Does Nodal Metastasis and Perineural Invasion Affect Local Control in Hyalinizing Clear Cell Carcinoma of the Oral Cavity? A Case Report with Long Term Follow-Up

Abstract

Hyalinizing clear cell carcinoma (HCCC) is a rare, low-grade neoplasm accounting for approximately 1% of salivary gland neoplasms. Histologically, it is characterized by a monomorphous population of clear cells arranged in sheets, nests, or cords, lacking ductal structures. Until recently, clear cell carcinoma of the oral cavity (OC) represented a diagnosis of exclusion when other head and neck pathologic entities such as epithelial-myoepithelial carcinoma or mucoepidermoid carcinoma could be ruled out, making definitive diagnosis by light microscopy and immunoprofiling a challenge. As a result, initial biopsies are often misclassified, and could result in under- or overtreatment. More recently, the presence of the EWSR1-ATF1 gene fusion has been adopted to definitively diagnose HCCC. Typically, HCCC demonstrates clinical indolence and responds well to curative surgical excision alone for localized disease, with adjuvant radiotherapy (RT) reserved for high risk features including perineural invasion, lymphovascular invasion, and regional cervical metastasis. The literature, however, lacks consensus regarding the role of adjuvant radiotherapy. In this article, we report a case of HCCC in a rare site involving the ventral tongue, with high risk features of perineural invasion and cervical nodal metastasis. The patient underwent surgical excision alone; declining adjuvant radiotherapy despite the high risk features, and was alive with no evidence of disease at the 42-month mark. Furthermore, we provide an update on the current prognostic indicators for HCCC, and emphasize the need for chromosomal analysis to achieve a definitive diagnosis.

Introduction

Hyalinizing clear cell carcinoma (HCCC), formerly known as clear cell carcinoma (CCC), not otherwise specified (NOS), is a rare salivary gland neoplasm most often arising from minor salivary glands in the oral cavity (OC), followed by the oropharynx, and major salivary glands [1–3]. Occurring most commonly in the sixth to eighth decades of life, these indolent neoplasms may not become symptomatic until months to years of painless, non-ulcerated growth [3]. Previously considered a diagnosis of exclusion, the use of immunohistochemistry and fluorescent in situ hybridization (FISH) for the Ewing sarcoma breakpoint region-1 (EWSR1) can now distinguish HCCC from other more aggressive, clear cell-containing malignancies including mucoepidermoid carcinoma, acinic cell carcinoma, epithelial-myoepithelial carcinoma and renal cell carcinoma [4–6]. For the clinician, treatment decisions are obfuscated when the histopathologic diagnosis is equivocal, and, in the case of misdiagnosis as high-grade clear cell variants such as mucoepidermoid carcinoma, leads to unnecessary surgery and/or radiotherapy (RT), and psychological distress for the patient. While the general treatment of HCCC is wide local excision, the addition of a cervical lymph node dissection may be useful if a nodal metastasis is suspected. The role of adjuvant radiotherapy also remains unclear [7]. Because there is currently no histological grading system or specific treatment guidelines for HCCC due to the short follow-up period in the literature, median of 24 months, and wide time range of recurrence (4–288 months), patients diagnosed with HCCC require ongoing long-term surveillance [7, 8].

Based on review of the available English language literature, our case represents one of two of HCCC in the oral tongue and not only presents with nodal metastasis and perineural invasion (PNI), but also responded well to surgical excision only without radiotherapy, with follow-up of 42 months exceeding that seen in the published studies of larger series. This case is the second report in the literature of oral cavity HCCC with high risk features, cured with surgery alone, with adequate follow-up.

Case Report

A 70-year-old white male presented with a slowly-growing, 2 cm painless exophytic ulcerative lesion on the midline ventral tongue which had been present for up to 6 months prior to being noted on a routine dental examination. An outside biopsy demonstrated moderately differentiated invasive squamous cell carcinoma (SCCa) with clear cell features. On oncologic workup per NCCN guidelines, there was no evidence of regional or distant metastasis noted on clinical examination and computed tomography (CT) imaging. The clinical appearance was that of an exophytic and ulcerative, well-circumscribed lesion located at the midline of the ventral tongue. On palpation, the senior author suspected tumor depth of invasion of at least 3–4 mm, and offered the patient elective neck dissections. This is the standard for lesions with up to 30% risk of occult disease, and in this case were performed bilaterally given the midline position of the tumor and chance for bilateral metastasis seen in squamous cell carcinoma. Selective neck dissection with removal of levels I–IV bilaterally and preservation of non-lymphatic structures (sternocleidomastoid muscle, internal jugular vein, spinal accessory nerve) was completed [9–11]. Intraoperative frozen sections demonstrated a low-grade carcinoma with a differential diagnosis including low-grade mucoepidermoid carcinoma, and polymorphous low-grade adenocarcinoma,however squamous cell carcinoma was not ruled out. A midline partial glossectomy was performed with 1.0–1.5 cm margins (Fig. 1a) and primary closure achieved after negative margins were confirmed (Fig. 1b). The neck dissection was completed in standard fashion. The patient had an uneventful postoperative course. At 2-month follow-up both the primary site and neck had healed well without any deficits. The tongue was freely mobile with good anteroposterior projection, maintaining good speech and swallowing function (Fig. 1c).

