Cancerous lesions in the vicinity of dental implants: a systematic review


Introduction: The massive diffusion of dental implant treatments in the last decades leads to the appearance of complications, most of them inflammatory, although important complications have been described as malignant lesions in the vicinity of dental implants. The objective of this article is to describe the cases described in the literature of oral squamous cell carcinoma (OSCC) or clinical variants and metastases, in the vicinity of dental implants and to analyze the possible etiological agents involved. Material and methods: The criteria used were those described in the PRISMA® Declaration for performing systematic reviews. An electronic search was performed on MEDLINE (via PubMed) using the terms MeSH: “dental implants” AND “squamous cell carcinoma” OR “dental implant complications” AND “squamous cell carcinoma”. Results: Thirty-eight articles describing a total of 76 cases of OSCC or clinical variants, as well as metastasis in the vicinity of dental implants, were included. Conclusions: It is not possible to establish a cause-effect relationship between dental implants and the development of OSCC. Its clinical appearance can be confused with periimplantitis, so that, in cases of sudden onset, which do not respond to conventional treatment and/or have associated alterations in sensitivity, a biopsy should be performed.


Introduction
The sixth most common cancers are located within the head and neck region, 48% of which in the oral cavity, representing 3% of all cancers [1]. 90% of cancers of the oral cavity are histopathologically oral squamous cell carcinomas (OSCC) [2,3], malignant neoplasms derived from the stratified squamous epithelium of the oral mucosa [4][5][6]. Men are more affected by OSCC than women (the ratio by gender being 1.5:1), especially during their sixties; however, the incidence in people younger than 45 is increasing [7], which could be related to human papilloma virus (HPV) infections.
The pathogenesis of OSCC is a multistep process, beginning with irreversible changes of the squamous epithelium, including epithelial dysplasia and further progression into an intraepithelial neoplasia/in-situ carcinoma [8]. This develops into an invasive carcinoma due to the cumulation of genetic alterations, triggered by prolonged exposure to carcinogens [9]. Risk factors such as the following have been cited as part of its multifactorial aetiology: tobacco smoking and alcohol consumption, precursor lesions or potentially malignant disorders (such as leukoplakias, erythroplakias, erythroleukoplakias and/or oral lichen planus), chronic inflammation due to mechanical irritation (trauma) or bacterial infections (poor oral hygiene and periodontitis), viral infections (such as HPV), genetic predisposition, and/or having received treatment for previous malignant lesions [10][11][12].
Primary cancers that metastasize to the oral cavity are estimated to represent 1%, and in two thirds of cases, until the metastasis is not diagnosed, primary tumours go unnoticed [5,13]. The most frequent primary tumours that metastasise to the oral cavity in women are those located in the breast (42%), while in men, the most frequent are those in the lung (22.3%) and prostate (12%) [5].
Today, dental implants stand as the most effective treatment option for the rehabilitation of partially dentate and edentulous patients, and in certain cases, they are the only option. However, as the popularity of using dental implants has increased over recent decades, alongside has the number of complications. Although most are inflammatory in nature, other important complications have also been described, such as cases of malignant lesions in close proximity to dental implants. The number of cases cited in the literature over the past decades has notably increased [14]. Bhatavadekar [15] estimated the standardized incidence rate of the risk of suffering OSCC following the placement of a dental implant to be 0.00017 per million inhabitants per year. Despite this being a low figure, it is presumable that the number of cases diagnosed and/or described in the literature is less than the real number of patients affected.
Being familiar with the levels of prevention of OSCC is of utmost importance, being of interest implementing in primary centres control and maintenance prevention initiatives for patients susceptible of suffering of OSCC. Primary prevention is essential, reducing or eliminating the risk factors clearly linked to carcinogenesis, especially tobacco smoking and alcohol consumption, as well as improving diet; and secondary prevention, which involves the early detection and diagnosis of potentially malignant disorders. Tertiary prevention aims to avoid the recurrence or appearance of new primary tumours, as well as reducing morbidity.
The objective of this article is to carry out a systematic review of the existing literature in order to collect the cases of OSCC described or clinical variants, as well as metastasis, in the vicinity of dental implants and analyse the possible etiological agents involved.

