Tongue carcinoma in young adults: a review of the literature
A recent reduction in the number of smoke-related tumours has been observed thanks to the diffusion of anti-tobacco campaigns carried out in the majority of developed countries. Nevertheless, as demonstrated by recent global epidemiologic studies, squamous cell carcinoma of the mobile tongue appears to be progressively increasing in incidence, particularly among young adults and especially in females. The driving mechanism responsible for such changes is still to be precisely defined. Several genetic studies have compared the mutational pat- tern of tongue squamous cell carcinoma in young adults to that of more elderly patients, without identifying significant differences that may help in better characterising this subgroup of subjects. Tongue squamous cell carcinomas in young adults have been historically considered as particularly aggressive clinical entities, with a high risk of loco-regional relapse, survival rates inferior to those of the general head and neck cancer group and need for a more aggressive therapy. However, considering the most recent studies, prognostic results in this patient group are heterogeneous and it is not possible to confirm this tendency. Thus, it is not justified to embrace different therapeutic approaches according to patient age. Eventually, an additional element to consider when examining young subjects affected by tongue cancer is the possibility of genetic predisposition. Alterations affecting pathways involved in DNA repair, surveillance of genetic stability or regulation of cellular growth may determine an increased likelihood of developing head and neck cancers.
Head and neck tumours are typically characterised by a peak of incidence in the elderly and a strong correlation with chronic exposure to risk factors such as smoking and alcohol abuse 1. This is a consequence of the typical pattern of initiation and neoplastic progression which is encountered in these anatomical sites, characterised by an incremental accumulation of mutations leading, in the long-run, to frank neoplastic transformation 2. In this sense, the adult-elderly age has a central role in allowing the accumulation of a sufficient rate of genetic alterations to promote carcinogenesis 3 4.
Nevertheless, around 5% of patients suffering from head and neck cancer are diagnosed before they reach the age of 45 years. The incidence in this subgroup appears to be consistently growing, a trend that is in contrast to what observed in the general population. This phenomenon seems to be mainly linked to an increase of oropharyngeal carcinomas (mainly human papilloma virus [HPV]-related) and of the oral cavity (apparently not related to any known virus), paralleled by a reduction of laryngeal and hypopharyngeal tumours 5.
The aim of this review is to collect evidence published in the recent literature and pertinent to such an epidemiological trend, with special emphasis on oral tongue squamous cell carcinoma (SCC) in the young population. Due to the remarkable heterogeneity of studies assessing the molecular, epidemiological and clinical characteristics of tongue SCC, a non-structured review on all evidence available to date will be presented for each of these aspects.
In oral cavity SCCs, gender distribution is different according to the age of onset taken into consideration: while in the elderly, males represent over 70% of cases, this percentage falls to 50-65% under 45 years of age. This difference has increased evidence when considering that the majority of non-smokers and non-drinkers young patients are females 6 7. These epidemiological features may thus represent the expression of aetiopathogenic differences distinguishing the young subgroup of patients from the elderly.
Thanks to the diffusion of anti-tobacco campaigns carried out in the majority of developed countries, it has been possible to observe a recent reduction in the number of smoke-related tumours. Nevertheless, oral cavity SCC, in particular of the mobile tongue, as well as those of the oropharynx, appear to have a progressively increasing incidence 8. While in oropharyngeal carcinomas this may be explained by an increasing exposure to the HPV, in the oral cavity, a specific pathogenic role for viral infections has yet to be demonstrated 9-11.
Tongue SCCs demonstrate an increasing incidence in the young population, as previously reported in the 1990s by a group of American authors 12-14. These data were further confirmed by a global study specifically investigating tongue carcinomas and including a total of 22 tumour registries. This study highlighted a yearly increase ranging from 0.4% to 3.3% that was significantly higher in young patients (#x003C; 45 years) in 14 out of 22 registries 8. In this comprehensive analysis, the greatest increase observed in females, particular in the younger cohort, was found only in some registries, while in others a homogenous distribution or even a male preponderance was shown 2 6 9 10 14 15. In this context, it is however important to point out the difficulties in sorting out precise data for particular subgroups with homogenous clinical characteristics. Considering these recent epidemiological changes, some authors have hypothesised the involvement of a viral agent in the carcinogenetic process which may be known or yet to be defined, even though no specific viral genetic material has been consistently isolated in tongue SCCs of young patients 16 17.
