Abstract
Purpose
To examine trends, characteristics and outcomes of women who develop both ovarian and breast cancers.
Methods
This is a retrospective study examining the Surveillance, Epidemiology, and End Results Program from 1973 to 2013. Among ovarian cancer (n = 133,149) and breast cancer (n = 1,143,219) cohorts, women with both diagnoses were identified and temporal trends, tumor characteristics and survival were examined.
Results
There were 6446 women with both malignancies, representing 4.8% of the ovarian cancer cohort and 0.6% of the breast cancer cohort. Women with ovarian cancer who had secondary breast cancer were younger than those without secondary breast cancer early in the study period (52.3 versus 59.2 in 1973) but older in more recent years (68.5 versus 62.1 in 2013, P < 0.001). The number of breast cancer survivors who developed postcedent ovarian cancer decreased from 1.5 to 0.2% from 1979 to 2008 (relative risk reduction 90.0%, P < 0.05). Similarly, the number of ovarian cancer survivors who developed postcedent breast cancer decreased from 7.2 to 2.0% from 1973 to 2008 (relative risk reduction 72.4%, P < 0.05). Tumor characteristics were more likely to be favorable in women with ovarian cancer who developed postcedent breast cancer but unfavorable in those who had antecedent breast cancer (all, P < 0.05). Women with ovarian cancer who had secondary breast cancer had superior cause-specific survival compared to those who did not develop breast cancer regardless of breast cancer timing (P < 0.05).
Conclusion
Our study demonstrated that the demographics of women who develop breast cancer and ovarian cancer have changed over time and diagnosis of secondary breast cancer after ovarian cancer has decreased.
Keywords: Ovarian cancer, Breast cancer, Secondary cancer, Trend, Survival
Introduction
While breast cancer is the most common female malignancy in the United States projected to affect nearly 268,600 women in 2019, ovarian cancer is the most deadly gynecologic malignancy, with approximately 22,500 new diagnoses and 14,000 deaths also estimated in 2019 [1]. These two malignancies have been inexorably linked since the landmark discovery of the breast cancer susceptibility genes, BRCA1 and BRCA2, in the mid-1990s. Up to 24% of ovarian cancer cases and about 5% of breast cancer cases are attributable to BRCA mutations [2]. And importantly, these women are at risk of developing not only primary ovarian cancer or breast cancer but also secondary breast cancer or secondary ovarian cancer, respectively, during their lifetime [3–14].
While various studies have previously examined secondary primary cancer (SPC) following both ovarian cancer and breast cancer, trends of secondary breast cancer and secondary ovarian cancer in this population have been understudied. Given the increasing utilization of genetic testing in ovarian cancer and breast cancer patients and recommendations for prophylactic intervention in those found to be at risk, it is important to examine potential changes in the trends of secondary breast cancer and ovarian cancer [2, 15]. Additionally, increasing life expectancy and obesity rates in the United States are both important factors that have undoubtedly influenced the epidemiology of these cancers [16, 17].
Heterogeneity in the SPC literature also exists with regard to the impact of the sequence of breast cancer and ovarian cancer diagnoses; some studies have examined breast cancer after ovarian cancer [3–9], while others have analyzed breast cancer before ovarian cancer [10–14]. To date, a comprehensive analysis examining both sequences of cancer diagnoses is missing.
Based on the above, we hypothesized that characteristics and survival outcomes of women with ovarian cancer who later developed breast cancer have changed over time and differ based on the timing of the secondary breast cancer diagnosis. The objective of this study was to examine trends, characteristics, and outcomes of women with ovarian cancer who develop secondary breast cancer.
Materials and methods
Data and eligibility
The National Cancer Institute’s Surveillance, Epidemiology, and End Results (SEER) Program was utilized for this retrospective observational cohort study. This database is the largest population-based tumor registry in the United States, covering ~ 34.6% of the population in the most recent areas [18]. Since its inception in 1973, the database spans over 40 years and is recognized as a powerful resource for trend and incidence analyses. Data entry and quality control for this database are managed by registered and trained personnel [19]. The University of Southern California institutional review board deemed this study exempt because of the use of publicly available deidentified data. Primary incident ovarian cancer and breast cancer cases diagnosed between 1973 and 2013 were eligible for study, being followed over time. Multiple entries in the database and borderline ovarian tumors were excluded from the analysis.
Study approach
The SEER*Stat 8.3.2 (IMS Inc., Calverton, MD, USA) was utilized to abstract the cases from the database. Two independent datasets were generated for ovarian cancer and breast cancer cohorts. Then, study identification numbers assigned by the program were reviewed in each cohort and the secondary entry cases were excluded to retain only the primary entry cases. The two datasets were then linked and cases with the same study identification number were considered SPC cases [20, 21].
Clinico-pathological variables
The following information was abstracted from the database: patient demographics at diagnosis (age, year and month, race/ethnicity, registration area, and marital status), tumor characteristics (cancer stage, histologic type, tumor grade, and tumor size), treatment types, and survival (follow-up, vital status, and cause of death).
Study definition
Among women with both ovarian cancer and breast cancer, the time interval between the two diagnoses was examined. If the time interval was ≤ 2 months, the case was defined as synchronous SPC [22]. If the SPC was diagnosed > 2 months before or after the index cancer, the case was defined as antecedent or postcedent SPC, respectively.
