Breast Cancer Stage at Diagnosis and Survival Among Patients with Prior Breast Implants
Total Page:16
File Type:pdf, Size:1020Kb
Breast Cancer Stage at Diagnosis and Survival among Patients with Prior Breast Implants Dennis Deapen, Dr.P.H., Ann Hamilton, Ph.D., Leslie Bernstein, Ph.D., and Garry S. Brody, M.D. Los Angeles, Calif. Longstanding concern exists regarding the potential indicated by clinical stage at diagnosis,8 reports for women with breast implants to experience delayed of case series are mixed, with both favorable9–12 detection of breast cancer. Furthermore, survival among 13,14 cosmetic breast implant patients who subsequently de- and unfavorable results. It has been specu- velop breast cancer is a concern. Since 1976, this institu- lated that breast cancers may be more aggres- tion has monitored cancer incidence in a cohort of 3182 sive among patients with breast implants than women who underwent cosmetic breast augmentation be- among those without,15 resulting in a poorer tween 1959 and 1981. The distributions of stage at diag- survival rate among implant patients. In a co- nosis and survival of the 37 women who subsequently developed in situ or invasive breast cancer were compared hort of more than 3000 augmentation mam- with the observed population distributions. The distribu- maplasty patients, we have observed 37 inci- tion of stage at diagnosis for cosmetic breast implant pa- dent breast cancer patients who were tients who subsequently developed breast cancer was vir- diagnosed before 1993. We compared the tually identical to that of all breast cancer patients in Los 5-year and individual year survival intervals of Angeles County who were of the same age and race, and were diagnosed during the same time period. Further- these patients with those derived from popula- more, the 5-year survival rate of the 37 patients did not tion-based figures obtained by the National differ from that which would be expected based on rates Cancer Institute’s SEER registries, adjusting established by the U.S. National Cancer Institute’s Sur- for breast cancer stage and age at diagnosis. veillance, Epidemiology, and End Results (SEER) pro- gram. These results suggest that cosmetic breast implant pa- MATERIALS AND METHODS tients are not at increased risk of delayed detection of breast cancer, nor do they suffer a poorer prognosis when Since 1976, we have been monitoring cancer breast cancer does occur. Although the number of breast incidence in a cohort of 3182 non–Spanish- cancer patients in this study is small, the results are highly surnamed white women in Los Angeles County consistent with the existing epidemiologic evidence re- who had cosmetic breast augmentation during lated to breast cancer detection and survival among breast the period 1959 to 1981.2 The cohort was es- implant patients. Although breast implant patients should continue appropriate breast cancer screening behavior, tablished by abstracting medical records from there seems to be no cause for alarm. (Plast. Reconstr. the practices of 35 board-certified plastic sur- Surg. 105: 535, 2000.) geons. Patients with prophylactic subcutaneous mastectomy or breast carcinoma before im- plantation were excluded, as were nonresi- Recent studies have consistently shown that dents of Los Angeles County. risk of breast cancer is not increased among Cancer incidence has been determined by women with breast implants.1–7 Furthermore, record linkage of the cohort with the Los An- although population-based studies show that geles County Cancer Surveillance Program, the the diagnosis of breast cancer is not delayed, as population-based cancer registry that has cov- From the Department of Preventive Medicine and the Division of Plastic and Reconstructive Surgery at the School of Medicine of the University of Southern California. Received for publication June 14, 1999. Presented at the Epidemiology of Medical Devices in Women meeting, sponsored by the Food and Drug Administration Office of Women’s Health and Epidemiology Branch, in Rockville, Maryland, on May 5, 1998. Drs. Deapen and Brody have served as consultants to manufacturers of breast implants and have provided breast implant-related testimony. The ideas and opinions expressed herein are those of the authors, and no endorsement by the State of California Department of Health Services or by the Public Health Institute is intended or should be inferred. 