Improving the safety of breast implants: implant-associated lymphoma

A likely causal link between breast implants and lymphoma highlights the importance of a prospective registry

Breast devices, including implants and tissue expanders, are classified as class III (high risk) medical devices by the Therapeutic Goods Administration, and are subject to the highest level of regulatory control. They have been associated with highly publicised health scares in the past, particularly, the Poly Implant Prothèse crisis.1 More recently, breast implants have again created national concern, with the Therapeutic Goods Administration confirming in late 2016 that there were 46 reports of breast implant-associated anaplastic large cell lymphoma (BIA-ALCL) in Australia, including three cases that resulted in death. This number has since increased to 53.2 Most breast implants are used in young women and in women who have had breast cancer, thus long term exposure to these devices can be anticipated. It is therefore imperative to identify serious adverse effects at the earliest opportunity. The Australian Breast Device Registry is ideally positioned to do this, but it requires sufficient resources and engagement to ensure that it remains fit for purpose.

Breast implant-associated anaplastic large cell lymphoma

BIA-ALCL is a rare T cell derived lymphoma within the non-Hodgkin lymphoma group. Patients present with unilateral swelling, pain or enlargement of an implanted breast3 on average 7–10 years after initial implant placement. It is expected that with improved recognition, BIA-ALCL will be more readily diagnosed and many more cases will emerge.

This type of lymphoma presents in two ways. The seroma type (or in situ disease) consists of a malignant effusion with or without the inner lining of the capsule involved. This form appears to have an indolent course, with cure obtained by removal of the implant and capsule. The mass type (or infiltrative disease) is less common, but has a worse prognosis and is usually treated with surgery and adjuvant chemotherapy. It is unclear whether both are on the spectrum of lymphoproliferative disorders or if they are separate diseases.4

The 53 confirmed cases of BIA-ALCL in Australian women were identified between 2007 and 2017. The cases occurred 3–14 years after implantation (range, 1–37 years), with a median interval of about 8 years. The Therapeutic Goods Administration updated its website on 20 December 2016 confirming the likelihood of a causal link, rather than just an association, between BIA-ALCL and breast implants — the first regulatory authority in the world to do so.2 A detailed case series across Australia and New Zealand found that all cases of BIA-ALCL occurred in women with exposure to implants with surface rendering that was either textured or polyurethane, with none to date recorded in women who had only ever had implants with smooth surfaces.5

The most common complication associated with breast implants is capsular contracture, in which there is thickening and tightening of the fibrous capsule that forms around the implant, resulting in hardening and alteration in shape. Textured surface implants are commonly used because they facilitate tissue ingrowth that may result in a reduced rate of capsular contracture compared with smooth-walled devices.6 There is a wide spectrum of surface renderings and their topographies are classed as “microtextured” or “macrotextured”, with the latter having the largest surface area. The macrotextured implants have come into question with BIA-ALCL; one leading hypothesis implicates bacterial biofilm growing on the more deeply textured implants,7 which in turn increases lymphocyte activation8 leading to T cell transformation. The finding of mutations in the janus kinase–signal transducer and activator of transcription 3 pathway suggests that a deficient response to chronic inflammation may also contribute to T cell transformation.9

Clinicians should be aware of BIA-ALCL and request an ultrasound for all patients with a breast implant presenting with an acutely swollen breast. If fluid is present around the implant, ultrasound-guided aspiration of this fluid should be performed and submitted for cytology and for CD30 and anaplastic lymphoma kinase staining. Once confirmed, magnetic resonance imaging (MRI) assessment of the breast can determine whether, in addition to a seroma, lymphadenopathy and any solid tumours are present, which would require a biopsy, and assists with surgical planning. We suggest that, in Australia, the Medicare Benefits Schedule should have a provision for MRIs to be reimbursed for cases of proven BIA-ALCL for this purpose.

