Abstract
There should be a heightened awareness of a possible relationship between the development of cancer and breast implants. To date there are case reports about a possible association between the development of ALCL and breast implants. The number of cases are few and our knowledge of the pathogenesis is little. Further investigation is needed to understand the possible link between breast implants and ALCL found in the breast.
Introduction
A recent warning from France’s National Cancer Institute (INCa) and the French Health Minister Marisol Touraine in March 2015 saying that experts identified 18 cases of ALCL in France since 2011 linked to silicone breast implants make this topic of particular relevance. Given their rarity the Institute stated there was no need for women to remove their breast implants. The BfArM (Federal Institute for Drugs and Medical Devices) immediately responded stating that to date there is no confirmed connection between breast implants and carcinogenesis in Europe. Half of Europe’s known cases of breast-implant associated ALCL occurred in France [4]. This might be linked to more detailed follow up documentation after breast implants.
ALCL is an aggressive non-Hodgkin T-cell lymphoma comprising about two to eight percent of all non-Hodgkin lymphomas in adolescence [40]. The US National Cancer Institute estimates that one in 500.000 women per year is diagnosed with ALCL in general [23, 24], but only three in 100 million women per year are diagnosed with ALCL in the breast as its incidence is extremely rare in breast tissue.
Histopathologically when ALCL is diagnosed the lymphoma cells are nearly exclusively found in the capsule surrounding the implants. Thompson et al. propose that the malignant cells are contained within the effusion and adherent to the capsule in a sero-fibrinous fluid without any cell-invasion beyond the fibrous capsule into the breast parenchyma [43].
ALCL can either be cutaneous or systemic and is characterized by large, anaplastic CD30(+)-tumor cells. ALCL can be subdivided into anaplastic lymphoma kinase (ALK) positive or negative forms. Like the cutaneous form of ALCL, breast implant-associated ALCL (BIA-ALCL) in most cases is ALK-negative [35], possibly because of the fact that breast and skin both have the ectoderm as common embryological origin. The presence of ALCL in periprosthetic capsule tissue should always result in a total capsulectomy for further histopathological examination.
Albeit primary breast lymphoma is rare, representing less than 1% of breast tumors, it is B-cell lymphoma that is found the most in those cases (90% of primary breast lymphomas) [41]. Nevertheless, recent publications suggest a slight, but existing context between the incidence of ALCL and breast implants.
Since 1962/1963 breast implants have been used for breast augmentation for reconstructive or aesthetic reasons [7, 26]. In 2014 47.905 breast augmentations and 4.404 breast reconstructions with implants were performed in Germany. In the USA the amount was even higher with 297.297 aesthetic augmentations and 54.243 breast reconstructions with implants according to the statistics of the International Society of Aesthetic Plastic Surgeons (ISAPS) [18]. Some studies suggest, that the development of ALCL in the breast is more common in patients with breast implants. First concerns about a possible causality of ALCL and breast implants arose when Duvic et al. in 1995 published a case report about three women with polyurethane silicone foam-covered, polyurethane-coated silicone and silicone implants [10]. In 1997, Keech and Creech presented a similar of a women developing CD30(+)-T-cell lymphoma in the breast after breast augmentation with saline-filled implants [22]. De Jong et al. published a study in 2008 stating the odds ratio for ALCL associated with breast implants was 18.2 (95% CI, 2.1–156.8) [20]. McLaughlin et al. however found no increased risk for the development of breast cancer in Swedish women with cosmetic breast implants. In their epidemiologic study based on data from the Swedish Inpatient Registry they followed 3486 women who had undergone cosmetic breast implantation between 1965 and 1993 long term until December 31, 2002 [29].
To date, there is only one documented case reporting death due to BIA-ALCL [6].
