Declining Second Primary Ovarian Cancer After First Primary Breast Cancer

Sara J Schonfeld; Amy Berrington de Gonzalez; Kala Visvanathan; Ruth M Pfeiffer; William F Anderson

doi:10.1200/JCO.2012.43.2757

. 2013 Jan 2;31(6):738–743. doi: 10.1200/JCO.2012.43.2757

Declining Second Primary Ovarian Cancer After First Primary Breast Cancer

Sara J Schonfeld ¹, Amy Berrington de Gonzalez ¹, Kala Visvanathan ¹, Ruth M Pfeiffer ¹, William F Anderson ^1,^✉

PMCID: PMC3574269 PMID: 23284037

Abstract

Purpose

Although ovarian cancer incidence rates have declined in the United States, less is known of ovarian cancer trends among survivors of breast cancer. Therefore, we examined second primary ovarian cancers after first primary breast cancer.

Methods

Data were obtained from the Surveillance, Epidemiology, and End Results program (1973 to 2008). Standardized incidence ratios (SIRs) were calculated as the observed numbers of ovarian cancers among survivors of breast cancer compared with the expected numbers in the general population. Absolute rates were measured as the incidence rates for second primary ovarian cancer by year of diagnosis of the first primary breast cancer adjusted for age of breast cancer diagnosis and years since diagnosis.

Results

SIRs for second primary ovarian cancer were elevated over the entire study period (SIR, 1.24; 95% CI, 1.2 to 1.3), whereas the absolute rates declined with an estimated annual percentage change near 1% (−1.34% to −0.09% per year). Secular trends for second ovarian cancers were similar after estrogen receptor (ER) –positive and ER-negative breast cancers, whereas the age-specific patterns varied significantly by ER expression (P for interaction < .001). The largest SIR was among women age less than 50 years with ER-negative breast cancer (SIR, 4.35; 95% CI, 3.5 to 5.4).

Conclusion

Persistently elevated SIRs along with decreasing absolute rates over the entire study period suggest that ovarian cancers in both the general population and survivors of breast cancer are declining in parallel, possibly because of common risk factor exposures. Analytic studies are needed to further assess the parallel overall trends and the age-specific interaction by ER expression.

INTRODUCTION

Ovarian cancer incidence rates have declined in the US general population for at least three decades,^1,2 possibly because of increased usage of combined oral contraceptives³ and/or other unknown factors. However, less is known of second primary ovarian cancer after breast cancer (ie, one of the more common new malignancies among an expanding population of survivors of breast cancer).^4,5 Therefore, we analyzed incidence trends for second primary ovarian cancer after first primary breast cancer using nationally representative data from the National Cancer Institute's Surveillance, Epidemiology, and End Results (SEER) program.

The incidence rates for second primary cancers generally are expressed as standardized incidence ratios (SIRs),⁶ defined as the ratio of the observed to expected cancers. However, SIRs do not capture important trends in the absolute rates of second primary cancers that might vary by age at diagnosis of the first primary cancer (ie, characterizing carcinogenic exposures accumulated over a lifetime),⁷ year of first primary cancer diagnosis (ie, proxy for changing screening, treatment, or practice patterns),⁸ and/or time since first primary cancer diagnosis (ie, latency exclusion period). Second primary ovarian cancers after breast cancer might also vary by estrogen receptor (ER) expression of the first primary breast cancer.

Therefore, we examined SIRs as well as absolute rates for second primary ovarian cancer after breast cancer overall (1973 through 2008) and after ER-positive or ER-negative cancers (1990 through 2008). To our knowledge, this is the first study to examine both the SIRs and absolute rates for second primary ovarian cancers among survivors of breast cancer overall and stratified by ER expression. This unique study was feasible because SEER is one of the few large-scale and population-based cancer registries that records both multiple primary cancers and ER expression for breast cancer with meticulous data collection and standards.

METHODS

Study Participants and Follow-Up

We obtained breast cancer case and population data from the National Cancer Institute's SEER 9 Registries Database, November 2010 submission (1973 to 2008).⁹ SEER's nine tumor registries (Atlanta, Connecticut, Detroit, Hawaii, Iowa, New Mexico, San Francisco-Oakland, Seattle-Puget Sound, and Utah) collectively represent approximately 10% of the US general population. We selected women who were diagnosed with invasive breast cancer between the ages of 20 and 84 years to avoid under-reporting of second cancers among older persons as a result of underdiagnosis, competing risks from comorbid conditions, and/or shortened life expectancies.¹⁰ We excluded patients with cancer who were diagnosed by autopsy or death certificate only. We observed the survivors of breast cancer for second primary ovarian cancers beginning 1 year after the date of breast cancer diagnosis (1-year latency exclusion period) to minimize misclassification of metastases or undetected synchronous cancers.⁶ Follow-up continued until date of diagnosis of any second cancer, death from any cause, date of last known vital status, age 85 years, or end of study (December 31, 2008), whichever occurred first.

The analytic cohort included 383,057 women with invasive breast cancer who survived for ≥ 1 year and who were diagnosed from 1973 to 2007 and observed through 2008. SEER did not record ER expression until 1990. From 1990 forward, there were 237,030 women with invasive breast cancer who survived ≥ 1 year and who were diagnosed from 1990 to 2007 and observed through 2008. We calculated SIRs and absolute incidence rates for second primary ovarian cancer by year or age at diagnosis of the first primary breast cancer overall and by ER expression. All P values were two-sided and considered statistically significant at the P < .05 level.

