Primary Squamous Cell Carcinoma of the Parotid Gland: Study and Review of the Literature

Abstract

Background/Aim

Primary squamous cell carcinoma of the parotid gland (pPSCC) is a rare tumor, accounting for less than 3% of all parotid carcinomas. There is no general consensus on the management of this tumor, since clinical experience for pPSCC is minimal. Moreover, pPSCC is often misdiagnosed for metastatic cutaneous carcinoma. In our study, we focused on evaluating its biological and clinical characteristics, treatment results and prognosis. We proposed an update on diagnostic and therapeutic management of pPSCC.

Patients and Methods

The retrospective cohort included 18 patients diagnosed and treated for pPSCC in three tertiary head and neck centers between 2008 and 2022. We retrospectively evaluated their prognosis and established a therapeutic recommendation after analyzing our own and previously published data.

Results

Fourteen of 18 tumors were diagnosed in stage IV. Five-year overall survival was 36 months. Six patients received palliative therapy. Twelve patients underwent parotidectomy, neck dissection, and adjuvant radiotherapy. Remission was achieved in 8 patients (follow-up interval 3-56 months). One patient died with recurrent disease. The others are alive and in complete remission.

Conclusion

The definitive diagnosis of pPSCC must meet the histological and clinical criteria. First of all, the metastatic origin of the tumor must be excluded. Five-year survival of this very aggressive tumor does not exceed 50%. Without surgery, the prognosis is poor. The best results, irrespective of tumor stage, are achieved with surgery. Therefore, a total parotidectomy, neck dissection (therapeutic or elective) and adjuvant radiotherapy are indicated for all resectable tumors.

Primary squamous cell carcinoma of the salivary gland represents less than 3% of the 28 histopathological types of salivary carcinomas listed in the current WHO classification (1-3). Up to 90% of these tumors affect the parotid gland (4). They may also arise from the submandibular gland. The reported incidence in the minor salivary glands is extremely low and could be underdiagnosed (5,6). The criteria for the salivary origin of squamous cell carcinoma (SCC) include tumor infiltration of the affected gland parenchyma with no evidence of cancerous or dysplastic changes in the overlying mucosa. This could be very problematic in minor salivary gland tumors due to their size and proximity of the mucosa (7,8).

The parotid gland SCC (PSCC) usually confirmed in intra- or paraparotid lymph nodes, is in 90% of cases of metastatic origin. The primary carcinoma is located mainly on the skin of the tributary areas of the face and frontal scalp (9). Infrequently, the primary tumor is located on the mucosa of the upper aerodigestive tract, commonly nasopharynx (10,11) and oropharynx (9,12). Exceptionally, kidney, lung, gastrointestinal, prostate, and breast tumors can also metastasize to the parotid lymph nodes (13,14). Primary PSCC (pPSCC) represents only about 10% of PSCC, and due to its low incidence, clinical experience with pPSCC is limited.

Our study aimed to analyze the clinical characteristics, diagnostic protocol, treatment results and prognosis of pPSCC.

Patients and Methods

Study design. The retrospective study included patients who met the clinical and histopathological criteria for pPSCC summarized by Edafe et al.: 1) clinically exclusion of any current or past duplex malignancy, 2) histologically a tumor infiltrating the parenchyma of the gland, without contact with skin or residual lymph node structure and 3) immunohistochemically exclusion of mucin in the tumor tissue (15) (Table I). Patients were diagnosed, treated, and followed up in three tertiary otorhinolaryngology and head and neck centers between 2008 and 2022.

Table I. Diagnostic clinical and pathological criteria for diagnosis of primary parotid squamous cell carcinoma suggested by Edafe et al. (15).

Set of patients. The set of 18 patients included 13 men and 5 women aged 33 to 90 (median 73) years. In all cases, cutaneous, or any other, recent, or current malignancy was excluded by clinical otolaryngological and dermatological examination, upper aerodigestive tract endoscopy, computer tomography (CT), and in 9 cases by whole-body combined positron emission tomography (PET)/CT scan. No patient had any history of other malignancy at the time of PSCC diagnosis or during the follow-up. Histopathological diagnosis was confirmed in all 12 surgically treated patients by examination of the surgical specimen, in 6 non-surgically treated patients by open (2 patients) or core cut biopsy (4 patients). The above-mentioned histological criteria were assessed in all patients. Staging of tumors was performed in all patients before starting treatment. The duration of the symptoms before establishing the diagnosis ranged from 6-310 (median 11.5) weeks. (Table II).

