15.1 Epidemiology, Etiology, and Genomics

Oral tongue cancer is the most common intraoral cancer in most countries worldwide, with squamous cell carcinoma continuing to be the most common type of tongue cancer. ¹ The average age of diagnosis is 64 years with most patients frequently diagnosed with early-stage lesions (T1 or T2). ¹ Epidemiologic studies suggest that black males have a higher incidence of oral tongue cancer and alongside low-income patients, are more often diagnosed with advanced disease. ²

Historically, alcohol and tobacco use has been associated with the development of oral cancer, with the risk of cancer related to the amount of tobacco or alcohol consumed. Synergistic effects exist between tobacco and alcohol such that combined use boosts the risk of developing cancer when compared to controls. ³ In smokers, the incidence of cancer in the head and neck is higher in males when compared to females. Use of smokeless tobacco and betel nut has also been implicated as etiologic factors in the development of oral tongue cancer, with smokeless tobacco imparting a fourfold increased risk of oral cavity cancer. ³ Even after diagnosis and treatment of oral cavity cancer, smoking cessation imparts further risk reduction of developing a second primary lesion. Moore demonstrated that the second primary rate after treatment of an oral cavity cancer went from 6% in former smokers to 40% in continuing smokers. ⁴ Yet, despite smoking cessation and improved public health awareness, epidemiologic data suggest increased incidence of oral tongue cancer in population subgroups, particularly in young patients. ^{5 , 6}

Though smoking and alcohol use remain significant risk factors for oral tongue cancer, epidemiologic studies reveal an increase in the number of nonsmoking nondrinking (NSND) patients diagnosed with oral tongue squamous cell carcinoma. ⁷ Oral trauma and inflammation have been proposed as risk factors for this subset of patients. ⁸ Indeed, the lateral aspect of the oral tongue is the most common site of oral tongue cancer and is twice as common in nonsmokers. ⁹ This suggests that dental trauma, bruxism, caries, or inflammation at the lateral tongue border leading to chronic inflammation may lead to malignant transformation. Furthermore, studies have shown a relationship between ill-fitting dentures and defective teeth and the development of oral cancer. ⁸ Taken together, this does support a role for chronic inflammation and the development of cancer in nonsmoking patients.

Base of tongue cancers associated with human papillomavirus (HPV) have been well described, yet studies by different groups have failed to demonstrate the same causative factors for oral tongue cancers. Prior studies have shown that HPV deoxyribonucleic acid (DNA) is present in a greater proportion of oral cavity specimens with squamous cell carcinoma than in matched noncarcinoma controls, suggesting that this may be an independent factor for the development of carcinoma. ¹⁰ HPV and its surrogate marker p16, have been identified in oral tongue squamous cell carcinoma in varying frequencies, from 11 to 55%. ¹¹ However, when defined as transcriptionally active HPV, the rate significantly decreases, suggesting that HPV may not be as significant in the etiology of these tumors as compared to oropharyngeal cancers. ^{12 , 13} Other studies suggest that in contrast to base of tongue cancers, different HPV subtypes (such as HPV beta-5) may be involved in oral tongue cancer, though the exact etiologic role remains to be determined. ⁹ Some preliminary studies have suggested that HPV status in oral tongue cancer may be associated with an overall worse prognosis though this remains an area of active investigation. ^{7 , 11} Furthermore, unlike the base of tongue and oropharyngeal subsites, HPV status in oral tongue cancer has yet to alter treatment strategies, specifically with regard to de-escalation or conservation therapy.

There has been increasing interest in studying the subset of young NSND patients developing oral tongue cancer, possible etiologic factors, and clinical implications. ¹¹ HPV detection remains low in the young NSND cohort as is the case for all oral tongue cancers, and when present it is predictive of an overall worse prognosis. ¹¹ This implies multiple transformative events leading to aggressive tumor biology. Other studies suggest this population has recurrence and survival similar to smokers, suggesting a unifying biology albeit through alternative molecular mechanisms. Genomically, the mutational landscape of tumors isolated from young NSND tongue cancer patients is similar to that of tumors isolated from older, smoking patients. ¹⁴ Mutation of p53 is relatively rare in nonsmokers when compared to smokers, as is the overall rate of loss of heterozygosity and allelic alterations, though it is common in young NSND patients. ¹⁵

Jeon et al have shown that young tongue cancer patients (< 40 years old) demonstrate a higher rate of distant metastasis and worse prognosis. ¹⁶ Other studies have shown that some young female patients with a high degree of immune checkpoint regulator PDL-1 expression had a decreased risk of recurrence and improved survival. ¹⁷ This expanding group of patients represents a unique subtype requiring further studies to better understand the epidemiologic risk factors and the molecular mechanisms to develop targeted therapy.

