Abstract
Background: A potential link between breast implants and anaplastic large-cell lymphoma (ALCL) has been suggested.
Methods: We examined lymphoma occurrence in a nationwide cohort of 19,885 Danish women who underwent breast implant surgery during 1973–2010. Standardized incidence ratios (SIR), with 95% confidence intervals (CI), for ALCL and lymphoma overall associated with breast implantation were calculated.
Results: During 179,246 person-years of follow-up, we observed 31 cases of lymphoma among cohort members. No cases of ALCL were identified. SIRs for ALCL and lymphoma overall were zero (95% CI, 0–10.3) and 1.20 (95% CI, 0.82–1.70), respectively.
Conclusions: In our nationwide cohort study, we did not find an increased risk of lymphoma in general, or ALCL in particular, among Danish women who underwent breast implantation. However, our evaluation of ALCL risk was limited by the rarity of the disease.
Impact: Our results do not support an association between breast implants and ALCL and are consistent with other studies on cancer risk and breast implants. Cancer Epidemiol Biomarkers Prev; 22(11); 2126–9. ©2013 AACR.
Introduction
A potential link between breast implants and anaplastic large-cell lymphoma (ALCL) of the breast has been suggested on the basis of several case reports (1), and a Dutch case–control study reported an OR of 18.2 for ALCL among women with silicone breast implants (2).
On the basis of the data from Danish population-based registries and private clinics, we evaluated the occurrence of ALCL and other lymphomas in a large cohort of Danish women who underwent breast implantation.
Materials and Methods
Study population
We established a nationwide cohort of Danish women who underwent breast implantation for cosmetic or reconstructive purposes during 1973–2010 (Fig. 1). The women were identified from the Danish National Registry of Patients, which contains data on all admissions to public hospitals since 1978 (3); The Danish Registry for Plastic Surgery of the Breast, which contains detailed information on 70% to 80% of all cosmetic or reconstructive breast implant procedures carried out during 1999–2009 (4), and a previously established cohort of 1,653 women who underwent breast implantation in private clinics during 1973–1995 (5). The vast majority of the breast implants were silicone-gel filled (4, 5).
Study subjects were followed for a lymphoma diagnosis from the date of first breast implant surgery (left-censored to 1978) until death, emigration, or end of study (December 31, 2010). Lymphoma diagnoses were ascertained from the Danish Cancer Registry and The Danish Lymphoma Group Registry (4, 5). Cancer diagnoses from 1978 onward are registered according to the International Classification of Diseases, version 10 (ICD-10) and the ICD for Oncology (ICD-O-1-3) for topography and morphology codes (6), thus, explaining the choice of 1978 as start of follow-up. A morphologic code for ALCL (ICD-O-3, 97143) was introduced in 1994 based on the Revised European-American Lymphoma Classification.
All cases of lymphoma following breast implant surgery were reviewed by an expert hematopathologist (coauthor K. Bendix) using World Health Organization 2008 criteria.
All data sources were linked using the civil registration number, which encodes gender and date of birth and permits unambiguous linkage between registries and databases in Denmark.
Statistical analyses
We computed standardized incidence ratios (SIR), and 95% confidence intervals (CI), for ALCL and lymphoma overall. For ALCL, the SIR analysis was left truncated to 1994 in accordance with the introduction of the specific histologic code for ALCL. For lymphoma overall, the complete study period (1978–2010) was applied. The expected numbers were estimated by multiplying the number of person-years among cohort members with the corresponding 5-year age- and calendar period-specific incidence rates among women in the general population.
Results
The study population consisted of 19,885 women. Left truncation to 1978 reduced the cohort to 19,639 women. During 1978–2010, we identified 31 cases (Fig. 1 and Table 1) of Hodgkin or non-Hodgkin lymphoma versus 26.6 expected, yielding a SIR for lymphoma overall of 1.20 (95% CI, 0.82–1.70). No cases of ALCL were identified, with 0.36 expected, yielding a SIR of 0 (95% CI, 0–10.3).