Peculiarities of tongue tumours in young adults
From a clinical standpoint, tongue SCCs in the under-45 year-old group, particularly if non-smokers and non-drinkers, are found to prevalently originate from the tongue edge 16. The concept of “field cancerisation”, typically seen in the general head and neck cancer group, in these patients appears less evident, with a pattern of relapse/second tumour generally characterised by the involvement of the residual tongue in continuity with the primary neoplastic lesion 16 18 19. However, no evidence of differences in histopathological features have been reported 20. Some authors have also described the relapse tendency in the first two years of follow-up, with a significant decrease in risk following this lapse of time 21.
Considering the varied epidemiological characteristics, some genetic studies have compared the mutational pattern of tongue SCCs in young adults to that of more elderly patients. In the vast majority of publications, the two types of tumours have been shown to be similar and generally comparable to head and neck carcinomas 22. The most frequently mutated genes have been reported to be TP53, CDKN2A, NOTCH1, CASP8, FAT1, PIK3CA and MLL2 in both groups of patients 23. In the literature, however, heterogeneous results have been reported in relation to the prevalence of p53 mutations in young patients, with a tendency to a significant increase in some studies 23 and a decrease in others, in particular for non-smokers 24 25. A recent analysis performed by Knopf and colleagues 26 has shed further light on the molecular mechanisms linked to the starting of tongue SCCs in patients below 45 years of age, highlighting a predominant alteration of the WNT-CTNNB1-STK11 and CDKN2A-HGF-MET pathways, as well as increased expression of ATM, BRCA1, E2F1, and FHIT. Such a biological molecular landscape may be associated with greater radio-sensitivity in this patient population, which has significant consequences from the therapeutic and prognostic points of view 27-30.
However, these differences in genes mutation/expression may be mainly related to heterogeneous study populations: when analysing the extensive “The Cancer Genome Atlas” (TCGA) database, it is not possible to confirm any significant difference in relation to age, even considering the most recent genomic data 31 32. Thus, tongue SCC in young patients appears to be comparable to those in elderly patients even when taking into account their genetic alterations.
Historically, tongue SCCs in young adults have been considered by many authors as particularly aggressive clinical entities, with a high risk of loco-regional relapse, and survival rates inferior to that of the general population and resulting need for a more aggressive course of therapy 12 33-35. However, considering the most recent studies, the prognostic results in this patient group are heterogeneous and it is not possible to confirm this tendency 12 13 19 24 33 35-41. This heterogeneity of results may also depend on the variable definition of “young age”, with a wide range of cut-offs going from 30 to 45 years of age. In the future, to make the data more homogeneous and comparable, it has been suggested that it might be useful to consider as young patients only those below the age of 30 42.