Statistical Analysis After Breast Cancer Overall (diagnosed from 1973 to 2007 and observed through 2008)

The SIR for second primary ovarian cancer was estimated as the ratio of the observed number of ovarian cancers after breast cancer compared with the expected number of ovarian cancers based on ovarian cancer incidence rates in the general population,⁶ with CIs calculated by Poisson exact methods for the ratio of observed to expected events.¹¹ Expected second primary ovarian cancers were obtained by multiplying incidence rates specific for race and 5-year attained age and calendar-year intervals by stratum-specific person-years at risk and summing across strata (SEER*Stat, version 7.0.6, http://seer.cancer.gov/seerstat/).¹²

Poisson regression models were used to calculate the absolute second primary ovarian cancer rates by year of diagnosis of the first primary breast cancer in continuous 1-year increments (YEAR) and expressed as rates per 10,000 woman-years.¹³ Second primary ovarian cancer rates were adjusted for age at diagnosis of the first primary breast cancer in 5-year age groups (age 20 to 24, 25 to 29, and so on, to age 80 to 84 years), with age 50 to 54 years as the referent age group, and time since breast cancer diagnosis (latency) in continuous 1-year increments. Flexible but smooth rates were obtained with regression splines where the knots were selected by Akaike's information criteria.^14,15 Our final model included a cubic function for year of first primary breast cancer diagnosis, a categorical function for age at first primary breast cancer diagnosis, and a linear term for latency. Log-linear trends in the absolute rates were summarized as the estimated annual percentage change (EAPC), calculated as the antilog for the regression coefficient for YEAR minus 1 times 100 {ie, EAPC = [exp(YEAR) – 1] × 100}. Poisson regression models were also used to calculate the absolute second primary ovarian cancer rates by age at diagnosis of the first primary breast cancer, adjusted for year of diagnosis of the first primary breast cancer and latency.

Statistical Analysis After ER-Positive, ER-Negative, and ER-Unknown Breast Cancer (diagnosed from 1990 to 2007 and observed through 2008)

SIRs are reported for second primary ovarian cancers after ER-positive, ER-negative, and ER-unknown breast cancers. The absolute rates were calculated after ER-positive and ER-negative expression. We also conducted sensitivity analyses to accommodate different ER missing patterns by allocating woman-years and tumors in the missing ER category to ER-positive and ER-negative cells in proportion to the observed age-specific and year-specific woman-years for the ER-positive or ER-negative cancers, as previously described.^16–18 We then recomputed the absolute incidence rates for the reconstructed ER-positive and ER-negative cancers.

RESULTS

One thousand five hundred forty-three second primary ovarian cancers were diagnosed after a first primary breast cancer between 1973 to 2007 and observed through 2008 with 3,060,703 person-years of follow-up (mean follow-up, 8 years; Table 1). Given that less than 10% of the ovarian cancers occurred among nonwhites (ie, others: 81 of 1,543 and blacks: 62 of 1,543), we combined all racial groups for our main analyses. In a sensitivity analysis, excluding the nonwhite patients did not significantly alter our results. Additionally, SEER did not record Hispanic origin until 1992. However, trends for second primary ovarian cancers were similar for non-Hispanic whites and Hispanic whites from 1992 through 2008, although the event rate was too low to reach statistical significance for Hispanic whites (ie, 61 second primary ovarian cancers after first primary breast cancers among Hispanic whites during this time period). For all racial groups combined, there were 384, 175, and 119 second primary ovarian cancers after ER-positive, ER-negative, and ER-unknown breast cancers from 1990 forward with 1,459,669 person-years of follow-up (mean follow-up, 6 years; Table 2).

Table 1.

SIRs for Second Primary Ovarian Cancer Among 1-Year Survivors of a First Primary Female Breast Cancer Diagnosed Between1973 and 2007 and Observed Through 2008 in SEER 9

Characteristic	No.^* of First Primary Breast Cancers	No. of Person-Years^†	Second Primary Ovarian Cancers
Characteristic	No.^* of First Primary Breast Cancers	No. of Person-Years^†	No.	SIR	95% CI
Total No.	383,057	3,060,703	1,543	1.24^‡	1.2 to 1.3
Race
White	326,939	2,654,581	1,400	1.23^‡	1.2 to 1.3
Black	30,984	210,576	62	1.12	0.9 to 1.4
Other	25,134	195,546	81	1.61	1.3 to 2.0
Age at breast cancer diagnosis, years
< 40	28,457	284,081	144	3.43^‡	2.9 to 4.0
40-49	74,577	731,646	321	1.58^‡	1.4 to 1.8
50-59	93,252	841,166	405	1.18^‡	1.1 to 1.3
60-69	92,200	753,598	397	1.01	0.9 to 1.1
70-84	94,571	450,212	274	1.06	0.9 to 1.2
Calendar year of breast cancer diagnosis
1973-1979	50,153	578,700	322	1.26^‡	1.1 to 1.4
1980-1984	42,227	455,544	230	1.16^‡	1.0 to 1.3
1985-1989	53,647	566,790	313	1.30^‡	1.2 to 1.4
1990-1994	59,298	567,248	276	1.21^‡	1.1 to 1.4
1995-1999	67,307	511,143	238	1.25^‡	1.1 to 1.4
2000-2004	69,690	323,909	150	1.36^‡	1.2 to 1.6
2005-2007	40,735	57,369	14	0.79	0.4 to 1.3
Time since breast cancer diagnosis, years
1-4	383,057	1,231,035	580	1.26^‡	1.2 to 1.4
5-9	244,027	922,950	480	1.29^‡	1.2 to 1.4
10-15	134,430	493,289	254	1.19^‡	1.0 to 1.3
≥ 15	69,802	413,429	229	1.18^‡	1.0 to 1.3

Open in a new tab

Abbreviations: SEER, Surveillance, Epidemiology, and End Results; SIR, standardized incidence ratio.

