Clinically presumed primary malignant melanoma of the parotid gland with extensive cervical lymph node metastasis: a case report

Abstract

Background

Primary malignant melanoma of the parotid gland (PMMPG) is an exceedingly rare neoplasm, accounting for < 0.7% of all parotid malignancies. Differentiating it from metastatic melanoma to the parotid is clinically critical but diagnostically challenging.

Case presentation

A 62-year-old Chinese man presented with a 6-month history of a painless, enlarging left parotid mass. Extensive physical examination and imaging (CT and MRI) of the head, neck, and chest revealed no extraparotid primary lesion. Core needle biopsy and immunohistochemistry (S100+, HMB45+, MelanA+, Ki-67 ~ 50%) confirmed malignant melanoma. He underwent total parotidectomy with comprehensive neck dissection (levels I–V). The marginal mandibular branch was sacrificed because it was encased by tumor; other branches were preserved. Pathology confirmed a 32 × 25 × 23 mm melanoma with invasion into adjacent salivary tissue and metastasis in 14 of 48 cervical nodes; intraparotid nodes were negative. Postoperatively, the patient developed incomplete eyelid closure and lower lip weakness (planned nerve sacrifice plus traction neuropraxia). Adjuvant radiotherapy was advised, but the patient was lost to follow-up.

Conclusions

Clinically presumed primary parotid melanoma should be considered in a parotid mass without an identifiable primary site. Complete surgical resection with appropriate neck dissection remains the mainstay. A high Ki-67 index and extensive nodal metastasis predict poor prognosis, underscoring the need for adjuvant therapy and close surveillance.

Background

Malignant melanoma of the parotid gland is rare, with primary origin accounting for < 0.7% of parotid malignancies [1]. Most melanomas in the parotid are metastatic from cutaneous primaries of the head and neck (scalp, temple, ear) [2–4]. Distinguishing primary from metastatic disease is crucial because prognosis and management differ substantially. Diagnosis is challenging due to non-specific presentation (painless mass), overlapping imaging features with benign tumors, and the necessity to rigorously exclude an occult extraparotid primary. Definitive diagnosis relies on histopathology and immunohistochemistry. Surgery is the main treatment, but adjuvant strategies remain controversial owing to rarity [3]. The prognosis is generally poor because of high rates of local recurrence and nodal/distant spread.

Case presentation

A 62-year-old Chinese man presented with a 6-month history of a gradually enlarging, non-tender mass below the left earlobe. He denied facial weakness, trismus, or dysphagia. Medical history included hypertension (6 years), diabetes mellitus (10 years), and cerebral infarction (5 years), all medically controlled. He was a non-smoker and non-drinker; family history was negative for malignancy.

Physical examination revealed a firm, immobile, non-tender mass in the left parotid region, with normal overlying skin. Facial nerve function was intact. Comprehensive examination of the oral cavity, nasopharynx, oropharynx, larynx, facial skin, scalp, and neck showed no pigmented lesions, suspicious nevi, or mucosal melanotic patches. No cervical lymphadenopathy was palpable.

Ultrasound showed a hypoechoic, lobulated, irregular mass (approx. 32.7 × 26.8 × 27.1 mm) with heterogeneous echogenicity, posterior acoustic enhancement, and punctate/short linear vascular signals on color Doppler, suggesting malignancy but not specific for melanoma.

Contrast-enhanced CT (neck and chest) revealed an oval, ill-defined left parotid mass (35 × 27 mm) with patchy hypodense areas; solid components measured 48 HU (non-contrast), 66 HU (arterial), 76 HU (venous), and 81 HU (delayed), with central non-enhancing necrosis. A “vascular rim sign” was noted. No chest metastases were found.

MRI showed an oval mass (32 × 28 mm) with peripheral slight hypointensity and central hyperintensity on T1-weighted images, and peripheral hyperintensity with marked central hypointensity on T2-weighted images.

Core needle biopsy revealed diffuse proliferation of medium-to-large atypical cells with intracytoplasmic pigment, fine chromatin, and prominent nucleoli. Immunohistochemistry (IHC) on the biopsy: CKpan-, S100+, HMB45+, MelanA+, CK7-, P40-, P53 (wild-type), Ki-67 ~ 50%, P16-, CyclinD1 (partial+), consistent with malignant melanoma (Fig. 1).

Fig. 1.