In 2011 the FDA (U.S. Food and Drug Administration) published a statement indicating a possible association between ALCL in women with breast implants -silicone gel- and saline-filled [45]. There were about 34 confirmed international case reports describing this phenomenon throughout the world until 2010 and additional cases identified by international regulatory agencies, scientific experts and breast implant manufacturers [45]. However, the number of cases of ALCL and breast implants reported in the literature has increased during the recent years. This indicates that this disease could have been underdiagnosed in the past [9]. Till 2015 the FDA received 258 medical device reports of anaplastic large cell lymphoma in women with breast implants, reporting of three deaths [44]. FDA’s current estimate is that there have been 100–250 known cases of ALCL in women with breast implants worldwide [44]. All published information to date still suggests that women with breast implants may have a very low but increased risk of developing ALCL, a low incidence considering to five – ten million women with breast implants. According to the FDA this risk might be very low but none the less significant.
Breast implant-related ALCL is exceedingly rare and a causal link remains highly controversial. Its genesis, its context, its risk factors and the immunologic responses related to different breast implant types are heavily discussed and hypothesized. Risk factors might be postoperative late-onset seroma [27, 32], infection [3, 33], hematoma [15, 28], capsular fibrosis [36], autoimmune-related reactions or chronic antigen stimulation. But a review of the latest literature also demonstrates, that there are primarily indolent clinical cases as well [19]. Release of toxic metabolites by different breast implant types and materials has been reported since their launch in the early 1960s, such as 2,4-Toluenediamine for polyurethane-coated breast implants [8] or siloxanes (D4, D5 and D6) for silicone gel-filled implants [8, 13]. Recent studies suggest that the implant surface and material can have an effect on the formation of capsular fibrosis [34, 37].
All these aspects are suspected to trigger the emergence of cancer, but there is no confirmation in comprehensive studies so far.
Although extremely rare, the association of implants with primary breast anaplastic T-cell lymphoma could represent a new clinicopathologic subtype of neoplasm within the CD30-positive lymphoproliferative diseases and their therapy. Therefore further monitoring and research is necessary.
Purpose of this study was to compare the latest published findings of ALCL and breast implants with our retrospective analysis of 296 capsular tissues of 227 patients with uni- or bilateral capsular fibrosis, focusing on histopathological findings in order to identify a possible association or risk factors.
This investigation was conducted using current pathological diagnostic methods in order to gather additional information on possible risk factors and to identify a specific type of implant associated with a lower or higher risk of developing ALCL.
Methods
Between 01/2000 and 06/2015 there were 2.100 women with breast implants who had been re-examined clinically during a follow-up. Of these women 229 had to undergo surgical revision. We retrospectively examined capsular tissue of 307 histological samples of 229 patients who had undergone removal or exchange of breast implants because of capsular fibrosis between 01/2000 and 06/2015. Additionally, those results were compared to pre-existing files of histopathological findings (Fig. 1).
This study included uni- or bilateral capsular fibrosis after aesthetic or reconstructive augmentation with different types, shells and materials of breast implants and breast tissue expanders. Excluding criteria was the appearance of anaplastic large cell lymphoma in the past medical history before the first breast augmentation. Clinical data, which included age, gender, implant material, placement of the implant, reason for augmentation and for revision were available for all patients.
Implant removal or exchange was prompted because of the formation of capsular fibrosis (Baker score I - IV). Capsulectomy was performed in any patient. For our study we used Baker’s clinical classification system (Table 1). Missing information was titled unknown.
The pathological diagnosis was established by capsulectomy taking samples of capsular tissue for standard histologic and immunohistochemical analysis (Fig. 1). Cell blocks were prepared by standard protocols for H&E-sections and some of them additionally for immunostaining. Pre-existing histologic examinations were performed by fixing the biopsies in paraffin-embedded, 10% -buffered formalin and cutting the fibrous capsules into small pieces (20-10 mm). Subsequently, the interior side of the capsule facing the implant was applied to a foil before placing it in liquid nitrogen (–196°C). Finally, the pieces were cut again into 4-5 μm units and stained with hematoxylin and eosin [H&E]. For standard histomorphologic examinations a light microscope (Axiostar Plus, Carl Zeiss®, Goettingen, Germany) with 40 × objective magnification (CP-Achromat 40×/0,65; ∞/0,17) was used. Detailed analyses of various cell types or structures (Table 2) were performed. About these findings we are going to publish another paper showing detailed information. Histological signs of cellular inflammation such as plasma cells, foam cells, lymphocytes, the number of foreign body giant cells, presence of silicone/amorphous material as well as the presence of synovia-like metaplasia were scored separately on a four-point scale depending on the amount of counted cells. Two pathologists checked all histological analyses. In this paper we focus on the appearance of anaplastic lymphoma cells. As we found no evidence of atypical lymphocytes in all 307 samples, we performed an immunohistochemical staining with a random selection to re-check our findings. We felt there was no indication to perform this staining on all histological specimens in the absence of atypical lymphocytes.
