Lancet Oncol. 2024;doi:10.1016/S1470-2045(23)00510-7.
- University of Alberta. Researchers develop tool to predict likelihood of premature menopause in childhood cancer survivors (press release). Available at: https://www.ualberta.ca/folio/2024/03/researchers-develop-tool-to-predict-likelihood-of-premature-menopause.html. Posted March 18, 2024. Accessed April 5, 2024.
For more information:
Yan Yuan, PhD, can be reached at yyuan@ualberta.ca.
Collapse
Healio Interviews
Click Here to Manage Email Alerts