No. of breast cancer survivors at the beginning of the study period.

^†

Begins 12 months after breast cancer diagnosis.

^‡

P < .05.

Table 2.

Second Primary Ovarian Cancer Among 1-Year Survivors of a First Primary Female Breast Cancer Diagnosed Between 1990 and 2007 and Observed Through 2008 by ER Expression of the Breast Cancer in SEER 9

Characteristic	Second Primary Ovarian Cancers in Survivors of ER-Positive Breast Cancer (n = 157,081; 66%)			Second Primary Ovarian Cancers in Survivors of ER-Negative Breast Cancer (n = 46,345; 20%)			Second Primary Ovarian Cancers in Survivors of ER-Unknown Breast Cancer (n = 33,604; 14%)
Characteristic	No.	SIR	95% CI	No.	SIR	95% CI	No.	SIR	95% CI
Total No.	384	1.05	0.9 to 1.2	175	1.98^*	1.7 to 2.3	119	1.3^*	1.1 to 1.5
Age at breast cancer diagnosis, years
< 50	62	1.21	0.9 to 1.6	89	4.35^*	3.5 to 5.4	28	1.96^*	1.3 to 2.8
50-59	106	1.21	1.0 to 1.5	33	1.26	0.9 to 1.8	24	1.10	0.7 to 1.6
60-69	117	0.99	0.8 to 1.2	32	1.30	0.9 to 1.8	28	0.93	0.6 to 1.3
70-84	99	0.92	0.7 to 1.1	21	1.22	0.8 to 1.9	39	1.50^*	1.0 to 2.0
Calendar year of breast cancer diagnosis
1990-1994	135	0.97	0.8 to 1.1	82	2.18^*	1.7 to 2.7	59	1.18	0.9 to 1.5
1995-1999	145	1.11	0.9 to 1.3	56	1.82^*	1.4 to 2.4	37	1.26	0.9 to 1.7
2000-2007	104	1.10	0.9 to 1.3	37	1.84^*	1.3 to 2.5	23	1.72^*	1.1 to 2.6
Time since breast cancer diagnosis, years
1-4	205	1.11	1.0 to 1.3	76	1.82^*	1.4 to 2.3	56	1.36^*	1.0 to 1.8
5-9	128	1.01	0.8 to 1.2	60	1.98^*	1.5 to 2.6	44	1.34	1.0 to 1.8
≥ 10	51	0.93	0.7 to 1.2	39	2.36^*	1.7 to 3.2	19	1.02	0.6 to 1.6

Open in a new tab

NOTE. There were 237,030 survivors of breast cancer at the beginning of the study period, which began 12 months after breast cancer diagnosis. The frequency distribution for ER expression among women in the analytic cohort (N = 237,030) was similar to frequency distribution for ER expression among all women with invasive breast cancer in the SEER 9 database from 1990 to 2008 (n = 343,789).

Abbreviations: ER, estrogen receptor; SEER, Surveillance, Epidemiology, and End Results; SIR, standardized incidence ratio.

P < .05.

SIRs

For the entire study period (Table 1), higher rates of second primary ovarian cancers occurred among breast cancer survivors than in the general female population in SEER (SIR, 1.24; 95% CI, 1.2 to 1.3). The SIRs decreased with increasing age at breast cancer diagnosis but did not vary substantially by year of breast cancer diagnosis or time since breast cancer diagnosis (latency).

SIRs for second primary ovarian cancers were significantly increased after ER-negative breast cancer (SIR, 1.98; 95% CI, 1.7 to 2.3) and ER-unknown breast cancer (SIR, 1.3; 95% CI, 1.1 to 1.5; Table 2). The largest SIR was observed among women diagnosed with ER-negative breast cancer before age 50 years (SIR, 4.35; 95% CI, 3.5 to 5.4). SIRs for ER-negative tumors were statistically significant irrespective of year of breast cancer diagnosis or time since breast cancer diagnosis (latency). After ER-positive breast cancer, there was no overall increase (SIR, 1.05; 95% CI, 0.9 to 1.2) but some suggestion of an elevated risk for women diagnosed before age 60 years.

Absolute Rates by Year of Breast Cancer Diagnosis

For the full follow-up period (Fig 1A), the absolute rate for second primary ovarian cancers decreased from a peak in 1978 of 5.1 per 10,000 woman-years (95% CI, 4.2 to 6.2 per 10,000 woman-years) to a low value in 2007 of 2.5 per 10,000 woman-years (95% CI, 1.6 to 3.9 per 10,000 woman-years). This decline occurred with an EAPC near 1% (ie, EAPC of −0.71% per year; 95% CI, −1.34% to −0.09% per year). The smoothed rates in Figure 1 provided a good fit to the empirical single-year rates (Appendix Fig A1, online only), guiding the eye and giving a good a sense of the overall trend.¹⁹ We observed similar trends among women with breast cancer that was diagnosed before and after age 50 years (graphs not shown).

Absolute ovarian cancer incidence trends were generally one- to two-fold higher after ER-negative than ER-positive breast cancer, with declining trends that did not reach statistical significance (Fig 1B), irrespective of ER expression (P for interaction or heterogeneity by ER = .40). Rates after ER-negative cancers decreased with an EAPC of −1.72% per year (95% CI, −4.17% to 0.79% per year). Rates after ER-positive cancers were stable until the year 2000 and then declined with an overall EAPC of −1.59% per year (95% CI, −5.43% to 2.4% per year). These ovarian cancer trends after ER-positive and ER-negative cancers did not vary substantially with correction for ER-unknown tumors (Appendix Fig A2A, online only).