Whole-exome sequencing and banked databases such as the Cancer Genome Atlas (TCGA) database have expanded our genomic understanding of oral tongue cancer. Oral tongue cancer predominantly represents HPV-negative disease and often carries p53 loss and CDKN2A inactivation. While these tumors are quite heterogeneous, NOTCH1 was noted to be a frequently mutated gene as well as copy number changes in NOTCH2/NOTCH3. ^{14 , 18} Numerous signaling pathway alterations are also encountered, including those involved with growth and cell cycle regulation, oxidative stress, and differentiation. ¹⁴ Clinically distinct behavior has been noted in tumors with and without alterations in the epidermal growth factor receptor (EGFR) pathway. Src family kinases are also altered in relation to EGFR signaling and represent a therapeutic target in oral tongue cancer. ¹⁸ Translational studies have also implicated dysregulation of immune checkpoints in head and neck cancer. PD-1 blockade increases T-cell proliferation and IFN-y production in tumors, and when combined with a toll-like receptor agonist enhanced granulocyte-macrophage vaccine (TEGVAX) causes regression of HPV-negative head and neck cancer growth in vivo. ¹⁹ Though preliminary, these studies have the potential to provide the basis for molecularly tailored therapy for head and neck cancer patients.

15.2 Clinical Presentation

Cancer of the oral tongue most commonly presents as pain followed by an ulcer or swelling on the tongue. Early, small lesions may go undetected until a critical size is reached that prompts symptoms and medical visits. As tumor size evolves and adjacent structures become involved, other complaints such as bleeding, tongue immobility, dysarthria, and dysphagia may develop. Regional disease may present as a neck mass. Certain clinical clues such as complaints of ear pain, dysphagia, voice changes, and trismus may suggest involvement of other subsites, such as the oropharynx.

An examination of the oral tongue should include evaluation of sensation and mobility. Any deficits should prompt concern for perineural invasion of the lingual and hypoglossal nerves. The fixed tongue should also trigger investigation into mandibular, floor of the mouth, or retromolar trigone involvement, significantly altering both staging and treatment. Furthermore, trismus may be a harbinger of pterygoid or masseter involvement and should be investigated carefully. ²⁰

Premalignant lesions such as leukoplakia, erythroplakia, submucous fibrosis, lichen planus, and others may be noted by the patient or other practitioners during routine examinations. These have various underlying etiologies such as ill-fitting dentures, smoking, as well as inherited conditions such as xeroderma pigmentosa, and should be treated as premalignant lesions until proven otherwise. Rate of conversion to malignancy is variable and may be compounded by exposures, heritable disorders, and comorbidities. ²¹ Typically, erythroplakia has a higher transformation rate to malignancy (reported from 15 to 50%) when compared to leukoplakia, the most common lesion of this class. ^{21 , 22} A full discussion precludes the scope of this chapter; however, these lesions remain a critical biologic entity to recognize. Various treatment strategies have been employed and include surgical excision as well as topical and systemic agents. Complete surgical excision is often superior to biopsy so as to avoid the possibility of “geographic miss.” Should a lesion prove to contain high-grade dysplasia or microinvasive carcinoma, excision remains the gold standard. Adjuncts such as laser treatment of wide fields have also been performed invariably with success. ²¹ Recurrence rates have also been noted to occur between 10 and 30% following appropriate treatments and emphasize ongoing multidisciplinary surveillance, including the patient self-surveillance. ²³ Photodynamic therapy, cryotherapy, retinoic acid, COX-2 (cyclooxygenase-2) inhibitors, and even metformin have been trialed with varying success in treatment and suppression of premalignant lesions and may be considered as treatment adjuncts.