High-grade serous ovarian cancer was defined as moderately and poorly differentiated serous histology in this study [23]. Cancer stage was based on the American Joint Committee on Cancer surgical-pathological staging classification schema [24]. Histologic subtype was based on the International Classification of Disease for Oncology 3rd edition site/histology validation and the World Health Organization histological classification codes [20, 21]. Grouping of clinico-pathological factors was based on our prior studies [20, 21].
Cause-specific survival (CSS) was defined as the time between diagnosis of the index cancer and death from the index cancer, which was determined by the SEER Cause Specific Death Classification, “death attributable to this cancer diagnosis.” If the SEER Death Classification code specified “death attributable to causes other than this cancer diagnosis,” the case was not interpreted as CSS. Overall survival (OS) was defined as the time between diagnosis of the index cancer and death from any cause. Cases without these survival events were censored at the last available follow-up. Cause of death was also linked with state mortality records and the National Death Index for external verification [25].
Study objectives
The primary objective of the study was to examine trends of secondary breast cancer among women with ovarian cancer. The secondary objective was to examine clinico-pathological characteristics and survival of women with ovarian cancer and secondary breast cancer. Secondary ovarian cancer was also examined in the breast cancer cohort.
Statistical consideration
Differences in continuous variables were assessed by the Student’s t test or one-way ANOVA as appropriate. Differences in ordinal/categorical variables were assessed by the Chi-squared test. The Kaplan–Meier method was used to construct survival curves and differences were assessed by the log-rank test. In multivariable analysis, Cox proportional hazard regression models were computed to estimate hazard ratio (HR) and 95% confidence interval (CI). The association between SPC and survival was adjusted for patient demographics, tumor characteristics, and treatment types.
Temporal trends were assessed using the Joinpoint Trend Software (version 4.4.0.0, National Cancer Institute, Bethesda, MD, USA). Patient age at the index cancer diagnosis or calendar years were used as time increments (percent frequency with CI or mean with standard error). Linear segmented regression was used for temporal trends and log transformation was performed to determine the annual percent change of the slope with a 95% CI [26].
In sensitivity analyses, the study population was limited to only epithelial ovarian cancer alone or high-grade serous ovarian cancer alone because hereditary breast–ovarian cancer syndrome is typically associated with these histologies [2]. Postcedent breast cancer was examined among women with ovarian cancer without synchronous/antecedent breast cancer and who did not die of ovarian cancer within a follow-up of ≥ 5 (and ≥ 10) years, termed cancer survivors. Among cancer survivors, trends in standardized incidence ratio (SIR) for breast cancer and ovarian cancer were also examined based on observed and expected events in a standardized population [27]. This was based on the rationale that diagnosis of postcedent SPC is dependent on duration of follow-up and a short follow-up period may result in lead time bias [28]. Trends of secondary ovarian cancer were also examined among women with breast cancer to assess if the results in the primary cohort of ovarian cancer were validated. Last, trends of mortality from SPC were examined in each cohort.
A P value of < 0.05 was considered statistically significant (two-tailed hypotheses). Statistical Package for Social Sciences (SPSS, version 24.0, IBM Corp, Armonk, NY, USA) was used for the statistical analysis. The STROBE guidelines were consulted to report the observational study [29].
Results
Ovarian cancer cohort
Among 133,481 ovarian cancer cases during the study period, 332 secondary entries were excluded and the remaining 133,149 women represented the ovarian cancer cohort. There were 6446 women who had both breast cancer and ovarian cancer, representing 4.8% (95% CI 4.7–5.0) of the ovarian cancer cohort. There were 6395 women with an available time interval between the two diagnoses and the most common sequence of diagnoses in women with ovarian cancer was antecedent breast cancer (n = 3973, 62.1%) followed by postcedent breast cancer (n = 1821, 28.5%) and synchronous breast cancer (n = 601, 9.4%).
Compared to women with ovarian cancer who did not have secondary breast cancer, women with ovarian cancer and secondary breast cancer were younger early in the study period (mean, 52.3 versus 59.2 in 1973) but older after the early–mid-1990s (68.5 versus 62.1 in 2013) (P < 0.05; Fig. 1a). Breast cancer diagnosis was more likely to antecede ovarian cancer diagnosis after age 38 (P < 0.001; Fig. 1c).
Fig. 1.
Trends for age at cancer diagnosis. a, b Age of ovarian cancer diagnosis and age of breast cancer diagnosis per calendar year are shown. Y-axes are truncated to 80 years. Women without secondary primary cancer is shown in blue lines and those who had diagnosis of secondary breast cancer or secondary ovarian cancer were shown in red lines. Dots represent observed value for mean age at cancer diagnosis. Bars represent standard error. Lines represent modeled value per a linear segmented regression model. c, d Boxplots for trends in time sequence of the two cancers based on calendar year for ovarian cancer and breast cancer per age at the index cancer diagnosis
Trends of SPC were examined by patient age (Fig. 2a). Among women with ovarian cancer, the number of women who developed secondary breast cancer increased significantly from 0.6 to 6.1% between age 21 and 45 (P < 0.05); the peak age for secondary breast cancer was 45. When women with ovarian cancer were stratified by the chronology of breast cancer diagnosis, postcedent breast cancer diagnosis peaked at age 45 while diagnosis of antecedent breast cancer continued to increase with increasing age (all, P < 0.05).