Absolute Rates by Age at Breast Cancer Diagnosis

For the full follow-up period (Fig 1C), absolute incidence rates for second primary ovarian cancers showed bimodal peaks between 5 and 6 per 10,000 woman-years near age 20 and 80 years with an overall EAPC of 0.78% per year (95% CI, 0.35% to 1.22% per year). Age-specific rates for second primary ovarian cancers varied by ER expression (Fig 1D, P for interaction or heterogeneity by ER < .001). Ovarian cancers after ER-positive tumors increased steadily with advancing aging (EAPC, 1.99% per year; 95% CI, 1.11% to 2.87% per year). However, ovarian cancers after ER-negative tumors seemed to have two peaks among younger and older women, with a nonsignificant EAPC of −0.40% (95% CI, −1.63% to 0.84%). The ER-positive and ER-negative age-specific trends did not differ substantially when woman-years and counts for the ER-unknown categories were reassigned to the ER-positive or ER-negative categories (Appendix Fig A2B).

To assess the contemporary risk for second primary ovarian cancer from 2000 forward, we estimated ovarian cancer incidence rates by age at breast cancer diagnosis and by ER expression (Table 3). Second primary ovarian cancer rates per 10,000 woman-years were 1.7 (95% CI, 0.8 to 2.6), 3.5 (95% CI, 2.8 to 4.2), and 4.5 (95% CI, 3.4 to 5.7) in women diagnosed with ER-positive breast cancer at age 30 to 39, 50 to 59, and ≥ 70 years, respectively. After ER-negative breast cancer, estimated rates for corresponding age groups were 6.2 (95% CI, 3.6 to 8.8), 4.5 (95% CI, 3.0 to 6.1), and 7.0 (95% CI, 3.7 to 10.4) per 10,000 woman-years, respectively.

Table 3.

Estimated Annual Ovarian Cancer Incidence Rates by Age at First Primary Breast Cancer (diagnosed between 2000 and 2007 and observed through 2008) and ER Expression of the Breast Cancer in SEER 9

Age at Breast Cancer Diagnosis (years)	ER-Positive Breast Cancer		ER-Negative Breast Cancer
Age at Breast Cancer Diagnosis (years)	Incidence Rate per 10,000 Woman-Years^*	95% CI	Incidence Rate per 10,000 Woman-Years^*	95% CI
20-29	1.2	0 to 3.3	3.2	0 to 7.2
30-39	1.7	0.8 to 2.6	6.2	3.6 to 8.8
40-49	2.5	1.8 to 3.1	6.0	3.9 to 8.1
50-59	3.5	2.8 to 4.2	4.5	3.0 to 6.1
60-69	4.4	3.4 to 5.4	4.1	2.5 to 5.7
≥ 70	4.5	3.4 to 5.7	7.0	3.7 to 10.4

Open in a new tab

Abbreviations: ER, estrogen receptor; SEER, Surveillance, Epidemiology, and End Results.

Estimated using Poisson regression model adjusted for time since breast cancer diagnosis.

In a series of sensitivity analyses, the secular trends did not change materially if the latency exclusion period for second primary ovarian cancer after a first primary breast cancer was reduced from 12 to 6 months (EAPC, −0.69% per year; 95% CI, −1.30% to −0.09% per year) or expanded from 12 to 24 months (EAPC, −0.74% per year; 95% CI, −1.42% to −0.05% per year). Additionally, we observed similar age-specific patterns when the attained age of second primary ovarian cancer diagnosis was substituted for age at first primary breast cancer diagnosis.

Finally, we assessed incidence rates for all first primary ovarian cancers in the general SEER 9 population from 1973 through 2008 (Fig 2). Incidence rates were age adjusted to the 2000 US standard population by the direct method. Age-adjusted rates declined steadily, with an EAPC of −0.68% per year (95% CI, −0.80% to −0.56% per year; Fig 2A). Age-specific incidence increased with advancing age (Fig 2B).

DISCUSSION

Our three main findings describe the SIRs for second primary ovarian cancers among survivors of breast cancer, the absolute rates for second primary ovarian cancers by year of breast cancer diagnosis, and the absolute rates for second primary ovarian cancers by age at breast cancer diagnosis. Our report supports and expands previous studies demonstrating elevated SIRs for ovarian cancer after breast cancer, especially among younger survivors of breast cancer.^20–24 This is generally believed to reflect common risk factors such as shared genetic and hormonal exposures,²⁵ which is also consistent with the reciprocal increase in the SIRs for second primary breast cancer after first primary ovarian cancer among young women.²⁶ We further show that the increased second primary ovarian cancer rate was largely confined to ovarian cancer after ER-negative cancers. The elevated rates among survivors of ER-negative breast cancer may be related to germline mutations of the BRCA1 gene,^27–29 an established risk factor for both ovarian cancer and early-onset ER-negative breast cancer. Accordingly, the largest SIR was observed among women diagnosed with an ER-negative breast cancer before age 50 years.