Table II. Clinicopathological characteristics of patients with primary parotid squamous cell carcinoma.

One patient was diagnosed with pPSCC in clinical stage II, 3 in stage III, and 14 in stage IV. Tumors were localized exclusively in the superficial lobes in 7 patients. In one case, it was limited to the deep lobe. Out of 10 tumors affecting both lobes, 9 showed extraglandular growth. Preoperative facial nerve palsy was present in 7 cases (House-Brackmann scale (HB) II in 3 cases, HB III, and HB IV each in 2 patients). Regarding histological grading, 6 carcinomas were moderately differentiated (G2), and 12 were poorly differentiated (G3).

Treatment. Six patients with unresectable carcinomas were treated with radiotherapy (RT). Out of remaining 12, one underwent partial superficial conservative parotidectomy (PE), and 6 had total conservative PE. Radical total PE with resection of the facial nerve was performed in 5 patients and extended PE in 3 cases where the tumors infiltrated surrounding structures. Comprehensive neck dissection (ND) was performed in 7 cN+ patients and elective ND in 5 cN0 patients. Radiation therapy or concurrent chemoradiotherapy (1 patient) was applied postoperatively in all patients. The clinicopathological characteristics, therapy and follow-up of the entire patient group are shown in Table III.

Table III. Treatment of patients with primary parotid squamous cell carcinoma.

Statistics. IBM SPSS Statistics Version 23 statistical software (IBM Corp., Armonk, NY, USA) was used to analyze the data. Disease free interval (DFI), overall survival (OS) and comparison of OS were evaluated using Kaplan-Meier analysis and Log Rank (Mantel-Cox) test. All tests were considered significant at the 0.05 level.

Results

Clinical findings. Microscopic extraglandular spread of the tumor was detected in 6 of 12 surgically treated patients. Positive resection margins (always microscopically positive, i.e., R1) were demonstrated in only 3 cases of extended total PE.

Seven pPSCC were classified as cN+, and suspicious nodes were only found in levels II and III based on imaging methods. The presence of neck metastases was histologically confirmed in 5 (70%) cases. The number of positive nodes, located in levels II-III in 2 cases and in levels I-V in 3 patients, ranged from 3 to 26 (median 19) nodes. Microscopically, an extracapsular spread was excluded in all of them. Elective ND was performed in 5 cN0 patients. Microscopic metastases were detected in two cases (i.e., 40%). In each there was a single metastatic lymph node in level II.

Treatment outcomes. Complete remission was achieved in 8 of the 12 surgically treated patients. Except for one patient who died of relapse after 21 months, all remained disease-free for a follow-up period of 3-56 (median 24) months. One patient died two years after termination of treatment from cardiac disease. Three of the 4 patients in whom remission was not achieved after completing combined treatment died in 2-12 (median 3) months. One was lost to follow-up after six months.

Remission was not achieved in any of the 6 non-surgically treated patients, 4 of whom died in 3-12 (median 11) months. Two are still alive with tumor persistence 3 and 13 months after completion of their treatment (Table IV).

Table IV. List of patients with primary parotid squamous cell carcinoma, tumor TNM stage, treatment modality and results.

G: Tumor grade; TPE: total parotidectomy; ND: neck dissection; RT: radiation therapy; CHT: chemotherapy; adj: adjuvant; pal: palliative; I, II, III, IV, V: levels of the neck nodes; CR: complete remission; D: death; PPE: partial parotidectomy; RTPE: total radical parotidectomy; P: persistence of tumor; R: tumor recurrence.

Prognosis prediction. The prediction of both 3- and 5-year survival in the entire group of patients with pPSCC reached 36%, and the median survival time was 24.3 (95%CI=18.5-42.3) months. Kaplan Meier survival analysis of the whole set is illustrated in Graph 1A (Figure 1).