535 536 PLASTIC AND RECONSTRUCTIVE SURGERY, February 2000 ered Los Angeles County since 1972. Diagnos- TABLE I tic medical records, including the pathology Type of First Implant among All Augmentation reports, have been collected for these patients. Mammaplasty Cohort Members, All Breast Cancer Patients These records also provide documentation of from the Cohort, and Deceased Breast Cancer Patients the date and stage at diagnosis; stage was clas- from the Cohort sified as SEER summary stage, i.e., in situ, lo- 16 calized, and regional/distant disease. Vital Breast Breast status was determined for each breast cancer All Implant Cancer Cancer Patients Patients Deaths patient through routine cancer registry fol- Implant low-up activities, as well as by cohort follow-up Type n % n % n % methods utilizing public records. In addition, a Gel 2374 75 27 73 4 67 Saline 294 9 4 11 2 33 linkage of all cohort members was performed Gel/saline 66 2 1 3 0 with the National Center for Health Statistics’ Other 33 1 3 8 0 National Death Index for the years 1979 to Unknown 415 13 2 5 0 TOTAL 3182 100 37 100 6 100 1992. Death certificates were obtained for all women who were reported to be deceased. The Los Angeles Cancer Surveillance Pro- percent of the entire cohort received silicone gram began collecting summary information gel implants (Table I). Six of the 37 patients on disease stage for cancer patients in Los died from breast cancer; 4 of these had silicone Angeles County in 1976. To estimate the ex- gel filled implants and two had saline filled pected distribution of stage at diagnosis, the implants. percentage distribution of breast cancer by The average age at implant of the 37 patients stage at diagnosis among cohort members was was 38.1 years, and the average age at breast standardized to the age distribution of all cancer diagnosis was 50.3 years (Table II). breast cancers occurring among non–Spanish- These patients have been followed, on average, surnamed white women 35 to 74 years old in for 18.7 years since their implant surgery, with Los Angeles County during the years 1976 to an average of 6.6 of these years occurring after 1991. the breast cancer diagnosis. The postdiagnosis Observed cumulative survival rates for the follow-up period ranged from 0.2 to 17.3 years, breast cancer patients with prior augmentation with 19 of 37 patients having more than 5 years mammaplasty were estimated annually of postdiagnosis follow-up. through year 5, overall and by stage of disease, Five patients (14 percent) were diagnosed by using the product limit method, with 95 with in situ breast cancer, 19 (51 percent) had percent confidence intervals for the 5-year sur- localized disease, and 13 (35 percent) were vival rate estimated by using Greenwood’s for- diagnosed with regional or distant disease. mula for the standard error.17 Expected sur- When compared with the distribution of sum- vival rates for these patients were generated by mary stage at diagnosis for all Los Angeles indirectly adjusting for age, year of diagnosis, County non–Spanish-surnamed, white, female and stage on the basis of the relative survival breast cancer patients 35 to 74 years old diag- rates for white women provided by the Na- nosed during 1976 to 1991, the distribution of tional Cancer Institute’s SEER Program (Lynn stage at diagnosis among the implanted pa- Ries, personal communication). To do this, the tients was very similar to the expected distribu- SEER rates within each of the specific catego- tion (Table III). Of the 37 patients, 24 were ries have been weighted by the age-, stage-, and diagnosed before the 1988 publication of spe- year of diagnosis–specific percentages of the cialized imaging techniques for the augmented augmentation mammaplasty breast cancer pa- breast.18 tients. Of the six patients who died, four deaths occurred within 2 years of the breast cancer RESULTS diagnosis (at 3 months, 4 months, 1.1 years, Of 3182 women in the cohort, 37 were diag- and 1.8 years, respectively); the other two nosed with in situ or invasive breast cancer deaths occurred 7.5 and 9.8 years after diagno- between 1975 and 1992. Among these, 73 per- sis, respectively. Among the four patients who cent had received silicone gel filled devices at died within 2 years of their diagnosis, two had first implantation (the reminder received sa- distant disease at diagnosis and two had re- line, double-lumen, or other implant types); 75 gional disease. The two patients whose deaths Vol. 105, No. 2 / BREAST CANCER IN PATIENTS WITH BREAST IMPLANTS 537 TABLE II Characteristics of and Length of Follow-Up for Augmentation Mammaplasty Cohort Members Diagnosed with Breast Cancer, by Stage of Disease at Diagnosis Mean Age Mean Interval Mean Age at Mean Interval at Implant Implant to Diagnosis Diagnosis to Last Parameter n (years) Diagnosis (years) (years) Follow-Up (years) All patients 37 38.1 12.2 50.3 6.6 Stage In situ 5 35.7 13.9 49.6 4.5 Local 19 37.8 12.0 49.9 8.6 Regional or distant 13 39.5 11.6 51.1 4.6 Vital status at last follow-up Alive 31 37.4 12.5 49.9 7.2 Deceased 6* 42.1 10.1 52.2 3.5 Vital status at last follow-up for patients with regional or distant stage Alive 7 37.4 12.9 50.2 5.5 Deceased 6* 42.1 10.1 52.2 3.5 * Two of these deaths occurred after 5 years of follow-up.