Search strategy
An electronic search in MEDLINE (via PubMed), without temporal restriction updated to March 2020, using a combination of the following MeSH terms (Medical Subjects Headings) and Boolean operators, was performed: "dental implants" AND "squamous cell carcinoma" OR "dental implant complications" AND "squamous cell carcinoma".
The present review followed the PRISMA® Statement guidelines (Preferred Reporting Items for Systematic Reviews and Meta-analysis). The objective of this systematic review was to answer the following "PICO" (P = patient / problem / population; I = intervention; C = comparison; OR = outcome) question (Tab. I).
What is the relationship between peri-implant oral cancer in patients with/without a history of potentially malignant disorders, oral cancer or cancer in other locations?
Moreover, a search in Google Scholar was also carried out, using the references of the articles which complied with inclusion criteria, for publications of potential interest that did not appear in the initial search.
Before the beginning of the study, a consensus was reached among all the authors, and a series of inclusion and exclusion criteria were defined.

Inclusion criteria
The studies to be selected had to fulfil the following criteria: (a) studies in humans; (b) articles published in English, Spanish or French; (c) systematic reviews with or without meta-analyses; and, (d) case reports and series.

Exclusion criteria
Studies that met the following exclusion criteria were disregarded: (a) experimental laboratory studies; (b) animal studies; (c) studies which main topic was not the description of the presence of metastasis, OSCC or clinical variants in the peri-implant mucosa; (d) duplicate articles; (e) books or book chapters; (f) author comments; and (g) letters to the Editor.

Bias risk assessment
Bias risk assessment was assessed independently by two reviewers (AOSP and VSS). Disagreements were resolved by consensus between the two reviewers or the intervention of a third author (MVMM).

Results
The initial search strategy on MEDLINE via PubMed yielded 232 articles for screening by title and abstract, out of which, under consensus, 160 were excluded due to lack of compliance with inclusion criteria and after duplicates removed. 30 articles were selected for full-text reading, out of which 4 were excluded for not describing malignancies in the peri-implant mucosa in the form of metastasis, OSCC, or other histopathological variants, as well as another case report for not providing sufficient information, resulting in 26 articles obtained from MEDLINE.
The references of the aforementioned articles were analysed for an additional electronic search on Google Scholar, adding 12 more articles that fulfilled inclusion criteria. Finally, 38 articles were included in the present review, describing cases of malignancies in the peri-implant mucosa in the form of metastasis, OSCC, or other histopathological variants (Fig. 1). In general, the available literature is limited to clinical cases or series of small cases, which represents a limitation for the present study (Tab. II). The 38 included studies describe a total of 76 cases in the peri-implant mucosa, in particular, 69 cases of OSCC [5,6, (of which 12 were recurrent) [16][17][18]20,26,27,30,35,39], two lung metastases [36,42] and three breast metastases [13,43,44]. Two cases of verrucous carcinomas were also described [32].
The mean age of the patients was 67.48 ± 9.99 years. More women (N = 47) were affected than men (N = 29). The mandibular arch (N = 67) was more affected than the maxillary arch (N = 7), and in one patient, both were affected [13], especially in posterior areas. In most cases, the implants were restored by overdentures (N = 36) and, to a lesser extent, by fixed prostheses (N = 34). In 6 cases, the authors did not describe restorations [30,32,35].

Discussion
The development of OSCC in proximity to dental implants is a rare event and so far, documented only by case reports or small case series. However, dental implants are becoming an increasingly popular option in the rehabilitation of partially dentate or edentulous patients, therefore, even a phenomenon that until now has been poorly documented can obtain clinical relevance as a consequence of a large number of implants that are placed [6].
It is difficult to establish a causal relation between the placement of a dental implant and the development of a malignant lesion in its vicinity [48], however, several risk factors have been described in the literature:

Relationship between chronic inflammation and cancer
The most plausible theory was described by Virchow in 1863, the first to describe the role of chronic inflammation in cancer [49]. Inflammation is a protective response of the tissues to irritation or infections, but in some cases, inflammation itself can trigger a malignant transformation, while, in other cases, it is activated and modulated by genetic factors and/or carcinogenic risk factors, linked to lifestyle (tobacco and alcohol), mechanical irritants (such as repeated trauma due to ill-fitting prostheses) [10][11][12], bacterial infections (periodontitis) [6] and even viral infections (HPV) [10][11][12]. The inflammatory response contributes to tumour progression, increasing the proliferation and survival of malignant cells, stimulating neoangiogenesis, and reducing antitumor immunity [50,51]. Peri-implant inflammation initially affects the peri-implant soft tissues, causing mucositis. When this inflammation is prolonged over time, it becomes chronic and causes peri-implant bone loss [18,39,40]. If this inflammation is sustained, it may have the sufficient potential to induce cell proliferation and prolong cell survival by activating oncogenes and inactivating tumour suppressor genes, causing genetic instability and an increased risk of cancer. In addition, certain inflammation pathways have an effect on carcinogenesis [4,6,21,52]. On the other hand, the peri-implant mucosa is a tissue that suffers chronic inflammation. In this regard, Silva et al. [53] carried out a study in which biopsies of the peri-implant mucosa were done in 20 patients 12 weeks following implant placement. The authors observed that, in patients without clinical signs of inflammation, the peri-implant mucosa had a mild inflammatory infiltrate. Berglundh and Donati [54] also described the presence of a chronic inflammatory infiltrate, dominated by plasma cells and B lymphocytes, in patients without clinical signs of inflammation.
Titanium has been considered an inert material, but recently, cases of type I or IV allergic reactions have been described, acting as irritant factors triggering the inflammatory process, leading to implant failure [55,56]. Sicilia et al. [57] performed a study on 1,500 patients with implants, observing positive allergic reactions in 0.6%.
In cases of dental implant mobility, at a microscopic level, metal particles can be released, potentially activating an inflammatory response, although its influence on carcinogenesis is unclear. Hydroxyapatite-coated implants aggravate this problem due to the different elasticity modulus between the coating and the underlying titanium [15]. Moxley et al. [29] presented a case in which two transmandibular posts were inserted. Two years following placement, they observed metal particles surrounding both posts, which had induced periimplant mucositis and, a year after, one of the implants presented an OSCC in its vicinity.
Another factor that can trigger inflammation is corrosion, defined as "the deterioration of metal as a result of an electrochemical attack of the oral environment". Most implants are made of titanium or titanium alloys, highly resistant to corrosion, due to the stability of the titanium dioxide (TiO 2 ) layer that they present [15,55,58]. When this layer is deteriorated or removed, for example, following the mechanical debridement of bacterial plaque, by the presence of fluoride ions, by the friction of the implant against the supporting bone, or by an acidic environment within the oral cavity due to inflammation, titanium can suffer corrosion [58,59]. Corrosion also occurs between titanium implants and other metal alloys used in other dental procedures. Alloys composed of a less noble metal act as an anode, and the titanium of the implant as a cathode, transferring electrons by metallic contact, and releasing ions into the crevicular space, activating the host's immune response [55,59]. In addition, corrosion can also occur within the implant itself, because of cracks or microscopic holes along its surface acting as an anode, while the rest of the implant acts as a cathode. Some authors have associated the release of corrosive products with carcinogenesis and mutagenesis in the oral cavity [59]. In this regard, in 2006, the Agency for Research on Cancer considered TiO 2 as a possible carcinogenic agent in humans [55,59]. Despite this, Doran et al. [60] concluded that titanium alloys are currently the safest for the manufacturing of dental implants, and that only vanadium, a minor component of their alloy, is toxic. In addition, the accumulation of a high concentration of metal ions in the oral cavity can act as an immunosuppressant agent locally, or they can be metabolized, generating cytotoxicity or potentially mutagenic reactive products [55,58].

OSCC recurrence
In many cases, regions of the jaws that have undergone resective surgery due to a previous OSCC are subsequently rehabilitated by implants which, along with other risk factors such as tobacco, alcohol, or the presence of potentially malignant disorders, could trigger the appearance of a new tumour. In this regard, locoregional recurrences have been described from 25 to 50% [61,62] and, orally, from 15 to 20% [24,39]. On the other hand, during radiotherapy, radiation can disperse, increasing dosage in front of the implant and decreasing it posteriorly. This theory, however, is difficult to associate with the recurrence of OSCC [27].