Figure 1. Kaplan-Meier plot of the overall survival of the entire patient cohort with primary parotid squamous cell carcinoma (A) and comparison between the subsets after different treatment (B).

A prognostic difference was evaluated by comparing groups according to their therapy. A significantly better prognosis was associated with surgical treatment (p=0.003, Log Rank Mantel-Cox). Median overall survival was 28.3 (95%CI=25.9-54.4) months for the subset of patients treated with surgery and adjuvant (chemo)radiotherapy but one patient underwent surgery only. Kaplan Meier survival analysis of non-surgically treated patients revealed median overall survival of 11.1 (95%CI=6.2-12.3) months (Figure 1).

Discussion

Both metastatic and primary PSCC grow aggressively. A published independent study included less than 30 patients. In our PubMed database search over the last 30 years, we found only 10 articles, which included a total of 153 cases of pPSCC (1,3,5,16-21). Of all the included studies, only 3 reported on patient prognosis (1,3,21). The remainder did not analyze treatment results at all (5,9,10,16-20,22) or did not distinguish between metastatic SCC and other histopathological parotid carcinomas (9,20,22-28). One study evaluated the prognosis of SCC in a group of patients that combined parotid and submandibular glands (29). The incidence of pPSCC represents 1% to 3% of all parotid malignancies, e.g., 10/1,000,000 newly diagnosed cases a year. In contrast to increasing incidence of metastatic forms (annual increment of 2.5%), a stagnant trend is reported for pPSCC (22,30). Both variants of PSCC are identical in histomorphological and immunohistochemical characteristics (3). Which is why, some authors even deny the existence of pPSCC (10,31). Its diagnosis can be established by exclusion of a previously, recently or subsequently manifesting primary tumor by clinical examination and appropriate imaging methods only (3,15) (Table I).

Primary PSCC affects men twice as often as women, with the highest incidence in the seventh decade of life. It is rare under the age of 20 (5,21,29), which corresponds to the epidemiological data of our group, in which men predominated by two-thirds and the average age was 73 years. Similarly, to other high risk salivary carcinomas, pPSCC clinically manifests by rapidly growing parotid gland swelling (5). In our group, the duration of symptoms generally did not exceed several months, with a median of 12 weeks. Histologically, an intermediate degree of differentiation usually prevails. Poorly differentiated SCC is generally less common (21). However, in our patients, the low histological grade was confirmed in two-thirds of the tumors. In approximately one-third of patients, the cancer caused facial nerve palsy (21), which was present in 7 (39%) patients in our group. The aggressive behavior of pPSCC was also confirmed by the fact that it was detected in almost 80% of cases in advanced stages and 75% of patients present with clinical metastases in the neck nodes (21). It is consistent with our experience as 14 (78%) of the 18 patients were diagnosed in stage IV and 12 (67%) with regional metastases (cN+).

The prognosis of pPSCC is generally poor. A 5-year OS was calculated at 36%. The results correspond to those of other authors, who reported OS and disease-free interval values in the range of 33%-65% (1,20,21,23,32-34), 35%-50% respectively (1,20,35). According to several studies, the prognosis between metastatic and primary PSCC does not differ significantly (1,10,21). The prognosis of pPSCC, like in other salivary carcinomas, is mainly influenced by its clinical stage. In Xiao’s study which included 29 cases, the median survival for initial (I+II) and advanced (III+IV) stage tumors was 60, respectively 13 months (21). In our set, with the absolute predominance of stage IV, its value of median survival reached 24 months. Radical surgery plays a decisive role in the therapy of pPSCC. A complete tumor resection must be performed, if possible, as positive resection margins significantly worsen the prognosis (24-26,36). Our study confirms that complete resection improves prognosis. None of the three patients with R1 resection achieved tumor remission despite subsequent RT. The prognostic difference between patients treated with surgery and those treated with radiotherapy only is significant. Adjuvant RT was applied in all patients after PE, with one exception. Therefore, total PE, extended PE in selected cases resp., is fully indicated; it was performed in 11 patients. In two of them, the T2 stage tumor was located only in the superficial lobe, without microscopic spread to the deep lobe or intraparenchymal lymphatic metastases. The exact incidence of these adverse events is unknown in pPSCC, but it can be assumed to be like other high-risk salivary carcinomas. Therefore, we take a cautious approach to superficial PE. Due to the significant risk of local recurrence, which occurs in approximately 1/3 of patients (21) with pPSCC, postoperative RT is recommended for all stages (1,10,21). Its effectiveness has been significantly confirmed by several retrospective studies, which, however, also included metastatic forms (16,17,20,25,31,36-38). Some articles exclusively evaluating pPSCC did not unequivocally confirm the benefit of RT, which their authors explain by selection bias and a small number of probands (1,10). In 4 of 11 total PE, the entire extratemporal ramification of the facial nerve had to be sacrificed due to tumor infiltration. In two patients only selected branches were resected, the buccal branch in one case and the zygomatic branch in another, while the unaffected branches were preserved. The indication for resection of an intact facial nerve ramification is controversial, as there is no clear evidence that its resection improves prognosis, even in metastatic PSCC (38-41). Such a radicality is fully justified in patients with intraoperatively confirmed nerve infiltration, as it represents one of the general indicators of a poor prognosis (38).