Additionally, we observed a significant decrease of almost 1% per year in the absolute rates of ovarian cancer after breast cancer overall. Decreasing absolute rates of ovarian cancer among survivors of breast cancer along with persistently elevated SIRs over the entire study period suggest that ovarian cancers in the surviving breast cancer population and general population are declining in tandem. Indeed, second primary ovarian cancers and all primary ovarian cancers decreased at similar rates (Figs 1A and 2A). These parallel trends may reflect changes in shared risk factor exposures in the general population such as the increased usage of combined oral contraceptives.³

Although secular trends for second primary ovarian cancers were similar after ER-positive and ER-negative breast cancers (Fig 1B), ER expression was an age-specific effect modifier (Fig 1D). Second primary ovarian cancer rates increased with advancing age at breast cancer diagnosis after ER-positive breast cancer, which resembles the age-specific pattern for ovarian cancers in the general population (Fig 2B). However, ovarian cancers seemed to peak among younger and older women after ER-negative breast cancers. Close to age 70 years, rates were similar after ER-positive and ER-negative cancers. However, these age-specific patterns should be interpreted with caution because of the small number of second primary ovarian cancers for some age groups.

Recognized limitations associated with the use of registry-based data include the lack of detailed breast cancer treatment data as well as information regarding lifestyle³⁰ and/or genetic cancer risk factors. Given that SEER does not generally collect complete treatment information, it was not possible to directly assess the impact of ovarian ablation or adjuvant systemic hormonal therapy and/or chemotherapy.^31,32 However, the parallel trends for second and first primary ovarian cancers suggest that it is unlikely that these therapies were the sole cause of the observed decline. Despite these limitations and to the best of our knowledge, this is the first robust analysis of both SIRs and absolute rates for second primary ovarian cancer among survivors of breast cancer and also the first to note the age-specific effect by ER expression.

In summary, the news overall is good; results show that ovarian cancer rates among the surviving breast cancer population and general population have declined in sync since the late 1970s. However, we also demonstrate that there is greater risk for ovarian cancer after ER-negative than ER-positive breast cancer from 1990 forward (Table 2) and have further quantified this difference for a more recent time period (ie, from 2000 through 2008; Table 3). Clinicians might use Table 3 to assess the contemporary risk for ovarian cancer after an ER-positive or ER-negative breast cancer. The distinct age-specific patterns by ER expression likely reflect differences in the risk factor profiles for second primary ovarian cancers in these two distinct types of breast cancer. However, additional analytic studies with treatment information, hormonal exposures, genetic risk factors, and ER data are needed to further investigate our descriptive results.

Acknowledgment

We thank the reviewers for their helpful comments, which greatly improved the content of this article.

Appendix

Fig A1. — The cubic functions for the hazard rates for second ovarian cancers are superimposed on the empirical single-year rates as a piecewise constant for the risk of second primary ovarian cancer. The cubic function provided a flexible and smoothed fit to the empirical rates, guiding the eye and giving a good sense of the overall trend. Absolute rates for second primary ovarian cancer (A) by year of first primary breast cancer diagnosis adjusted for age at breast cancer diagnosis and years since breast cancer diagnosis (ie, latency); (B) by year of first primary breast cancer diagnosis adjusted for age at breast cancer diagnosis and years since breast cancer diagnosis (ie, latency), stratified by estrogen receptor (ER) expression; (C) by age at first primary breast cancer diagnosis adjusted for year of breast cancer diagnosis and years since breast cancer diagnosis (ie, latency); and (D) by age at first primary breast cancer diagnosis adjusted for year of breast cancer diagnosis and years since breast cancer diagnosis (ie, latency), stratified by ER expression.

Fig A2. — Absolute incidence rates for second primary ovarian cancers after a first primary breast cancer by year of diagnosis, age at diagnosis, and estrogen receptor (ER) –positive and ER-negative expression corrected for missing/unknown ER data. Absolute rates for second primary ovarian cancers (A) by year of first primary breast cancer diagnosis adjusted for age at breast cancer diagnosis and years since breast cancer diagnosis (ie, latency), stratified by ER expression and (B) by age at first breast cancer diagnosis adjusted for year of breast cancer diagnosis and years since breast cancer diagnosis (ie, latency), stratified by ER expression. EAPC, estimated annual percentage change.

Footnotes

Supported in part by the Intramural Research Program of the National Institutes of Health, National Cancer Institute.

The authors had full access to all of the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis.

Authors' disclosures of potential conflicts of interest and author contributions are found at the end of this article.

AUTHORS' DISCLOSURES OF POTENTIAL CONFLICTS OF INTEREST

The author(s) indicated no potential conflicts of interest.

AUTHOR CONTRIBUTIONS

Conception and design: Sara J. Schonfeld, Amy Berrington de Gonzalez, Ruth M. Pfeiffer, William F. Anderson

Collection and assembly of data: Sara J. Schonfeld, Amy Berrington de Gonzalez, William F. Anderson

Data analysis and interpretation: Sara J. Schonfeld, Amy Berrington de Gonzalez, Kala Visvanathan, Ruth M. Pfeiffer, William F. Anderson