Fig. 1.

Tumor on ventral surface of the gone (a) treated with wide local excision and primary closure (b). Patient regained excellent speech and swallowing function at 2-months postoperatively (c)

Final pathology results by hematoxylin and eosin staining (H&E) demonstrated hyalinizing clear cell carcinoma of minor salivary gland origin (Fig. 2a) with perineural invasion (Fig. 2b) and metastasis to 1 of 26 cervical lymph nodes. FISH analysis for EWSR1 revealed positive chromosomal anomaly at 22q12 in 91% of cells (Fig. 3). The final pathological staging was pT1N1 hyalinizing clear cell carcinoma.

Fig. 2.

a The tumor is present in cords and nests of epithelial and clear cells with interspersed hyalinized stroma. Focal invasion of the overlying epithelium is seen (yellow arrow). b Perineural invasion of a small nerve (yellow arrow)

Fig. 3.

This image is of the EWSR1 break apart probe showing positive gene rearrangement in the tumors cells using FISH

After discussion at the institutional Multispecialty Head and Neck Tumor Board, 60 cGy of postoperative radiation therapy over 30 fractions was recommended to the primary site and bilateral necks due to perineural invasion and a positive nodal status. After reviewing the final histopathology, and expected survival benefit for adjuvant radiotherapy, the patient declined further treatment. The patient continues to be followed intermittently due to social issues, and is alive without clinical evidence of disease at the 42-month mark.

Discussion

Although rare, cases of hyalinizing clear cell carcinoma continue to be reported in the literature due to limited evidence on overall disease-specific survival associated with various treatment modalities. Specifically, although there remains limited evidence on the curative effects of radiotherapy and or chemotherapy as mono- or adjuvant therapy, the recommendation of adjuvant radiotherapy can be reserved for cases with an increased likelihood of disease recurrence, i.e., those with high risk features such as PNI, LVI, or nodal metastasis [7, 8]. While the majority of HCCC have been reported to occur in the oral cavity followed by the oropharynx, there continues to be an increasing number of case reports of HCCC described in the lung, trachea, paranasal sinus, and thymus [12–18]. Within the oral cavity, the palate is the most common site, followed by the maxilla, then floor of mouth [1]. Because oral tongue HCCC is commonly reported together with tongue base (BOT) tumors, the true incidence of oral tongue HCCC is unknown [7, 8]. Abergotti et al. reported 8 cases of HCCC involving the oral tongue [1].

The uniqueness of this case report is the primary location along with its pathological characteristics and response to treatment. Our case presented as a lesion on the ventral surface of the oral tongue. When reviewing the literature, most articles that categorize the primary location of the tongue do not distinguish the BOT (oropharynx) from the oral tongue (oral cavity). This is a key distinction because oral tongue HCCC, although lower in incidence, has been reported with a higher rate of LVI, PNI, and nodal metastasis (Table 1). Therefore, the clinicopathologic behavior of HCCC may vary based on location and respond differently to treatment modalities, as seen in HPV-driven SCCa of the oropharynx. HCCC, an indolent tumor, also lacks radiosensitivity and would be expected to respond poorly to single-modality radiation therapy. Of the more recent review articles, only one distinguished between BOT and oral tongue HCCC [1]. This issue results in a fundamental reporting error, making either major salivary gland the second most common primary location of HCCC reported in the literature [1, 7, 8, 19, 20], however when the BOT and oral tongue are reported as separate entities, oropharynx is the second most common location following the oral cavity [1].

Table 1.