Imaging and pathological findings of the presumed primary parotid melanoma. A Coronal contrast-enhanced CT (MPR) shows an ill-defined left parotid mass. B Coronal STIR MRI demonstrates the extent of the lesion. C Axial T1-weighted MRI. D Axial T2-weighted MRI at the same level. E Color Doppler ultrasound shows a hypoechoic, vascularized mass. F HMB45 immunohistochemical staining of core biopsy tissue (×100) confirms melanocytic differentiation

Based on the following findings, a multidisciplinary team opted for total parotidectomy with neck dissection: (i) tumor size > 3 cm and imaging features (ill-defined margins, necrosis, heterogeneous enhancement) suggestive of malignancy; (ii) core biopsy confirmed melanoma with high Ki-67 (50%) indicating aggressive biology; (iii) no extraparotid primary was identified after extensive workup, supporting a clinically presumed primary parotid origin; and (iv) the known high risk of regional lymphatic spread in this entity warranted comprehensive nodal clearance. The patient underwent total parotidectomy with comprehensive neck dissection (levels I–V). Intraoperatively, the tumor was black, firm, and located in the lower pole. The facial nerve trunk and branches were identified. The marginal mandibular branch was completely encased by the tumor and was intentionally sacrificed to achieve negative margins; the zygomatic, buccal, and cervical branches were preserved.

Postoperative histopathology of the parotid specimen confirmed melanoma (32 × 25 × 23 mm) with ill-defined borders, invading adjacent salivary tissue; no intravascular tumor thrombi or perineural invasion. Cervical nodes showed metastasis at levels I (1/5), II (5/12), III (5/15), IV (2/7), and V (1/9); intraparotid nodes were negative. IHC for HMB45 and SOX-10 was positive.

At the first follow-up (Day 26), the patient had incomplete left eyelid closure, mouth deviation to the right, and air leakage on cheek puffing, attributed to permanent marginal mandibular branch palsy and transient neuropraxia of other branches from traction edema. The wound healed well. Adjuvant radiotherapy was recommended, but the patient was lost to follow-up thereafter; further treatment and survival data are unavailable. Key events are summarised in Table 1.

Table 1.

Timeline

Time point	Event	Key clinical findings & decision rationale
Day 0	Outpatient presentation	Chief complaint: 6-month history of painless, enlarging left infra-auricular mass. Physical exam: firm, immobile mass; normal facial nerve function; no cutaneous or mucosal primary lesion identified after thorough head & neck examination
Day 1	Admission & core needle biopsy (CNB)	Ultrasound-guided CNB performed to obtain tissue for histopathology and immunohistochemistry (IHC), given suspicion of parotid malignancy
Day 2	Contrast-enhanced CT (neck & chest)	CT revealed a 35 × 27 mm ill-defined left parotid mass with heterogeneous enhancement and central necrosis. No distant metastases detected in chest
Day 4	MRI (head & neck)	MRI showed mixed T1/T2 signals, suggestive of melanotic content or hemorrhage. Margins were better defined than on CT
Day 8	Pathology confirmation & surgical decision	CNB pathology: malignant melanoma (S100+, HMB45+, MelanA+, Ki-67 ~ 50%, CKpan-). Multidisciplinary team decision: proceed with total parotidectomy + neck dissection due to high proliferative index and risk of nodal metastasis
Day 9	Total parotidectomy + radical neck dissection (Levels I-V)	Intraoperative finding: black, firm tumor in lower pole encasing the marginal mandibular branch → nerve sacrificed for negative margins. Other facial nerve branches preserved
Day 16	Drainage tube removed	Postoperative recovery uneventful; drain output < 20 mL/24 h
Day 17	Final pathology & discharge	Final pathology: 32 × 25 × 23 mm melanoma with ill-defined borders, invading adjacent salivary tissue.Metastasis in 14/48 cervical nodes (Levels I-V); intraparotid nodes negative. Patient discharged with wound healed
Day 26	First postoperative follow-up (1 week post-discharge)	Findings: incomplete left eyelid closure, mouth droop to right, cheek puffing with air leak → marginal mandibular + zygomatic/buccal neuropraxia (due to planned sacrifice and traction edema). Wound well-approximated
Day 34	Second postoperative follow-up	Facial nerve dysfunction persistent.Advised adjuvant radiotherapyto reduce locoregional recurrence risk. Patient subsequently lost to follow-up; further treatment (including possible immunotherapy) and survival data unavailable

Discussion

PMMPG is exceptionally rare [1, 2, 4]. Most parotid melanomas are metastatic from cutaneous primaries of the head and neck, as the parotid contains intraglandular lymph nodes draining these areas [4]. A systematic review found the parotid as the most common site of major salivary gland metastases (93.6%), with primary melanoma most often in the head and neck (65%) [5].

Distinguishing primary from metastatic disease is clinically important, though both have aggressive behaviour. Woodward’s criteria for primary parotid melanoma include: (i) tumor centre within parotid parenchyma; (ii) not arising from an intraparotid lymph node; (iii) no primary melanoma on thorough skin/mucosal/eye/meningeal examination; (iv) no prior melanoma excision [2]. Our patient met these clinical criteria, but whole‑body PET‑CT was not performed, so we cannot completely exclude an occult distant primary. Therefore, we designate this case as a clinically presumed primary parotid melanoma, rather than a definitively proven one. This cautious wording is important given the absence of PET‑CT.