Manuscript writing: All authors

Final approval of manuscript: All authors

REFERENCES

1.Howlader N, Noone AM, Krapcho M, et al. Bethesda, MD: National Cancer Institute; 2011. Ovary, SEER Cancer Statistics Review 1975-2008. [Google Scholar]
2.Siegel R, Naishadham D, Jemal A. Cancer statistics, 2012. CA Cancer J Clin. 2012;62:10–29. doi: 10.3322/caac.20138. [DOI] [PubMed] [Google Scholar]
3.Gnagy S, Ming EE, Devesa SS, et al. Declining ovarian cancer rates in U.S. women in relation to parity and oral contraceptive use. Epidemiology. 2000;11:102–105. doi: 10.1097/00001648-200003000-00004. [DOI] [PubMed] [Google Scholar]
4.Centers for Disease Control and Prevention (CDC): Cancer survivors–United States, 2007. MMWR Morb Mortal Wkly Rep. 2011;60:269–272. [PubMed] [Google Scholar]
5.Curtis RE, Ron E, Hankey BF, et al. New malignancies following breast cancer. In: Curtis RE, Freedman DM, Ron E, et al., editors. New Malignancies Among Cancer Survivors: SEER Cancer Registries, 1973-2000. Bethesda, MD: National Cancer Institute, NIH publication 05-5302; 2006. pp. 181–205. [Google Scholar]
6.Curtis RE, Ries LA. Methods. In: Curtis RE, Freedman DM, Ron E, et al., editors. New Malignancies Among Cancer Survivors: SEER Cancer Registries, 1973-2000. Bethesda, MD: National Cancer Institute, NIH publication 05-5302; 2006. pp. 9–14. [Google Scholar]
7.Peto J. Cancer epidemiology in the last century and the next decade. Nature. 2001;411:390–395. doi: 10.1038/35077256. [DOI] [PubMed] [Google Scholar]
8.Bray F, Møller B. Predicting the future burden of cancer. Nat Rev Cancer. 2006;6:63–74. doi: 10.1038/nrc1781. [DOI] [PubMed] [Google Scholar]
9.Surveillance, Epidemiology, and End Results (SEER): SEER 9: SEER*Stat Database. http://seer.cancer.gov/
10.Fraumeni JF, Jr, Curtis RE, Edwards BK, et al. Introduction. In: Curtis RE, Freedman DM, Ron E, et al., editors. New Malignancies Among Cancer Survivors: SEER Cancer Registries, 1973-2000. Bethesda, MD: National Cancer Institute, NIH publication 05-5302; 2006. pp. 1–7. [Google Scholar]
11.Liddell FD. Simple exact analysis of the standardised mortality ratio. J Epidemiol Community Health. 1984;38:85–88. doi: 10.1136/jech.38.1.85. [DOI] [PMC free article] [PubMed] [Google Scholar]
12.SEER*Stat: Surveillance Research Program, National Cancer Institute SEER*Stat software version 7.0.6. http://seer.cancer.gov/seerstat/
13.Nichols HB, Berrington de González A, Lacey JV, Jr, et al. Declining incidence of contralateral breast cancer in the United States from 1975 to 2006. J Clin Oncol. 2011;29:1564–1569. doi: 10.1200/JCO.2010.32.7395. [DOI] [PMC free article] [PubMed] [Google Scholar]
14.Akaike H. Information theory and an extension of the maximum likelihood principle. In: Petrov BN, Csaki F, editors. Second International Symposium on Information Theory. Budapest, Hungary: Akademiai Kiado; 1973. pp. 267–281. [Google Scholar]
15.Rosenberg PS. Hazard function estimation using B-splines. Biometrics. 1995;51:874–887. [PubMed] [Google Scholar]
16.Anderson WF, Katki HA, Rosenberg PS. Breast cancer incidence in the United States: Current and future trends. J Natl Cancer Inst. 2011;103:1397–1402. doi: 10.1093/jnci/djr257. [DOI] [PMC free article] [PubMed] [Google Scholar]
17.Pfeiffer RM, Mitani A, Matsuno RK, et al. Racial differences in breast cancer trends in the United States (2000–2004) J Natl Cancer Inst. 2008;100:751–752. doi: 10.1093/jnci/djn112. [DOI] [PubMed] [Google Scholar]
18.Gail MH, Anderson WF, Garcia-Closas M. Absolute risk models for subtypes of breast cancer. J Natl Cancer Inst. 2007;99:1657–1659. doi: 10.1093/jnci/djm228. [DOI] [PubMed] [Google Scholar]
19.Steenland K. Smoothing is soothing, and splines are fine. Occup Environ Med. 2005;62:141–142. doi: 10.1136/oem.2004.017194. [DOI] [PMC free article] [PubMed] [Google Scholar]
20.Brown LM, Chen BE, Pfeiffer RM, et al. Risk of second non-hematological malignancies among 376,825 breast cancer survivors. Breast Cancer Res Treat. 2007;106:439–451. doi: 10.1007/s10549-007-9509-8. [DOI] [PubMed] [Google Scholar]
21.Mellemkjaer L, Friis S, Olsen JH, et al. Risk of second cancer among women with breast cancer. Int J Cancer. 2006;118:2285–2292. doi: 10.1002/ijc.21651. [DOI] [PubMed] [Google Scholar]
22.Schaapveld M, Visser O, Louwman MJ, et al. Risk of new primary nonbreast cancers after breast cancer treatment: A Dutch population-based study. J Clin Oncol. 2008;26:1239–1246. doi: 10.1200/JCO.2007.11.9081. [DOI] [PubMed] [Google Scholar]
23.Berkowitz Z, Rim SH, Peipins LA. Characteristics and survival associated with ovarian cancer diagnosed as first cancer and ovarian cancer diagnosed subsequent to a previous cancer. Cancer Epidemiol. 2011;35:112–119. doi: 10.1016/j.canep.2010.07.001. [DOI] [PubMed] [Google Scholar]
24.Hall HI, Jamison P, Weir HK. Second primary ovarian cancer among women diagnosed previously with cancer. Cancer Epidemiol Biomarkers Prev. 2001;10:995–999. [PubMed] [Google Scholar]
25.Ng AK, Travis LB. Subsequent malignant neoplasms in cancer survivors. Cancer J. 2008;14:429–434. doi: 10.1097/PPO.0b013e31818d8779. [DOI] [PubMed] [Google Scholar]
26.Freedman DM, Curtis RE, Travis LB, et al. New malignancies following cancer of the uterine corpus and ovary. In: Curtis RE, Freedman DM, Ron E, et al., editors. New Malignancies Among Cancer Survivors: SEER Cancer Registries, 1973-2000. Bethesda, MD: National Cancer Institute, NIH publication 05-5302; 2006. pp. 231–255. [Google Scholar]
27.Foulkes WD, Metcalfe K, Sun P, et al. Estrogen receptor status in BRCA1- and BRCA2-related breast cancer: The influence of age, grade, and histological type. Clin Cancer Res. 2004;10:2029–2034. doi: 10.1158/1078-0432.ccr-03-1061. [DOI] [PubMed] [Google Scholar]
28.Tung N. Management of women with BRCA mutations: A 41-year-old woman with a BRCA mutation and a recent history of breast cancer. JAMA. 2011;305:2211–2220. doi: 10.1001/jama.2011.678. [DOI] [PubMed] [Google Scholar]
29.Lynch HT, Silva E, Snyder C, et al. Hereditary breast cancer: Part I. Diagnosing hereditary breast cancer syndromes. Breast J. 2008;14:3–13. doi: 10.1111/j.1524-4741.2007.00515.x. [DOI] [PubMed] [Google Scholar]
30.Persson I: Estrogens in the causation of breast, endometrial and ovarian cancers: Evidence and hypotheses from epidemiological findings. J Steroid Biochem Mol Biol. 2000;74:357–364. doi: 10.1016/s0960-0760(00)00113-8. [DOI] [PubMed] [Google Scholar]
31.Early Breast Cancer Trialists' Collaborative Group: Ovarian ablation in early breast cancer: Overview of the randomised trials. Lancet. 1996;348:1189–1196. [PubMed] [Google Scholar]
32.Early Breast Cancer Trialists' Collaborative Group (EBCTCG) Peto R, Davies C, et al. Comparisons between different polychemotherapy regimens for early breast cancer: Meta-analyses of long-term outcome among 100,000 women in 123 randomised trials. Lancet. 2012;379:432–444. doi: 10.1016/S0140-6736(11)61625-5. [DOI] [PMC free article] [PubMed] [Google Scholar]