In patients with previously treated head and neck cancer, diagnosis of recurrent or new lesions of the oral tongue may prove difficult. Clinically, during surveillance, worsening or new symptoms similar to their original presentation should tip the clinician to a through physical examination for new lesions. Pain is often the most sensitive indicator of recurrence and patients should be queried as to whether they are taking pain medications as this may mask new-onset pain. Some symptoms may be difficult to separate from treatment sequelae—such as numbness from prior surgery, or mucositis and pain from radiation. Patients previously treated for oral tongue cancer can most commonly fail locoregionally and these sites should be examined closely. ²⁴

15.3 Diagnosis and Evaluation

Initial diagnosis for oral tongue cancers centers on tissue biopsy, which can often be performed with local anesthesia in the clinic. Histologic characteristics such as degree of keratinization, nuclear polymorphism, and perineural and perivascular invasion combined with tumor biomarkers such as p53, Ki-67, cyclin-D1, WGFR, STAT3, VEGF, and HOXB7 aid in diagnosis and treatment. ²⁵

While classic squamous cell carcinoma can be readily identified on histopathology, certain controversies exist. In patients with prior surgical or radiation treatment of an oral tongue lesion, there may be extensive atypia, metaplasia, and inflammation, resulting in difficulty in diagnosing persistent, recurrent, or new disease. Another diagnostic conundrum may be microinvasive disease. Should adequate biopsies be performed, microinvasive disease may alter management in terms of extent of resection and need for treatment of the neck, as depth of invasion has been correlated with regional disease. ²⁶ Necrotizing sialometaplasia can be mistaken for squamous cell carcinoma and requires close consultation with head and neck pathologists to differentiate this benign condition from malignancy. ²⁷

For surgical planning, tumor mapping may be performed by either in-office examination or operative endoscopy. For small, accessible lesions, office examination may be sufficient to plan for surgical resection, with careful attention being paid to the floor of the mouth, base of the tongue, retromolar trigone, oropharynx, and mandibular periosteum. Palpation should be performed to identify deeply invasive lesions or those that cross the midline as resection may not only impart substantial functional deficits but also require increasingly advanced reconstructive efforts.

For tumors located posteriorly or for patients who do not tolerate a full bimanual office examination, operative endoscopy may be necessary. Historically, panendoscopy was performed to evaluate for synchronous lesions. While this has been demonstrated to be safe, ^{28 , 29} recent reviews have found a relatively low rate of synchronous primary lesions in patients with early-stage oral tongue cancers. ³⁰ With an increased proportion of NSND patients, the role of panendoscopy and synchronous primary rate may be evolving. In practice, endoscopy can be combined at the time of surgical intervention.

Evaluation of the neck should also be completed during diagnostic workup. In-office examination of the neck may reveal palpable adenopathy, particularly in levels 1, 2, and 3, which represent the first echelon of lymphatic drainage from the oral tongue and thus are the most frequently involved regional metastatic site. ³¹ Imaging is a useful adjunct for staging and surgical planning for both primary and regional sites. CT and MRI have distinct advantages and may be complementary. While CT scan offers excellent bony definition, MRI offers improved resolution of soft tissue and may be used to help assess for perineural invasion, changes in bone marrow signal suggestive of invasion, and intracranial involvement. ³² Ultimately, individual patient disease factors may dictate which imaging modality is best suited to answer a given clinical question. Both CT and MRI may also be used to evaluate the neck for regional metastasis, although ultrasound has been demonstrated to have improved accuracy in assessment of lymph nodes with greater detail of architecture. ³³ Furthermore, image-guided biopsy of abnormal lymph nodes may be easily performed at the time of ultrasound examination. ³³

PET/CT can be used for metastatic staging of oral tongue cancer patients; however, studies specifically evaluating the utility of PET/CT in early oral tongue cancer are limited. Studies examining both early- and late-stage head and neck cancer patients did show downstream alterations in treatment in a significant proportion of both patient groups, though this study included subsites with much higher risk of metastatic disease. ³⁴ While routine PET/CT may be of limited utility for early-stage primary disease, the utility of PET/CT in far advanced, recurrent, or persistent disease may be extremely helpful. ³⁵ Current National Comprehensive Cancer Network (NCCN) guidelines recommend PET scan for patients with stage III/IV disease, as prior studies demonstrate clear upstaging and alteration of this group of patients. ³⁵ PET/CT should also be considered in patients undergoing primary nonsurgical treatment or in patients who are to receive adjuvant therapy to have an index marker of uptake before and after treatment. ³⁴