Pathoanatomic characteristics of lymphoma cases (n = 31)
| Patient | Year of breast implant surgery | Year of lymphoma diagnosis | Lymphoma subtype | Localization | CD30 and ALK status |
|---|---|---|---|---|---|
| 1 | 1977 | 1993 | Follicular lymphoma, grade 2 | Neck and mediastinum | CD30−, ALK− |
| 2 | 1978 | 1998 | Peripheral T-cell lymphoma, NOS | Vulva, labia minor, larynx, skin lower back, and bone marrow | CD30−, ALK− |
| 3 | 1981, 2001 | 2007 | Follicular lymphoma, grade 1 | Lymph node neck | CD30−, ALK− |
| 4 | 1983 | 2010 | Diffuse large B-cell lymphoma | Mediastinum | CD30−, ALK− |
| 5 | 1985 | 1996 | Follicular lymphoma, grade 3a | Lymph node neck | Partially CD30+, ALK− |
| 6 | 1985 | 1994 | Diffuse large B-cell lymphoma | Retroperitoneum | Insufficient amount of tissue for CD30 and ALK staining. |
| 7 | 1986 | 2006 | Diffuse large B-cell lymphoma | Neck, tonsils, and skin of front-side thorax | CD30−, ALK− |
| 8 | 1987 | 2010 | Diffuse large B-cell lymphoma | Gluteal region | Partially CD30+, ALK− |
| 9 | 1988 | 1996 | Follicular lymphoma, grade 1 | Lymph node neck, right side | CD30−, ALK− |
| 10 | 1988 | 1996 | Diffuse large B-cell lymphoma | Skin right arm and lower back | CD30−, ALK− |
| 11 | 1989 | 2000 | Splenic marginal zone lymphoma | Retroperitoneum | CD30−, ALK− |
| 12 | 1989 | 2000 | Hodgkin lymphoma, NS | Left inguinal lymph node | CD30+, ALK− |
| 13 | 1989 | 2000 | Diffuse large B-cell lymphoma | Mediastinum | Partially CD30+, ALK− |
| 14 | 1990 | 1993 | Follicular lymphoma | Skin of leg | NA |
| 15 | 1991 | 2005 | Follicular lymphoma, grade 2 | Ovaries, cervix uteri, small bowel, and uterus | CD30−, ALK− |
| 16 | 1994 | 1999 | Marginal zone B-cell lymphoma | Lymph node and skin, neck left side | NA |
| 17 | 1994 | 1998 | B-cell lymphoma | Lymph node neck | NA |
| 18 | 1995 | 2003 | Hodgkin lymphoma, NS | Lymph node neck, left side | CD30+, ALK− |
| 19 | 1995 | 1997 | Follicular lymphoma, grade 1 | Lymph node neck, right side | CD30−, ALK− |
| 20 | 1998 | 2004 | Follicular lymphoma, grade 1 | Lymph node neck | CD30−, ALK− |
| 21 | 1999 | 2001 | Follicular lymphoma, grade 2 | Lymph node, right axillary and neck | CD30−, ALK− |
| 22 | 2000 | 2005 | Primary mediastinal B-cell lymphoma | Lymph node anterior mediastinum | Partially CD30+, ALK− |
| 23 | 2000 | 2008 | Diffuse large B-cell lymphoma | Cervix uteri | Partially CD30+, ALK− |
| 24 | 2000 (left breast) | 2004 | Follicular lymphoma, grade 2 | Right breast and axillary lymph node | CD30−, ALK− |
| 25 | 2001 | 2008 | Small lymphocytic lymphoma | Lymph node neck, left side | CD30−, ALK− |
| 26 | 2001 | 2007 | Hodgkin lymphoma, NS | Right axillary lymph node | CD30+, ALK− |
| 27 | 2002 | 2010 | Splenic B-cell lymphoma/leukemia, unclassifiable | Spleen and bone marrow | CD30−, ALK− |
| 28 | 2003 | 2007 | Small lymphocytic lymphoma | Lymph node neck, left side | CD30+, ALK− |
| 29 | 2003 | 2008 | Marginal zone B-cell lymphoma | Ventricle | CD30−, ALK− |
| 30 | 2003 | 2005 | Diffuse large B-cell lymphoma | Right tonsil | CD30−, ALK− |
| 31 | 2004 | 2010 | Follicular lymphoma, grade 3A | Retroperitoneum | CD30−, ALK− |