Fig. 2.
Age-specific trends of secondary primary cancer. a, b Age-specific trends are shown for proportions of secondary breast cancer among women with ovarian cancer and secondary ovarian cancer among women with breast cancer are shown. Age was based on the index cancer diagnosis. Y-axes are truncated to 8% (a) and 1% (b), respectively. Dots represent observed value for percentage at cancer diagnosis. Bars represent 95% confidence interval. Lines represent modeled value per a linear segmented regression model
Trends of SPC were examined per calendar year (Figure S1a). The number of women with ovarian cancer and secondary breast cancer increased from 2.9 to 6.6% between 1973 and 1998, corresponding to a relative risk increase of 126% (P < 0.05). The number was then statistically similar afterwards (P > 0.05). When stratified by the timing of secondary breast cancer diagnosis in relation to the ovarian cancer diagnosis, the number of women with ovarian cancer and antecedent breast cancer increased from 2.3 to 4.1% between 2001 and 2013 (81.0% relative risk increase, P < 0.05).
SPC trends were then examined in ovarian cancer survivors. Among the survivors without synchronous/antecedent breast cancer (n = 30,170, median follow-up 11.6 years; Fig. S2), the number of women who developed postcedent breast cancer decreased from 7.2 to 2.0% between 1973 and 2008 (relative risk reduction 72.4%, P < 0.05; Fig. 3a). Similar results were observed for 10-year cancer survivors. When standardized for the incidence in the general population (Fig. 3b), decreasing trends of secondary breast cancer among ovarian cancer survivors were observed during the study period (both, P < 0.001). Mortality from secondary breast cancer in women with ovarian cancer decreased during the study period (Fig. 3c).
Fig. 3.
Trends of postcedent secondary primary cancer among cancer survivors and mortality rate from secondary primary cancer. a Women who survived from the index cancer for 5 years or longer without antecedent or synchronous secondary primary cancer were examined for postcedent diagnosis of secondary primary cancer. b Standardized incidence ratio for postcedent secondary cancer is shown for breast and ovarian cancer survivors. c *Deaths from secondary ovarian cancer among women with breast cancer and deaths from breast cancer among women with ovarian cancer are shown over time. Dots represent observed value for percentage at cancer diagnosis. Bars represent 95% confidence interval. Lines represent modeled value per a linear segmented regression model. SPC secondary primary cancer, OC ovarian cancer, BC breast cancer
Patient demographics were examined among women with ovarian cancer. Irrespective of the timing of secondary breast cancer diagnosis (Table S1), women with ovarian cancer and secondary breast cancer were older and more likely to be white and married (all, P < 0.05). Ovarian tumors in the secondary breast cancer group were less likely to be stage IV but more likely to be high grade and serous histology (all, P < 0.05). However, when stratified by the chronology of SPC in relation of ovarian cancer diagnosis (Table 1), clinico-pathological characteristics were largely different across the groups. Women with ovarian cancer in the antecedent/synchronous breast cancer groups were older and more likely to have serous, high-grade, and advanced-stage disease. On the contrary, women with ovarian cancer and postcedent breast cancer were younger with endometrioid or mucinous tumors and low-grade, early-stage disease (all, P < 0.05).
Table 1.
Clinico-pathological characteristics of ovarian cancer based on timing of secondary breast cancer (n = 133,098)
| Characteristic | No breast cancer | Postcedent breast cancer | Synchronous breast cancer | Antecedent breast cancer | P value |
|---|---|---|---|---|---|
| Number | n = 126,703 | n = 1821 | n = 601 | n = 3973 | |
| Age (continuous) | 62.0 (± 16.2) | 57.4 (± 12.6) | 64.1 (± 12.9) | 66.5 (± 13.0) | < 0.001 |
| Year of diagnosis | < 0.001 | ||||