For the immunohistochemical staining paraffin sections were placed on slides and dried for two hours at 60°C. Those slides were exposed to dewaxing and antigen retrieval with Ventana’s CC1 retrieval solution for half an hour at 100°C on an automated immunostainer (Ventana Medical Systems Inc.®, Tucson, Arizona). As suggested in national and international recommendations for the diagnosis of anaplastic large cell lymphoma [14, 16], a selection of primary antibodies can be used to screen for ALCL. We used CD30 and ALK-1, which were applied to the sections at 37°C for about 30 minutes. Antigen was visualized by using the ultraView Universal DAB Detection Kit (Ventana Medical Systems Inc.®, Tucson, Arizona).
All findings were compared to the pre-existing histological results.
The diagnoses were based on clinical, histopathological and immunohistochemical criteria and were made by experienced plastic surgeons and pathologists at the time of diagnosis of capsular fibrosis as well as for the retrospective examinations.
Statistical analysis
Statistical analysis was performed using a Mann-Whitney rank sum test. Continuous variables and all data are described using mean±SD. A p-value of less than 0.05 was considered statistically significant. Paired comparisons t-test for dependent samples and two-sample t-test assuming unequal variances were used. All analyses were performed with the SPSS statistic software for Windows (version 20.0; SPSS, Chicago, IL). The study was approved by the local ethic committee (Reference No.: 15-101-0024).
Results
In total, 296 histological samples of capsular tissue of 227 patients were analyzed.
Clinical findings
The womens’ average age was 42.91±12.66 years (median: 43.83 years) at the time of breast implant surgery and 51.40±11.40 years (median: 52.37 years) at breast implant removal or explantation. In 51% implantation was performed for reconstructive reasons, in 48% for aesthetic reasons, in 1% for other reasons (Fig. 2). The main reason for implant removal or explantation with subsequent capsulectomy was capsular fibrosis (n = 173), followed by regular exchange of tissue expander to insert breast implant (n = 29), implant rupture (n = 27) and other indications (Fig. 3).
Implants’ residing time was clustered into groups (Fig. 4). Mean duration of implant residing time was 8.49±8.90 years (median: 5.83 years). The strong variance highlighted by the SD is based upon a significant subset with rather prolonged residing time of the implant (19.93% of our observations have an implant longevity of over 15 years).
To determine the clinical degree of capsular fibrosis the four-grade Baker scale was used and compared with different implant surfaces. Of the 68 smooth-surfaced breast implants three patients were allocated to Baker 1, 12 to Baker 2, 33 to Baker 3 and 20 to Baker 4. Of the 93 textured-surfaced implants two patients showed Baker 1, eight Baker 2, 48 Baker 3 and 35 Baker 4 (Fig. 5).
In 135 cases the implant surface was not documented and could not be retraced.
Histologic and immunohistochemical findings
The histologic features of all cases are summarized (Table 2). In 296 capsular tissue specimens no atypical lymphoid cells, including ‘hallmark’ cells, were found in standard histologic staining. This was confirmed by the sample of immunohistochemical analysis, demonstrating the absence of malignant cells by negative staining for CD30 and ALK-1. No ALCL was identified in any of the cases.
A mixture of inflammatory cells - such as infiltrates of typical lymphocytes, plasma cells and foam cells - was present in almost every capsular tissue specimen. We looked at mean and standard deviation (SD) of those cell types for the different implants’ residing time-clusters, but saw no statistical significance (Fig. 6 and Table 3).