| Patient | Year of breast implant surgery | Year of lymphoma diagnosis | Lymphoma subtype | Localization | CD30 and ALK status |
|---|---|---|---|---|---|
| 1 | 1977 | 1993 | Follicular lymphoma, grade 2 | Neck and mediastinum | CD30−, ALK− |
| 2 | 1978 | 1998 | Peripheral T-cell lymphoma, NOS | Vulva, labia minor, larynx, skin lower back, and bone marrow | CD30−, ALK− |
| 3 | 1981, 2001 | 2007 | Follicular lymphoma, grade 1 | Lymph node neck | CD30−, ALK− |
| 4 | 1983 | 2010 | Diffuse large B-cell lymphoma | Mediastinum | CD30−, ALK− |
| 5 | 1985 | 1996 | Follicular lymphoma, grade 3a | Lymph node neck | Partially CD30+, ALK− |
| 6 | 1985 | 1994 | Diffuse large B-cell lymphoma | Retroperitoneum | Insufficient amount of tissue for CD30 and ALK staining. |
| 7 | 1986 | 2006 | Diffuse large B-cell lymphoma | Neck, tonsils, and skin of front-side thorax | CD30−, ALK− |
| 8 | 1987 | 2010 | Diffuse large B-cell lymphoma | Gluteal region | Partially CD30+, ALK− |
| 9 | 1988 | 1996 | Follicular lymphoma, grade 1 | Lymph node neck, right side | CD30−, ALK− |
| 10 | 1988 | 1996 | Diffuse large B-cell lymphoma | Skin right arm and lower back | CD30−, ALK− |
| 11 | 1989 | 2000 | Splenic marginal zone lymphoma | Retroperitoneum | CD30−, ALK− |
| 12 | 1989 | 2000 | Hodgkin lymphoma, NS | Left inguinal lymph node | CD30+, ALK− |
| 13 | 1989 | 2000 | Diffuse large B-cell lymphoma | Mediastinum | Partially CD30+, ALK− |
| 14 | 1990 | 1993 | Follicular lymphoma | Skin of leg | NA |
| 15 | 1991 | 2005 | Follicular lymphoma, grade 2 | Ovaries, cervix uteri, small bowel, and uterus | CD30−, ALK− |
| 16 | 1994 | 1999 | Marginal zone B-cell lymphoma | Lymph node and skin, neck left side | NA |
| 17 | 1994 | 1998 | B-cell lymphoma | Lymph node neck | NA |
| 18 | 1995 | 2003 | Hodgkin lymphoma, NS | Lymph node neck, left side | CD30+, ALK− |
| 19 | 1995 | 1997 | Follicular lymphoma, grade 1 | Lymph node neck, right side | CD30−, ALK− |
| 20 | 1998 | 2004 | Follicular lymphoma, grade 1 | Lymph node neck | CD30−, ALK− |
| 21 | 1999 | 2001 | Follicular lymphoma, grade 2 | Lymph node, right axillary and neck | CD30−, ALK− |
| 22 | 2000 | 2005 | Primary mediastinal B-cell lymphoma | Lymph node anterior mediastinum | Partially CD30+, ALK− |
| 23 | 2000 | 2008 | Diffuse large B-cell lymphoma | Cervix uteri | Partially CD30+, ALK− |
| 24 | 2000 (left breast) | 2004 | Follicular lymphoma, grade 2 | Right breast and axillary lymph node | CD30−, ALK− |
| 25 | 2001 | 2008 | Small lymphocytic lymphoma | Lymph node neck, left side | CD30−, ALK− |
| 26 | 2001 | 2007 | Hodgkin lymphoma, NS | Right axillary lymph node | CD30+, ALK− |
| 27 | 2002 | 2010 | Splenic B-cell lymphoma/leukemia, unclassifiable | Spleen and bone marrow | CD30−, ALK− |
| 28 | 2003 | 2007 | Small lymphocytic lymphoma | Lymph node neck, left side | CD30+, ALK− |
| 29 | 2003 | 2008 | Marginal zone B-cell lymphoma | Ventricle | CD30−, ALK− |
| 30 | 2003 | 2005 | Diffuse large B-cell lymphoma | Right tonsil | CD30−, ALK− |
| 31 | 2004 | 2010 | Follicular lymphoma, grade 3A | Retroperitoneum | CD30−, ALK− |