| Before 1980 | 9949 (7.9%) | 261 (14.3%) | 46 (7.7%) | 81 (2.0%) | |
| 1980–1989 | 16,311 (12.9%) | 412 (22.6%) | 65 (10.8%) | 401 (10.1%) | |
| 1990–1999 | 23,304 (18.4%) | 469 (25.2%) | 114 (19.0%) | 862 (21.7%) | |
| 2000–2009 | 55,149 (43.5%) | 632 (34.7%) | 262 (43.6%) | 1745 (43.9%) | |
| 2010 or later | 21,990 (17.4%) | 57 (3.1%) | 114 (19.0%) | 884 (22.3%) | |
| Registry area | < 0.001 | ||||
| West | 65,595 (51.8%) | 894 (49.1%) | 298 (49.6%) | 2126 (53.5%) | |
| Central | 29,477 (23.3%) | 466 (25.6%) | 145 (24.1%) | 990 (24.9%) | |
| East | 31,631 (25.0%) | 461 (25.3%) | 158 (26.3%) | 857 (21.6%) | |
| Race/ethnicity | < 0.001 | ||||
| White | 97,265 (76.9%) | 1,482 (81.4%) | 481 (80.0%) | 3260 (82.1%) | |
| Black | 9365 (7.4%) | 107 (5.9%) | 44 (7.3%) | 215 (5.4%) | |
| Hispanic | 10,993 (8.7%) | 106 (5.8%) | 38 (6.3%) | 271 (6.8%) | |
| Asian | 7199 (5.7%) | 102 (5.6%) | 28 (4.7%) | 183 (4.6%) | |
| Others/unknown | 1604 (1.3%) | 24 (1.3%) | 10 (1.7%) | 44 (1.1%) | |
| Marital status | < 0.001 | ||||
| Single | 20,294 (16.0%) | 272 (14.9%) | 92 (15.3%) | 423 (10.6%) | |
| Married | 61,591 (48.6%) | 1064 (58.4%) | 291 (48.4%) | 2063 (51.9%) | |
| Others | 40,466 (31.9%) | 431 (23.7%) | 201 (33.4%) | 1342 (33.8%) | |
| Unknown | 4352 (3.4%) | 54 (3.0%) | 17 (2.8%) | 145 (3.6%) | |
| Stage | < 0.001 | ||||
| I | 26,485 (20.9%) | 737 (40.5%) | 114 (19.0%) | 625 (15.7%) | |
| II | 8910 (7.0%) | 216 (11.9%) | 49 (8.2%) | 339 (8.5%) | |
| III | 38,155 (30.1%) | 442 (24.3%) | 187 (31.1%) | 1559 (39.2%) | |
| IV | 39,832 (31.4%) | 324 (17.8%) | 187 (31.1%) | 1111 (28.0%) | |
| Unknown | 13,321 (10.5%) | 102 (5.6%) | 64 (10.6%) | 339 (8.5%) | |
| Grade | < 0.001 | ||||
| 1 | 9022 (7.1%) | 233 (12.8%) | 41 (6.8%) | 163 (4.1%) | |
| 2 | 16,723 (13.2%) | 324 (17.8%) | 69 (11.5%) | 435 (10.9%) | |
| 3 | 48,495 (38.3%) | 676 (37.1%) | 239 (39.8%) | 1903 (47.9%) | |
| Unknown | 52,463 (41.4%) | 588 (32.3%) | 252 (41.9%) | 1472 (37.1%) | |
| Histology | < 0.001 | ||||
| Serous | 52,920 (41.8%) | 781 (42.9%) | 277 (46.1%) | 2005 (50.5%) | |
| Clear cell | 5280 (4.2%) | 101 (5.5%) | 24 (4.0%) | 134 (3.4%) | |
| Endometrioid | 11,961 (9.4%) | 324 (17.8%) | 48 (8.0%) | 298 (7.5%) | |
| Mucinous | 9279 (7.3%) | 202 (11.1%) | 37 (6.2%) | 175 (4.4%) | |
| Other epithelial | 24,411 (19.3%) | 217 (11.9%) | 122 (20.3%) | 807 (20.3%) | |
| SCST | 2038 (1.6%) | 45 (2.5%) | 6 (1.0%) | 32 (0.8%) | |
| MOGCT | 3295 (2.6%) | 28 (1.5%) | 3 (0.5%) | 9 (90.2%) | |
| Others | 17,519 (13.8%) | 123 (6.8%) | 84 (14.0%) | 513 (12.9%) | |
| Tumor size (cm) | < 0.001 | ||||
| < 4 | 15,740 (12.4%) | 294 (16.1%) | 71 (11.8%) | 577 (14.5%) | |
| 4–10 | 20,172 (15.9%) | 276 (15.2%) | 97 (16.1%) | 830 (20.9%) | |
| > 10 | 22,170 (17.5%) | 297 (16.3%) | 131 (21.8%) | 624 (15.7%) | |
| Unknown | 68,621 (54.2%) | 954 (52.4%) | 302 (50.2%) | 1942 (48.9%) |
Among 6446 women who had both breast and ovarian cancer diagnosis, there were 6395 women who had information for time interval between the two diagnoses. Mean (± standard deviation) or number (percent per column) is shown. Percentages per row for year of diagnosis are shown in Supplemental Table S5
SCST sex cord-stromal tumors, MOGCT malignant ovarian germ cell tumors
The median follow-up of censored cases was 6.0 years (interquartile range 2.3–12.1). There were 73,303 deaths from ovarian cancer and 90,633 deaths from any cause. In multivariable analysis (Table 2), not stratifying by the chronology of the SPC and ovarian cancer diagnoses, women with ovarian cancer and secondary breast cancer had improved CSS and OS compared to those without secondary breast cancer (both, P < 0.001). However, when analyzed by the chronology of the SPC and ovarian cancer diagnoses, all women with ovarian cancer and secondary breast cancer still had improved CSS, but only women with postcedent breast cancer (adjusted-HR 0.44, 95% CI 0.41–0.47) and not those with synchronous/antecedent breast cancer had improved OS compared to women without secondary breast cancer.
Table 2.