The reported incidence of occult neck metastases of pPSCC in locally advanced (T3/4) pPSCC is 45%-60% (1,10,21). In T1/2 tumors, it is 28%, according to the only study performed to date (1). Therefore, elective ND is indicated even in these early stages. A positive effect of elective ND can be expected, as it was reported in a study by Phisterer et al., which included primary and metastatic PSCC of all clinical stages. ND significantly improved the prognosis, and its omission was associated with a threefold higher risk of death regardless of adjuvant RT (9). The indication of elective ND is also supported by our results, wherein in 2 of 5 cN0 cases (40%), we detected occult metastases, always limited to neck level II. However, according to the recommendations in the literature, resection should also include region III, as recommended in other parotid carcinomas (23,42) (Figure 2).

Figure 2. Primary parotid squamous cell carcinoma - proposed diagnostic and therapeutic management. ENT: Otorhinolaryngology; PET/CT: positron emission tomography/computed tomography; SCC: squamous cell carcinoma; cN0: clinically excluded neck metastases; cN+: clinically present neck metastases; TPE: total parotidectomy; ND - neck dissection, RT: radiation therapy; CHRT: chemoradiotherapy; II, III, V: levels of the neck nodes.

A low number of cases in our cohort limits statistical analysis. However, our set of patients is quite comparable to those of previously published works (1,3,5,16-21), as the general incidence of this type of tumor is very low. As the study retrospective in its design, it was not possible to find out why PET/CT was not performed in some patients. Anyway, any other primary tumor in the area of ​​the upper aerodigestive tract and relevant regions of the skin was excluded by clinical examination at the time of diagnosis and during regular follow-up in all patients. In general, we recommend performing PET/CT in these patients to rule out a possible distant metastasis.

Conclusion

Primary PSCC is a sporadic disease with a stationary incidence over the past decades. Unlike parotid metastases of cutaneous SCC, which are more frequent, and their incidence has been permanently increasing. Diagnosis of pPSCC is determined by thorough exclusion of metastatic origin from any other primary malignancy. Due to its biologically aggressive behavior and poor prognosis, its 5-year survival generally does not exceed 50%. Primary PSCC requires multimodal treatment, including total PE with ND (therapeutic or elective) and adjuvant radiotherapy for all stages of the disease. Radiation monotherapy has only palliative value, resulting in a poor prognosis.

Funding

The study was supported by the Ministry of Health, Czech Republic – conceptual development of research organization (FNOL, 00098892) and internal research IGA LF 2023-08.

Conflicts of Interest

The Authors declare no conflicts of interest regarding the publication of this article.

Authors’ Contributions

Authors contributed to the following roles: Zuzana Horáková: conceptualization, data curation, project administration, writing of original draft; Ivo Stárek: conceptualization, review and editing; David Kalfert: data curation, project administration, review and editing; Lucie Dostalova: data curation, project administration; Ivan Kalivoda: data curation, project administration; Richard Salzman: supervision, review and editing.

Acknowledgements

We would like to thank Katherine Vomacka for English language revision.