[B1] 1.Howlader N, Noone AM, Krapcho M, et al. Bethesda, MD: National Cancer Institute; 2011. Ovary, SEER Cancer Statistics Review 1975-2008. [Google Scholar]

[B2] 2.Siegel R, Naishadham D, Jemal A. Cancer statistics, 2012. CA Cancer J Clin. 2012;62:10–29. doi: 10.3322/caac.20138. [DOI] [PubMed] [Google Scholar]

[B3] 3.Gnagy S, Ming EE, Devesa SS, et al. Declining ovarian cancer rates in U.S. women in relation to parity and oral contraceptive use. Epidemiology. 2000;11:102–105. doi: 10.1097/00001648-200003000-00004. [DOI] [PubMed] [Google Scholar]

[B4] 4.Centers for Disease Control and Prevention (CDC): Cancer survivors–United States, 2007. MMWR Morb Mortal Wkly Rep. 2011;60:269–272. [PubMed] [Google Scholar]

[B5] 5.Curtis RE, Ron E, Hankey BF, et al. New malignancies following breast cancer. In: Curtis RE, Freedman DM, Ron E, et al., editors. New Malignancies Among Cancer Survivors: SEER Cancer Registries, 1973-2000. Bethesda, MD: National Cancer Institute, NIH publication 05-5302; 2006. pp. 181–205. [Google Scholar]

[B6] 6.Curtis RE, Ries LA. Methods. In: Curtis RE, Freedman DM, Ron E, et al., editors. New Malignancies Among Cancer Survivors: SEER Cancer Registries, 1973-2000. Bethesda, MD: National Cancer Institute, NIH publication 05-5302; 2006. pp. 9–14. [Google Scholar]

[B7] 7.Peto J. Cancer epidemiology in the last century and the next decade. Nature. 2001;411:390–395. doi: 10.1038/35077256. [DOI] [PubMed] [Google Scholar]

[B8] 8.Bray F, Møller B. Predicting the future burden of cancer. Nat Rev Cancer. 2006;6:63–74. doi: 10.1038/nrc1781. [DOI] [PubMed] [Google Scholar]

[B9] 9.Surveillance, Epidemiology, and End Results (SEER): SEER 9: SEER*Stat Database. http://seer.cancer.gov/

[B10] 10.Fraumeni JF, Jr, Curtis RE, Edwards BK, et al. Introduction. In: Curtis RE, Freedman DM, Ron E, et al., editors. New Malignancies Among Cancer Survivors: SEER Cancer Registries, 1973-2000. Bethesda, MD: National Cancer Institute, NIH publication 05-5302; 2006. pp. 1–7. [Google Scholar]

[B11] 11.Liddell FD. Simple exact analysis of the standardised mortality ratio. J Epidemiol Community Health. 1984;38:85–88. doi: 10.1136/jech.38.1.85. [DOI] [PMC free article] [PubMed] [Google Scholar]

[B12] 12.SEER*Stat: Surveillance Research Program, National Cancer Institute SEER*Stat software version 7.0.6. http://seer.cancer.gov/seerstat/

[B13] 13.Nichols HB, Berrington de González A, Lacey JV, Jr, et al. Declining incidence of contralateral breast cancer in the United States from 1975 to 2006. J Clin Oncol. 2011;29:1564–1569. doi: 10.1200/JCO.2010.32.7395. [DOI] [PMC free article] [PubMed] [Google Scholar]

[B14] 14.Akaike H. Information theory and an extension of the maximum likelihood principle. In: Petrov BN, Csaki F, editors. Second International Symposium on Information Theory. Budapest, Hungary: Akademiai Kiado; 1973. pp. 267–281. [Google Scholar]