Implants’ residing time-clusters compared to inflammatory cells is shown in Fig. 7.
Discussion
In this report we collate the present knowledge about a possible association between breast implants and the generation of ALCL with our finding in 296 capsular tissues of 227 patients. Overall, the evidence regarding breast implant-related ALCL is inconclusive.
Recent scandals about ALCL or PIP (Poly Implant Prothèse)-breast implants led to increased concerns about safety and oncogenesis in general, which underscores the importance of a detailed and widespread scientific research. Although the complication of developing ALCL is very infrequent, increased vigilance, awareness and screening methods need to be done.
Highlighting that non-Hodgkin lymphomas involving the breast in general amount to 1.7% to 2.2% of extranodal non-Hodgkin lymphomas, resulting in 0.38% to 0.7% of all non-Hodgkin diseases [2], ALCL in breast tissue really is very rare [25, 42]. Some studies suggested, that its occurrence could be slightly promoted by a certain type of implant. Some prior investigations have been published, reporting histologic findings of lymphomas involving the breast, most of them limited by their small sample size.
This investigation is one of the few larger studies to retrospectively examine a possible association between breast implants and ALCL including 296 capsular tissue specimens.
Of course our study has several limitations. Currently, new data about the expression of specific biomarkers could be connected with BIA-ALCL has been published [21]. It would be interesting to analyze, whether T-cell expansion is observed in the infiltrate around breast implants or if there is an interaction of T cells with breast implants.
Whereas Roden et al. in 2008 postulated ALCL to be seroma-associated [38], new evidence highlighted the term effusion-associated [1, 46]. Miranda et al. recently published a long-term follow-up of 60 women with BIA-ALCL stating that ALCL could be more dangerous for women who had a solid ALCL tumor-mass than for patients with ALCL cells in their surrounding effusion, resulting in a different therapeutical interventions after implant removal [30]. Brody et al. collected 173 cases reporting about ALCL in breast implant patients [5]. In conclusion they found no correlation to a specific type of implant-filling or indication for implant, but they found that all patients had received at least one textured-surface device.
In 2015 Hu et al. published an animal study, where they propose a possible connection between chronic bacterial biofilm around breast implants and T-cell hyperplasia as a possible mechanism for development of ALCL [17]. Some studies found ALCL-tumor cells in the scar formation, some in the capsule and some in the peri-implant fluid. These different locations indicate the possibility of a multifactorial pathogenesis similar to the development of capsular fibrosis. Almost all studies name similar risk factors for ALCL as for capsular fibrosis [37], e.g. late onset seroma or infection. This enhances the question about a possible connection between capsular fibrosis andALCL.
These results originate from the latest findings in research and underscore the importance of advanced studies in this field.
Plastic surgeons and gynecologists should be aware of complications and risks due to breast implant augmentation. Clearly more comparative long-term studies on breast implant safety are needed.
We could not establish an association between different types of breast implants and ALCL, as there were no anaplastic large-cell lymphoma T-cells or any malignant tumor cells in the examined samples of capsular tissue. A limiting factor in our study is, that we could not investigate peri-implant fluid due to retrospective examination.
Further histologic evaluations of capsular tissue and peri-implant fluid as well as basic scientific studies concerning different shells and materials of breast implants need to be done.
In agreement to Eaves et al. in 2011, we emphasize the need of additional scientific studies concerning this topic, as the evidence to date regarding a causal connection between breast implants and ALCL is inconclusive [11].
Although breast implant-associated ALCL represents an infinitesimal small number in the field of lymphoma [39], further scientific analyses are required to understand this disease.
Additionally, all concerns and discussions underscore the importance of a common national and international breast implant registry to quickly monitor, evaluate and control complications vigilantly. Beyond an international registry an international multicentric prospective study – powered for the rare occurrence of ALCL – should be emphasized. Otherwise a causal context will not be ensured.
Funding
There were no sources of funding for research and publication.