Associations of secondary primary breast cancer and ovarian cancer survival based on histology types
| Characteristic | Cause-specific survival |
Overall survival |
||
|---|---|---|---|---|
| HR (95% CI) | P value | HR (95% CI) | P value | |
| All histology | ||||
| Secondary breast cancer | ||||
| No | 1 | 1 | ||
| Yes (any timing) | 0.56 (0.54–0.59) | < 0.001 | 0.75 (0.73–0.78) | < 0.001 |
| Secondary breast cancer | ||||
| No | 1 | 1 | ||
| Postcedent | 0.21 (0.18–0.23) | < 0.001 | 0.44 (0.41–0.47) | < 0.001 |
| Synchronous | 0.64 (0.56–0.73) | < 0.001 | 0.96 (0.87–1.06) | 0.41 |
| Antecedent | 0.78 (0.74–0.81) | < 0.001 | 0.96 (0.93–1.01) | 0.06 |
| Epithelial | ||||
| Secondary breast cancer | ||||
| No | 1 | 1 | ||
| Yes (any timing) | 0.57 (0.55–0.60) | < 0.001 | 0.75 (0.73–0.78) | < 0.001 |
| Secondary breast cancer | ||||
| No | 1 | 1 | ||
| Postcedent | 0.21 (0.18–0.23) | < 0.001 | 0.44 (0.41–0.47) | < 0.001 |
| Synchronous | 0.71 (0.62–0.81) | < 0.001 | 1.02 (0.92–1.13) | 0.72 |
| Antecedent | 0.80 (0.76–0.84) | < 0.001 | 0.96 (0.93–1.01) | 0.14 |
| High-grade serous | ||||
| Secondary breast cancer | ||||
| No | 1 | 1 | ||
| Yes (any timing) | 0.62 (0.58–0.66) | < 0.001 | 0.75 (0.71–0.79) | < 0.001 |
| Secondary breast cancer | ||||
| No | 1 | 1 | ||
| Postcedent | 0.23 (0.19–0.27) | < 0.001 | 0.39 (0.34–0.44) | < 0.001 |
| Synchronous | 0.72 (0.57–0.91) | 0.006 | 0.99 (0.82–1.19) | 0.88 |
| Antecedent | 0.84 (0.79–0.91) | < 0.001 | 0.96 (0.90–1.02) | 0.19 |
Multivariable analysis with Cox proportional hazard regression models for the results. Covariates adjusted for the association of secondary breast cancer and ovarian cancer survival included patient age (continuous), registry area (West, Central, and East), race/ethnicity (White, Black, Hispanic, Asian, and others/unknown), marital status (single, married, others, and unknown), histology (serous, clear cell, endometrioid, mucinous, other epithelial, sex cord-stromal tumors, malignant germ cell tumors, and others), cancer stage (I, II, III, IV, and unknown), tumor grade (1, 2, 3, and unknown), and tumor size (< 4, 4–10, > 10 cm, and unknown)
HR hazard ratio, CI confidence interval
When the study population was limited to epithelial tumors only (n = 109,403 including 5552 secondary breast cancer, 5.1%) or high-grade serous tumors only (n = 37,466 including 2164 secondary breast cancer 5.8%), similar results as obtained in analysis of the whole cohort were observed. Women with ovarian cancer and postcedent breast cancer had more favorable patient characteristics, tumor factors, and prognosis (all, P < 0.05; Table 2 and Table S2–3). Diagnosis of postcedent breast cancer in long-term ovarian cancer survivors decreased over time in both subgroups (epithelial ovarian cancer, 7.5–2.1% between 1973 and 2008, relative risk reduction 71.0%; high-grade serous ovarian cancer, 9.5–3.1% between 1995 and 2008, relative risk reduction 67.5%, all, P < 0.05; Fig. S3).
Breast cancer cohort
Among 1,210,332 breast cancer cases during the study period, 67,113 secondary entries were excluded and the remaining 1,143,219 women represented the breast cancer cohort. There were 6446 women who had both breast cancer and ovarian cancer, representing 0.6% (95% CI 0.6–0.6) of the breast cancer cohort. Women with breast cancer and secondary ovarian cancer were younger than those without secondary ovarian cancer early in the study period (mean, 54.1 versus 60.1 in 1973) but older after the mid-late 1990s (64.6 versus 61.8 in 2013) (P < 0.05; Fig. 1b). Ovarian cancer diagnosis was more likely to antecede breast cancer diagnosis after age 67 (P < 0.001; Fig. 1d).
Trends of SPC were examined by patient age among the breast cancer cohort (Fig. 2b. The peak age for secondary ovarian cancer was 75 (P < 0.05). Antecedent ovarian cancer diagnosis peaked at age 64, diagnosis of postcedent ovarian cancer decreased after age 30, and diagnosis of synchronous ovarian cancer increased with increasing age (all, P < 0.05). Trends of SPC were examined per calendar year (Fig. S1b). The number of women with breast cancer and secondary ovarian cancer decreased from 0.9 to 0.3% between 1994 and 2013, corresponding to a relative risk reduction of 69.3% (P < 0.05). The number of women with breast cancer and antecedent ovarian cancer increased from 0.1 to 0.2% between 2001 and 2013 (99.5% relative risk increase), while the number of women with breast cancer and synchronous ovarian cancer decreased by 14.4% between 1973 and 2013 (all, P < 0.05).
SPC trends were then examined in breast cancer survivors. Among the survivors without synchronous/antecedent ovarian cancer (n = 571,677, median follow-up 10.8 years; Figure S2), the number with postcedent ovarian cancer decreased from 1.5 to 0.2% between 1979 and 2008 (relative risk reduction 89.9%, P < 0.05; Fig. 3a). Similar results were observed for 10-year cancer survivors. When standardized for the incidence in the general population (Fig. 3b), decreasing trends of secondary ovarian cancer among breast cancer survivors were observed during the study period (both, P < 0.001). Mortality from secondary ovarian cancer in women with breast cancer decreased during the study period (Fig. 3c).