[B15] 15.Rosenberg PS. Hazard function estimation using B-splines. Biometrics. 1995;51:874–887. [PubMed] [Google Scholar]

[B16] 16.Anderson WF, Katki HA, Rosenberg PS. Breast cancer incidence in the United States: Current and future trends. J Natl Cancer Inst. 2011;103:1397–1402. doi: 10.1093/jnci/djr257. [DOI] [PMC free article] [PubMed] [Google Scholar]

[B17] 17.Pfeiffer RM, Mitani A, Matsuno RK, et al. Racial differences in breast cancer trends in the United States (2000–2004) J Natl Cancer Inst. 2008;100:751–752. doi: 10.1093/jnci/djn112. [DOI] [PubMed] [Google Scholar]

[B18] 18.Gail MH, Anderson WF, Garcia-Closas M. Absolute risk models for subtypes of breast cancer. J Natl Cancer Inst. 2007;99:1657–1659. doi: 10.1093/jnci/djm228. [DOI] [PubMed] [Google Scholar]

[B19] 19.Steenland K. Smoothing is soothing, and splines are fine. Occup Environ Med. 2005;62:141–142. doi: 10.1136/oem.2004.017194. [DOI] [PMC free article] [PubMed] [Google Scholar]

[B20] 20.Brown LM, Chen BE, Pfeiffer RM, et al. Risk of second non-hematological malignancies among 376,825 breast cancer survivors. Breast Cancer Res Treat. 2007;106:439–451. doi: 10.1007/s10549-007-9509-8. [DOI] [PubMed] [Google Scholar]

[B21] 21.Mellemkjaer L, Friis S, Olsen JH, et al. Risk of second cancer among women with breast cancer. Int J Cancer. 2006;118:2285–2292. doi: 10.1002/ijc.21651. [DOI] [PubMed] [Google Scholar]

[B22] 22.Schaapveld M, Visser O, Louwman MJ, et al. Risk of new primary nonbreast cancers after breast cancer treatment: A Dutch population-based study. J Clin Oncol. 2008;26:1239–1246. doi: 10.1200/JCO.2007.11.9081. [DOI] [PubMed] [Google Scholar]

[B23] 23.Berkowitz Z, Rim SH, Peipins LA. Characteristics and survival associated with ovarian cancer diagnosed as first cancer and ovarian cancer diagnosed subsequent to a previous cancer. Cancer Epidemiol. 2011;35:112–119. doi: 10.1016/j.canep.2010.07.001. [DOI] [PubMed] [Google Scholar]

[B24] 24.Hall HI, Jamison P, Weir HK. Second primary ovarian cancer among women diagnosed previously with cancer. Cancer Epidemiol Biomarkers Prev. 2001;10:995–999. [PubMed] [Google Scholar]

[B25] 25.Ng AK, Travis LB. Subsequent malignant neoplasms in cancer survivors. Cancer J. 2008;14:429–434. doi: 10.1097/PPO.0b013e31818d8779. [DOI] [PubMed] [Google Scholar]

[B26] 26.Freedman DM, Curtis RE, Travis LB, et al. New malignancies following cancer of the uterine corpus and ovary. In: Curtis RE, Freedman DM, Ron E, et al., editors. New Malignancies Among Cancer Survivors: SEER Cancer Registries, 1973-2000. Bethesda, MD: National Cancer Institute, NIH publication 05-5302; 2006. pp. 231–255. [Google Scholar]

[B27] 27.Foulkes WD, Metcalfe K, Sun P, et al. Estrogen receptor status in BRCA1- and BRCA2-related breast cancer: The influence of age, grade, and histological type. Clin Cancer Res. 2004;10:2029–2034. doi: 10.1158/1078-0432.ccr-03-1061. [DOI] [PubMed] [Google Scholar]

[B28] 28.Tung N. Management of women with BRCA mutations: A 41-year-old woman with a BRCA mutation and a recent history of breast cancer. JAMA. 2011;305:2211–2220. doi: 10.1001/jama.2011.678. [DOI] [PubMed] [Google Scholar]

[B29] 29.Lynch HT, Silva E, Snyder C, et al. Hereditary breast cancer: Part I. Diagnosing hereditary breast cancer syndromes. Breast J. 2008;14:3–13. doi: 10.1111/j.1524-4741.2007.00515.x. [DOI] [PubMed] [Google Scholar]

[B30] 30.Persson I: Estrogens in the causation of breast, endometrial and ovarian cancers: Evidence and hypotheses from epidemiological findings. J Steroid Biochem Mol Biol. 2000;74:357–364. doi: 10.1016/s0960-0760(00)00113-8. [DOI] [PubMed] [Google Scholar]

[B31] 31.Early Breast Cancer Trialists' Collaborative Group: Ovarian ablation in early breast cancer: Overview of the randomised trials. Lancet. 1996;348:1189–1196. [PubMed] [Google Scholar]

[B32] 32.Early Breast Cancer Trialists' Collaborative Group (EBCTCG) Peto R, Davies C, et al. Comparisons between different polychemotherapy regimens for early breast cancer: Meta-analyses of long-term outcome among 100,000 women in 123 randomised trials. Lancet. 2012;379:432–444. doi: 10.1016/S0140-6736(11)61625-5. [DOI] [PMC free article] [PubMed] [Google Scholar]

PERMALINK

Declining Second Primary Ovarian Cancer After First Primary Breast Cancer

Sara J Schonfeld

Amy Berrington de Gonzalez

Kala Visvanathan

Ruth M Pfeiffer

William F Anderson