Pre-existence of potentially malignant disorders
In several of the described cases, the patients presented oral mucosal lesions, some of them initially benign, but following the insertion of a dental implant, became malignant. Either new lesions or recurrences may appear in the periimplant mucosa in patients with a history of cancer or potentially malignant disorders [4,55].
In this regard, the pathogenesis of oral lichen planus (OLP) is not fully understood, but it is considered a potentially malignant disorder. According to Gonzáles-Moles et al. [63] the rate of malignant transformation of oral lichen planus in different studies carried out between 1924 and 2007 is established within a margin between 0% and 12.5%. It is considered a chronic inflammatory autoimmune disorder and, if we consider the effect of dental implants on the immune response and the development of chronic inflammation, it is reasonable to suggest that they could trigger the malignant transformation of this type of lesions. This malignant transformation is more likely to occur in atrophic and/or erosive lichen planus lesions [64,65]. In OLP, the response to chronic inflammation and the healing process can increase the number of genetic mutations [64]. This hypothesis seems to be confirmed by studies that illustrate the relationship between T-cell inflammatory mediators and carcinogenesis [66]. These studies show that the macrophage migration inhibitory factor (MIF), released by T-cells and macrophages, decreases p53 transcription, a protein considered as a tumour suppressant gene. Protein p53 plays an important role in carcinogenesis, eliminating mutant cells by apoptosis. In this regard, mutant p53 have been found in over 60% of cancers. P53 blockage by MIF (or other mediators) increases the risk of developing cancer in cases of OLP [67]. Despite this, further studies are needed to confirm this hypothesis [4,40].
Other disorders classified as potentially malignant are erythroplakia, erythroleukoplakia and leukoplakia [68]. Non-homogenous nodular or speckled leukoplakia present a higher risk of malignancy [65] along with verrucous proliferative leukoplakia, which is considered an early form of verrucous carcinoma [68]. Likewise, erythroplakia have a high rate of malignant transformation (14-50%). Another entity associated with a high rate of malignancy is submucosal oral fibrosis, which is estimated to be around 7-30%. Other malignancy risk factors to be considered in these patients are female gender, long-lasting lesions, presence of leukoplakia unrelated to smoking, lesion size greater than 200 mm 2 , nonhomogeneous forms and presence of epithelial dysplasia [65].

Migration of malignant cells through the peri-implant groove
The potential route through which OSCC could spread is the invasion of the alveolar ridge, as well as the cortical bone [39,69]. Brown [69] studied other possible routes of dissemination, such as the periodontal ligament. However, as implants lack periodontal ligament, they do not fit into this theory. It is not known if the presence of dental implants could influence or modify the pattern of mandibular invasion. Most of the described cases emphasize that dental implants have not been clearly identified as a potential route of invasion of OSCC towards the jaws [39]. Only Schache et al. [24] suggest that the placement of dental implants could enable OSCC dissemination towards the alveolar bone, through the bone-implant interface. Therefore, further histological studies are needed to understand the possible dissemination routes, and their relationship, if any, with implants [5].

Metastatic lesions in soft peri-implant tissues
Some cases of metastasis in the vicinity of dental implants have been described [13,36,[42][43][44]. It is important to consider that the pathogenesis of metastasis to the maxilla and mandible is not clear. One theory that could explain its pathogenesis is the vascular theory. The premolar and molar mandibular region represent a region rich in bone marrow. In this situation, blood flow decreases and causes a change in blood circulation, which could result in a greater trapping of metastatic cells [5,70]. Local factors, such as trauma or surgical procedures, could further increase this process, due to the trap of tumour cells during the formation of clots in recent wounds and to growth factors locally released by regenerated tissues, which would stimulate the formation and proliferation of malignant tumour cells. Chronic inflammation also promotes metastasis locally because circulating tumour cells can become trapped within the rich capillary network of tissues that exhibit chronic inflammation. Favia et al. [43] describe a case of breast cancer metastasis around dental implants, in a patient under treatment with intravenous bisphosphonates (4 mg of zoledronic acid per month, for 33 months), who developed a periimplant medication-related osteonecrosis of the jaws (MRONJ), not linked to the insertion of it. Upon histopathological analysis, metastatic deposits were observed within the peri-implant necrotic bone. Orhan et al. [44] describe another case of metastasis in a 69-year-old patient treated for breast cancer 18 years earlier undergoing annual check-ups, with no signs of disease recurrence. Two implants were placed in the mandibular premolar area, and 3 months later, the patient developed numb chin syndrome, along with chest pain, arthralgia, back pain, vomiting and perspiration that, once studied, confirmed a recurrence of breast cancer, with metastasis in different locations, among them, around the implants.
It is not possible to establish a link between dental implants and cancerous lesions. However, before considering implant placement, a careful assessment of the patient's medical history and potential risk factors is essential, particularly a possible history of cancer.

Conclusions
Implant treatment is not to be contraindicated in patients at risk, but any risk factors present at the time must be comprehensively controlled, as well as implementing a stricter recall protocol. Upon the appearance of lesions compatible with periimplantitis, a differential diagnosis should be made with OSCC and, in cases of the sudden appearance of hyperplastic or osteolytic lesions, unresponsive to conventional treatment and/ or anaesthesia or paraesthesia, a biopsy is crucial.
Due to the increasing popularity of dental implants, it is essential that, in turn, dentists who review these patients in the future are aware of the possible complications that may arise, among which the presence of malignant lesions could be found.

Conflicts of interests:
The authors declare that they have no conflicts of interest in relation to the publication of this article.