Discussion
The most salient finding of our study is that diagnosis of postcedent breast cancer decreased among women with ovarian cancer during the study period. This is particularly noteworthy as the incidence of breast cancer in the general US population has been stable if not slightly increasing over the past few decades [30]. We also found that postcedent ovarian cancer has decreased in long-term breast cancer survivors, which validates prior literature [9].
Decreasing rates of postcedent breast cancer and postcedent ovarian cancer seen in our study may be due to increasing utilization of genetic testing for women with newly diagnosed ovarian cancer, allowing for prophylactic intervention [15]. Women found to have hereditary familial breast–ovarian cancer syndrome are often recommended oral contraceptive use and lifestyle modifications as well as to undergo prophylactic mastectomy or risk reducing salpingo-oophorectomy due to the high risk of developing subsequent ovarian or breast cancer [2, 31–34]. For instance, women with BRCA1-2 mutation have 6.8–12.7% of developing subsequent ovarian cancer after breast cancer [2]. Our study found that the rate of decrease in postcedent breast/ovarian cancer diagnosis was most marked in the late 1990s, concomitant with the discovery of the BRCA gene [35, 36]. Thus, this may partly support the link between increasing utilization of genetic testing and a decreasing number of postcedent breast/ovarian cancer.
We also found that antecedent breast cancer among women with ovarian cancer as well as antecedent ovarian cancer in women with breast cancer have increased during the study period. Improved cancer treatments that increase life expectancy and an aging US population with higher chances of developing SPC may be responsible for the increase in antecedent cancers. Our analysis supports this as the age of women with SPC increased compared to the age of those without SPC during the study period.
The heterogeneity of prior studies regarding the effect of the sequence of ovarian cancer and breast cancer diagnoses was addressed in our study with comprehensive analyses stratifying by the chronology of breast cancer and ovarian cancer diagnoses. Among women who developed both malignancies, two-thirds had breast cancer prior to ovarian cancer, emphasizing the importance of genetic testing and preventive strategies in women with breast cancer.
This study found that among women with ovarian cancer and secondary breast cancer, tumor characteristics and outcomes were largely different based upon the sequence of the two malignancies [3–14]. Specifically, women with ovarian cancer and postcedent breast cancer had favorable tumor factors and outcomes whereas those with antecedent breast cancer had high-risk tumor characteristics and poorer survival. While the favorable survival in the ovarian cancer with postcedent breast cancer group is somehow expected as breast cancer diagnosis was captioned among ovarian cancer survivors, it will be of interest if genetic, epigenetic, and patient factors as well as treatment response differ among these groups.
Strengths of this study include that this is a population-based study linking multiple datasets with comprehensive analyses for women with ovarian cancer and secondary breast cancer. This study is likely the first to demonstrate decreasing trends of postcedent breast cancer in women with ovarian cancer. Multiple sensitivity analyses enhance the robustness of analysis.
Limitations of this study are as follows. First, this is a retrospective study and there may be unmeasured confounding factors. For instance, because this database only covers a fraction of US geographic area and does not have information regarding patient relocation, some SPC events occurring outside of database-covered areas may have been missed. Another major limitation is that genetic testing information is not captured in this database, which is particularly relevant as ovarian cancer associated with hereditary breast–ovarian cancer syndrome has similar tumor characteristics but superior survival compared to wild-type ovarian cancer [37]. For instance, differences in the results across the registry area may be due to the differences in availability of/access to genetic testing. Further study is warranted to characterize the differences in ovarian cancer with secondary breast cancer taking into account genetic testing results.
Similarly, information regarding prophylactic interventions such as mastectomy, salpingo-oophorectomy, or oral contraceptive use was not available in this database. As this database does not have recurrence information, complete risk assessment for oncologic outcome was also not possible.
In summary, while postcedent breast cancer following ovarian cancer is not rare and approximately 1 in 20 women with ovarian cancer will develop breast cancer, the number of breast cancer diagnoses following ovarian cancer seems to be decreasing. These results, as well as the recent decrease in secondary malignancy in uterine cancer, are encouraging for patients and care providers with regard to screening for secondary primary cancers during follow-up [38, 39] and further study is warranted to validate this study in different populations.
Supplementary Material
Acknowledgments
Funding Ensign Endowment for Gynecologic Cancer Research (KM).
Footnotes
Electronic supplementary material The online version of this article (https://doi.org/10.1007/s00404-020-05508-3) contains supplementary material, which is available to authorized users.
Conflict of interest Consultant, Clovis Oncology and Tesaro, research funding, Merck (JDW); consultant, Quantgene (LDR); honorarium, Chugai, textbook editorial expense, Springer, and investigator meeting attendance expense, VBL therapeutics (KM); none for others.
Ethical approval This article used publicly available deidentified data and does not contain any studies with human participants performed by any of the authors.
Informed consent Not applicable for this study.
