Risk of Breast Cancer and Gynecologic Cancers in a Large Population
Risk of Breast Cancer and Gynecologic Cancers in a Large Population
Infertility is considered to influence the risk of breast cancer and gynecologic cancers. To assess this association, the authors used data from a large cohort of 54,362 women with a diagnosis of infertility who were referred to Danish fertility clinics between 1963 and 1998. Through 2003, 1,975 cancers were identified by linkage to the Danish Cancer Registry. Cancer risk was assessed through standardized incidence ratios (SIRs) and corresponding 95% confidence intervals, using general and parity-specific cancer incidence rates in the general population of Denmark as a reference. After adjustment for parity status, significantly increased SIRs were observed for breast (SIR = 1.08, 95% confidence interval: 1.01, 1.16) and ovarian (SIR = 1.46, 95% confidence interval: 1.24, 1.71) cancer. The risk of breast cancer increased with follow-up time. Similar risk patterns were observed for the different histologic types of breast cancer and all nonmucinous types of ovarian cancer, whereas the risk of mucinous ovarian cancers seemed not to be increased. These data thus suggest higher risks of breast and ovarian cancer among infertile women. However, since these results could not distinguish the effects of underlying infertility from the effects of fertility treatment, additional studies are needed to disentangle the effects of these two factors.
Breast cancer and gynecologic cancers are known to be of multifactorial etiology. One of the most well-established risk factors for breast, ovarian, and endometrial cancer is parity: Nulliparous women have an increased risk in comparison with parous women. A diagnosis of infertility and infertility treatment are related factors also considered to influence risk of these cancers. The question of whether infertility and infertility treatment increase the risk of breast and gynecologic cancers is a matter of great public health concern, considering the large and constantly growing number of women requesting infertility treatment and the high incidence of these cancers in most Western countries.
A number of epidemiologic studies have examined the risks of breast, ovarian, endometrial, and cervical cancer in populations of infertile women. Results have been inconsistent. The majority of studies have found no convincing relation to risk, but the results in some may have been influenced by various methodological limitations, such as low statistical power due to small numbers of cases, short follow-up periods, loss to follow-up, and inability to control for potential confounders.
Concerning confounders, it is of particular importance to take parity into account, as the frequency of nulliparity is higher among infertile women than among fertile women. Otherwise, the cancer risk among infertile women will be overestimated as a consequence of the increased risk in nulliparous women. The availability of appropriate comparison groups is therefore of fundamental importance in cohort studies examining the association between infertility and cancer. Researchers in several cohort studies have used standardized incidence ratios (SIRs) to compare cancer risk in infertile women with that in the general population. However, use of general-population cancer incidence rates for comparisons makes it difficult to adjust for differences in cancer predictors, except age and calendar time. None of the previous studies using SIRs to compare cancer risk in cohorts of infertile women with that in the general population have been able to stratify or adjust for parity; therefore, the SIRs in these studies may have overestimated cancer risk and are thus difficult to interpret.
In Denmark, however, we have the opportunity to calculate parity-specific and parity-adjusted SIRs in cohorts of infertile women, since parity status for the infertility cohort and parity-specific cancer incidence rates for the general population can be calculated on the basis of information contained in the Danish Civil Registration System and the Danish Cancer Register, both of which are nationwide population-based registries. We have established a cohort including 54,362 Danish women diagnosed with infertility during the period 1963-1998. To our knowledge, this infertility cohort involves the largest number of breast and gynecologic cancer cases compiled to date. Furthermore, practically no women from the cohort are lost to follow-up, and there is almost complete ascertainment of cancer diagnoses, since the personal identification number assigned in Denmark allows precise linkage between the infertility cohort and the Danish population-based registries. Through the use of these data, we designed the present study to evaluate the risk (measured as parity-specific and parity-adjusted SIR estimates) of breast and gynecologic cancers after a diagnosis of infertility.
Infertility is considered to influence the risk of breast cancer and gynecologic cancers. To assess this association, the authors used data from a large cohort of 54,362 women with a diagnosis of infertility who were referred to Danish fertility clinics between 1963 and 1998. Through 2003, 1,975 cancers were identified by linkage to the Danish Cancer Registry. Cancer risk was assessed through standardized incidence ratios (SIRs) and corresponding 95% confidence intervals, using general and parity-specific cancer incidence rates in the general population of Denmark as a reference. After adjustment for parity status, significantly increased SIRs were observed for breast (SIR = 1.08, 95% confidence interval: 1.01, 1.16) and ovarian (SIR = 1.46, 95% confidence interval: 1.24, 1.71) cancer. The risk of breast cancer increased with follow-up time. Similar risk patterns were observed for the different histologic types of breast cancer and all nonmucinous types of ovarian cancer, whereas the risk of mucinous ovarian cancers seemed not to be increased. These data thus suggest higher risks of breast and ovarian cancer among infertile women. However, since these results could not distinguish the effects of underlying infertility from the effects of fertility treatment, additional studies are needed to disentangle the effects of these two factors.
Breast cancer and gynecologic cancers are known to be of multifactorial etiology. One of the most well-established risk factors for breast, ovarian, and endometrial cancer is parity: Nulliparous women have an increased risk in comparison with parous women. A diagnosis of infertility and infertility treatment are related factors also considered to influence risk of these cancers. The question of whether infertility and infertility treatment increase the risk of breast and gynecologic cancers is a matter of great public health concern, considering the large and constantly growing number of women requesting infertility treatment and the high incidence of these cancers in most Western countries.
A number of epidemiologic studies have examined the risks of breast, ovarian, endometrial, and cervical cancer in populations of infertile women. Results have been inconsistent. The majority of studies have found no convincing relation to risk, but the results in some may have been influenced by various methodological limitations, such as low statistical power due to small numbers of cases, short follow-up periods, loss to follow-up, and inability to control for potential confounders.
Concerning confounders, it is of particular importance to take parity into account, as the frequency of nulliparity is higher among infertile women than among fertile women. Otherwise, the cancer risk among infertile women will be overestimated as a consequence of the increased risk in nulliparous women. The availability of appropriate comparison groups is therefore of fundamental importance in cohort studies examining the association between infertility and cancer. Researchers in several cohort studies have used standardized incidence ratios (SIRs) to compare cancer risk in infertile women with that in the general population. However, use of general-population cancer incidence rates for comparisons makes it difficult to adjust for differences in cancer predictors, except age and calendar time. None of the previous studies using SIRs to compare cancer risk in cohorts of infertile women with that in the general population have been able to stratify or adjust for parity; therefore, the SIRs in these studies may have overestimated cancer risk and are thus difficult to interpret.
In Denmark, however, we have the opportunity to calculate parity-specific and parity-adjusted SIRs in cohorts of infertile women, since parity status for the infertility cohort and parity-specific cancer incidence rates for the general population can be calculated on the basis of information contained in the Danish Civil Registration System and the Danish Cancer Register, both of which are nationwide population-based registries. We have established a cohort including 54,362 Danish women diagnosed with infertility during the period 1963-1998. To our knowledge, this infertility cohort involves the largest number of breast and gynecologic cancer cases compiled to date. Furthermore, practically no women from the cohort are lost to follow-up, and there is almost complete ascertainment of cancer diagnoses, since the personal identification number assigned in Denmark allows precise linkage between the infertility cohort and the Danish population-based registries. Through the use of these data, we designed the present study to evaluate the risk (measured as parity-specific and parity-adjusted SIR estimates) of breast and gynecologic cancers after a diagnosis of infertility.