Abbreviations: NA, tissue sample not available for review; NOS, not otherwise specified; NS, nodular sclerosis.
Discussion
In this nationwide cohort study, we found no cases of ALCL among Danish women who underwent breast implant surgery. Our cohort approach based on virtually complete data basically eliminated selection and information bias. Furthermore, the pathologic review precluded misclassification of ALCL. Still, although our cohort was large, the risk estimate for ALCL suffered from limited statistical precision because of the rarity of the disease.
Previous epidemiologic studies (7) did not indicate an increased risk of non-Hodgkin lymphoma associated with silicone breast implants (7, 8). An important contribution of our study has been to supplement these results with information on lymphoma subtype.
The Dutch study reporting a strong association between silicone breast implants and ALCL had some important limitations (2). First, for 3 of 5 cases of ALCL observed among women with breast implants, the latency period was short, for example, 1 patient developed ALCL within 1 year after implantation, a time frame most likely too short to induce malignant transformation. Second, whereas all cases (n = 11) in the study had ALCL of the breast, all controls (n = 35) had breast lymphomas other than ALCL, thus hampering identification of the underlying study base. Third, 3 patients had disseminated disease, questioning that they were primary breast lymphomas. Finally, all 3 women with ALCL and breast implants had saline-filled implants. In our opinion, no valid conclusion can be drawn from the Dutch study with regard to a potential excess risk of ALCL among women with silicone breast implants.
In conclusion, in this first study addressing the suggested association between silicone breast implants and ALCL in a well-defined study population, we found no evidence of an association between breast implantation and risk of ALCL or lymphoma overall. On the basis of the available evidence, it seems reasonable to conclude that a potential association between silicone breast implants and ALCL is not a major safety issue. Because of the rarity of ALCL, a small to moderate excess risk is difficult to show. Therefore, additional studies with even larger sample size, longer follow-up and attention to possible host-related risk factors may be warranted.
Disclosure of Potential Conflicts of Interest
No potential conflicts of interest were disclosed.
Authors' Contributions
Conception and design: M.Ø. Vase, S. Friis, H.T. Sørensen, F. d'Amore
Development of methodology: S. Friis, H.T. Sørensen
Acquisition of data (provided animals, acquired and managed patients, provided facilities, etc.): M.Ø. Vase, S. Friis
Analysis and interpretation of data (e.g., statistical analysis, biostatistics, computational analysis): M.Ø. Vase, S. Friis, A. Bautz, K. Bendix, H.T. Sørensen
Writing, review, and/or revision of the manuscript: M.Ø. Vase, S. Friis, K. Bendix, H.T. Sørensen, F. d'Amore
Administrative, technical, or material support (i.e., reporting or organizing data, constructing databases): S. Friis, A. Bautz
Study supervision: S. Friis, H.T. Sørensen, F. d'Amore
RSS Feeds