Clinical behavior of published HCCC cases by location

Location	Freq (%)	Publications	Cases (n)	Size range (cm)	PNI (%)	LVI (%)	Bone invasion (%)	Nodal metastasis (%)	Distant metastasis (%)
BOT	17.9	[1, 3–5, 8, 20, 21, 28, 30, 37–48, 50–53]	40	0.8–6	18	5	0	18	0
Bronchus	3.6	[12, 14, 16, 54, 55]	8	1.6–3.5	13	0	0	13	0
Buccal/cheek	4.9	[1, 3, 4, 16, 20, 32, 41, 56–58]	11	1.5–4.9	0	9	0	0	0
FOM	6.7	[4, 8, 20, 28, 35, 41, 51, 53, 58–63]	15	0.5–5	33	13	7	13	0
Gingiva	0.9	[3, 48]	2	0.4	0	0	50	0	0
Hypopharynx	0.9	[8, 64]	2	2–4.8	0	0	0	50	0
Larynx	1.3	[4, 28, 65]	3	1.3–2	0	0	0	0	0
Lip	1.3	[4, 8]	3	1.5	0	67	0	0	0
Mandible	3.1	[4, 8, 20, 28, 45, 66]	7	0.6–4.8	29	0	43	14	0
Maxillaa	5.8	[1, 4, 33, 34, 58, 66–70]	13	0.5–6.5	0	7	38	0	0
Nasal cavity	0.9	[1, 71]	2	3	0	0	0	0	0
Nasopharynx	5.5	[1, 3–5, 30, 72–77]	12	1.7–4.9	25	0	25	0	0
Neck	0.4	[41]	1	–	–	–	–	100	-
Oral	0.4	[51]	1	1.2	–	–	–	-	0
Orbit/lacrimal gland	0.4	[20]	1	4.5	0	0	0	0	0
Oropharynx	2.2	[4, 16, 78, 79]	5	2–3.5	0	0	0	0	0
Palateb	26	[1, 3–5, 7, 8, 20, 21, 28–31, 47, 51, 53, 56–58, 80–92]	58	0.4–5	19	3	10	7	3
Paranasal sinus	1.3	[17, 93]	3	6	33	33	100	0	0
Parotid	4.9	[4, 5, 28, 36, 45, 51, 94, 95]	11	1–10	27	9	0	0	0
Sinonasal tract	0.4	[96]	1	2	–	–	–	0	-
Sublingual	0.4	[7]	1	2.5	0	0	0	0	0
Submandibular	0.4	[21]	1	4	100	0	–	0	-
Thymus	0.9	[18]	2	–	–	–	–	-	-
Tongue	5.4	[4, 20, 28, 49, 51, 82, 97–101]	12	1.5–5	25	17	0	25	0
Tonsil	2.2	[1, 51, 103–105]	5	2.3–3	0	0	0	20	0
Trachea	1.3	[3, 13, 106]	3	1.3–2.5	0	0	0	-	0

BOT base of tongue, FOM floor of mouth, PNI perineural invasion, LVI lymphovascular invasion

^aMaxilla: includes cases in maxillary sinus

^bPalate: includes hard and soft palate

With EWSR1-ATF1 gene fusion presenting in the majority of HCCC cases, there is now an increased role for diagnostic molecular markers to differentiate malignant salivary gland neoplasms with clear cell features [4–6, 21]. Additional gene fusions such as MYB-NFIB and CRTC1-MAML2 are specifically associated with adenoid cystic carcinoma and mucoepidermoid carcinoma, respectively [22, 23]. Two additional mutations, CREM and CREB1, genes belonging to the same family of transcription factors as ATF1, were also found translocated with EWSR1 in cases of HCCC, clear cell carcinoma of the soft tissue, and a distinct group of myxoid mesenchymal neoplasms [24–26]. As the list of specific and recurrent gene translocations associated with solid neoplasms continue to evolve, the once homogenous precursor neoplasms will be differentiated by unique and specific genetic alteration(s). By understanding the mechanisms of these genetic translocations in tumorigenesis, new approaches for early diagnosis and targeted therapy can be developed. Overall, the diagnosis of HCCC should take into consideration not only the molecular characteristics of the disease, but also the immunophenotypic, histomorphological, and clinicopathologic features. This is of note, since gene mutations in head and neck cancer overall are rare compared to other body sites [27].