The risk of developing BIA-ALCL is low, as it is estimated to be between 1/1000 to 1/10 000 women with implants.2 Spontaneous case reports have formed the bulk of the known cases of BIA-ALCL. This method of post-market vigilance is significantly limited as the population from which the reports arise is not defined, resulting in poor estimates of risk. Estimates of differential risk were calculated with the Australian and New Zealand case series based on historical implant sales. Compared with smooth implants, the odds ratio was 14 for developing BIA-ALCL with textured implants, and 11 with polyurethane implants.5

The likelihood that indolent disease, in particular, may previously have been managed without a diagnosis of BIA-ALCL further complicates risk calculations. Internationally, there are regional differences in the incidence of BIA-ALCL, and it is unclear why some areas, such as Europe, South America and China, have few cases compared with Australia.3

Advising the many women with textured or polyurethane implants in situ of the potential risks is therefore difficult. However, given that the condition is extremely rare and most have an indolent course, in the absence of symptoms there is no need to recommend implant removal.2

The Australian Breast Device Registry

The emergence of BIA-ALCL highlights the importance of systematic post-market surveillance of breast devices. The Australian Breast Device Registry has a pivotal role to play in furthering our understanding of BIA-ALCL, as well as monitoring the occurrence of other known and unanticipated adverse events. The Department of Health has contracted Monash University to operate the Australian registry — an opt-out clinical quality registry for all recipients of breast devices, including breast implants for cosmetic or reconstructive purposes, tissue expanders and acellular dermal matrices. The registry was established following an Australian Senate inquiry recognising the need for a better audit trail in light of the Poly Implant Prothèse crisis,1 and the design of the registry has addressed many of the weaknesses of its predecessor.10

National roll-out of the registry commenced in 2015, and to date over 300 surgeons have contributed data from over 17 000 patients. When complete, it aims to capture over 95% of the estimated 20 000 breast device operations occurring in Australia annually, including implant, explant and repositioning. The registry will be the national data centre for the majority of cases of BIA-ALCL going forward, building on the existing case series,5 and will be able to provide systematic information regarding the patient-specific and implant-related characteristics of BIA-ALCL and any other reported adverse events. The large registry dataset may also help inform and evaluate strategies proposed to reduce risk from breast implant devices. In the case of BIA-ALCL, this includes evaluating improvements in surgical techniques using a 14-point plan11 to reduce the risk of biofilm formation, and considering a recommended lifespan in vivo of less than the median interval to develop BIA-ALCL for textured and polyurethane implants.

Another role for the registry will be to proactively monitor the breast implant revision rate and reason for revision (eg, rupture and capsular contracture) in order to identify underperforming devices earlier than is currently achievable with spontaneous reporting, an approach successfully demonstrated by the Australian Orthopaedic Association National Joint Replacement Registry.12 Patient-reported outcome measures will also be collected, and may provide a means for early detection of complications. Feedback will also be provided to practitioners, which has been shown to improve performance,13 and device information will be available to consumers.

Following on from other successful international registry consortia, with orthopaedic and vascular networks being the exemplars,14 the International Collaboration of Breast Registries Activities network brings together countries working on breast implant safety;15 the Australian Breast Device Registry dataset is provided to these countries to facilitate international data harmonisation. The Netherlands has adopted the dataset in the Dutch Breast Implant Registry, and work is currently underway to amplify the dataset using cloud technology to combine the Australian and Dutch de-identified datasets.

The urgent need for collaboration

Previous implant-related health scares escalated rapidly because of a lack of data.1 The existence of the Australian Breast Device Registry has meant that BIA-ALCL is the first implant-related concern to be dealt with prospectively, and international collaborations have the potential to rapidly amplify the datasets. Data quality is key to ensuring that the registry can be used to accurately determine implant-related risks, as well as to identify problematic devices. We suggest three key strategies to ensure adequate case ascertainment.

First, contribution to the registry should be an essential requirement of membership to the collaborating societies: the Australian Society of Plastic Surgeons, the Australasian College of Cosmetic Surgery and Breast Surgeons of Australia and New Zealand, to ensure population representation nationally.

Second, the Therapeutic Goods Administration classification as a class III device should obligate manufacturers to financially contribute to an appropriate registry for post-market surveillance. Such models are already occurring in the US for cardiac valves and are critical to ensure the ongoing financial sustainability of a national registry.

Third, and crucially, industry should be mandated to provide ongoing accurate sales data to the registry to validate its case capture rate to ensure that it is a reliable source of information. In this way, the Australian Breast Device Registry will provide — and will be recognised internationally — for its comprehensive capture of critical information regarding the long term safety and risks associated with breast implants in Australia.