Female adolescents and young adults (AYA, 15–39 years) with cancer are a unique population because this age range encompasses most of the reproductive life span. Advancements in cancer treatments have contributed to significant increases in survival times; nonetheless, quality of life among survivors is also a priority. Uncertainty about future fertility has been described as a distressing factor for this population.
While some studies have reported a decrease in pregnancy rates in survivors of certain cancers, to our knowledge the risk of infertility after cancer diagnosis has not been studied at the population level in AYA.1-4 To this end, we conducted a population-based study in the province of Ontario, Canada, to evaluate the risk of infertility in AYA survivors of selected cancers.
Using health administrative databases, we identified 15,107 female survivors of brain, breast, haematological, head and neck, thyroid, melanoma, colorectal, or urological cancer, who were diagnosed with cancer at a mean age of 31.2 years (standard deviation [SD]: 6.3). These women were compared to 64,315 cancer-free women. Both groups were followed-up for approximately 14 years. Infertility diagnosis after 1 year of cancer was identified using physicians’ billing codes (International Classification of Diseases [ICD]-9 code: 628). Women with infertility previous to cancer diagnosis were excluded. Log-binomial regression models were used adjusting for sociodemographic factors (adjusted relative risk [aRR]).
Overall, the frequency of infertility diagnosis was higher in cancer survivors compared to unexposed women (12.0% versus 9.4%; p<0.001), at a mean age of 34.5 years (SD: 5.7) in survivors and 34.9 years (SD: 5.5) in unexposed women (p<0.001). Survivors of breast (aRR: 1.38; 95% confidence interval [CI]: 1.23–1.55), haematological (aRR: 1.42; 95% CI: 1.28–1.59), thyroid (aRR: 1.17; 95% CI: 1.08–1.27), and melanoma (aRR: 1.13; 95% CI: 0.99–1.30) had a higher risk of infertility diagnosis than women without cancer. These associations were stronger in nulliparous women (i.e., no previous pregnancies) compared to women with previous children. In addition, we conducted multiple sensitivity analyses that reaffirmed our results.
Our study had limitations. Firstly, the accuracy of infertility diagnosis using ICD-9 codes in administrative datasets has not been validated. Secondly, there are nonbiologic factors that could influence the likelihood of seeking a fertility assessment that may not be captured in our study. Finally, information regarding cancer stage or treatment was not available for this analysis.
We concluded that AYA with cancer should have access to specialists in reproductive health for surveillance, and that prospective research studies should be conducted to monitor the reproductive function of this population. While the evidence is stronger for women diagnosed with breast and haematological cancer, our finding of a potential effect of thyroid cancer or melanoma needs to be further studied before any conclusion can be made. Factors other than the type of treatment could play a role on the risk of infertility, including the distress caused by cancer diagnosis, or the pathophysiology of cancer itself, including genetic factors that could be associated to both cancer diagnosis and infertility.