When evaluating nodal metastasis and recurrent disease, a literature review by Daniele et al. showed only 3 of 152 cases of HCCC involving the tongue (excluding BOT) that demonstrated metastatic or recurrent disease. Only one of those 3 cases presented with positive nodal metastasis, did not receive adjuvant radiotherapy, and was alive without disease 11 years postoperatively [28]. We also performed an updated PubMed search using the same keywords as Daniele et al. from the years 2015 to 2020. We specifically screened the new cases involving the oral cavity and salivary glands. Twenty-nine cases were reported with 25 cases involving the oral cavity and 4 involving the major salivary gland, but no new reports associated with the oral tongue [1, 5, 7, 8, 21, 29–36]. Therefore, our case is the second report of oral tongue HCCC in the literature, with the high risk features of perineural invasion and regional metastasis. Interestingly, he was treated with surgery alone and remains without evidence of disease at 42 months postoperatively. A detailed list of reported HCCC cases by location is illustrated in Tables 1 and 2.

Table 2.

Treatment and outcomes of published HCCC cases by location

Location	Exc (%)	Exc and LND (%)	Exc, LND, and RT (%)	Exc and RT (%)	RT only (%)	Exc and ChemoRT (%)	Othera (%)	Follow-up range (months)	NED (%)	Recurrence (%)
BOT	58	10	8	5	3	3	3	3–124	65	18
Bronchus	50	50	0	0	0	0	0	10–181	100	0
Buccal/cheek	73	9	0	18	0	0	0	5–132	64	27
FOM	67	13	20	0	0	0	0	2.5–108	73	7
Gingiva	50	50	0	0	0	0	0	12–31	100	0
Hypopharynx	0	50	0	0	0	0	0	62	100	0
Larynx	67	0	0	33	0	0	0	24–55	100	0
Lip	100	0	0	0	0	0	0	2–73	100	0
Mandible	71	29	0	0	0	0	0	3–78	71	29
Maxilla	46	7	7	7	0	0	0	3–36	38	23
Nasal cavity	50	0	0	50	0	0	0	7	0	50
Nasopharynx	50	0	8	25	0	0	1	2.5–144	58	8
Neck	–	100	0	0	0	0	0	–	–	–
Oral	100	0	0	0	0	0	0	–	–	0
Orbit/lacrimal gland	100	0	0	0	0	0	0	–	100	0
Oropharynx	80	0	0	0	0	20	0	12–81	60	20
Palate	67	2	7	10	0	0	3	6–348	60	19
Paranasal sinus	34	0	0	67	0	0	0	4–9	67	0
Parotid	100	0	0	0	0	0	0	1–195	55	18
Sinonasal tract	–	–	–	–	–	–	–	48	100	0
Sublingual	100	–	–	–	–	–	–	12	100	–
Submandibular	–	–	–	–	–	–	–	–	100	0
Thymus	–	–	–	–	–	–	–	–	–	–
Tongue	67	8	0	17	0	0	8	8–132	42	8
Tonsil	80	0	0	20	0	0	0	4–72	40	20
Trachea	0	33	33	33	0	0	33	12–52	67	0

BOT base of tongue, FOM floor of mouth, Exc excision, LND lymph node dissection, RT radiotherapy, ChemoRT chemoradiotherapy, NED no evidence of disease

^aOther; includes no treatment or debulking for unresectable tumors

Conclusion

Hyalinizing clear cell carcinoma (HCCC) of the oral cavity remains a relatively uncommon malignant salivary gland neoplasm, but it continues to be frequently presented in the literature due to its unpredictable treatment response and challenging pathological diagnosis. Recent changes in molecular diagnosis have generated a renewed interest in HCCC, as well as clinicians’ demands for updated treatment guidelines. HCCC more commonly presents in white females during the sixth decade of life and is discovered either incidentally or during a routine examination as a painless mass. The differential diagnosis for HCCC varies widely due to the microscopically varied phenotype of clear cell salivary gland tumors, however, the use of FISH analysis for EWSR1 gene rearrangement is shown to be highly specific. As more cases of accurately-diagnosed HCCC are reported with long-term outcomes, clinicians can select treatment with optimal disease-specific survival, individualizing therapy while minimizing morbidity. The current case is one of two in the literature where high-risk features, typically warranting radiotherapy, did not impact survival following single modality surgical treatment. Additional publications are needed for this rare entity, now that molecular diagnosis is widely available, to characterize the benefit of postoperative RT versus observation.

Funding

This study did not receive any funding source.

Compliance with Ethical Standards

Conflict of interest

The authors declare that they have no conflicts of interest.

Ethics Approval

This study was performed in accordance with the principles of the Declaration of Helsinki. Approval was deemed exempt from the University of Alabama at Birmingham Institutional Review Board (IRB).

Informed Consent

Informed consent was obtained from all individual participants included in the study. Patients signed informed consent regarding publishing their data and photographs.