References
- 1.Siegel RL, Miller KD, Jemal A. Cancer statistics, 2019. CA Cancer J Clin 2019;69:7–34. [DOI] [PubMed] [Google Scholar]
- 2.Practice Bulletin No 182: Hereditary Breast and Ovarian Cancer Syndrome (2017) Obstet Gynecol 130:e110–e126. [DOI] [PubMed] [Google Scholar]
- 3.Molina-Montes E, Requena M, Sanchez-Cantalejo E, Fernandez MF, Arroyo-Morales M, Espin J, Arrebola JP, Sanchez MJ (2015) Risk of second cancers cancer after a first primary breast cancer: a systematic review and meta-analysis. Gynecol Oncol 136:158–171 [DOI] [PubMed] [Google Scholar]
- 4.Mellemkjaer L, Friis S, Olsen JH, Scelo G, Hemminki K, Tracey E, Andersen A, Brewster DH, Pukkala E, McBride ML, Kliewer EV, Tonita JM, Kee-Seng C, Pompe-Kirn V, Martos C, Jonasson JG, Boffetta P, Brennan P (2006) Risk of second cancer among women with breast cancer. Int J Cancer 118:2285–2292 [DOI] [PubMed] [Google Scholar]
- 5.Kirova YM, De Rycke Y, Gambotti L, Pierga JY, Asselain B, Fourquet A (2008) Second malignancies after breast cancer: the impact of different treatment modalities. Br J Cancer 98:870–874 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.McGuire V, Whittemore AS, Norris R, Oakley-Girvan I (2000) Survival in epithelial ovarian cancer patients with prior breast cancer. Am J Epidemiol 152:528–532 [DOI] [PubMed] [Google Scholar]
- 7.Li Z, Wu Q, Song J, Zhang Y, Zhu S, Sun S (2018) Risk of second primary female genital malignancies in women with breast cancer: a SEER Analysis. Horm Cancer 9:197–204 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Hall HI, Jamison P, Weir HK (2001) Second primary ovarian cancer among women diagnosed previously with cancer. Cancer Epidemiol Biomarkers Prev 10:995–999 [PubMed] [Google Scholar]
- 9.Schonfeld SJ, de Gonzalez BA, Visvanathan K, Pfeiffer RM, Anderson WF (2013) Declining second primary ovarian cancer after first primary breast cancer. J Clin Oncol 31:738–743 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Domchek SM, Jhaveri K, Patil S, Stopfer JE, Hudis C, Powers J, Stadler Z, Goldstein L, Kauff N, Khasraw M, Offit K, Nathanson KL, Robson M (2013) Risk of metachronous breast cancer after BRCA mutation-associated ovarian cancer. Cancer 119:1344–1348 [DOI] [PubMed] [Google Scholar]
- 11.Gangi A, Cass I, Paik D, Barmparas G, Karlan B, Dang C, Li A, Walsh C, Rimel BJ, Amersi FF (2014) Breast cancer following ovarian cancer in BRCA mutation carriers. JAMA Surg 149:1306–1313 [DOI] [PubMed] [Google Scholar]
- 12.Vencken PM, Kriege M, Hooning M, Menke-Pluymers MB, Heemskerk-Gerritsen BA, van Doorn LC, Collee MM, Jager A, van Montfort C, Burger CW, Seynaeve C (2013) The risk of primary and contralateral breast cancer after ovarian cancer in BRCA1/BRCA2 mutation carriers: Implications for counseling. Cancer 119:955–962 [DOI] [PubMed] [Google Scholar]
- 13.McGee J, Giannakeas V, Karlan B, Lubinski J, Gronwald J, Rosen B, McLaughlin J, Risch H, Sun P, Foulkes WD, Neuhausen SL, Kotsopoulos J, Narod SA (2017) Risk of breast cancer after a diagnosis of ovarian cancer in BRCA mutation carriers: Is preventive mastectomy warranted? Gynecol Oncol 145:346–351 [DOI] [PubMed] [Google Scholar]
- 14.Hunter MI, Ziogas A, Flores F, Brewster WR (2007) Epithelial ovarian cancer and low malignant potential (LMP) tumors associated with a lower incidence of second primary breast cancer. Am J Clin Oncol 30:1–7 [DOI] [PubMed] [Google Scholar]
- 15.Chen Z, Kolor K, Grosse SD, Rodriguez JL, Lynch JA, Green RF, Dotson WD, Bowen MS, Khoury MJ (2018) Trends in utilization and costs of BRCA testing among women aged 18–64 years in the United States, 2003–2014. Genet Med 20:428–434 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Health, United States (2016) Centers for Disease Control and Prevention. https://www.cdc.gov/nchs/data/hus/hus16.pdf#053. Accessed 30 May 2019
- 17.An Aging World: 2015. International Population Reports. United States Census Bureau; https://www.census.gov/content/dam/Census/library/publications/2016/demo/p95-16-1.pdf. Accessed 30 May 2019 [Google Scholar]
- 18.The Surveillance, Epidemiology, and End Results (SEER) Program of the National Cancer Institute, https://seer.cancer.gov/. Accessed 30 May 2019
- 19.National Cancer Registrars Association. http://www.ncra-usa.org/i4a/pages/index.cfm?pageid=1%3e. Accessed 30 May 2019