Imaging provides supportive clues. Melanotic melanomas often show T1 hyperintensity and T2 hypointensity on MRI due to melanin and haemorrhage [4]; our patient showed mixed signals, possibly from cystic degeneration and haemorrhage. CT helped assess local extent and exclude chest metastases. Definitive diagnosis rests on histopathology and IHC. S100 is highly sensitive but not specific; HMB45 and MelanA are more specific for melanocytic differentiation, though they can be negative in desmoplastic subtypes [6]. SOX10 has high sensitivity and specificity (100% and 93–100%) [7]. PRAME is a newer specific marker [6], not used here but potentially useful in future cases. Our patient’s IHC (S100 +, HMB45+, MelanA+, SOX10+, CKpan‑, P40‑) effectively excluded carcinoma and other salivary tumours. The high Ki‑67 (50%) correlates with aggressive behaviour and poor prognosis [8], as does the extensive nodal metastasis (14/48).

Complete resection with negative margins is the mainstay, typically via superficial or total parotidectomy depending on tumour location [3]. Given the high risk of regional metastasis, elective neck dissection is recommended for larger or node‑positive tumours. In our patient, the comprehensive neck dissection (levels I–V) was chosen preoperatively based on the aggressive histology and risk of occult spread, which was confirmed pathologically. The marginal mandibular branch was intentionally sacrificed because it was encased by the tumour; any attempt to spare it would have left microscopic residual disease. The transient weakness of other branches was due to traction neuropraxia. This distinction—between permanent sacrifice and temporary paresis—is critical for patient counselling.

Adjuvant radiotherapy is often advised for high‑risk features (positive/close margins, high grade, nodal involvement). Immunotherapy (for example, pembrolizumab, nivolumab) has emerged as effective adjuvant treatment for high‑risk melanoma, with case reports suggesting benefit in parotid melanoma [9]. Zhao et al. reported a patient who received surgery plus immunotherapy/chemotherapy/radiotherapy and achieved > 19 months progression‑free survival [3]. However, evidence remains limited. BRAF mutation testing, which guides targeted therapy (dabrafenib/trametinib), was not performed due to economic constraints—a limitation in resource‑limited settings. The prognosis of primary parotid melanoma is poor; a systematic review reported 1‑ and 5‑year survival rates of 79% and 45% for metastatic melanoma of major salivary glands [5], and another review noted 54% metastasis and 45% mortality in PMMPG [10]. Our patient’s high Ki‑67 and extensive nodal spread predict early distant recurrence, emphasizing the need for close surveillance and consideration of aggressive adjuvant therapy.

Conclusions

Clinically presumed primary parotid melanoma is a rare but highly aggressive tumour that should be considered in the differential diagnosis of parotid masses when no extraparotid primary is found. Definitive diagnosis requires immunohistochemistry. Total parotidectomy with appropriate neck dissection is the primary treatment, and facial nerve sacrifice may be necessary if invaded. Adjuvant radiotherapy and emerging immunotherapies may benefit high‑risk patients, but evidence is limited. High Ki‑67 and extensive nodal metastasis portend a poor prognosis. Reporting such cases raises clinical awareness and may improve early recognition and management.

Limitations

A few limitations should be acknowledged. First, whole-body PET-CT was not available, so the occult extraparotid primary could not be completely excluded. Thus the diagnosis is clinically presumed, not proven Second, BRAF mutation testing was not done, which restricts the discussion of targeted therapy. Third, after adjuvant radiotherapy recommendation, the patient was lost to follow-up, which prevented the assessment of long-term outcomes. These limitations are the real-world constraints in resource-limited situations.

Patient perspective

Based on the Day 26 follow-up visit “I thought it was just a lump; I didn’t expect surgery. Now eating and drinking are difficult. I need to think about further treatment.”

Abbreviations

CNB

Core needle biopsy

CT

Computed tomography

HU

Hounsfield units

IHC

Immunohistochemistry

MRI

Magnetic resonance imaging

PET-CT

Positron emission tomography-computed tomography

PMMPG

Primary malignant melanoma of the parotid gland

Author contributions

Xinran Lin performed the literature review, imaging analysis, and initial draft writing. Yafang Zhou contributed to clinical documentation and manuscript revision. Minge Li contributed to data curation and manuscript editing. Yujie Ma contributed to data curation and manuscript editing. Xuewen Yang supervised the case management and gave final approval of the manuscript. All authors read and approved the final manuscript.

Funding

This work was supported by the Yinchuan Science and Technology Innovation Team Training Program (Grant No. 2023CXTD29).

Data availability

No datasets were generated or analysed during the current study.

Declarations

Ethics approval and consent to participate

Not applicable. According to the policies of our institution, a single case report does not require ethical approval.

Consent for publication

Written informed consent was obtained from the patient for publication of this case report and any accompanying images. A copy of the written consent is available for review by the Editor-in-Chief of this journal.

Competing interests

The authors declare no competing interests.