In the specific field of tongue SCCs, the analysis of a wide case series of 276 patients (66 below and 210 over the age of 45) performed by Knopf and colleagues 26 has revealed a significant prognostic advantage in younger patients. Regardless of this consideration, it is crucial to interpret this result as likely influenced by tumour extension, and thus T category, which is significantly lower in patients below 45 years of age (without differences in N and M categories). This selection bias is partly mitigated by case-matched studies characterised by a homogenous patient distribution in relation to tumour characteristics and comorbidities. Considering these types of studies, a broad cohort of 185 patients below the age of 45 affected by head and neck SCC, 80 of which of the oral cavity, compared with an analogous group of elderly subjects, demonstrated better results in terms of overall survival and second tumour occurrence in young patients 13. Focusing the analysis on tongue carcinomas, 20 cases under-45 years compared with the same number of older subjects, matched by stage and gender, highlighted better overall and disease-related survivals in the younger subgroup 43. On the other hand, some studies have demonstrated the absence of prognostically significant differences, taking into account 87 patients ≤ 45 years of age in a study by Lassig and colleagues 44 and 31 patients ≤ 40 years analysed by Pytynia and coworkers 45, both case-matched and including head and neck tumours in general. Finally, Blanchard and co-authors 46 have recently confirmed the absence of significant differences in 50 patients ≤ 40 years affected by tongue carcinomas versus older subjects matched by stage at presentation. It is nevertheless important to point out the outlier results on tongue SCC reported by at least three different studies 47-49, which showed a significant increase of relapses in young adults and a related negative impact on overall survival within this cohort. A non-significant trend towards poorer prognosis has also been reported by the group led by Kourelis 50 which considered only patients with early stage tongue SCCs. In particular, in young women, there has also been evidence of a significant increase in the risk of relapse compared to matched cases 50.
Genetically-determined conditions of increased susceptibility to tongue cancer
An additional element to take into consideration when examining patients affected by tongue SCC in the young age is the possibility of some sort of genetic predisposition 2. Alterations affecting pathways involved in DNA repair, surveillance of genetic stability or regulation of cellular growth may determine an increased likelihood of developing head and neck cancers. This is clearly exemplified in rare hereditable syndromes linked to specific genetic mutations leading to a dramatically increased incidence of head and neck cancers and tumours in general. These are represented, in particular, by Fanconi anaemia, xeroderma pigmentosum, Bloom, Li-Fraumeni, familial atypical multiple mole melanoma, and ataxia telangiectasia syndromes. In all these diseases, the increased risk in developing tongue SCC (that may reach an incidence up to 700-1000 greater than that for the general population) can be associated with that reported for other neoplasms and syndromic presentations. The suspicion of a genetically-determined alteration should arise from a detailed family history of the proband positive for: 1) first-degree relative with the same kind of tumour and same clinical presentation; 2) two or more first-degree relatives with a tumour in the same location; 3) two or more first-degree relatives affected by rare tumours.
It is, however, important to note the existence of increased susceptibility to head and neck cancers even in the absence of syndromic presentations. Retrospective studies have, in fact, demonstrated a relative risk that is increased 3.6-fold in subjects with first-degree relatives affected by other head and neck cancers 51-53. In addition, young adults affected by tumours in these locations have been shown to be increasingly susceptible to genotoxic stress as shown by the bleomycin test, strongly suggesting the development of an increased number of chromosomal alterations in respect to older subjects 53. In a similar manner, young adults affected by such neoplasms have shown an increased vulnerability to the negative effects of smoking (compared to healthy volunteers), a feature likely linked to an increased prevalence in this population of single nucleotide polymorphisms in genes involved in the repair of alcohol- and smoke-related mutations and in regulation of the cell cycle 54-56.
Tongue SCC in young adults represents a rare disease, characterised by an incidence that has been recently reported as increasing, particularly in females. However, no specific aetiopathogenic agent has been identified to date that may provide a univocal explanation for such a trend. As of today, it is not possible to demonstrate significant differences with respect to the general head and neck cancer patients group in terms of prognosis and, thus, a comprehensive therapeutic approach according to current guidelines should be recommended. In this view, thanks to the overall fitness of young patients, the “optimal treatment” according to tumour stage and characteristics may be frequently obtained, while it is essential to avoid unjustified overtreatments on the basis of patient’s age alone.
In consideration of the patients’ life expectancy, functional impairment and risk of recurrence, a long-lasting follow-up should be considered, with the aim of allowing an early diagnosis of persistent/recurrent disease, monitoring and optimising the patient’s quality of life, and promoting awareness of adjunctive risk factors such as tobacco or alcohol consumption.
Patients presenting with associated syndromic characteristics, a suggestive family history and age of onset below 20-30 years might also require detailed genetic counselling to better assess the general disease framework.