- 20.Matsuo K, Machida H, Stone RL, Soliman PT, Thaker PH, Roman LD, Wright JD (2017) Risk of subsequent ovarian cancer after ovarian conservation in young women with stage I endometrioid endometrial cancer. Obstet Gynecol 130:403–410 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Matsuo K, Machida H, Blake EA, Holman LL, Rimel BJ, Roman LD, Wright JD (2018) Trends and outcomes of women with synchronous endometrial and ovarian cancer. Oncotarget 9:28757–28771 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Italian Cancer Figures (2014) Report 2013: multiple tumours. Epidemiol Prev 37:1–152 [PubMed] [Google Scholar]
- 23.Bodurka DC, Deavers MT, Tian C, Sun CC, Malpica A, Coleman RL, Lu KH, Sood AK, Birrer MJ, Ozols R, Baergen R, Emerson RE, Steinhoff M, Behmaram B, Rasty G, Gershenson DM (2012) Reclassification of serous ovarian carcinoma by a 2-tier system: a Gynecologic Oncology Group Study. Cancer 118:3087–3094 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Edge SB, Byrd DR, Compton CC, Fritz AG, Greene FL, Trotti A (eds) (2010) AJCC cancer staging manual, 7th edn Springer, New York [Google Scholar]
- 25.National Death Index. https://www.cdc.gov/nchs/ndi/. Accessed 30 May 2019
- 26.Kim HJ, Fay MP, Feuer EJ, Midthune DN (2000) Permutation tests for joinpoint regression with applications to cancer rates. Stat Med 19:335–351 [DOI] [PubMed] [Google Scholar]
- 27.Cancer stat facts. National Cancer Institute Surveillance Epidemiology End Result Program https://seer.cancer.gov/statfacts/. Accessed 30 May 2019 [Google Scholar]
- 28.Rosso S, De Angelis R, Ciccolallo L, Carrani E, Soerjomataram I, Grande E, Zigon G, Brenner H (2009) Multiple tumours in survival estimates. Eur J Cancer 45:1080–1094 [DOI] [PubMed] [Google Scholar]
- 29.von Elm E, Altman DG, Egger M, Pocock SJ, Gotzsche PC, Van-denbroucke JP (2007) Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) statement: guidelines for reporting observational studies. BMJ 335:806–808 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30.Cancer stat facts: female breast cancer. National Cancer Institute, Surveillance, Epidemiology, and End Results Program; https://seer.cancer.gov/statfacts/html/breast.html. Accessed 30 May 2019 [Google Scholar]
- 31.Kotsopoulos J, Lubinski J, Gronwald J, Cybulski C, Demsky R, Neuhausen SL, Kim-Sing C, Tung N, Friedman S, Senter L, Weitzel J, Karlan B, Moller P, Sun P, Narod SA (2015) Factors influencing ovulation and the risk of ovarian cancer in BRCA1 and BRCA2 mutation carriers. Int J Cancer 137:1136–1146 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32.Primas H, Kroiss R, Kalteis K, Rappaport C, Muhr D, Primas C, Kubista E, Horvat R, Oefner P, Singer WC, The Austrian Hereditary B, Ovarian Cancer Group T (2012) Impact of lifestyle factors on preneoplastic changes in prophylactic oophorectomies of BRCA mutation carriers. Eur J Cancer Prev 21(199):204. [DOI] [PubMed] [Google Scholar]
- 33.Eisen A, Lubinski J, Klijn J, Moller P, Lynch HT, Offit K, Weber B, Rebbeck T, Neuhausen SL, Ghadirian P, Foulkes WD, Gershoni-Baruch R, Friedman E, Rennert G, Wagner T, Isaacs C, Kim-Sing C, Ainsworth P, Sun P, Narod SA (2005) Breast cancer risk following bilateral oophorectomy in BRCA1 and BRCA2 mutation carriers: an international case-control study. J Clin Oncol 23:7491–7496 [DOI] [PubMed] [Google Scholar]
- 34.Walker JL, Powell CB, Chen LM, Carter J, Bae Jump VL, Parker LP, Borowsky ME, Gibb RK (2015) Society of Gynecologic Oncology recommendations for the prevention of ovarian cancer. Cancer 121:2108–2120 [DOI] [PubMed] [Google Scholar]
- 35.Wooster R, Bignell G, Lancaster J, Swift S, Seal S, Mangion J, Collins N, Gregory S, Gumbs C, Micklem G (1995) Identification of the breast cancer susceptibility gene BRCA2. Nature 378:789–792 [DOI] [PubMed] [Google Scholar]
- 36.Hall JM, Lee MK, Newman B, Morrow JE, Anderson LA, Huey B, King MC (1990) Linkage of early-onset familial breast cancer to chromosome 17q21. Science 250:1684–1689 [DOI] [PubMed] [Google Scholar]
- 37.Boyd J, Sonoda Y, Federici MG, Bogomolniy F, Rhei E, Maresco DL, Saigo PE, Almadrones LA, Barakat RR, Brown CL, Chi DS, Curtin JP, Poynor EA, Hoskins WJ (2000) Clinicopathologic features of BRCA-linked and sporadic ovarian cancer. JAMA 283:2260–2265 [DOI] [PubMed] [Google Scholar]
- 38.Matsuo K, Mandelbaum RS, Machida H, Yoshihara K, Muggia FM, Roman LD, Wright JD (2019) Decreasing secondary primary uterine cancer after breast cancer: a population-based analysis. Gynecol Oncol (In Press) [DOI] [PMC free article] [PubMed] [Google Scholar]
- 39.Matsuo K, Mandelbaum RS, Machida H, Yoshihara K, Muggia FM, Roman LD, Wright JD (2019) Decreasing trends of secondary primary colorectal cancer among women with uterine cancer: a population-based analysis. J Clin Med 8:714. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.